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/W4323850632	https://www.imrpress.com/journal/CEOG/50/3/10.31083/j.ceog5003052/pdf	English	null	Vitamin D Deficiency in a Cohort of Neapolitan Pregnant Women: Do We Really Live in the City of the Sun?	Clinical and experimental obstetrics & gynecology	2,023	cc-by	3,746	Abstract Background: Insufficient serum 25-hydroxyvitamin D [25(OH)D] levels are a global public health issue, and pregnant women are a significant at-risk group. We aimed to assess maternal serum 25(OH)D concentrations in a cohort of Neapolitan pregnant women and the association with dietary habits, to define which patients are at increased risk of hypovitaminosis and how we can identify them. Methods: This was a prospective observational study. We included 103 pregnant women attending a routine third trimester obstetric examination. Information on obstetrical history and socio-demographic characteristics were obtained through interviews and medical records. Vitamin D intake was assessed using a food frequency questionnaire. Serum 25(OH)D concentration was measured by DiaSorin Liaison and the cut-off value for deficiency was set at 20 ng/mL. Results: Among the 103 pregnant women recruited, 71 (68.9%) were Vitamin D deficient (mean value 12.6 ± 0.5 ng/mL). No statistically significant differences were found between the women involved in the study for: maternal age, pre-pregnancy body mass index (BMI), gestational weight gain, and gestational age at investigation. Women with hypovitaminosis reported a significantly lower intake of milk/yoghurt, cheeses/dairy products and fish. Maternal hypovitaminosis D significantly correlated with low birth weight. Conclusions: Our study confirms that Vitamin D deficiency is a common finding also in sunny areas. Patients with low level of Vitamin D reported a lower intake of food rich of this micronutrient. An accurate anamnesis can be an easy way to identify pregnant women at risk of hypovitaminosis D for whom screening and supplementation can be suitable. Keywords: Vitamin D; sun exposure; dietary intake; pregnancy; obstetric outcomes g Vitamin D Deficiency in a Cohort of Neapolitan Pregnant Women: Do We Really Live in the City of the Sun? Laura Sarno1,, Annunziata Carlea1, Serena Cabaro2, Carmen Imma Aquino3, Dalila Mantelli1, Giuseppe Perruolo2, Pietro Formisano2, Maurizio Guida1, Daniela Terracciano2 1Department of Neurosciences, Reproductive Science and Dentistry, University of Naples Federico II, 80131 Naples, Italy 2Department of Translational Medical Sciences, University of Naples Federico II, 80131 Naples, Italy 3Department of Gynecology and Obstetrics, University of Piemonte Orientale, Ospedale Maggiore della Carità, 28100 Novara, Italy Correspondence: laura.sarno@unina.it (Laura Sarno) Academic Editor: George Daskalakis Submitted: 29 October 2022 Revised: 8 December 2022 Accepted: 14 December 2022 Published: 10 March 2023 1Department of Neurosciences, Reproductive Science and Dentistry, University of Naples Federico II, 80131 Naples, Italy 2Department of Translational Medical Sciences, University of Naples Federico II, 80131 Naples, Italy 3Department of Gynecology and Obstetrics, University of Piemonte Orientale, Ospedale Maggiore della Carità, 28100 Novara, Italy *Correspondence: laura.sarno@unina.it (Laura Sarno) Academic Editor: George Daskalakis Submitted: 29 October 2022 Revised: 8 December 2022 Accepted: 14 December 2022 Published: 10 March 2023 Submitted: 29 October 2022 Revised: 8 December 2022 Accepted: 14 December 2022 Published: 10 March 2023 Copyright: © 2023 The Author(s). Published by IMR Press. This is an open access article under the CC BY 4.0 license. Publisher’s Note: IMR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. MR Press stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Clin. Exp. Obstet. Gynecol. 2023; 50(3): 52 https://doi.org/10.31083/j.ceog5003052 3. Results At enrollment, patients were asked to sign a written consent. Moreover, they were invited to answer a specific questionnaire about dietary habits. From January to May 2019, 125 pregnant women were eligible for the study, but only 103 women consented to par- ticipate. Among them, 71 (68.9%) women presented a de- ficiency of Vitamin D (<20 ng/mL). The questionnaire was made according to Italian di- etary habits. The Italian Society for Osteoporosis, Min- eral Metabolism and Bone Diseases (SIOMMMS) stated that the main sources of Vitamin D are represented by fish and dairy products [8]. Our questionnaire consisted of four questions: (1) Do you eat fatty fish (salmon, sar- dine, tuna, mackerel, swordfish, snapper)? (2) Do you drink milk/yoghurt? (3) Do you eat dairy products? (4) Do you eat fruits and vegetables? Moreover, patients were asked to report how often they eat/drink each type of food. We used the same categories reported in the short Vitamin D ques- tionnaire (VDQ) that has been used in previous studies [11]. Compared to VDQ, we included the use of dairy products instead of margarine, because of its uncommon use in Italy and we ask patients to report also about fruits and vegeta- bles. Maternal characteristics of women presenting hypovi- taminosis and those with normal Vitamin D levels are sum- marized in Table 1. No differences were reported in maternal age, GA at investigation, maternal BMI, GWG, smoking and attitude to physical exercise between the two groups. Based on ma- ternal questionnaires, no significant difference was noticed regarding the intake of fruits and vegetables. Women with hypovitaminosis D reported a signifi- cantly lower intake of milk/yoghurt, cheeses/dairy products and fish: eight patients (25%) with normal levels of Vita- min D and only 7 (9.9%) in the group with hypovitaminosis reported consumption of milk and yoghurt more than once a day (p < 0.05); 24 (75%) women with normal Vitamin D level and only 30 (42.3%) in hypovitaminosis group re- ported an intake of cheese or other dairy products more than once a week (p < 0.05), while 18 patients (56.3%) with nor- mal Vitamin D level and 29 (40.9%) with hypovitaminosis consumed fish more than once a week (p < 0.05) (Table 2). We did not include mushrooms, since they are not common in our diet and their consumption is seasonal and in general it is limited to autumn. 1. Introduction of severe postpartum hemorrhage [7]. In Italy, while it has been reported that the inci- dence of Vitamin D deficiency is high (up to 86%) among women above 70 years old [8], there is few evidence about its incidence among young people. According to the BONTOURNO study, in a cohort of 608 pre-menopausal women, almost one-third of women presented Vitamin D deficiency [9]. However, no recent studies can confirm this data and we do not really know how common is Vitamin D deficiency among our pregnant women. Vitamin D deficiency is a widespread health problem and mostly affects children, adolescents and elders, also in well sun exposed countries, especially in the Middle East [1,2]. It is a common finding during pregnancy. Less than 50% of world population presents an adequate Vita- min D status (>20 ng/mL) [3], and Vitamin D deficiency is described in 33% of American and 24% Canadian preg- nant women [1,4]. Another recent review including 13 tri- als from seven countries evidenced a prevalence of Vita- min D deficiency from 39.4% to 76.5%, with the highest prevalence among Chinese (100%), Turkish (95.6%), Ira- nian (89.4%) and Pakistanian (89.0%) women [5]. Currently, in our setting, Vitamin D dosage and sup- plementation are not routinely recommended during preg- nancy, but it has to be decided case by case according to the risk of having hypovitaminosis [10]. However, it has not clearly established how to define such a risk. Deficiency of Vitamin D seems to be related to poorer outcomes during pregnancy, neonatal and infant develop- ment [6]. Therefore, we decided to introduce a questionnaire re- garding dietary habits and sun exposure to verify if we could have relevant information to easily identify patients at high- risk of vitamin deficiency that could benefit from supple- mentation. Firstly, we assess the incidence of Vitamin D deficiency in a cohort of Neapolitan pregnant women; sec- ondarily, we compared dietary habits between women with Maternal Vitamin D serum level is strictly related to fetal and neonatal Vitamin D status [1,2]. Moreover, sev- eral studies found that Vitamin D supplementation during pregnancy may reduce the risk of preeclampsia, gestational diabetes, low birthweight and could also decrease the risk D [25(OH)D] was measured by DiaSorin Liaison and the cut-off values for deficiency were fixed at 20 ng/mL. Val- ues were given as the sum of 25(OH)D3 and 25(OH)D2. 2. Materials and Methods This was an observational prospective study. We in- cluded pregnant women attending a routine third trimester obstetric examination at the Department of Mother & Child of University Hospital Federico II of Naples, Southern Italy, from January to May 2019. Exclusion criteria were: age <18 years, twin pregnancy, Vitamin D supplementation. Comparison between groups was performed by the Student’s t test and the Mann-Whitney test for continuous variables and the Fisher’s exact test and the Chi-square test for categorical variables. p value < 0.05 was considered statistically significant. This study was conducted according to the Declaration of Helsinki and all procedures were approved by the local Ethics Committee. 1. Introduction normal levels of Vitamin D and those with a deficiency, to define the dietary habits of patients with Vitamin D defi- ciency. Statistical analysis was performed using the statistical analysis program SPSS 20.0 (Chicago INC, Chicago, IL, USA). Data are presented as mean ± standard deviation for continuous variables, and number (percentage) for categor- ical variables. 3. Results Moreover, we asked the patient about the frequency of physical activity, since it has been reported that it can affect Vitamin D concentration [12], and the personal perception of sun exposure during the summer period, when the sun exposure should be longest and the production and storage of Vitamin D should be more influenced by sunlight [13]. About sun exposure, 28 patients with hypovitaminosis (39.4%) and 19 (59.4%) with normal level of Vitamin D reported a sun exposure above 30 minutes per day during the previous summer season (p = 0.06). Regarding obstetric outcomes, no significant differ- ence was found in gestational age at delivery between pa- tients with hypovitaminosis and those with normal Vitamin D levels (38 ± 1.5 weeks vs. 39 ± 2 weeks; p = 0.65). However, in the hypovitaminosis group the mean birth- weight was significantly lower compared to controls (2845 ± 151 grams vs. 3049 ± 118 grams; p < 0.05). No differ- ence was found for the other considered clinical conditions (preeclampsia, gestational diabetes, preterm birth). Data related to general and demographic information, age, gestational age (GA) at investigation, pre-gestational body mass index (BMI), gestational weight gain (GWG), obstetric history, family history of diseases, complications and therapies during pregnancy were also collected and recorded on a dedicated dataset. After delivery, obstetric and neonatal outcomes was assessed from medical records. For all enrolled patients, a blood sample was collected. We asked all women to respect a 12-h fast before blood col- lection. Blood was centrifuged for 10 minutes within 0.5–2 h of sampling. Serum was extracted and stored at –80 °C until analysis by liquid chromatography tandem mass spec- trometry. The serum concentration of 25-hydroxyvitamin 4. Discussion In our prospective observational study, we found that almost 70% of pregnant women present low levels of Vita- 2 Table 1. Vitamin D serum level and characteristics of study population. Vitamin D deficiency Normal Vitamin D level p value n = 71 (68.9%) n = 32 (31.1%) Vitamin D serum level main value (ng/mL) 12.6 ± 0.5 26.4 ± 5.8 <0.0001 Age (years) 32 ± 0.6 32.5 ± 1.1 0.2 BMI (kg/m2) 26.7 ± 1.5 25.6 ± 1.5 0.6 GWG (kg) 10.9 ± 0.8 10.6 ± 4.5 0.1 GA at investigation (weeks) 30.0 ± 1.5 30.0 ± 2.0 0.8 Smoking (yes) 9 (12.7) 2 (6.3) 0.5 Physical activity† 6 (8.5) 3 (9.4) 1 †Percentage of women reporting physical exercises at least twice per week. BMI, Body Mass Index; GWG, Gestational Weight Gain; GA, Gestational Age. Bold face data, p value< 0.0001. Table 1. Vitamin D serum level and characteristics of study population. Table 2. Dietary intake in patients with hypovitaminosis and with normal Vitamin D level. Table 2. Dietary intake in patients with hypovitaminosis and with normal Vitamin D level. Dietary intake Vitamin D deficiency Normal Vitamin D level p value n = 71 (68.9%) n = 32 (31.1%) Fish >once a week 29 (40.8) 18 (56.3) <0.05 Once a week 19 (26.8) 8 (25) <once a week 23 (32.4) 6 (20.7) Milk/yoghurt >once a day 7 (9.9) 8 (25) <0.05 Once a day 45 (63.4) 18 (56.3) <once a day 19 (26.8) 6 (20.7) Cheese/dairy products >once a week 30 (42.3) 24 (75) <0.05 Once a week 29 (40.8) 6 (18.8) <once a week 12 (16.9) 2 (6.2) Vegetables >once a day 45 (63.4) 22 (68.8) 0.6 Once a day 14 (19.7) 7 (21.9) <once a day 12 (16.9) 3 (9.4) Fruits >once a day 29 (40.8) 14 (43.8) 0.7 Once a day 35 (49.3) 17 (53.1) <once a day 7 (9.9) 1 (3.1) Bold face data, p value< 0.05. min D during the third trimester of pregnancy; this is an un- expected high percentage, considering that a previous study reported that the incidence of vitamin D deficiency in fer- tility period has been estimated to be in 1:3 women in Italy [9]. Our finding was consistent with the results of a sys- tematic review by Karras et al. Acknowledgment Not applicable. References [1] El Koumi MA, Ali YF, Abd El Rahman RN. Impact of maternal vitamin D status during pregnancy on neonatal vitamin D status. The Turkish Journal of Pediatrics. 2013; 55: 371–377. [1] El Koumi MA, Ali YF, Abd El Rahman RN. Impact of maternal vitamin D status during pregnancy on neonatal vitamin D status. The Turkish Journal of Pediatrics. 2013; 55: 371–377. To the best of our knowledge, this is the first study reporting the prevalence of hypovitaminosis D among Neapolitan pregnant women. The small sample size rep- resents the main limitation of the study. In future, it will be interesting compare our sample considering different gesta- tional age and different sun exposed population. Moreover, we reported sun exposure based on self-reported personal perception and referring to the previous summer season, while it would be more accurate to use objective methods, such as UV radiations dosimeters [13]. However, our aim was to understand which information directly obtained from the patients could be helpful in identify the risk of Vitamin D deficiency. [2] Sachan A, Gupta R, Das V, Agarwal A, Awasthi PK, Bhatia V. High prevalence of vitamin D deficiency among pregnant women and their newborns in northern India. The American Journal of Clinical Nutrition. 2005; 81: 1060–1064. [3] van Schoor N, Lips P. Global Overview of Vitamin D Status. Endocrinology and Metabolism Clinics of North America. 2017; 46: 845–870. [3] van Schoor N, Lips P. Global Overview of Vitamin D Status. Endocrinology and Metabolism Clinics of North America. 2017; 46: 845–870. [4] Palacios C, Gonzalez L. Is vitamin D deficiency a major global public health problem? The Journal of Steroid Biochemistry and Molecular Biology. 2014; 144: 138–145. [5] van der Pligt P, Willcox J, Szymlek-Gay EA, Murray E, Wors- ley A, Daly RM. Associations of Maternal Vitamin D Defi- ciency with Pregnancy and Neonatal Complications in Develop- ing Countries: A Systematic Review. Nutrients. 2018; 10: 640. ing Countries: A Systematic Review. Nutrients. 2018; 10: 64 [6] Møller UK, Streym S, Mosekilde L, Heickendorff L, Flyvbjerg A, Frystyk J, et al. Changes in calcitropic hormones, bone mark- ers and insulin-like growth factor I (IGF-I) during pregnancy and postpartum: a controlled cohort study. Osteoporosis Inter- national. 2013; 24: 1307–1320. Author Contributions Due to the lack of clear evidence about the relation- ship between Vitamin D and obstetric outcomes, there are no official recommendations about Vitamin D intake: the World Health Organization (WHO) does not indicate it as routine antenatal care [16]. Also, according to our national guidelines, Vitamin D supplementation in pregnancy is not routinely recommended [10]. Extraction and drafting of the manuscript—LS, CIA, AC; Analysis of data, manuscript revision—LS, MG, SC, CIA, DM, GP, PF, DT; Design and revision, statistical analysis—LS, MG, SC, DM, GP, PF, DT. All authors read and approved the final manuscript. This research received no external funding. This research received no external funding. Several findings suggest that supplementation around 1000 IU/day may be optimal to obtain a Vitamin D serum level greater than 50 nmol/L (20 ng/mL) [18–21]. Availability of Data and Materials All authors are at disposal for any additional data. 4. Discussion [14], reporting that, de- spite a good sun exposure, Vitamin D deficiency seems to be quite high in the Mediterranean region: they reported a prevalence ranging from 22.7% to 90.3%. According to that, apart from sun exposure that was lower in the group with hypovitaminosis D in our population, we should con- sider other factors, such as dietary habits, to identify pa- tients at risk of deficiency. Despite our study has been conducted in a very sunny area, our percentage is very similar to the one reported in some Northern European countries, like Germany (77%) and in high-risk ethnicities, like South-Asian women, re- porting a prevalence of Vitamin D hypovitaminosis that can be up to 96% [7]. Indeed, according to the self-reported perception of sun exposure, even if patients with Vitamin D deficiency reported a lower sun exposure, in all the cohort less than half of women reported a sun exposure above 30 Regarding dietary habits, we found that the consump- tion of milk/yoghurt, cheese, and dairy products signifi- cantly differed between the two groups and it could affect Vitamin D serum levels, if we consider that has been stated 3 that the main sources of Vitamin D in the food are repre- sented by fish and dairy products [8]. plementation is recommended only in high-risk subjects but no tools are provided to identify at-risk patients, a good anamnesis addressing nutritional habits and personal sun exposure perception could be an easy way to identify pa- tients at risk of Vitamin D deficiency. Several studies reported an association between ad- verse obstetric outcomes and hypovitaminosis D [15]. In our study, patients with low level of Vitamin D had signif- icantly lower birthweight compared to those with a normal level. However, this data must be confirmed with larger studies and considering other possible actors linked to lower birthweight (i.e., smoke): no definitive conclusion can be drawn on the basis of this result. The authors declare no conflict of interest. This dosage is supposed to increase blood 25- hydroxyvitamin D by 1.2 nmol/L for every mcg (40 IU) of vitamin D3 given orally to individuals with low 25- hydroxyvitamin D levels [22]. Conflict of Interest The authors declare no conflict of interest. Ethics Approval and Consent to Participate However, considering the high percentage of hypovi- taminosis D in our setting and the long-term effect of this condition during women’s life, we cannot ignore that preg- nancy could be an important moment to screen for vita- min deficiency; on the contrary, a universal screening to all pregnant women could be difficult to realize. All subjects gave their informed consent for inclusion before they participated in the study. The study was con- ducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of the University of Naples Federico II (prot. n. 80/19). Vitamin D supplementation has been considered a valid intervention to correct deficiency in a specific pop- ulation, such as pregnant women [17], but it has been im- portant to stress with our patients that changing the lifestyle is the best way to prevent such a deficit. 5. Conclusions Despite we conducted this observational study in a sunny area, we demonstrated a high percentage of Vita- min D deficiency among pregnant women during the third trimester and a low sun exposure, based on what emerged from the questionnaires. In a setting where Vitamin D sup- [7] Palacios C, Kostiuk LK, Peña-Rosas JP. Vitamin D supplemen- tation for women during pregnancy. The Cochrane Database of Systematic Reviews. 2019; 7: CD008873. 4 [8] Adami S, Romagnoli E, Carnevale V, Scillitani A, Giusti A, Rossini M, et al. Guidelines on prevention and treatment of vitamin D deficiency. Italian Society for Osteoporosis, Min- eral Metabolism and Bone Diseases (SIOMMMS). Reumatismo. 2011; 63: 129–147. (In Italian) pact on maternal-fetal outcomes in pregnancy: A critical review. Critical Reviews in Food Science and Nutrition. 2018; 58: 755– 769. [16] WHO Recommendations on Antenatal Care for a Positive Preg- nancy Experience. World Health Organization: Geneva. 2016. [9] Adami S, Bertoldo F, Braga V, Fracassi E, Gatti D, Gandolini G, et al. 25-hydroxy vitamin D levels in healthy premenopausal women: association with bone turnover markers and bone min- eral density. Bone. 2009; 45: 423–426. [17] NICE Group. Nice Guidelines. Vitamin D: supplement use in specific populations groups. 2017 Available at: https: //www.nice.org.uk/guidance/ph56/resources/vitamin-d-suppl ement-use-in-specific-population-groups-pdf-1996421765317 (Accessed: 29 October 2022). [10] Linee Guida Gravidanza fisiologica, Istituto Superiore di Sanità. 2011. Available at: http://www.salute.gov.it/imgs/C_17_pubbli [10] Linee Guida Gravidanza fisiologica, Istituto Superiore di Sanità. 2011. Available at: http://www.salute.gov.it/imgs/C_17_pubbli cazioni_1436_allegato.pdf (Accessed: 29 October 2022). [18] Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ. Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol [published correction appears in American Journal of Clinical Nutrition. 2003; 78: 1047]. Amer- ican Journal of Clinical Nutrition. 2003; 77: 204–210. cazioni_1436_allegato.pdf (Accessed: 29 October 2022) [11] Bärebring L, Amberntsson A, Winkvist A, Augustin H. Valida- tion of Dietary Vitamin D Intake from Two Food Frequency Questionnaires, Using Food Records and the Biomarker 25- Hydroxyvitamin D among Pregnant Women. Nutrients. 2018; 10: 745. [19] Hollis BW, Wagner CL. Vitamin D requirements during lacta- tion: high-dose maternal supplementation as therapy to prevent hypovitaminosis D for both the mother and the nursing infant. The American Journal of Clinical Nutrition. 2004; 80: 1752S– 1758S. [12] Hauta-Alus HH, Holmlund-Suila EM, Rita HJ, Enlund-Cerullo M, Rosendahl J, Valkama SM, et al. Season, dietary factors, and physical activity modify 25-hydroxyvitamin D concentra- tion during pregnancy. European Journal of Nutrition. 2018; 57: 1369–1379. [20] Hollis BW. 5. Conclusions Vitamin D requirement during pregnancy and lac- tation. Journal of Bone and Mineral Research. 2007; 22: V39– V44. [13] Webb AR, Kift R, Durkin MT, O’Brien SJ, Vail A, Berry JL, et al. The role of sunlight exposure in determining the vitamin D status of the U.K. white adult population. The British Journal of Dermatology. 2010; 163: 1050–1055. [21] Vieth R, Chan PC, MacFarlane GD. Efficacy and safety of vi- tamin D3 intake exceeding the lowest observed adverse effect level. The American Journal of Clinical Nutrition. 2001; 73: 288–294. [14] Karras S, Paschou SA, Kandaraki E, Anagnostis P, Annweiler C, Tarlatzis BC, et al. Hypovitaminosis D in pregnancy in the Mediterranean region: a systematic review. European Journal of Clinical Nutrition. 2016; 70: 979–986. [22] Dawson-Hughes B, Heaney RP, Holick MF, Lips P, Meunier PJ, Vieth R. Estimates of optimal vitamin D status. Osteoporosis In- ternational. 2005; 16: 713–716. [15] Agarwal S, Kovilam O, Agrawal DK. Vitamin D and its im- 5 5
https://openalex.org/W2520939316	https://nottingham-repository.worktribe.com/preview/804203/1-s2.0-S1053811916304530-main%20-%20MEG.pdf	English	null	Measurement of dynamic task related functional networks using MEG	NeuroImage	2,017	cc-by	14,324	1. Introduction have been based on an assumption of stationarity; i.e. a connection between two regions is characterised by a single parameter derived over many minutes (or even hours) of data. However, the brain is a dynamic system and efﬁcient function likely relies on the formation and dissolution of many hierarchical networks at rapid time scales, which support ongoing cognition. The characterisation of such transient networks represents a key goal in neuroscience. Increasing evidence (Allen et al., 2014; Baker et al., 2014; Baker et al., 2012; Chang and Glover, 2010; Chang et al., 2013; de Pasquale et al., 2010; de Pasquale et al., 2015; Hutchison et al., 2013; Karahanoğlu and Van De Ville, 2015; O’Neill et al., 2015b; Smith et al., 2012; Yaesoubi et al., 2015) suggests that such transient networks might be measured using neuroimaging. In this paper, we introduce a method which tracks the formation of functional electrophysiological networks, and use it to image the network formations associated with self-initiated movement, and working memory. Measurement of statistical interdependencies between neuroi- maging signals has revealed a number of robust networks of func- tional connectivity in the brain (Beckmann et al., 2005; Corbetta, 1998; Fox and Raichle, 2007; Fox et al., 2005; Friston, 1994; Raichle et al., 2001; Smith et al., 2009). These networks, each with their own characteristic spatial signature, are thought to govern core mental processes with some supporting sensory integration and others as- sociated with cognition or attention. Most networks are observed even in subjects at rest and are thus termed resting state networks (RSNs). Characterisation of RSNs is an emerging focus: not only does it offer new insight into how spatially separate regions integrate, RSNs (and functional connections in general) have been shown to be compromised in a variety of pathologies (Brookes et al., 2016; Friston, 1998; Guggisberg et al., 2008; Kessler et al., 2014; Palaniyappan and Liddle, 2012; Schnitzler and Gross, 2005; Stufﬂebeam et al., 2011; Tewarie et al., 2014; van Dellen et al., 2012) highlighting their clinical importance. To date, the majority of functional connectivity studies In order to track network dynamics effectively, we require a modality which can match the rapid timescales of the brain. a r t i c l e i n f o Article history: Received 15 May 2016 Accepted 29 August 2016 Keywords: Network Dynamics Magnetoencephalography MEG Sternberg task The characterisation of dynamic electrophysiological brain networks, which form and dissolve in order to support ongoing cognitive function, is one of the most important goals in neuroscience. Here, we in- troduce a method for measuring such networks in the human brain using magnetoencephalography (MEG). Previous network analyses look for brain regions that share a common temporal proﬁle of ac- tivity. Here distinctly, we exploit the high spatio-temporal resolution of MEG to measure the temporal evolution of connectivity between pairs of parcellated brain regions. We then use an ICA based procedure to identify networks of connections whose temporal dynamics covary. We validate our method using MEG data recorded during a ﬁnger movement task, identifying a transient network of connections linking somatosensory and primary motor regions, which modulates during the task. Next, we use our method to image the networks which support cognition during a Sternberg working memory task. We generate a novel neuroscientiﬁc picture of cognitive processing, showing the formation and dissolution of multiple networks which relate to semantic processing, pattern recognition and language as well as vision and movement. Our method tracks the dynamics of functional connectivity in the brain on a timescale commensurate to the task they are undertaking. Article history: Received 15 May 2016 Accepted 29 August 2016 Keywords: Network Dynamics Magnetoencephalography MEG Sternberg task Article history: Received 15 May 2016 Accepted 29 August 2016 y g & 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). n Correspondence to: Sir Peter Mansﬁeld Imaging Centre, School of Physics and Astronomy, University of Nottingham, University Park, Nottingham, NG7 2RD, UK. E-mail address: matthew.brookes@nottingham.ac.uk (M.J. Brookes). http://dx.doi.org/10.1016/j.neuroimage.2016.08.061 1053-8119/& 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). NeuroImage journal homepage: www.elsevier.com/locate/neuroimage mage.2016.08.061 ublished by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). http://dx.doi.org/10.1016/j.neuroimage.2016.08.061 1053-8119/& 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 1. Introduction MEG data were recorded using a 275-channel CTF MEG system (MISL; Coquitlam, BC, Canada) in synthetic 3rd order synthetic gradiometer conﬁguration; at a sampling rate of 600 Hz. Subjects were positioned supine. To ascertain the location of the head within the MEG helmet, three head position indicator (HPI) coils where attached to the subject at the nasion and preauricular points. These were energised periodically in order to track con- tinuously the subjects head position. To allow coregistration of brain anatomy to the MEG sensor geometry, a measurement of the locations of the HPI coils relative to the scalp surface was created with a 3D digitiser (Polhemus; Colchester, VT). Anatomical images were acquired using either a 3T or 7T Philips Achieva MRI scanner (MPRAGE; 1 mm3 resolution). Coregistration of MEG data to ana- tomical MRI was then achieved by matching the digitised head surface to the equivalent surface extracted from the MRI. g In this paper, we undertake analysis of the timecourses of dy- namic connectivity. Our method is based upon measurement of envelope correlation within small time-windows of data and be- tween pairs of brain regions deﬁned via cortical parcellation (Allen et al., 2014; Bola and Sabel, 2015; Colclough et al., 2015; Finn et al., 2015; Hassan et al., 2015; Hillebrand et al., 2012; Smith et al., 2015; Tewarie et al., 2014). By computing dynamic connectivity in a sliding window, we facilitate estimation of timecourses showing the evolution of functional connectivity at an intermediate time- scale on the order of a few seconds. These connectivity time- courses are then analysed using independent component analysis (ICA). Past studies have shown ICA to be a valuable tool to eluci- date networks: for example, timecourses of brain ‘activity’ can be acquired from multiple voxels and decomposed into a smaller number of temporally independent components, with a single component representing temporal signatures at multiple voxels; assessment of which voxels contribute to each component yields images of regions that share a temporal proﬁle of activity (Brookes et al., 2011b). Here, distinct from this, having characterised the timecourses of connectivity, we decompose those timecourses in such a way that a single component represents an independent temporal signature shared by multiple connections. Assessment of the speciﬁc connections contributing to a component allows elu- cidation of networks of dynamically changing connectivities which share a single temporal signature. (I.e. 1. Introduction This temporal richness has been exploited via the introduction of many techniques to quantify connectivity (Scholvinck et al., 2013) and correlation between the amplitude envelopes of band limited oscillations has shown that static networks, similar to RSNs observed using fMRI, can be observed (Baker et al., 2014; Baker et al., 2012; Brookes et al., 2011a; Brookes et al., 2011b; de Pasquale et al., 2010; Hipp et al., 2012; Hipp and Siegel, 2015; Liu et al., 2010; Luckhoo et al., 2012; O’Neill et al., 2015b; Wens et al., 2014b). More im- portantly, the temporal richness facilitates measurement of con- nectivity in short time windows, making MEG the method of choice for capturing transiently active networks; indeed exploitation of the spatiotemporal resolution of MEG has already shown that functional connectivity changes on the time scale of seconds (Brookes et al., 2014; O’Neill et al., 2015b) and even milliseconds (Baker et al., 2014). However, this area remains in its infancy; few methods are available and some remain limited by, for example, a-priori selection of brain regions or the signiﬁcant problem of signal leakage between regions (Brookes et al., 2012b; Colclough et al., 2015; Wens et al., 2015). Further, few studies have attempted to probe the evolution of dy- namic networks during a task. press with the index ﬁnger of their non-dominant hand. Sub- jects were instructed to press the button infrequently; they were told that a button press should be executed approximately once every 30 seconds, but precise timekeeping was not im- portant and they should not count the time between presses. A subset of these data has been used in prior publications (O’Neill et al., 2015b; Vidaurre et al., 2016). Dataset 2 – Sternberg task: 19 healthy participants (10 male, aged 2573 years) performed a Sternberg working memory task. Two example visual stimuli (abstract geometric shapes) were presented on a screen; each stimulus was shown for 0.6 s with 1 s between onsets. Following this, a period of 7 seconds was left, known as the maintenance phase, before a third (probe) stimulus was presented. If the probe stimulus matched either of the two example stimuli, the subject was told to execute a button press with their right index ﬁnger. Subjects received immediate feedback as to whether their response was correct. Trials were separated by 30 seconds of rest, where subjects ﬁxated on a cross. 30 trials were presented to every subject. 1. Introduction Func- tional Magnetic Resonance Imaging (fMRI) has provided signiﬁcant evidence of non-stationary connectivity (Hutchison et al., 2013), but its reliance on the blood oxygenation level dependant (BOLD) re- sponse means the fastest time scales are obfuscated by the latency and longevity of the haemodynamic response. In contrast, magne- toencephalography (MEG; Cohen, 1968, 1972) detects changes in Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 2 extracranial magnetic ﬁelds induced by dendritic currents in the brain. Because signals are generated from electrical activity in neu- rons, it is possible to observe brain activity on a millisecond time- scale. This temporal richness has been exploited via the introduction of many techniques to quantify connectivity (Scholvinck et al., 2013) and correlation between the amplitude envelopes of band limited oscillations has shown that static networks, similar to RSNs observed using fMRI, can be observed (Baker et al., 2014; Baker et al., 2012; Brookes et al., 2011a; Brookes et al., 2011b; de Pasquale et al., 2010; Hipp et al., 2012; Hipp and Siegel, 2015; Liu et al., 2010; Luckhoo et al., 2012; O’Neill et al., 2015b; Wens et al., 2014b). More im- portantly, the temporal richness facilitates measurement of con- nectivity in short time windows, making MEG the method of choice for capturing transiently active networks; indeed exploitation of the spatiotemporal resolution of MEG has already shown that functional connectivity changes on the time scale of seconds (Brookes et al., 2014; O’Neill et al., 2015b) and even milliseconds (Baker et al., 2014). However, this area remains in its infancy; few methods are available and some remain limited by, for example, a-priori selection of brain regions or the signiﬁcant problem of signal leakage between regions (Brookes et al., 2012b; Colclough et al., 2015; Wens et al., 2015). Further, few studies have attempted to probe the evolution of dy- namic networks during a task. extracranial magnetic ﬁelds induced by dendritic currents in the brain. Because signals are generated from electrical activity in neu- rons, it is possible to observe brain activity on a millisecond time- scale. 2.2 Pre-processing and Source Reconstruction MEG data were initially inspected visually. Any trials deemed to contain excessive interference, for example generated by muscles or eye movement, were removed. In addition, datasets in which the subject's head moved more than 5mm (Euclidean distance) from its starting position were excluded. A schematic of the sub- sequent data processing pipeline is given in Fig. 1. Following pre-processing, data were analysed using beam- forming. The cortex was parcellated using the Automated Anato- mical Labelling (AAL) atlas (Tzourio-Mazoyer et al., 2002) which had been modiﬁed by removing subcortical ROIs to leave 78 re- gions (Gong et al., 2009), and was transformed to each individual's brain geometry using FMRIB Linear Image Registration Tool (FLIRT) (Jenkinson et al., 2012). In order to obtain a representative time- series for every region, the centre of mass of each region was deﬁned and used as a single representative location (Fig. 1 – step 1). MEG data were frequency ﬁltered 1-150 Hz and source loca- lised using an adaptive beamformer (Robinson and Vrba, 1998; Van Veen et al., 1997) in order to derive 78 source timecourses per subject, one for each AAL region (Fig. 1 – step 2). For beamforming, data covariance was deﬁned in a frequency window spanning 1-150 Hz and a time window covering the entire experiment (Brookes et al., 2008). The covariance matrix was regularised using the Tikhonov method with the regularisation parameter set such that the regularised covariance matrix would have a condition number of 100. Forward ﬁelds were based upon dipole approx- imations (Sarvas, 1987) and a multiple local spheres head model (Huang et al., 1999). Dipole orientation was determined using a non-linear search for the optimal signal to noise ratio (SNR) Dataset 1 – Self Paced Motor task: 15 volunteers (9 male, aged 25 7 4 years (mean 7 SD)) were asked to execute a button 1. Introduction we assume that multiple connections modulated in the same way in time are related functionally.) Our method characterises both the way in which connectivity evolves in time, and the spatial signatures of that evolution. In this way, we can uniquely track the dynamic beha- viour of networks, on a timescale which is commensurate to the task they are undertaking. l, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), oimage.2016.08.061i Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i 2.1 Data Acquisition Note that the leakage reduction step was applied on each window se- parately (separate orthogonalisation for each i), rather than on the whole time series. This is because previous work (O’Neill et al., 2015b) has shown that leakage depends on signal to noise ratio, which changes in different time windows. et al., 2012b; Hipp et al., 2012; O’Neill et al., 2015a; Wens et al., 2015). Most methods rely on the fact that leakage manifests as a zero-time lag linear summation of underlying signals and for this reason orthogonalisation of beamformer projected signals (e.g. orthogonalisation of the rows of Q i) results in the effective re- moval of leakage, albeit at the expense of genuine zero-lag con- nectivity. An elegant means to achieve orthogonalisation si- multaneously over a set of multiple brain regions was recently proposed by Colclough et al. (2015). Here, signals from all nn re- gions are symmetrically orthogonalised within a single computa- tion. The full mathematical details of this procedure can found elsewhere (Colclough et al., 2015). Brieﬂy, the method involves two steps: First, a set of orthonormal time-courses, closest to the data matrix Q i and for which there is a simple analytic solution, is found. Second, the solution is ﬁnessed by iteratively adjusting the lengths and orientations of the corrected vectors until the solution is as close as possible to the uncorrected timecourses. The result is a set of matrices, Oi, whose rows contain the orthogonalised (windowed) time series for all 78 AAL regions (Fig. 1 – step 4). Note that the leakage reduction step was applied on each window se- parately (separate orthogonalisation for each i), rather than on the whole time series. This is because previous work (O’Neill et al., 2015b) has shown that leakage depends on signal to noise ratio, which changes in different time windows. 2.3 Dynamic Functional Connectivity Analysis We aimed to undertake a dynamic, all-to-all, functional con- nectivity analysis. This means that connectivity between all pos- sible pairs of AAL regions is estimated, as a function of time, using a sliding window approach. 2.1 Data Acquisition Two separate MEG datasets were acquired. Both were approved by the University of Nottingham Medical School Research Ethics Committee. Dataset 1 – Self Paced Motor task: 15 volunteers (9 male, aged 25 7 4 years (mean 7 SD)) were asked to execute a button Fig. 1. A schematic diagram describing the fundamental processing pipeline. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 3 Fig. 1. A schematic diagram describing the fundamental processing pipeline. (Robinson and Vrba, 1998; Sekihara et al., 2004). This process creates a source space data matrix, Q of dimension × n n n s, where nn is the number of AAL regions (78) and ns is the number of samples. (Robinson and Vrba, 1998; Sekihara et al., 2004). This process creates a source space data matrix, Q of dimension × n n n s, where nn is the number of AAL regions (78) and ns is the number of samples. number of methods for leakage reduction now in place (Brookes et al., 2012b; Hipp et al., 2012; O’Neill et al., 2015a; Wens et al., number of methods for leakage reduction now in place (Brookes et al., 2012b; Hipp et al., 2012; O’Neill et al., 2015a; Wens et al., 2015). Most methods rely on the fact that leakage manifests as a zero-time lag linear summation of underlying signals and for this reason orthogonalisation of beamformer projected signals (e.g. orthogonalisation of the rows of Q i) results in the effective re- moval of leakage, albeit at the expense of genuine zero-lag con- nectivity. An elegant means to achieve orthogonalisation si- multaneously over a set of multiple brain regions was recently proposed by Colclough et al. (2015). Here, signals from all nn re- gions are symmetrically orthogonalised within a single computa- tion. The full mathematical details of this procedure can found elsewhere (Colclough et al., 2015). Brieﬂy, the method involves two steps: First, a set of orthonormal time-courses, closest to the data matrix Q i and for which there is a simple analytic solution, is found. Second, the solution is ﬁnessed by iteratively adjusting the lengths and orientations of the corrected vectors until the solution is as close as possible to the uncorrected timecourses. The result is a set of matrices, Oi, whose rows contain the orthogonalised (windowed) time series for all 78 AAL regions (Fig. 1 – step 4). Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i 2.1 Data Acquisition Previous work (Baker et al., 2014; Hipp et al., 2012) has shown that functional connectivity is dependent on frequency band studied; for the self-paced motor study we employed a 13-30 Hz frequency window and for the Sternberg task, we employed a 4-30 Hz frequency window as it has been shown multiple frequency bands contribute to working memory (Brookes et al., 2012a). After frequency ﬁltering, Q was segmented into overlapping time windows (Fig. 1 – step 3): we denote the data in a single window, Q i, which has dimensions δ × n f n . Here, i denotes window number, δ is the window width in seconds, and f is sampling frequency. In everything that follows δ ¼ 6 s; the window was shifted in time by 0.5 s for each window number (i). In the self-paced motor task, time windows were centred between t ¼ 12s and t ¼ 12 s (where t represents window centre relative to the button press). There were 49 time windows per trial. In the Sternberg task, time windows were centred between t ¼ 13s and t ¼ 25 s (t represents window centre relative to trial onset). There were 75 time windows per trial. Within each window, we measured connectivity between all pairs of AAL regions. Following leakage correction, the amplitude envelopes of the windowed timecourses were found using Hilbert transformation. This resulted in a set of matrices Ei whose rows contained the amplitude envelopes of orthogonalised neural oscillations (i.e. the envelope of the rows of Oi; Fig. 1 – step 5). Following this, Pearson correlation between amplitude envelopes was measured to form connectivity matrices, Ri, such that In MEG, a signiﬁcant confound for source space connectivity is that the ill-posed inverse problem, coupled with inaccuracies in the forward solution, cause a degree of spatial blurring and mis- localisation of sources. This means that two beamformer derived timecourses (e.g. from two regions) may exhibit signiﬁcant cor- relation, purely due to ‘signal leakage’. Without careful control, this artifactually inﬂates estimated connectivity between regions (Maldjian et al., 2014). Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i Ρ =AX, T  ( ) 2 where the rows of the × n NN ic s matrix X represent temporally independent signatures of functional connectivity, collapsed across all connections. The mixing matrix, A, has dimension × n n c ic and each column represents the contribution of each in- dividual connection to the independent component. The ‘hat’ notation in Eq. 2 denotes that Ρ^ is an estimate of Ρ based upon the derived independent components. Here, Ρ was formed by concatenating all time windows, including all trials and subjects. The ICA decomposition (including prewhitening) was performed using the fastICA method (Hyvarinen, 1999) using a deﬂation approach with = n 10 ic . The spatial signature of each derived in- dependent component was reconstructed based upon the col- umns of A (Fig. 1 – step 9). where the rows of the × n NN ic s matrix X represent temporally independent signatures of functional connectivity, collapsed across all connections. The mixing matrix, A, has dimension × n n c ic and each column represents the contribution of each in- dividual connection to the independent component. The ‘hat’ notation in Eq. 2 denotes that Ρ^ is an estimate of Ρ based upon the derived independent components. Here, Ρ was formed by concatenating all time windows, including all trials and subjects. The ICA decomposition (including prewhitening) was performed using the fastICA method (Hyvarinen, 1999) using a deﬂation approach with = n 10 ic . The spatial signature of each derived in- dependent component was reconstructed based upon the col- umns of A (Fig. 1 – step 9). An independent component was deemed signiﬁcant if, at any one time point in the trial average, the associated column of ̅X ﬂuctuated such that it fell outside a threshold deﬁned by the null distributions for both tests (sign-ﬂip and randomised onset). The threshold for sig- niﬁcance was deﬁned at 0.05, however this was corrected in three ways: First a 2-tailed distribution was allowed, meaning that ﬂuctua- tions in the columns of ̅X could be both greater than, or less than the null distributions. Second, we Bonferroni corrected for multiple com- parisons across the 10 independent components (for both tasks). Third, we Bonferroni corrected across independent temporal degrees of freedom. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ In other words, Ri represents an × n n n n adjacency matrix representing connectivity between all AAL region pairs, in time window i (Fig. 1 – step 6). This process was repeated for all i, resulting in a set of N matrices per subject (one for each time window used; N represents the number of windows per subject). These matrices were concatenated both in time and over all subjects (Fig. 1 – step 7) to form an adjacency tensor, R, with dimensions × × n n NN n n s, where Ns represents the number of subjects. 1. In the ﬁrst case, a ‘sham’ matrix, ̃Xflip, was generated in ex- actly the same way as ̅X, but prior to averaging over trials, half of the subjects were selected randomly and the sign of their contribution to the independent components in X was ‘ﬂipped’ (i.e. multiplied by 1). We reasoned that, if the independent components were not related to the tasks, this sign ﬂipping would have no effect on the magnitude of the trial averaged timecourses, and therefore the magnitudes of G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 4 1. In the ﬁrst case, a ‘sham’ matrix, ̃Xflip, was generated in ex- actly the same way as ̅X, but prior to averaging over trials, half of the subjects were selected randomly and the sign of their contribution to the independent components in X was ‘ﬂipped’ (i.e. multiplied by 1). We reasoned that, if the independent components were not related to the tasks, this sign ﬂipping would have no effect on the magnitude of the trial averaged timecourses, and therefore the magnitudes of ﬂuctuations in ̃Xflip and ¯X would match. However, if the in- dependent components contained trial-onset-locked in- creases or decreases in connectivity, which were robust across subjects, then these would be maintained in ¯X but diminished in ̃Xflip. This procedure was repeated 6435 times for the self-paced data (reﬂecting C7 15 where ( )) ( ) ( ) = ! ! − ! C n k n k / k n and 92,378 times for the Sternberg data ( ) C9 19 , giving all possible realisations of ̃Xflip, with a different set of subjects selected each time. In this way an empirical null distribution was constructed against which the magni- tude of signal ﬂuctuations in ¯X was tested. Note that this ‘sign-ﬂipping’ permutation approach has been employed in previously published work (Hunt et al., 2012; Winkler et al., 2014). where eik represents the vector of timecourse measurements in the kth row of Ei and ( ) r x y , represents the Pearson correlation coefﬁcient between x and y. In other words, Ri represents an × n n n n adjacency matrix representing connectivity between all AAL region pairs, in time window i (Fig. 1 – step 6). This process was repeated for all i, resulting in a set of N matrices per subject (one for each time window used; N represents the number of windows per subject). These matrices were concatenated both in time and over all subjects (Fig. 1 – step 7) to form an adjacency tensor, R, with dimensions × × n n NN n n s, where Ns represents the number of subjects. where eik represents the vector of timecourse measurements in the kth row of Ei and ( ) r x y , represents the Pearson correlation coefﬁcient between x and y. Ρ =AX, T  Note that the 6 s sliding window used to estimate con- nectivity means that the number of temporal degrees of freedom in the averaged trial is substantially less than the number of time points. Here, we assumed that a single temporal degree of freedom was ad- ded each time the window shifts by more than half of its width (i.e. when adjacent windows share less than 50% overlap). This meant a total of 8 temporal degrees of freedom in the self-paced data and 12 in the Sternberg data. Thresholds were therefore set at ( ( × × 0.05/ 2 10 8) ¼ 0.0003) for the self-paced experiment and ( ( × × 0.05/ 2 10 12) ¼ 0.0002) for Sternberg experiment. [It is im- portant to note that these Bonferroni corrections are rigorous, but nevertheless are likely conservative; in future uses of this technique methods based on, for example, maximal statistics (Sekihara et al., 2005) might offer a less conservative approach.] 2.5 Testing for task-modulated networks The above analyses yield a set of ( = ) n 10 ic networks, showing functional connections that share similar (independent) temporal proﬁles. The challenge now becomes to determine which of these represent genuine brain processes. The question of which in- dependent components to keep and which reﬂect only noise is a problem in all ICA based methods. Here for simplicity, we sought to determine which components were modulated signiﬁcantly by the tasks. Our procedure was based on previously described al- gorithms (Clare et al., 1999; Hunt et al., 2012; Winkler et al., 2014). We ﬁrst deﬁned a new matrix, ¯X, containing nic trial averaged independent component timecourses (i.e. ¯X is just X averaged over all trials in all subjects; i.e. the ‘bar’ notation represents a trial average.) The size of ¯X was × N n Trial c (where NTrial represents the number of time windows per trial; 49 for the self-paced task and 75 for the Sternberg task). Following this, we constructed two empirical null distributions: 2.4 Temporal ICA The adjacency tensor, R, measures the temporal evolution of functional connectivity between all pairs of AAL brain regions. We now seek to apply ICA to derive independent temporal signatures of connectivity. To be able to perform ICA we need to reduce to the dimensionality of the data (Fig. 1 – step 8). We begin by vectorising each × n n n n matrix Ri into a × n 1 n 2 row vector. Then, noting that the inherent diagonal symmetry in the adjacency matrix leads to re- dundancy, we remove that redundancy to generate the × n 1 c vector ρi, where = − nc n n 2 n n 2 is the total number of unique con- nections modelled in Ri. These multiple row vectors are then concatenated in time to generate a new matrix Ρ such that ⎡⎣ ⎤⎦ Ρ ρ ρ ρ = … , , , NN T 1 2 s . This means that each column of Ρ represents the timecourse of an individual connection between 2 AAL regions. The dimensionality of this matrix was further reduced by pre- whitening, and ICA was then used generate nic temporally in- dependent components. Mathematically, 2. In the second case, we reasoned that if no task induced re- sponse was expected, then the trial onset times would be meaningless. The trial averaging procedure was again repeated, however rather than the trial averaged data ( ̅X) deﬁned based on genuine trial onsets, a ‘sham’ averaged dataset, ̃Xonset, was deﬁned based on randomly selected ‘sham trial onsets’. 6000 realisations of ̃Xonset were created, again allowing for the gen- eration of a null distribution for each time point and each in- dependent component. Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i l, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), oimage.2016.08.061i 2.1 Data Acquisition Signal leakage has been well studied, with a ⎡ ⎣ ⎢ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥ ⎥ = ( ) ⋯ ( ) ⋮ ⋱ ⋮ ( ) ⋯ ( ) ( ) r e e r e e r e e r e e R , , , , , 1 i i i i in in i in in 1 1 1 1 n n n n ( ) 1 Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ supplementary material. that a network, representing primary somatosensory and motor regions, is modulated signiﬁcantly by the task. Given that the task requires both movement, and elicits a tactile response (since the subject will feel the button press), this network is plausible. A second component, representing the visual network, also showed signiﬁcant modulations around the time of the button press, this can be found in the supplementary material. Fig. 2 shows the results of our method applied to the self-paced data. Although 10 independent components were derived, here we present one of two networks that demonstrated signiﬁcant task modulation as an example. The other 9 networks are shown in supplementary material. Fig. 2A shows a matrix representation of the network. The ordering of the 78 AAL regions is overlaid for reference. Fig. 2B shows the same network represented in 3D and thresholded (70% of the maximum connection strength) for clarity. Both the matrix and 3D visualisation show clearly that the net- work is centred on the right primary somatosensory cortex and highlights strong connections both between sensory and motor areas, the supplementary motor regions and left primary sensor- imotor cortices. Fig. 2C shows the time evolution of this network, represented as the corresponding trial averaged independent component in ̅Xj. Time values on the x-axis represent the centre of the window with respect to the button press, which was at time zero and is shown by the vertical line. The grey area represents the null distribution generated by randomising the trial start times ( ̃Xonset). Signiﬁcant modulation of connectivity occurs during the task; although this begins 3 s prior to the button press, recall that the window size used was 6 s, generating inherent temporal uncertainty. Fig. 2D mirrors the results in Fig. 2C, but the grey area shows the empirical null distribution derived using the ‘sign-ﬂip’ analysis ( ̃Xflip). Again, the black line represents the average re- sponse across all subjects, and the grey distribution is the 95th percentile threshold for the null distribution. Overall, it is clear Fig. 3 shows the results of our method applied to the Sternberg dataset. Clearly, the increased cognitive load evoked by the Sternberg tasks elicits changes in a greater number of brain net- works, and this is shown by 9 of the 10 networks derived de- monstrating signiﬁcant task induced modulation. Fig. 3. Results In what follows, we demonstrate the utility of our method in real MEG data, however our methodology was also tested in si- mulation. These results can be found in the Appendix and G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 5 5 Fig. 2. Results of the self-paced experiment. A) Matrix representation of the network; the ordering of the 78 AAL regions is overlaid. Note that the values in the matrix are the ICA derived mixing coefﬁcients. B) 3D representation of the same network, thresholded for visualisation. Lines show connections, with thicker lines indicating stronger connections. Circles represent the summed magnitude of connectivity between that region and the rest of the brain. C) Time evolution of the network during the self-paced ﬁnger movement task, averaged across trials in all subjects (black line). Time represents the position of the centre of the 6 s window, relative to the button press at t ¼ 0 s The grey shaded region represents the null distribution based on a hypothesis that the response is not time locked to the button press. D) Sign-ﬂip analysis, again showing the mean response across all 15 subjects (black line). The grey shaded area represents the null distribution based on a null hypothesis that the modulation is driven by a small number of subjects. Signiﬁcance (pcorrectedo0.05) is attributed if the black line appears outside the null distribution in both C and D. Note that the network clearly represents the primary somatosensory, motor and supplementary motor regions and demonstrates signiﬁcant modulation with the task. (An interactive version of this Figure can be found at http://nottingham.ac.uk/ppzgo/ica_nets.). Fig. 2. Results of the self-paced experiment. A) Matrix representation of the network; the ordering of the 78 AAL regions is overlaid. Note that the values in the matrix are the ICA derived mixing coefﬁcients. B) 3D representation of the same network, thresholded for visualisation. Lines show connections, with thicker lines indicating stronger connections. Circles represent the summed magnitude of connectivity between that region and the rest of the brain. C) Time evolution of the network during the self-paced ﬁnger movement task, averaged across trials in all subjects (black line). Time represents the position of the centre of the 6 s window, relative to the button press at t ¼ 0 s. 3. Results The grey shaded region represents the null distribution based on a hypothesis that the response is not time locked to the button press. D) Sign-ﬂip analysis, again showing the mean response across all 15 subjects (black line). The grey shaded area represents the null distribution based on a null hypothesis that the modulation is driven by a small number of subjects. Signiﬁcance (pcorrectedo0.05) is attributed if the black line appears outside the null distribution in both C and D. Note that the network clearly represents the primary somatosensory, motor and supplementary motor regions and demonstrates signiﬁcant modulation with the task. (An interactive version of this Figure can be found at http://nottingham.ac.uk/ppzgo/ica_nets.). Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i supplementary material. 3 is laid out such that the columns represent: (A) a 3D network visualisation, (B) the average timecourse (19 subjects) alongside a null dis- tribution based upon ̃Xonset and (C) the average timecourse alongside a null distribution based upon ̃Xflip. The separate rows (I through IX) show the 9 networks which modulate signiﬁcantly. Unsurprisingly given the visual nature of the task, the four networks showing early task modulation all involve the visual areas. These are shown in rows I to IV of Fig. 3. Speciﬁcally, row I depicts a primary visual network whose connectivity increases during presentation of the two example stimuli (and also during the probe). Rows II and III show left and right lateralised con- nections between the primary visual areas and tempero-parietal regions, with both networks exhibiting an early increase in con- nectivity peaking immediately before presentation of the example stimuli. Row IV shows a visual to right motor cortex connection, G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 6 Fig. 3. Results of the Sternberg experiment. The separate columns show A) 3D network visualisation. B) The average timecourse across 19 subjects with null distribution based on randomised trial start times. C) Equivalent to B but null distribution based upon sign ﬂipping. Rows I to IX show the 9 networks which modulate signiﬁcantly with the task, including I) primary visual; II) Visual to left tempero-parietal; III) Visual to right tempero-parietal; IV) Visuomotor V) Somantic; VI) Language; VII) Reﬁned Visual to left tempero-parietal; VIII) Reﬁned visuomotor; IX) Sensorimotor. Note how the timings allow a temporal sequence of network involvement to be deduced. (An interactive version of this Figure can be found at http://nottingham.ac.uk/ppzgo/ica_nets.). Fig. 3. Results of the Sternberg experiment. The separate columns show A) 3D network visualisation. B) The average timecourse across 19 subjects with null distribution based on randomised trial start times. C) Equivalent to B but null distribution based upon sign ﬂipping. Rows I to IX show the 9 networks which modulate signiﬁcantly with the task, including I) primary visual; II) Visual to left tempero-parietal; III) Visual to right tempero-parietal; IV) Visuomotor V) Somantic; VI) Language; VII) Reﬁned Visual to left tempero-parietal; VIII) Reﬁned visuomotor; IX) Sensorimotor. Note how the timings allow a temporal sequence of network involvement to be deduced. (An interactive version of this Figure can be found at http://nottingham.ac.uk/ppzgo/ica_nets.). supplementary material. Row VII shows a reﬁned visual to tem- poral and parietal network, similar to that in III but this time peaking around the time of the probe stimulus. Row VIII again shows a visual to motor connection (similar to IV), and ﬁnally row IX shows the sensorimotor network which becomes most strongly connected around the time of the button press response (in agreement with our result in Fig. 2). It is noteworthy that the brain regions implicated in these networks incorporate the primary sensory cortices, association areas, and cognitive networks that supplementary material. would be associated with semantic processing, pattern recognition and verbalisation, and so these networks are plausible given the task. This is addressed further in our discussion. which demonstrates a signiﬁcant drop in connectivity during presentation of the example stimuli. Transient networks forming in later task phases are shown in rows V to IX. Row V shows a breakdown in connectivity during the task maintenance phase within a bilateral parietal, temporal and frontal network. Inter- estingly, this network captures some areas associated with the default mode network whose activity is known to decrease with a cognitive task. However, the network also captures areas asso- ciated with semantic processing and is thus termed the semantic network. Row VI highlights a left lateralised network that in- corporates regions of temporal, parietal and frontal cortex. The regions implicated are strongly associated with the production of language as well as shape and pattern recognition; this is con- sistent with peaks in connection strength occurring during pre- sentation of the stimuli. Row VII shows a reﬁned visual to tem- poral and parietal network, similar to that in III but this time peaking around the time of the probe stimulus. Row VIII again shows a visual to motor connection (similar to IV), and ﬁnally row IX shows the sensorimotor network which becomes most strongly connected around the time of the button press response (in agreement with our result in Fig. 2). It is noteworthy that the brain regions implicated in these networks incorporate the primary sensory cortices, association areas, and cognitive networks that which demonstrates a signiﬁcant drop in connectivity during presentation of the example stimuli. Transient networks forming in later task phases are shown in rows V to IX. Row V shows a breakdown in connectivity during the task maintenance phase within a bilateral parietal, temporal and frontal network. Inter- estingly, this network captures some areas associated with the default mode network whose activity is known to decrease with a cognitive task. However, the network also captures areas asso- ciated with semantic processing and is thus termed the semantic network. Row VI highlights a left lateralised network that in- corporates regions of temporal, parietal and frontal cortex. The regions implicated are strongly associated with the production of language as well as shape and pattern recognition; this is con- sistent with peaks in connection strength occurring during pre- sentation of the stimuli. Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016 euroimage.2016.08.061i Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks usi http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i 4. Discussion However this 73 s resolution allows us to investigate the functional connections which evolve at an in- termediate scale between the infra-slow connection evolution and the millisecond scale; this scale remains signiﬁcantly higher than would be possible using techniques such as fMRI where a 6 s window would not facilitate sufﬁcient data capture to accurately deﬁne con- nectivity. With these two considerations in mind it proves instructive to discuss the primary results of our method applied to the two da- tasets used. Fig. 2 shows clearly that a network of brain connections involving primary somatosensory and motor cortices, as well as supplementary motor areas, can be identiﬁed based upon our self- paced ﬁnger movement task. Furthermore, this network of connec- tions modulates signiﬁcantly with the button press. Although simple, this result conﬁrms the validity of our method by depicting clearly the primary sensorimotor and motor planning regions. The fact that only one other network (visual) modulates signiﬁcantly with the task also helps to verify that the statistical method used is capable of re- jecting those networks that do not show task modulation. The applications of these methods in the clinical domain are promising. It is well known that neural oscillations, upon which these connectivity metrics are based, are perturbed in a wide variety of developmental, psychotic and degenerative disorders. Similarly the efﬁcacy of aspects of cognition such as working memory are also signiﬁcantly reduced in many patients. It follows that the time-evolving networks of functional connectivity derived in the present paper may differ between control and patient groups, and such ﬁndings might offer a novel means to understand the neural substrates underlying cognitive decline in disease. Fu- ture studies will therefore likely be able to employ the metho- dology presented here to highlight dysconnectivity in disorders such as schizophrenia, where abnormal recruitment of brain re- gions might be expected. In the Sternberg task, the formation of networks encompassing visual (Fig. 3,I) and sensorimotor (IX) regions is consistent with the presentation of visual stimuli and execution of the motor response (Metzak et al., 2011; Woodward et al., 2013; Yamashita et al., 2015). Nodes in the occipital lobe typically include a lateral component which supports the notion that lateral occipital cortex (LOC) is specialised for object shape recognition (Corbetta et al., 1991; Grill-Spector et al., 2001; Haxby et al., 1994; Kourtzi and Kanwisher, 2001). 4. Discussion (2015) in a similar (abstract shape based) working memory task. These authors also considered a verbalisation strategy as the likely interpretation. Finally, two net- works (IV & VIII) show ipsilateral motor connectivity with an extended network of occipital and parietal nodes. This is unusual considering the expected motor response would be in the contralateral hemi- sphere. However, the 4Hz-30Hz frequency band used encompassed alpha and beta oscillations and it is possible that, to suppress ipsi- lateral motor activity, alpha oscillations are increased (Brinkman et al., 2014). Overall, the transient networks induced by the Sternberg task are plausible given the previous literature on working memory and sensory processes. The results generated by our method are of signiﬁcant neu- roscientiﬁc interest and warrant further discussion. However prior to this, two key points regarding the method should be understood: Firstly, the timecourses shown in Figs. 2 and 3 depict increases and decreases in connectivity. In other words, the peaks refer to points in time when two or more regions deﬁning a network are most corre- lated. Just because regions are not correlated at some particular point in time, does not necessarily mean that those regions are not engaged by the task. This is an important point since many of the regions implicated by our networks are likely to be engaged constantly throughout the Sternberg task, but may only connect to wider net- works at speciﬁc points in time. Second, recall that there is inherent temporal smoothness in the method. Despite the excellent temporal resolution of MEG, a reasonable data window is required in order to derive reliably each individual adjacency matrix Ri (see also below). Here we employ a 6 s window width, meaning that features in a timecourse have an inherent temporal uncertainty of 73 s. This means that, for example in the self-paced motor task where con- nectivity appears to increase before the button press, there is a degree of ambiguity; this could be representative of preparatory effects, or could result simply from the limited temporal resolution of the method. This temporal resolution is lower than other MEG based connectivity techniques, for example the Hidden Markov model in- troduced by Baker et al. (2014). 4. Discussion Other net- works encompass areas thought to be responsible for the higher level cognition required for successful completion of the Sternberg task. The Angular Gyrus (AG) is particularly evident in the majority of these networks. Structurally this region has been identiﬁed as a centrally connected hub serving multiple sub-networks. This hub has also been identiﬁed functionally in a variety of task-positive contexts ranging from semantic processing to numerical calculation. A uniﬁed account of AG function is presented by Seghier (2013) who suggests that the AG is an integration site receiving input from sensory (Demonet et al., 1992; Vandenberghe et al., 1996), memorial (Geschwind, 1965) and higher-level nodes. We speculate that the extent of our higher order networks is in agreement with this model of AG function. Notably, the dorso-lateral pre-frontal cortex (DLPFC) is recruited in network V, connecting bilaterally with the AG. The left and right DLPFC are well established in the literature as controlling executive-attention function 4. Discussion This paper has introduced a novel ICA based method which, when applied to MEG data, allows characterisation of transiently forming and dissolving electrophysiological networks in the brain, at time-scales much faster than could be achieved using fMRI. Previous MEG-ICA-network approaches typically look for brain regions whose activity, measured as a function of time, covaries. Here distinct from this, we measure the temporal evolution of functional connectivity between regions and use temporal ICA to cluster together connections that share similar temporal proﬁles. In this way, we identify networks of connections whose temporal dynamics covary, with no prior assumptions. This allows us to identify where and when signiﬁcant modulations in connectivity occur, We have veriﬁed our method in simulation (see appendix and supplementary material) and using a simple ﬁnger movement task. Moreover, we have shown that our method allows generation of a unique picture of cognitive processing, showing the formation G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 7 and dissolution of multiple brain networks required to allow subjects to complete a working memory task. in working memory (Barbey et al., 2013; Kane and Engle, 2002), with the right DLPFC being shown to be sensitive to shape in particular (Nystrom et al., 2000). This network also incorporates bilateral inferior temporal gyri, regions considered important for semantic processing (Vigneau et al., 2006). This leads us to name this network as a ‘se- mantic network’. This network was also seen in a recent study by Shine et al. (2015), who saw the connectivity between the DLPFC and ventral visual regions vary with cognitive load in a working memory task. A second cognitive network (VI) has been termed a ‘language network’. Although stimuli were abstract shapes, participant feedback suggests a ‘naming’ strategy was used in the majority of cases. If a verbalization strategy was employed by the participants to aid in memory encoding, then nodes of the language network may be im- plicated. Indeed, this left lateralised network is anchored in the AG with extensions to the inferior frontal gyrus (IFG), inferior temporal gyrus and a number of nodes spanning the inferior to superior pre- central gyrus. These regions are consistent with previous accounts of semantic cognition (Binder et al., 1997; Demb et al., 1995; Derrfuss et al., 2004; Kang et al., 1999; Vigneau et al., 2006). Furthermore, this effect was also seen by Caminiti et al. Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), image.2016.08.061i Methodological considerations In the present work we selected 10 components based on our previous experience, al- though varying this parameter in our current work made little difference to the overall results. Further, here we select which independent components to keep based upon those networks which modulate signiﬁcantly with the task. However, just because a network does not modulate with the task does not necessarily mean that this network is not genuinely representative of con- nectivity. Future work should therefore seek other methods to determine validity of networks, particularly if the present algo- rithm was to be used for resting state investigation. et al., 2016). However, other methods (for example those based on ﬁxed phase measurements between regions) are available; these should not be considered competitor techniques but rather ways to probe a different type of functional connectivity (Scholvinck et al., 2013). For example, using a time-varying multivariate au- toregressive model, it has been demonstrated that task-dependent brain states can be identiﬁed in a ﬁnger tapping task, and corre- spond to unique cross-spectral (i.e. coherence) patterns (Vidaurre et al., 2016). Although at present this method this is limited (computationally) to pairs of brain areas, whereas our method in this paper is whole-brain. The two methods may be combined in the future. Indeed, the adjacency matrices derived in our metho- dology could easily be substituted for similar adjacency matrices derived using any alternative metric (assuming sufﬁciently high signal to noise ratio), and transient networks probed. We note that there is signiﬁcant variability in the timecourse of connectivity across subjects. This is demonstrated in null dis- tributions formed based upon randomized sign ﬂip in half of the subjects; i.e. if connectivity change was equal in all subjects, the sign ﬂip would perfectly cancel any task response, and no variation over time would be seen in the in null distributions (grey shaded regions in Figs. 2D and 3C). The fact that null distributions follow, to a degree, the genuine timecourses shows, in the Sternberg and self-paced experiments, a marked variation in connectivity proﬁle over subjects. In fact, relatively poor within and between subject reliability of (static) MEG connectivity measurements has been shown previously. For example, Wens et al. (2014a) show that whilst group level static connectivity within several well-known distributed networks is stable, there is signiﬁcant variability at the individual subject level. Similarly Colclough et al. Methodological considerations (2016) tested the cross session repeatability of a large number of static functional connectivity measurements, showing clearly that although group level inference is reliable, network metrics can be very variable across individuals. In addition, Tewarie et al. (2016) used MEG networks to predict those observed in fMRI; whilst predictions were robust at the group level, they fared less well within in- dividuals. Interestingly, these variations across subjects may not be due to stochastic noise, but rather identiﬁable intrinsic processes which are subject speciﬁc (Finn et al., 2015). Given these previous ﬁndings of large inter-individual differences in static connectivity, it is not surprising that dynamic functional connectivity metrics presented here also exhibit relatively high inter-individual differ- ences. There are a number of possible explanations for this. Firstly, our measurement of connectivity itself (i.e. the dynamic adjacency matrices) are based only on 6 s of unaveraged MEG data. Given the relatively low SNR of MEG data it is possible that reliability is only realised with large quantities of data – hence the requirement for large subject cohorts. Second, source localisation could affect the robustness of connectivity; here we use beamforming alongside the AAL atlas, a technique well established by previously published work. However, a limitation is that if a speciﬁc region, e.g. left motor cortex, is mislocalised (e.g. due to a poor forward model in one subject) then the signal derived would no longer be re- presentative of that region. This potential confound would add markedly to variability over subjects. Thirdly, the reliability of the amplitude envelope correlation metric itself could be questioned. However, Colclough et al. (2016) showed that of all of the MEG based connectivity metrics, AEC fared well in terms of robustness over repeated measures. Finally, this variability could genuinely reﬂect the variability across individual subjects in terms of the neural network mechanisms used to carry out the tasks under- taken. Ultimately, if techniques like the one presented here are to be useful clinically, then we must derive means to ensure their robustness in individuals. Further effort is thus needed in this area. In addition to parameter selection, there are three other core components of the method that warrant discussion; namely, the choice of cortical parcellation, the underlying source space pro- jection method, and the choice of connectivity metric. First, re- garding the AAL parcellation, this was chosen based on its suc- cessful use in previous MEG investigations (e.g. Methodological considerations Our algorithm allows detection and characterisation of tran- siently forming task induced electrophysiological networks. In achieving this, two core parameters require setting, the window width (here 6 s) and the number of independent components (here 10). Both warrant further discussion. A judicious selection of window width is important, and represents a trade-off between temporal resolution and the accuracy of the derived adjacency matrices. Here, separate elements of the adjacency matrices are based upon temporal correlation of envelope signals within the window. It is well known that the accuracy of correlation between two variables (r) relates to the number of degrees of freedom (η) in the underlying data; speciﬁcally if one assumes no underlying genuine correlation between two timecourses then standard de- viation of correlation, σ η ( ) = r 1/ ; i.e. the variability (noise) in- herent in the adjacency tensor is increased as η is decreased. Further, the number of degrees of freedom in a windowed envel- ope timecourse is unrelated to the number of sample points (or sampling frequency). In fact, Fourier theory shows that for G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 8 envelope data, an upper limit on degrees of freedom is given by η δ = Bw , where δ is the window width and Bw represents band- width of the carrier signal (i.e. for a 13-30 Hz beta envelope, = B Hz 17 w ). This means that σ δ ( ) = r B 1/ w ; in other words, ad- jacency matrix noise is increased by either reducing bandwidth of the carrier signal, or the window width. Typically, bandwidth is set by the scientiﬁc question to be asked (e.g. one might be interested in beta band networks, such as in the self-paced motor study), and therefore δ must be set to reduce the random noise to an accep- table level. Here σ ( ) = r 0.1 for the self-paced data and σ ( ) = r 0.08 for the Sternberg data, which was deemed acceptable. Future studies should bear this calculation in mind. The selection of the number of independent components is less well prescribed; this is not a limitation of our algorithm directly, but rather is a funda- mental question for all ICA methodologies. Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i l, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), oimage.2016.08.061i 5. Conclusion adjacency matrix; 3D visualisations are shown in Fig. A1B. A timecourse of connectivity for each network was also generated; these comprised 60 minutes of periodically occurring Hanning windows. Amplitudes were set to 0.45, the full width half max- imum was 6 seconds, and onsets were set to the 3rd, 18th, 33rd and 48th second of every minute. A single trial was deﬁned as 1 minute in length. The trial averaged simulated data are shown in Fig. A1A (left hand panel). The timecourses were multiplied by their respective network adjacency matrices to create Ssim. The interference tensor, Isim was generated by source reconstructing 60 minutes of recorded empty room MEG data onto a simulated brain geometry, and measuring amplitude envelope connectivity as described in our methods section. This generated structured (but spurious) networks of inference. Both Ssim and Ssim were sampled at 2 Hz to reﬂect the analysis pipeline described above. The two tensors were combined thus, The characterisation of dynamic electrophysiological brain networks, which form and dissolve in order to support ongoing cognitive function, is one of the most important challenges in neuroscience. Here, we introduce an ICA based method for mea- suring such networks in the human brain using MEG. Previous MEG-ICA network analyses look for brain regions that share a common temporal proﬁle of activity. Here distinctly, we measure the temporal evolution of connectivity between region pairs and use ICA to identify clusters of connections that share an in- dependent temporal proﬁle. The validity of our method was de- monstrated in simulation and in a self-paced ﬁnger movement paradigm, showing that a sensorimotor network can be dis- tinguished. The broader applicability of our method was demon- strated by its application to a Sternberg task. We have shown that our method allows generation of a unique picture of cognitive processing, showing the formation and dissolution of the brain networks required to allow subjects to complete the task. This represents a signiﬁcant step forward in the characterisation of brain network connectivity and will prove to be a key tool in the future investigation of healthy brain networks, and their break- down in a variety of pathological conditions. ) ( = + + + ( ) R S 1 I 1 SNR , A1 sim sim 1 SNR sim ( ) A1 where SNR represents the effective signal to interference ratio of the simulated data (see supplementary material for details). G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 9 9 5. Conclusion In order to test the effectiveness of ICA at extracting the net- work time courses in the presence of interference, the SNR para- meter was allowed to vary and a temporal ﬁgure of merit derived to characterise how well a single component represents a simu- lated time course. This was deﬁned as the correlation between a simulated network and its best matching independent component, minus the mean correlations between the reconstructed time- course and all other simulated components. This value ranges between 0 and 1. Note that the ﬁgure of merit penalises both poor representation and degeneracy, i.e. a low value is computed if a single component represents multiple simulated networks. Appendix. Validation by simulation Fig. A1A shows the trial averaged independent components from of 4 out of 7 reconstructed components. The associated spatial patterns are shown in Fig. A1B. Note strong temporal and spatial resemblance to the simulated connectivity which is also reﬂected by ﬁgure of merit scores of 0.84, 0.91, 0.84 and 0.94 for the 4 simulated networks. Fig. A1C shows the results of varying the SNR of the simulated data be- tween 0 and 2.5 in steps of 0.02. 10 simulations were run for each SNR value. As would be expected, at high SNR the ﬁgure of merit scores are high, meaning that our simulated networks are reconstructed faith- fully. However, a sharp transition below a minimum ‘threshold’ SNR is observed below which simulated networks are irretrievable. Interest- ingly, a different threshold value can be observed clearly for each network, and this relates to the number of connections which exist in a network; the more connections present, the lower the SNR that can be tolerated. This is reﬂected in Fig. A1D, where the threshold (deﬁned as the SNR which generates a ﬁgure of merit score of 0.5) is plotted against the Frobenius norm of the network adjacency matrices (which here indirectly represents number of connections). Fig. A1D shows clearly that SNR is a monotonically decreasing function of norm, making the important point that networks with large spatial extent (lots of connections) can be reconstructed using lower SNR data. In other words, our method will preferentially reconstruct networks with larger spatial extent; future applications of this method should bear this potential confound in mind. The validation of our methodology provided in this paper is centred upon its application to real MEG data. However, it also proves instructive to examine the performance of our methodol- ogy in simulation. The performance of envelope correlation as a means to examine connectivity in short time windows has been addressed extensively in previous work (Brookes et al., 2014; O’Neill et al., 2015b) and will not be repeated here. However, the ability of ICA, applied to timecourses of connectivity, to extract the spatiotemporal signatures of multiple brain networks has not yet been validated. In what follows, we test the extent to which ICA characterises a set of simulated networks which are obfuscated by noise. Here, key ﬁndings are presented; a full description of the methods and extended results can be found in supplementary material. Acknowledgements This work was funded by a Medical Research Council (MRC) New Investigator Research Grant (MR/M006301/1) awarded to MJB and an MRC studentship awarded to GCO. We also acknowl- edge Medical Research Council Partnership Grant (MR/K005464/ 1). MWW is funded by the Wellcome Trust and supported by the National Institute for Health Research (NIHR) Oxford Biomedical Research Centre based at Oxford University Hospitals Trust (the views expressed are those of the author(s) are not necessarily those of the NHS, the NIHR or the Department of Health). Methodological considerations Tewarie et al., 2016). However, our method could be used with any cortical par- cellation, provided that the number of regions is sufﬁciently low, and those regions are sufﬁciently well separated to ensure that the windowed data matrices, Q i, are of full rank. (This is a requirement of the orthogonalisation procedure (Colclough et al., 2015).) It is noteworthy that the separate AAL regions vary markedly in size; our use of a single point location, based on the centre of mass of the region, may therefore mean that some regions are better re- presented than others. This suggests that brain regions that are poorly represented may be missing from the networks shown. For example, one would expect that areas in the ventral visual path- way (e.g., fusiform gyrus) to involved in our Sternberg task. However, they were not core to any of the networks shown. A likely reason is that they are missed by the cortical parcellation and single point (centre of mass) representation. The future use of brain parcellations based on functional MRI (Craddock et al., 2012), MEG, multmodal (Glasser et al., 2016), or even a-priori (literature based) knowledge of brain regions involved in a task may there- fore prove instructive. Secondly, for source localisation, we chose to employ a beamformer spatial ﬁltering procedure. Beamforming is a popular method of inverse solution and has been shown previously to be particularly useful in the characterisation of neural oscillations. Further, beamforming has been used success- fully in the measurement of static (Brookes et al., 2011a) and dy- namic (Baker et al., 2014) functional connectivity. The reasons for the success of this algorithm in such studies has been addressed at length in previous papers, and will not be repreated here. How- ever, we point out that other inverse solutions could be sub- stituted for beamforming in the present processing pipeline, and would likely generate similar results. Thirdly, we chose an envel- ope correlation procedure as our estimator of functional con- nectivity between regions. This procedure has been successful in elucidating electrophysiological networks of functional con- nectivity (Colclough et al., 2016), particularly in the study of the electrophysiological basis of haemodynamic networks (Tewarie Simulation Methods We simulated an adjacency tensor Rsim based on a weighted combination of “true” connectivity (Ssim) and interference (Isim). To generate Ssim, four spatially distinct networks were constructed based on a previous study (Brookes et al., 2015). The spatial pat- terns of connectivity, reﬂecting visual, sensorimotor, superior frontal and fronto-parietal networks, were each represented by an eill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), uroimage.2016.08.061i Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i References Bola, M., Sabel, B.A., 2015. Dynamic reorganization of brain functional networks during cognition. Neuroimage 114, 398–413. 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc. Natl. Acad. Sci. U S A 102, 9673–9678. Friston, K.J., 1994. Functional and effective connectivity in neuroimaging: A synthesis. Human. Brain Mapp. 2, 56–78. Friston, K.J., 1998. The disconnection hypothesis. Schizophr. Res 30, 115–125. Geschwind, N., 1965. Disconnexion syndromes in animals and man. I. Brain 88, 237–294. functional networks. Proc. Natl. Acad. Sci. U S A 102, 9673–9678. Friston, K.J., 1994. Functional and effective connectivity in neuroimaging: A synthesis. Human. Brain Mapp. 2, 56–78. Brinkman, L., Stolk, A., Dijkerman, H.C., de Lange, F.P., Toni, I., 2014. Distinct Roles for Alpha- and Beta-Band Oscillations during Mental Simulation of Goal-Di- rected Actions. J Neurosci. 34, 14783–14792. y pp , Friston, K.J., 1998. The disconnection hypothesis. Schizophr. Res 30, 115–125. Geschwind, N., 1965. Disconnexion syndromes in animals and man. I. Brain 88, 237–294. , J , yp p , Geschwind, N., 1965. Disconnexion syndromes in animals and man. I. Brain 88, 237–294. Brookes, M.J., Hale, J.R., Zumer, J.M., Stevenson, C.M., Francis, S.T., Barnes, G.R., Owen, J.P., Morris, P.G., Nagarajan, S.S., 2011a. Measuring functional con- nectivity using MEG: methodology and comparison with fcMRI. Neuroimage 56, 1082–1104. Glasser, M.F., Coalson, T.S., Robinson, E.C., Hacker, C.D., Harwell, J., Yacoub, E., Ugurbil, K., Andersson, J., Beckmann, C.F., Jenkinson, M., Smith, S.M., Van Essen, D.C., 2016. A multi-modal parcellation of human cerebral cortex. Nature. Brookes, M.J., Hall, E.L., Robson, S.E., Price, D., Palaniyappan, L., Liddle, E.B., Liddle, P. F., Robinson, S.E., Morris, P.G., 2015. Complexity measures in magnetoence- phalography: measuring “disorder” in schizophrenia. PLoS One 10, e0120991. Gong, G., He, Y., Concha, L., Lebel, C., Gross, D.W., Evans, A.C., Beaulieu, C., 2009. Mapping anatomical connectivity patterns of human cerebral cortex using in vivo diffusion tensor imaging tractography. Cereb. Cortex 19, 524–536. p g p y g p , Brookes, M.J., Liddle, E.B., Hale, J.R., Woolrich, M.W., Luckhoo, H., Liddle, P.F., Morris, Brookes, M.J., Liddle, E.B., Hale, J.R., Woolrich, M.W., Luckhoo, H., Liddle, P.F., Morris, P.G., 2012a. Task induced modulation of neural oscillations in electro- physiological brain networks. Neuroimage 63, 1918–1930. Grill-Spector, K., Kourtzi, Z., Kanwisher, N., 2001. The lateral occipital complex and its role in object recognition. Vision. Res 41, 1409–1422. References Brookes, M.J., O’Neill, G.C., Hall, E.L., Woolrich, M.W., Baker, A., Palazzo Corner, S., Robson, S.E., Morris, P.G., Barnes, G.R., 2014. Measuring temporal, spectral and spatial changes in electrophysiological brain network connectivity. Neuroimage 91, 282–299. Guggisberg, A.G., Honma, S.M., Findlay, A.M., Dalal, S.S., Kirsch, H.E., Berger, M.S., Nagarajan, S.S., 2008. Mapping functional connectivity in patients with brain lesions. Ann. Neurol. 63, 193–203. Hassan, M., Benquet, P., Biraben, A., Berrou, C., Dufor, O., Wendling, F., 2015. Dy- namic reorganization of functional brain networks during picture naming. Cortex 73, 276–288. Brookes, M.J., Tewarie, P.K., Hunt, B.A., Robson, S.E., Gascoyne, L.E., Liddle, E.B., Liddle, P.F., Morris, P.G., 2016. A multi-layer network approach to MEG con- nectivity analysis. Neuroimage 132, 425–438. Haxby, J.V., Horwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P., Grady, C.L., 1994. The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. J Neurosci. 14, 6336–6353. Brookes, M.J., Vrba, J., Robinson, S.E., Stevenson, C.M., Peters, A.M., Barnes, G.R., Hillebrand, A., Morris, P.G., 2008. Optimising experimental design for MEG beamformer imaging. Neuroimage 39, 1788–1802. Hillebrand, A., Barnes, G.R., Bosboom, J.L., Berendse, H.W., Stam, C.J., 2012. Fre- quency-dependent functional connectivity within resting-state networks: an atlas-based MEG beamformer solution. Neuroimage 59, 3909–3921. Brookes, M.J., Woolrich, M., Luckhoo, H., Price, D., Hale, J.R., Stephenson, M.C., Barnes, G.R., Smith, S.M., Morris, P.G., 2011b. Investigating the electro- physiological basis of resting state networks using magnetoencephalography. Proc. Natl. Acad. Sci. U S A 108, 16783–16788. Hipp, J.F., Hawellek, D.J., Corbetta, M., Siegel, M., Engel, A.K., 2012. Large-scale cortical correlation structure of spontaneous oscillatory activity. Nat. Neurosci. 15, 884–890. Brookes, M.J., Woolrich, M.W., Barnes, G.R., 2012b. Measuring functional con- nectivity in MEG: a multivariate approach insensitive to linear source leakage. Neuroimage 63, 910–920. Hipp, J.F., Siegel, M., 2015. BOLD fMRI Correlation Reﬂects Frequency-Speciﬁc Neuronal Correlation. Curr. Biol. 25, 1368–1374. g Caminiti, S.P., Siri, C., Guidi, L., Antonini, A., Perani, D., 2015. The neural correlates of spatial and object working memory in elderly and Parkinson’s disease subjects. Behav. Neurol. 2015, 123636. Huang, M.X., Mosher, J.C., Leahy, R.M., 1999. A sensor-weighted overlapping-sphere head model and exhaustive head model comparison for MEG. Phys. Med. Biol. 44, 423–440. Chang, C., Glover, G.H., 2010. Time-frequency dynamics of resting-state brain connectivity measured with fMRI. Neuroimage 50, 81–98. Hunt, L.T., Kolling, N., Soltani, A., Woolrich, M.W., Rushworth, M.F., Behrens, T.E., 2012. Mechanisms underlying cortical activity during value-guided choice. Nat. Neurosci. References de Pasquale, F., Della Penna, S., Sporns, O., Romani, G.L., Corbetta, M., 2015. A Dy- namic Core Network and Global Efﬁciency in the Resting Human Brain. Cereb Cortex. Allen, E.A., Damaraju, E., Plis, S.M., Erhardt, E.B., Eichele, T., Calhoun, V.D., 2014. Tracking whole-brain connectivity dynamics in the resting state. Cereb. Cortex 24, 663–676. Demb, J.B., Desmond, J.E., Wagner, A.D., Vaidya, C.J., Glover, G.H., Gabrieli, J.D., 1995. Semantic encoding and retrieval in the left inferior prefrontal cortex: a func- tional MRI study of task difﬁculty and process speciﬁcity. J Neurosci. 15, 5870–5878. Baker, A.P., Brookes, M.J., Rezek, I.A., Smith, S.M., Behrens, T., Probert Smith, P.J., Woolrich, M., 2014. Fast transient networks in spontaneous human brain ac- tivity. Elife (Camb.) 3, e01867. Demonet, J.F., Chollet, F., Ramsay, S., Cardebat, D., Nespoulous, J.L., Wise, R., Rascol, A., Frackowiak, R., 1992. The anatomy of phonological and semantic processing in normal subjects. Brain 115 (Pt 6), 1753–1768. Baker, A.P., Luckhoo, H., Brookes, M.J., Smith, P., Woolrich, M., 2012. Investigating the Temporal Dynamics of Resting State Brain Connectivity using Magne- toencephalography. Organization of Human Brain Mapping, Beijing, China. B b A K K i M G f J 2013 D l l f l ib i Derrfuss, J., Brass, M., von Cramon, D.Y., 2004. Cognitive control in the posterior frontolateral cortex: evidence from common activations in task coordination, interference control, and working memory. Neuroimage 23, 604–612. p g p y g pp g j g Barbey, A.K., Koenigs, M., Grafman, J., 2013. Dorsolateral prefrontal contributions to human working memory. Cortex 49, 1195–1205. Finn, E.S., Shen, X., Scheinost, D., Rosenberg, M.D., Huang, J., Chun, M.M., Papade- metris, X., Constable, R.T., 2015. Functional connectome ﬁngerprinting: identi- fying individuals using patterns of brain connectivity. Nat. Neurosci. 18, 1664–1671. Beckmann, C.F., DeLuca, M., Devlin, J.T., Smith, S.M., 2005. Investigations into resting-state connectivity using independent component analysis. Philos. Trans. R Soc. Lond. B Biol. Sci. 360, 1001–1013. Binder, J.R., Frost, J.A., Hammeke, T.A., Cox, R.W., Rao, S.M., Prieto, T., 1997. Human brain language areas identiﬁed by functional magnetic resonance imaging. J Neurosci. 17, 353–362. Fox, M.D., Raichle, M.E., 2007. Spontaneous ﬂuctuations in brain activity observed with functional magnetic resonance imaging. Nat. Rev. Neurosci. 8, 700–711. Fox, M.D., Snyder, A.Z., Vincent, J.L., Corbetta, M., Van Essen, D.C., Raichle, M.E., 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc. Natl. Acad. Sci. U S A 102, 9673–9678. References 15 (470-476), S471–S473. Chang, C., Liu, Z., Chen, M.C., Liu, X., Duyn, J.H., 2013. EEG correlates of time-varying BOLD functional connectivity. Neuroimage 72, 227–236. Hutchison, R.M., Womelsdorf, T., Allen, E.A., Bandettini, P.A., Calhoun, V.D., Corbetta, M., Della Penna, S., Duyn, J.H., Glover, G.H., Gonzalez-Castillo, J., Handwerker, D. A., Keilholz, S., Kiviniemi, V., Leopold, D.A., de Pasquale, F., Sporns, O., Walter, M., Chang, C., 2013. Dynamic functional connectivity: promise, issues, and in- terpretations. Neuroimage 80, 360–378. Clare, S., Humberstone, M., Hykin, J., Blumhardt, L.D., Bowtell, R., Morris, P., 1999. Detecting activations in event-related fMRI using analysis of variance. Magn. Reson Med. 42, 1117–1122. Cohen, D., 1968. Magnetoencephalography: evidence of magnetic ﬁelds produced by alpha-rhythm currents. Science 161, 784–786. Hyvarinen, A., 1999. Fast and robust ﬁxed-point algorithms for independent com- ponent analysis. IEEE Trans. Neural Netw. 10, 626–634. Cohen, D., 1972. Magnetoencephalography: detection of the brain’s electrical ac- tivity with a superconducting magnetometer. Science 175, 664–666. Jenkinson, M., Beckmann, C.F., Behrens, T.E., Woolrich, M.W., Smith, S.M., 2012. Fsl. Neuroimage 62, 782–790. Colclough, G.L., Brookes, M.J., Smith, S.M., Woolrich, M.W., 2015. A symmetric multivariate leakage correction for MEG connectomes. Neuroimage 117, 439–448. Kane, M.J., Engle, R.W., 2002. The role of prefrontal cortex in working-memory capacity, executive attention, and general ﬂuid intelligence: an individual-dif- ferences perspective. Psychon. Bull. Rev. 9, 637–671. Colclough, G.L., Tewarie, P.K., Brookes, M.J., Smith, S.M., Woolrich, M.W., 2016. How reliable are MEG resting-state connectivity metrics? Neuroimage 138, 284–293. Kang, A.M., Constable, R.T., Gore, J.C., Avrutin, S., 1999. An event-related fMRI study of implicit phrase-level syntactic and semantic processing. Neuroimage 10, 555–561. Karahanoğlu F I Van De Ville D 2015 Transient brain activity disentangles fMRI g, , , , , J , , , y implicit phrase-level syntactic and semantic processing. Neuroimage 10, 555–561. Karahanoğlu, F.I., Van De Ville, D., 2015. Transient brain activity disentangles fMRI resting-state dynamics in terms of spatially and temporally overlapping net- works. Nat. Commun. 6, 7751. Corbetta, M., 1998. Frontoparietal cortical networks for directing attention and the eye to visual locations: identical, independent, or overlapping neural systems? Proc. Natl. Acad. Sci. U S A 95, 831–838. Karahanoğlu, F.I., Van De Ville, D., 2015. Transient brain activity disentangles fMRI resting-state dynamics in terms of spatially and temporally overlapping net- works. Nat. Commun. 6, 7751. Kessler, D., Angstadt, M., Welsh, R.C., Sripada, C., 2014. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 11 Appendix A. Supporting information speed: functional anatomy by positron emission tomography. J Neurosci. 11, 2383–2402. Craddock, R.C., James, G.A., Holtzheimer 3rd, P.E., Hu, X.P., Mayberg, H.S., 2012. A whole brain fMRI atlas generated via spatially constrained spectral clustering. Hum. Brain Mapp. 33, 1914–1928. Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.neuroimage.2016. 08.061. de Pasquale, F., Della Penna, S., Snyder, A.Z., Lewis, C., Mantini, D., Marzetti, L., Belardinelli, P., Ciancetta, L., Pizzella, V., Romani, G.L., Corbetta, M., 2010. Temporal dynamics of spontaneous MEG activity in brain networks. Proc. Natl. Acad. Sci. U S A 107, 6040–6045. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 10 G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ A1. Simulation results. A) The trial averaged simulated timecourses (left) and the trial averaged independent components (right) (SNR ¼ 3). B) 3D visualisations of the ulated (top) and ICA reconstructed (bottom) network topographies (top row). C) Temporal ﬁgure of merit scores plotted as a function of SNR for all 4 simulated networks. e that accurate reconstructions are observed whilst SNR is above a critical threshold level; threshold differs for each network. D) The relation between SNR threshold ﬁned as when the ﬁgure of merit falls below 0.5) and Frobenius norm of the network adjacency matrix. Note that networks with a larger number of connections fare ter in low SNR conditions. Fig. A1. Simulation results. A) The trial averaged simulated timecourses (left) and the trial averaged independent components (right) (SNR ¼ 3). B) 3D visualisations of the simulated (top) and ICA reconstructed (bottom) network topographies (top row). C) Temporal ﬁgure of merit scores plotted as a function of SNR for all 4 simulated networks. Note that accurate reconstructions are observed whilst SNR is above a critical threshold level; threshold differs for each network. D) The relation between SNR threshold (deﬁned as when the ﬁgure of merit falls below 0.5) and Frobenius norm of the network adjacency matrix. Note that networks with a larger number of connections fare better in low SNR conditions. Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), image.2016.08.061i G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ 12 Yacoub, E.S., Ugurbil, K., 2012. Temporally-independent functional modes of t b i ti it P N tl A d S i U S A 109 3131 3136 Matter, and White Matter. J. Neurosci. 34, 16555–16566. Kourtzi Z Kanwisher N 2001 Representation of perceived o Matter, and White Matter. J. Neurosci. 34, 16555–16566. Kourtzi, Z., Kanwisher, N., 2001. Representation of perceived object shape by the human lateral occipital complex. Science 293, 1506–1509. Kourtzi, Z., Kanwisher, N., 2001. Representation of perceived object shape by the human lateral occipital complex. Science 293, 1506–1509. spontaneous brain activity. Proc. Natl. Acad. Sci. U S A 109, 3131–3136. Smith, S.M., Nichols, T.E., Vidaurre, D., Winkler, A.M., Behrens, T.E., Glasser, M.F., , , , , , , , , , , , , Ugurbil, K., Barch, D.M., Van Essen, D.C., Miller, K.L., 2015. A positive-negative mode of population covariation links brain connectivity, demographics and behavior. Nat. Neurosci. 18, 1565–1567. p p Liu, Z., Fukunaga, M., de Zwart, J.A., Duyn, J.H., 2010. Large-scale spontaneous ﬂuctuations and correlations in brain electrical activity observed with mag- netoencephalography. Neuroimage 51, 102–111. Luckhoo, H., Hale, J.R., Stokes, M.G., Nobre, A.C., Morris, P.G., Brookes, M.J., Woolrich, M.W., 2012. Inferring task-related networks using independent component analysis in magnetoencephalography. Neuroimage 62, 530–541. Stufﬂebeam, S.M., Liu, H., Sepulcre, J., Tanaka, N., Buckner, R.L., Madsen, J.R., 2011. Localization of focal epileptic discharges using functional connectivity magnetic resonance imaging. J Neurosurg. 114, 1693–1697. Maldjian, J.A., Davenport, E.M., Whitlow, C.T., 2014. Graph theoretical analysis of resting-state MEG data: Identifying interhemispheric connectivity and the default mode. Neuroimage 96, 88–94. Tewarie, P., Steenwijk, M.D., Tijms, B.M., Daams, M., Balk, L.J., Stam, C.J., Uitdehaag, B.M., Polman, C.H., Geurts, J.J., Barkhof, F., Pouwels, P.J., Vrenken, H., Hillebrand, A., 2014. Disruption of structural and functional networks in long-standing multiple sclerosis. Hum. Brain Mapp. 35, 5946–5961. B.M., Polman, C.H., Geurts, J.J., Barkhof, F., Pouwels, P.J., Vrenken, H., Hillebrand, A., 2014. Disruption of structural and functional networks in long-standing multiple sclerosis. Hum. Brain Mapp. 35, 5946–5961. Metzak, P., Feredoes, E., Takane, Y., Wang, L., Weinstein, S., Cairo, T., Ngan, E.T., Woodward, T.S., 2011. Constrained principal component analysis reveals func- tionally connected load-dependent networks involved in multiple stages of working memory. Hum. Brain Mapp. 32, 856–871. multiple sclerosis. Hum. Brain Mapp. 35, 5946–5961. Tewarie, P.K., Bright, M.G., Hillebrand, A., Robson, S.E., Gascoyne, L.E., Morris, P.G., Meier, J., Van Mieghem, P., Brookes, M.J., 2016. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ Med Biol. 32, 11–22. Wens, V., Bourguignon, M., Goldman, S., Marty, B., Op de Beeck, M., Clumeck, C., Mary, A., Peigneux, P., Van Bogaert, P., Brookes, M.J., De Tiege, X., 2014a. Inter- and Intra-Subject Variability of Neuromagnetic Resting State Networks. Brain Topogr. Schnitzler, A., Gross, J., 2005. Normal and pathological oscillatory communication in the brain. Nat. Rev. Neurosci. 6, 285–296. Scholvinck, M.L., Leopold, D.A., Brookes, M.J., Khader, P.H., 2013. The contribution of electrophysiology to functional connectivity mapping. Neuroimage 80, 297–306. Wens, V., Marty, B., Mary, A., Bourguignon, M., Op de Beeck, M., Goldman, S., Van Bogaert, P., Peigneux, P., De Tiege, X., 2015. A geometric correction scheme for spatial leakage effects in MEG/EEG seed-based functional connectivity map- ping. Hum. Brain Mapp. 36, 4604–4621. Seghier, M.L., 2013. The angular gyrus: multiple functions and multiple subdivi- sions. Neuroscientist 19, 43–61. Sekihara, K., Nagarajan, S.S., Poeppel, D., Marantz, A., 2004. Asymptotic SNR of scalar and vector minimum-variance beamformers for neuromagnetic source reconstruction. IEEE Trans. Biomed. Eng. 51, 1726–1734. Wens, V., Mary, A., Bourguignon, M., Goldman, S., Marty, B., Beeck, M.O., Bogaert, P. V., Peigneux, P., Tiege, X.D., 2014b. About the electrophysiological basis of resting state networks. Clin. Neurophysiol. 125, 1711–1713. Sekihara, K., Sahani, M., Nagarajan, S.S., 2005. A simple nonparametric statistical thresholding for MEG spatial-ﬁlter source reconstruction images. Neuroimage 27, 368–376. Winkler, A.M., Ridgway, G.R., Webster, M.A., Smith, S.M., Nichols, T.E., 2014. Per- mutation inference for the general linear model. Neuroimage 92, 381–397. W d d T S F d E M t k PD T k Y M h D S 2013 E h Shine, J.M., Koyejo, O., Bell, P.T., Gorgolewski, K.J., Gilat, M., Poldrack, R.A., 2015. Estimation of dynamic functional connectivity using Multiplication of Temporal Derivatives. Neuroimage 122, 399–407. Woodward, T.S., Feredoes, E., Metzak, P.D., Takane, Y., Manoach, D.S., 2013. Epoch- speciﬁc functional networks involved in working memory. Neuroimage 65, 529–539. Yaesoubi, M., Allen, E.A., Miller, R.L., Calhoun, V.D., 2015. Dynamic coherence ana- lysis of resting fMRI data to jointly capture state-based phase, frequency, and time-domain information. Neuroimage 120, 133–142. Smith, S.M., Fox, P.T., Miller, K.L., Glahn, D.C., Fox, P.M., Mackay, C.E., Filippini, N., Watkins, K.E., Toro, R., Laird, A.R., Beckmann, C.F., 2009. Correspondence of the brain’s functional architecture during activation and rest. Proc. Natl. Acad. Sci. U S A 106, 13040–13045. Yamashita, M., Kawato, M., Imamizu, H., 2015. Predicting learning plateau of working memory from whole-brain intrinsic network connectivity patterns. G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ Predicting haemodynamic net- works using electrophysiology: the role of non-linear and cross-frequency in- teractions. Neuroimage 130, 273–292. Nystrom, L.E., Braver, T.S., Sabb, F.W., Delgado, M.R., Noll, D.C., Cohen, J.D., 2000. Working memory for letters, shapes, and locations: fMRI evidence against sti- mulus-based regional organization in human prefrontal cortex. Neuroimage 11, 424–446. Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., Crivello, F., Etard, O., Delcroix, N., Mazoyer, B., Joliot, M., 2002. Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain. Neuroimage 15, 273–289. O’Neill, G.C., Barratt, E.L., Hunt, B.A., Tewarie, P.K., Brookes, M.J., 2015a. Measuring electrophysiological connectivity by power envelope correlation: a technical review on MEG methods. Phys. Med. Biol. 60, R271–R295. van Dellen, E., de Witt Hamer, P.C., Douw, L., Klein, M., Heimans, J.J., Stam, C.J., Reijneveld, J.C., Hillebrand, A., 2012. Connectivity in MEG resting-state net- works increases after resective surgery for low-grade glioma and correlates with improved cognitive performance. Neuroimage Clin. 2, 1–7. O’Neill, G.C., Bauer, M., Woolrich, M.W., Morris, P.G., Barnes, G.R., Brookes, M.J., 2015b. Dynamic recruitment of resting state sub-networks. Neuroimage 115, 85–95. Van Veen, B.D., van Drongelen, W., Yuchtman, M., Suzuki, A., 1997. Localization of brain electrical activity via linearly constrained minimum variance spatial ﬁl- tering. IEEE Trans. Biomed. Eng. 44, 867–880. Palaniyappan, L., Liddle, P.F., 2012. Does the salience network play a cardinal role in psychosis? An emerging hypothesis of insular dysfunction. J Psychiatry Neu- rosci. 37, 17–27. Vandenberghe, R., Price, C., Wise, R., Josephs, O., Frackowiak, R.S., 1996. Functional anatomy of a common semantic system for words and pictures. Nature 383, 254–256. Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A., Shulman, G. L., 2001. A default mode of brain function. Proc. Natl. Acad. Sci. U S A 98, 676–682. Vidaurre, D., Quinn, A.J., Baker, A.P., Dupret, D., Tejero-Cantero, A., Woolrich, M.W., 2016. Spectrally resolved fast transient brain states in electrophysiological data. Neuroimage 126, 81–95. Robinson, S., Vrba, J., 1998. Functional neuroimaging by synthetic aperture mag- netometry (SAM). In: Yoshimoto, T., Kotani, M., Kuriki, S., Karibe, H., Nakasato, N. (Eds.), Recent Advances in biomagnetism. Tohoku University Press, Sendai, pp. 302–305. Vigneau, M., Beaucousin, V., Herve, P.Y., Duffau, H., Crivello, F., Houde, O., Mazoyer, B., Tzourio-Mazoyer, N., 2006. Meta-analyzing left hemisphere language areas: phonology, semantics, and sentence processing. Neuroimage 30, 1414–1432. Sarvas, J., 1987. Basic mathematical and electromagnetic concepts of the biomag- netic inverse problem. Phys. References Modality-Spanning Deﬁcits in Attention-Deﬁcit/Hyperactivity Disorder in Functional Networks, Gray Corbetta, M., Miezin, F.M., Dobmeyer, S., Shulman, G.L., Petersen, S.E., 1991. Selec- tive and divided attention during visual discriminations of shape, color, and G.C. O’Neill et al. / NeuroImage ∎(∎∎∎∎) ∎∎∎–∎∎∎ Sci. Rep. 5, 7622. Smith, S.M., Miller, K.L., Moeller, S., Xu, J., Auerbach, E.J., Woolrich, M.W., Beckmann, C.F., Jenkinson, M., Andersson, J., Glasser, M.F., Van Essen, D.C., Feinberg, D.A., Please cite this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), http://dx.doi.org/10.1016/j.neuroimage.2016.08.061i l, G.C., et al., Measurement of dynamic task related functional networks using MEG. NeuroImage (2016), oimage.2016.08.061i this article as: O’Neill, G.C., et al., Measurement of dynamic task related functional networks using MEG. N oi.org/10.1016/j.neuroimage.2016.08.061i
https://openalex.org/W2895551815	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0204924&type=printable	English	null	Transcriptome analyses of Atlantic salmon muscle genes induced by a DNA vaccine against salmonid alphavirus, the causative agent of salmon pancreas disease (PD)	PloS one	2,018	cc-by	8,142	RESEARCH ARTICLE Editor: Tzong-Yueh Chen, National Cheng Kung University, TAIWAN Received: June 15, 2018 Accepted: September 17, 2018 Published: October 1, 2018 Copyright: © 2018 Sobhkhez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Funding: Borre Robertsen received funding from the Aquaculture programme of the Research Council of Norway (https://www.forskningsradet. no/en/Home_page/1177315753906), grant no. 254907/E40. NOFIMA AS (https://nofima.no/en/ research/) provided support in the form of salaries for author AK, but did not have any additional role Transcriptome analyses of Atlantic salmon muscle genes induced by a DNA vaccine against salmonid alphavirus, the causative agent of salmon pancreas disease (PD) Mehrdad Sobhkhez1, Aleksei Krasnov2, Børre RobertsenID1* Mehrdad Sobhkhez1, Aleksei Krasnov2, Børre RobertsenID1* 1 Norwegian College of Fishery Science, UiT The Arctic University of Norway, Tromsø, Norway, 2 Nofima AS, Norwegian Institute of Food, Fisheries & Aquaculture Research, Ås, Norway 1 Norwegian College of Fishery Science, UiT The Arctic University of Norway, Tromsø, Norway, 2 Nofima AS, Norwegian Institute of Food, Fisheries & Aquaculture Research, Ås, Norway a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 * borre.robertsen@uit.no * borre.robertsen@uit.no Abstract Salmonid alphavirus (SAV) is the causative agent of pancreas disease (PD) in farmed Atlan- tic salmon. A previous study showed that vaccination of pre-smolt salmon with a plasmid encoding the structural polypeptide of SAV gave protection against infection and develop- ment of PD accompanied by production of antibodies against the virus. In the present work we analyzed transcript responses in the muscle to vaccination with this plasmid (here named pSAV). The purpose was to shed light on how pSAV might initiate adaptive immune responses in the fish. The work was based on microarray and reverse transcription quantita- tive PCR analyses of muscle at the injection site 7 days after vaccination. The results showed that pSAV and pcDNA3.3 had similar abilities to up-regulate type I IFN stimulated genes. In contrast, pSAV caused higher up-regulation of IFNγ and several IFNγ inducible genes. Compared to pcDNA3.3, pSAV also gave larger increase in transcripts of marker genes for B-cells, T-cells and antigen presenting cells (APCs), which suggest attraction and role of these cells in the adaptive immune responses elicited by pSAV. Moreover, pSAV caused a stronger up-regulation of the chemokine CXCL10 and the proinflammatory cyto- kines IL-1ß and TNFα, which may explain attraction of lymphocytes and APCs. The present work shows that the expression profile of genes resulting from vaccination with pSAV is dif- ferent from the expression profiles obtained previously by vaccination of salmonids with DNA vaccines against infectious salmon anemia virus and infectious hematopoietic necrosis virus. OPEN ACCESS Citation: Sobhkhez M, Krasnov A, Robertsen B (2018) Transcriptome analyses of Atlantic salmon muscle genes induced by a DNA vaccine against salmonid alphavirus, the causative agent of salmon pancreas disease (PD). PLoS ONE 13(10): e0204924. https://doi.org/10.1371/journal. pone.0204924 Editor: Tzong-Yueh Chen, National Cheng Kung University, TAIWAN Received: June 15, 2018 Accepted: September 17, 2018 Published: October 1, 2018 Copyright: © 2018 Sobhkhez 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. Fish Atlantic salmon (Salmo salar L.) presmolts (30–40 g), Aquagen standard (Aquagen, Kyrksæ- terøra, Norway), were kept at 10˚C in 300 L tanks supplied with fresh water and were fed com- mercial dry food in Tromsø Aquaculture Research Station, Norway. For vaccination, fish were anesthetized with 0.005% benzocaine (ACD Pharmaceuticals, Norway). Different groups of fish were labeled by tattooing (2% alcian blue, Panjet inoculator). For harvest of muscle sam- ples, fish were euthanized by an overdose of benzocaine (0.01%). Ethics statement All fish experiments were approved by the Norwegian Animal Research Authority (NARA) and Norwegian “Regulation on Animal Experimentation” was observed during fish handling. NOFIMA AS is a non-profit research institution supported by grants from the following sources: The Research Council of Norway; The Ministry of Trade, Industry and Fisheries; The Norwegian Seafood Research Fund (FHF). 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. The purpose of the study was to find out whether expression of the structural proteins of SAV3 have unique effects on gene transcription that might shed light on how this DNA vaccine influ- ences the immune system. It is likely that the adaptive immune system reacts to antigens as soon as they are expressed. Day 7 post injection was thus chosen as time point for harvest of mRNA since it represents early high level expression of plasmid encoded genes in the muscle [5, 7, 8]. The data showed that vaccination with pSAV and the vector plasmid pcDNA3.3 gave simi- lar up-regulation of type I IFN stimulated genes (ISGs), which is different from the HE-plas- mid of SAV. In contrast, pSAV caused higher up-regulation of IFNγ and several IFNγ inducible genes compared to pcDNA3.3. Moreover, pSAV gave increase in transcripts of B- cells, T-cells and antigen-presenting cells (APCs) presumably due to attraction of these cells by up-regulation of inflammatory chemokines and cytokines. Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus infectious hematopoietic necrosis virus (IHN) in salmonids [1–3]. More recently, a DNA vac- cine against infectious salmon anemia virus (ISAV) has shown a high level of protection when injected together with a plasmid encoding type I IFN from Atlantic salmon [4]. The vaccine was a plasmid expressing the virus surface protein hemagglutininesterase (HE) and the IFN plasmid was shown to function as an adjuvant. A transcriptome analysis of genes induced at the muscle injection site by a plasmid encoding HE, plasmid vector without insert, and plasmid encoding IFNa was recently carried out [5]. The work showed that plasmid DNA induced a similar set of genes as the IFNa plasmid, but at a lower level. In contrast, the HE-plasmid induced lower levels of IFN-induced genes compared to plasmid without insert. This suggested that HE expressed by the plasmid, inhibited expression of IFN-induced genes thus explaining the need for IFNa plasmid as adjuvant. Salmonid alphavirus (SAV) is the causative agent of pancreas disease (PD) in farmed Atlantic salmon. A DNA vaccine against SAV was recently shown to give significantly higher pro- tection of Atlantic salmon against pancreas disease (PD) than a commercial vaccine based on inactivated virus [6]. This vaccine is a plasmid encoding the structural polyprotein C-E3-E2- 6K-E2 of SAV3 (here named pSAV), which upon uptake in cells results in expression of E2 on the surface of the cells. In contrast to the ISAV DNA vaccine, the SAV3 DNA vaccine did not need addition of IFN plasmid to achieve protection. In the present study we did transcriptome analyses of muscle at the injection site of pSAV, pcDNA3.3 and PBS 7 days after injection. This was per- formed by microarray analysis and reverse transcription quantitative PCR (RT-qPCR). in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. NOFIMA AS is a non-profit research institution supported by grants from the following sources: The Research Council of Norway; The Ministry of Trade, Industry and Fisheries; The Norwegian Seafood Research Fund (FHF). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction Virus diseases cause major problems in fish farming due to economic losses and suffering of infected fish. Traditional vaccines based on inactivated virus have so far been unable to provide sufficient protection against virus diseases of fish. DNA vaccination against virus has received high expectations due to the high level of protection obtained with the DNA vaccine against 1 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Treatment of fish with plasmids pcDNA3.3 encoding the open reading frame (ORF) for SAV3 complete structural polyprotein C- E3-E2-6K-E1 (pSAV) is based on the genome sequence of the SAV3 H10 isolate (GenBank 2 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Accession number: JQ799139) and has previously been named pCSP [6]. pcDNA3.3 without insert was used as control. The plasmids were purified by EndoFree plasmid purification kit from Qiagen. Presmolt salmon were injected intramuscularly (i.m.) approximately 1 cm below the dorsal fin with 15 μg of plasmid in 50 μl sterile PBS pH 7.4 or with 50 μl PBS only. Fish groups (N = 5) were injected with PBS (Group 1), pcDNA3.3 (Group 2) and pSAV (Group 3). Muscle samples from the injection site were harvested one week post injection and used for RNA isolation. Accession number: JQ799139) and has previously been named pCSP [6]. pcDNA3.3 without insert was used as control. The plasmids were purified by EndoFree plasmid purification kit from Qiagen. Presmolt salmon were injected intramuscularly (i.m.) approximately 1 cm below the dorsal fin with 15 μg of plasmid in 50 μl sterile PBS pH 7.4 or with 50 μl PBS only. Fish groups (N = 5) were injected with PBS (Group 1), pcDNA3.3 (Group 2) and pSAV (Group 3). Muscle samples from the injection site were harvested one week post injection and used for RNA isolation. Reverse transcription quantitative PCR (RT-qPCR) Muscle samples from the injection site were homogenized using TissueLyser II (QIAGEN) (30 sec, 30 1/s) and total RNA was extracted using the RNeasy1 Mini Kit (Qiagen). cDNA was pro- duced using high capacity cDNA Reverse Transcription kit (Applied Biosystems) (500 ng RNA in a 20 μl reaction). cDNA was diluted 10 times and 5 μl of cDNA was used per 20 μl PCR reaction, which also included primers for target gene (0.25 μM) and Fast SYBR1 Green Master Mix (Applied Biosystems). The samples were applied in duplicates and the measurement was done using ABI Prism 7500 FAST Cycler from Applied Biosystems (initial denaturation 95˚C: 20s and 40 cycles of 95˚C: 3s, and 60˚C: 30s). Primer sequences are included in S1 Table. Expression val- ues were normalized against the levels of EF1αβ and fold change in expression of different genes was calculated against that of PBS group using the method described by Pfaffl [9]. Unpaired t-test with two-tail distribution was used for statistical analysis, p 0.05. Microarray analysis The analyses were carried out essentially as described earlier [5]. Nofima’s Atlantic salmon oli- gonucleotide microarray SIQ-6 (GEO Accession no. GPL16555) was produced by Agilent Technologies in the 15 K x 8 format, all reagents and equipment were from the same source [10]. Analyses included three groups, four individuals per group at one time-point; totally twelve arrays were used. RNA was isolated as described for RT-qPCR. Total RNA (200 ng per reaction) was labelled with Cy3 using Low Input Quick Amp Labeling Kit and fragmented with Gene Expression Hybridization Kit. Hybridization was performed for 17 hours in an oven at 65˚C at rotation speed of 10 rounds per minute. Arrays were washed for one minute with Gene Expression Wash Buffer I at room temperature, and one minute with Gene Expres- sion Wash Buffer II at 37˚C and scanned. Data analyses were carried out with Nofima’s bioin- formatics package [10]. Global Normalization was performed by equalizing the mean intensities of all microarrays. Next, the individual values for each feature were divided to the mean value of all samples producing expression ratios (ER). The log2-ER were calculated and normalized with the locally weighted non-linear regression (Lowess). The data are presented as ratios of plasmid injected groups to PBS injected control. Differentially expressed genes were selected by criteria: p < 0.05 and log2-ER > \|0.8\| (1.74-fold). The microarray data pre- sented in this publication has been deposited in the NCBI´s Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) and is available under the accession number GSE115689. IFN response Because type I IFNs (IFN-I) enhance adaptive immune responses and function as adjuvants in the DNA vaccine against ISAV [4], we wanted to study whether pSAV had any effect on IFN-I induced genes (ISGs) during DNA vaccination. The microarray data showed that pcDNA3.3 and pSAV had similar ability to up-regulate typical ISGs such as ISG15, Mx and viperin (Fig 1), which was confirmed by RT-qPCR (Fig 2A). A similar pattern of response was seen for a several other 3 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus ISGs (Fig 1). In contrast pSAV compared to pcDNA3.3 gave significantly higher up-regulation of guanylate binding protein 1 (GBP1) interferon regulatory factor 1 (IRF1) IRF5 and IFN Fig 1. Expression of IFN stimulated genes (ISGs) in muscle in response to pcDNA3.3 and pSAV measured by microarray analysis. The ISGs include the main antiviral genes, interferon regulatory factors (IRFs), virus RNA sensors and tripartite motif-containing proteins (TRIMS). Groups of fish (N = 4) were injected with PBS, pcDNA3.3 and pSAV. RNA was isolated from muscle at the injection site 7 days post injection. Numbers represent fold changes compared to the PBS group. Symbols $ and denote significant difference (p 0.05) between PBS and plasmid groups, and between pcDNA3.3 and pSAV groups, respectively. https://doi.org/10.1371/journal.pone.0204924.g001 Fig 1. Expression of IFN stimulated genes (ISGs) in muscle in response to pcDNA3.3 and pSAV measured by microarray analysis. The ISGs include the main antiviral genes, interferon regulatory factors (IRFs), virus RNA sensors and tripartite motif-containing proteins (TRIMS). Groups of fish (N = 4) were injected with PBS, pcDNA3.3 and pSAV. RNA was isolated from muscle at the injection site 7 days post injection. Numbers represent fold changes compared to the PBS group. Symbols $ and denote significant difference (p 0.05) between PBS and plasmid groups, and between pcDNA3.3 and pSAV groups, respectively. Fig 1. Expression of IFN stimulated genes (ISGs) in muscle in response to pcDNA3.3 and pSAV measured by microarray analysis. The ISGs include the main antiviral genes, interferon regulatory factors (IRFs), virus RNA sensors and tripartite motif-containing proteins (TRIMS). Groups of fish (N = 4) were injected with PBS, pcDNA3.3 and pSAV. RNA was isolated from muscle at the injection site 7 days post injection. Numbers represent fold changes compared to the PBS group. IFN response Symbols $ and denote significant difference (p 0.05) between PBS and plasmid groups, and between pcDNA3.3 and pSAV groups, respectively. https://doi.org/10.1371/journal.pone.0204924.g001 ISGs (Fig 1). In contrast pSAV compared to pcDNA3.3 gave significantly higher up-regulation of guanylate binding protein 1 (GBP1), interferon regulatory factor 1 (IRF1), IRF5 and IFN- inducible protein IFI44 (Fig 1). This was confirmed by RT-qPCR for GBP1 and IRF1 (Fig 2B). 4 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 2. Expression of type I IFN inducible genes (A), IFNγ inducible genes (B), and type I IFN (IFNa) and type II IFN (IFNγ) genes (C) in response to vaccination with plasmids. Groups of fish (N = 5) were injected with PBS, pcDNA3.3 and pSAV. RNA was isolated from muscle at the injection site 7 days post injection and expression of different IFNs and ISGs was measured by RT-qPCR. Expression value of each gene was normalized against the levels of EF1αβ. Numbers represents fold increase in expression (mean ± SD) compared to the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi org/10 1371/journal pone 0204924 g002 Fig 2. Expression of type I IFN inducible genes (A), IFNγ inducible genes (B), and type I IFN (IFNa) and type II IFN (IFNγ) genes (C) in response to vaccination with plasmids. Groups of fish (N = 5) were injected with PBS, pcDNA3.3 and pSAV. RNA was isolated from muscle at the injection site 7 days post injection and expression of different IFNs and ISGs was measured by RT-qPCR. Expression value of each gene was normalized against the levels of EF1αβ. Numbers represents fold increase in expression (mean ± SD) compared to the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g002 https://doi.org/10.1371/journal.pone.0204924.g002 GBP-1 is an antiviral and antibacterial GTP-ase while IRF1 activates transcription of IFN-I and viperin [11–14]. IRF5 is involved in induction of IFN-I, chemokines and proinflammatory cyto- kines [15–17]. IFI44 has anti-proliferative functions [18]. Notably, GBP1 and IRF1 are more strongly induced by IFNγ (type II IFN) than by IFN-I in both mammals and Atlantic salmon [19, 20]. IFN response Indeed, RT-qPCR showed that pSAV up-regulated IFNγ more strongly than pcDNA3.3 (Fig 2C), which supports that GBP1 and IRF1 were induced by IFNγ in the pSAV group. RT-qPCR showed that the IFN-I IFNa was slightly up-regulated by pSAV, but not by pcDNA3.3 (Fig 2C). Attraction of B- and T-cells The microarray data showed that both pcDNA3.3 and pSAV enhanced transcripts encoding constant regions of IgD, IgM and IgT and segments of the heavy and light chains, totally 34 probes (Fig 3), the increase in the pSAV group was 1.95-fold greater compared to the pcDNA3.3 group. RT-qPCR analysis of membrane bound IgM (mIgM) and soluble IgM (sIgM) confirmed up-regulation by both plasmids, but only expression of sIgM was significantly higher in the pSAV group (Fig 4). p g g y g p g p g The microarray data showed that expression of T-cell receptor (TCR) alpha and beta chains were increased by both plasmids, but were significantly higher up-regulated by pSAV than by pcDNA3.3 (Fig 5). This was confirmed by RT-qPCR for TCRbeta (Fig 4). The microarray anal- ysis also showed that compared to pcDNA3.3, pSAV caused significant higher increase in tran- scripts for other typical T-cell markers (Fig 5). This was confirmed by RT-qPCR for CD4, CD8a, CD3z and granzyme K (Fig 4). Taken together these data suggests that B- and T-cells are attracted to the muscle injection site by plasmid DNA and that pSAV had a stronger attrac- tive effect than pcDNA3.3, especially on T-cells. CD8a, CD3z and granzyme K (Fig 4). Taken together these data suggests that B- and T-cells are attracted to the muscle injection site by plasmid DNA and that pSAV had a stronger attrac- tive effect than pcDNA3.3, especially on T-cells. Attraction of B- and T-cells Protection of salmon against SAV by vaccination with pSAV, could be due to both increased anti- body response and increased production of specific cytotoxic T cells. We thus wanted to find out 5 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 3. B cell specific genes, microarray results. Markers correspond to probes representing constant regions of IgM, IgT and IgD and segments of light and heavy chains (totally 34 probes). Data are log2-ER (expression ratio) to PBS control. Difference between the groups is significant (p<0.0001, paired t test). https://doi.org/10.1371/journal.pone.0204924.g003 Fig 3. B cell specific genes, microarray results. Markers correspond to probes representing constant regions of IgM, IgT and IgD and segments of light and heavy chains (totally 34 probes). Data are log2-ER (expression ratio) to PBS control. Difference between the groups is significant (p<0.0001, paired t test). Fig 3. B cell specific genes, microarray results. Markers correspond to probes representing constant regions of IgM, IgT and IgD and segments of light and heavy chains (totally 34 probes). Data are log2-ER (expression ratio) to PBS control. Difference between the groups is significant (p<0.0001, paired t test). Fig 3. B cell specific genes, microarray results. Markers correspond to probes representing constant regions of IgM, IgT and IgD and segments of light and heavy chains (totally 34 probes). Data are log2-ER (expression ratio) to PBS control. Difference between the groups is significant (p<0.0001, paired t test). https://doi.org/10.1371/journal.pone.0204924.g003 if pSAV vaccination was associated with increased attraction of B- and T-cells to the muscle injec- tion site, which should be revealed by gene markers for these cells. The microarray data showed that both pcDNA3.3 and pSAV enhanced transcripts encoding constant regions of IgD, IgM and IgT and segments of the heavy and light chains, totally 34 probes (Fig 3), the increase in the pSAV group was 1.95-fold greater compared to the pcDNA3.3 group. RT-qPCR analysis of membrane bound IgM (mIgM) and soluble IgM (sIgM) confirmed up-regulation by both plasmids, but only expression of sIgM was significantly higher in the pSAV group (Fig 4). if pSAV vaccination was associated with increased attraction of B- and T-cells to the muscle injec- tion site, which should be revealed by gene markers for these cells. Antigen presenting cells (APCs) and the MHC II antigen presentation pathway Since attraction of APCs might indicate that antigen presentation to T-cells occurs at the mus- cle injection site, we analysed expression of genes for surface markers of APCs. Extracellular protein antigens are taken up by APCs and processed by endolysosomal proteases into pep- tides, which are loaded onto MHC II molecules by displacing the peptide CLIP derived from the class II-associated invariant chain, which is also known as MHC II gamma chain or CD74 when it is surface expressed [26]. The microarray data showed that compared to pcDNA3.3, pSAV up-regulated more strongly MHC II alpha and beta chains and CD74 (Fig 8). RT-qPCR confirmed stronger up-regulation of MHC II by pSAV (Fig 7A). One of the most prominent genes that showed higher transcript levels with pSAV was high affinity Ig receptor I (FcγRI), which is encoded by FCGRI (Figs 7B and 8). This protein is expressed on monocytes, macro- phages and dendritic cells (DCs) and can be up-regulated by IFNγ [27, 28]. MHC I antigen presentation pathway Most cells have the ability to present antigen to cytotoxic T-cells through MHC I, which occurs by digestion of antigen by the proteasome, transportation of peptides into ER by TAP followed by loading of antigenic peptides onto MHC-I involving tapasin and other chaperone molecules [21]. As shown in Fig 6, the microarray data showed that pcDNA3.3 and pSAV up-regulated to a similar extent MHC I, beta-2-microglobulin and PSMB8, which is a subunit of the protea- some. This was confirmed by RT-qPCR for MHC I (Fig 7A). Other proteasome subunits (PSMB6B and PSMB7) were significantly higher up-regulated by pSAV than by pcDNA3.3 6 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 4. Expression of T- and B- cell marker genes in response to injection of plasmids measured by RT-qPCR. Treatment groups and experimental details as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ. Numbers represents fold increase in expression (mean ± SD) compared to the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). Fig 4. Expression of T- and B- cell marker genes in response to injection of plasmids measured by RT-qPCR. Treatment groups and experimental details as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ. Numbers represents fold increase in expression (mean ± SD) compared to the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g004 (Fig 6). Higher transcript values were also observed for TAP and tapasin in the pSAV group, but the differences between pSAV and pcDNA3.3 were not significant for these genes. Inter- estingly, PSMB6, PSMB7, TAP and tapasin are all IFNγ -induced genes [22–25]. (Fig 6). Higher transcript values were also observed for TAP and tapasin in the pSAV group, but the differences between pSAV and pcDNA3.3 were not significant for these genes. Inter- estingly, PSMB6, PSMB7, TAP and tapasin are all IFNγ -induced genes [22–25]. Effector molecules of APCs/phagocytic cells pSAV showed higher transcript levels of lysozyme G, cytochrome b-245 beta chain and neu- trophil cytosol factor (NCF1), also known as p47phox (Fig 8). Lysozymes are responsible for degradation of peptidoglycans in the cell wall of gram-positive and gram-negative bacteria. NCF1 is the cytoplasmic subunit of NADPH-oxidase complex in neutrophils and macro- phages. NCF1 has a central role in oxidative burst, which leads to production of superoxide anion and formation of antimicrobial reactive oxygen species [29]. Cytochrome b-245 is a 7 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 5. Microarray data of T cell marker genes. Data produced and presented as explained in Fig 1. membrane-bound component of the NADPH-oxidase complex [30]. RT-qPCR confirmed that NCF1 was significantly higher expressed in response to pSAV compared to pcDNA3.3 and PBS (Fig 7B). Altogether the increase in transcripts of this group of genes suggests attrac- tion of phagocytic cells like macrophages or neutrophils to the site of pSAV injection. Chemokines and pro-inflammatory cytokines Attraction of leukocytes to the plasmid injection site must be mediated by chemokines and proinflammatory cytokines [35]. Microarray data for chemokines, chemokine receptors and cytokine receptors that showed up-regulation with pcDNA3.3 and pSAV are presented in Fig 9. The data showed higher transcript values for the chemokines CCL4, CCL19 and CXCL10 in response to pSAV compared to pcDNA3.3, but the differences were not significant. In general, cytokines showed low upregulation in response to plasmids in the microarray analysis. RT- qPCR analyses showed that CXCL10 and the proinflammatory cytokines IL-1ß and TNFα were more strongly up-regulated by pSAV than by pcDNA3.3 (Fig 10). Up-regulation of CXCL10 may be mediated by IFNγ as suggested by previous work [19]. Of cytokine receptors, the microarray data showed higher up-regulation of IL2R and IL6R by pSAV than by pcDNA3.3. Other leukocyte markers Several other genes, which are commonly expressed by lymphocytes and other leukocytes were significantly higher up-regulated by pSAV than by pcDNA3.3 (S1 Fig). These included plastin-2, Wiskott-Aldrich syndrome protein (WASP) and Src kinase-associated phosphopro- tein 2 (Skap-2). Plastin-2 is specifically expressed in hematopoietic cell lineage and has impor- tant roles in processes related to neutrophil biology and function, stabilization of immune synapse (IS) between MHCII and TCR, T-cell activity as well as T-cell and B-cell motility and macrophage activity [31]. WASP is necessary for many lymphoid immune cell functions like phagocytosis, NK-cell cytotoxicity, TCR/BCR downstream signaling and monocyte chemo- taxis [32, 33]. Skap-2 is involved in macrophage cytoskeletal rearrangement [34]. 8 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 6. Microarray data showing expression of genes involved in MHC I antigen presentation. Data produced and t d l i d i Fi 1 Fig 6. Microarray data showing expression of genes involved in MHC I antigen presentation. Data produced and presented as explained in Fig 1. Discussion A previous study showed that i.m. injection of Atlantic salmon with a plasmid expressing the whole SAV3 structural polyprotein (here named pSAV) gave strong protection against 9 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus development of pancreas disease caused by SAV3 infection accompanied by production of Fig 7. Expression of genes involved in antigen presentation / phagocytosis in response to plasmids measured by RT-qPCR. Treatment groups and sampling for RNA extraction as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ and fold increase in expression of different genes was calculated against that of the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g007 Fig 7. Expression of genes involved in antigen presentation / phagocytosis in response to plasmids measured by RT-qPCR. Treatment groups and sampling for RNA extraction as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ and fold increase in expression of different genes was calculated against that of the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g007 https://doi.org/10.1371/journal.pone.0204924.g007 development of pancreas disease caused by SAV3 infection accompanied by production of IgM antibodies against the virus [6]. The present study was conducted to find out if transcrip- tome analysis of muscle at the injection site might reveal a pattern of gene expression that could shed light on the protective effect of the vaccine. DNA vaccination against a virus results in uptake of plasmid and expression of virus protein both by muscle cells and resident APCs [36]. Changes in immune gene expression at the muscle injection site might be due to innate responses to plasmid DNA and to both innate and adaptive immune responses to the expressed viral proteins. The transcriptome data in this work demonstrated that pSAV and pcDNA3.3 had similar abilities to up-regulate typical IFN-I inducible genes (ISGs). The IFN-I response to pSAV is thus likely to be a response to plasmid DNA rather than to the viral development of pancreas disease caused by SAV3 infection accompanied by production of IgM antibodies against the virus [6]. Discussion The present study was conducted to find out if transcrip- tome analysis of muscle at the injection site might reveal a pattern of gene expression that could shed light on the protective effect of the vaccine. DNA vaccination against a virus results in uptake of plasmid and expression of virus protein both by muscle cells and resident APCs [36]. Changes in immune gene expression at the muscle injection site might be due to innate responses to plasmid DNA and to both innate and adaptive immune responses to the expressed viral proteins. The transcriptome data in this work demonstrated that pSAV and pcDNA3.3 had similar abilities to up-regulate typical IFN-I inducible genes (ISGs). The IFN-I response to pSAV is thus likely to be a response to plasmid DNA rather than to the viral PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 10 / 16 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 8. Microarray data showing expression of marker genes for antigen presenting cells (APC) / phagocytic cells. Data produced and presented as explained in Fig 1. htt //d i /10 1371/j l 0204924 008 https://doi.org/10.1371/journal.pone.0204924.g008 proteins. The virus proteins expressed by pSAV do thus not seem to induce IFN-I such as the G-protein of the DNA vaccine against IHNV [5, 8]. Moreover, pSAV does not inhibit the IFN-I response unlike hemagglutininesterase in the DNA vaccine against ISAV [5, 8]. In fact, pSAV induced a distinct set of genes in the muscle compared to the other two DNA vaccines. Compared to pcDNA3.3, pSAV firstly caused higher up-regulation of IFNγ, which was sup- ported by higher up-regulation of several IFNγ inducible genes. Secondly, pSAV caused larger expression of marker genes for T-cells, B-cells and APCs, which might all be important for development of the adaptive immune responses against the SAV3. Thirdly, pSAV caused a stronger up-regulation of the chemokine CXCL10 and proinflammatory cytokines IL-1ß and TNFα which may explain attraction of lymphocytes and APCs. IFNγ response One of the most prominent responses to pSAV was increased expression of IFNγ (Fig 2C) and the IFNγ inducible genes IRF-1, GBP-1 and CXCL-10 (Figs 1, 2B and 9). This suggests that expression of the SAV3 structural proteins results in increased IFNγ production at the muscle injection site. IFNγ is regarded as a typical Th1 cytokine because it directs differentiation of naïve CD4+ cells toward a Th1 phenotype and is a major product of CD4+ T helper 1 (Th1) cells and CD8+ cytotoxic T cells [20]. Whether Th1 or cytotoxic T cells are the source of IFNγ is, however, uncertain since 7 days post vaccination is very early in the development of an adaptive immune response in cold water fish. More likely sources of IFNγ at this stage are nat- ural killer (NK) cells, APCs and natural killer T (NKT) cells [20, 37]. Little is yet known about NK and NKT cells in fish, however. Whichever source, enhanced production of IFNγ might PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 11 / 16 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus Fig 9. Microarray data of chemokines, cytokines and their receptors. Data produced and presented as for Fig 1. https://doi.org/10.1371/journal.pone.0204924.g009 Fig 9. Microarray data of chemokines, cytokines and their receptors. Data produced and presented as for Fig 1. https://doi.org/10.1371/journal.pone.0204924.g009 Microarray data of chemokines, cytokines and their receptors. Data produced and presented as for Fig 1. Fig 9. Microarray data of chemokines, cytokines and their receptors. Data produced and presented as for Fig 1. https://doi.org/10.1371/journal.pone.0204924.g009 Fig 9. Microarray data of chemokines, cytokines and their receptors. Data produced and presented as for Fig 1. https://doi.org/10.1371/journal.pone.0204924.g009 Fig 10. Expression of chemokines and inflammatory cytokines in response to plasmids measured by RT-qPCR. Treatment groups and sampling for RNA extraction as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ and fold change in expression of different genes was calculated against that of the PBS group. Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g010 Fig 10. Expression of chemokines and inflammatory cytokines in response to plasmids measured by RT-qPCR. Treatment groups and sampling for RNA extraction as described in Fig 2. Expression value of each gene was normalized against the levels of EF1αβ and fold change in expression of different genes was calculated against that of the PBS group. PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 IFNγ response Bars not sharing common letter indicate significant difference (p 0.05). https://doi.org/10.1371/journal.pone.0204924.g010 https://doi.org/10.1371/journal.pone.0204924.g010 12 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus increase antigen presentation as discussed below, and thus influence the differentiation of T- cells at the injection site. Attraction of B- and T-cells The data for pcDNA3.3 suggests that B-cells and T-cells are attracted to the plasmid injection site by recognition of plasmid DNA by muscle cells as described in previous work [4]. Higher mean transcript values for B-cell markers (IgM, IgD and IgT) were observed for the pSAV group compared to pcDNA3.3, but the differences for the individual genes were not signifi- cant. On the other hand, both the microarray and RT-qPCR analysis suggest increased attrac- tion of T-cells to the pSAV injection site compared to pcDNA3.3, due to significantly increased transcript levels of TCR chains, CD4, CD8 and CD3zeta. Attraction of cytotoxic T- cells by pSAV is suggested by increased expression of CD8a and granzyme K. Taken together, it is interesting to note that the T-cell response to pSAV appears to be more prominent than the B-cell response. Whether cytotoxic T-cells are involved in the protective immune response to SAV infection should thus be investigated in future studies. APCs and genes involved in antigen presentation Professional APCs include dendritic cells, macrophages and B-cells [26]. Participation of APCs in initiation of the adaptive immune response against SAV3 structural proteins is sup- ported by the fact that compared to pcDNA3.3, pSAV increased more strongly expression of MHC II, CD74 and FcγRI (Figs 7 and 8). These molecules are constitutively expressed in pro- fessional APCs, but may be up-regulated by IFNγ in professional and non-professional APCs [20, 38]. pSAV may thus have a positive effect on antigen presentation through the MHC II pathway. In addition, pSAV gave higher transcript values for several genes that are involved in the MHC I antigen presentation pathway (PSMB6, PSMB7, LMP-7, TAP1A, TAP2B and tapasin) and which may be induced by IFNγ [20]. This suggests a more efficient presentation of virus antigens through the MHC-I pathway possibly leading to cytotoxic T-cell production against the virus. Notably, pSAV and pcDNA3.3 showed similar increases in MHC I heavy chain, which suggests that MHC I expression is mainly triggered by plasmid DNA. MHC I is expressed by most cells in mammals, but can be up-regulated by IFN-I and IFNγ [39, 40]. Taken together, pSAV appears to attract APCs and/or induce genes in resident APCs, which provide a more efficient presentation of virus antigen to T-cells through both MHC pathways. Chemokines and pro-inflammatory cytokines Attraction of leukocytes to the plasmid injection site must be mediated by chemokines and/or proinflammatory cytokines [35]. The RT-qPCR analyses suggest that CXCL10, IL-1ß and TNFα, are likely to be involved in attraction of T-cells, B-cells and APCs since they are more strongly up-regulated by pSAV than by pcDNA3.3. Up-regulation of CXCL10 may be medi- ated by IFNγ as suggested by previous work [19]. IL-1ß and TNFα are key pro-inflammatory cytokines [35, 41]. Supporting information S1 Fig. Microarray data showing expression of various leukocyte marker genes. Data pro- duced and presented as explained in Fig 1. (TIF) S1 Fig. Microarray data showing expression of various leukocyte marker genes. Data pro- duced and presented as explained in Fig 1. S1 Table. a. List of primers used for RT-qPCR with SYBR Green. b. List of primers used for RT-qPCR with TaqMan. (DOCX) S1 Table. a. List of primers used for RT-qPCR with SYBR Green. b. List of primers used for RT-qPCR with TaqMan. Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus independent or whether cytotoxic T-cells are involved in protection against SAV3 is not known. This study thus motivates future studies of cytotoxic T-cell responses to the DNA vac- cine against SAV. The data support the hypothesis that the muscle injection site is a meeting place for B-cells, T-cells and APCs that is important for initiation of the adaptive immune response by the DNA vaccine against SAV3. In mammals, activation of B- and T-cells might occur at the muscle injection site, but plasmid transfected APCs also travel to lymph nodes for expression of anti- gen and activation of B- and T -cells [36]. While fish do not possess lymph nodes, it cannot be excluded that plasmid-transfected APCs travel to the head kidney and/or spleen for activation of lymphocytes. With these data, it can be concluded that the DNA vaccines against the viruses IHNV, ISAV and SAV, each have unique transcript profiles at the muscle injection site 7 days after vaccination. The G-protein vaccine against IHNV shows a strong IFN-I stimulatory profile while the ISAV HE vaccine shows a profile of IFN-I inhibition [5, 8]. In contrast, the SAV3 vaccine shows neither positive nor negative effects on IFN-I induced genes while it shows a clear IFNγ stimulatory profile. Author Contributions Conceptualization: Børre Robertsen. Formal analysis: Mehrdad Sobhkhez, Aleksei Krasnov. Funding acquisition: Børre Robertsen. Investigation: Mehrdad Sobhkhez, Aleksei Krasnov. Methodology: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. Project administration: Børre Robertsen. Writing – original draft: Mehrdad Sobhkhez, Børre Robertsen. Writing – review & editing: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. Conceptualization: Børre Robertsen. Formal analysis: Mehrdad Sobhkhez, Aleksei Krasnov. Funding acquisition: Børre Robertsen. Funding acquisition: Børre Robertsen. Methodology: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. Methodology: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. Project administration: Børre Robertsen. Writing – original draft: Mehrdad Sobhkhez, Børre Robertsen. Writing – original draft: Mehrdad Sobhkhez, Børre Robertsen. Writing – review & editing: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. Writing – review & editing: Mehrdad Sobhkhez, Aleksei Krasnov, Børre Robertsen. PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Conclusion While previous work showed that vaccination of Atlantic salmon with pSAV leads to produc- tion of antibodies against SAV, the present study suggests that pSAV vaccination also results in a strong T-cell response due to attraction of T-cells, APCs and up-regulation of IFNγ. Whether the antibody response to the SAV3 surface protein E2 is T-dependent or T-cell 13 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 4. Chang CJ, Sun B, Robertsen B. Adjuvant activity of fish type I interferon shown in a virus DNA vaccina- tion model. Vaccine. 2015; 33(21):2442–8. https://doi.org/10.1016/j.vaccine.2015.03.093 PMID: 25865470. 4. Chang CJ, Sun B, Robertsen B. Adjuvant activity of fish type I interferon shown in a virus DNA vaccina- tion model. Vaccine. 2015; 33(21):2442–8. https://doi.org/10.1016/j.vaccine.2015.03.093 PMID: 25865470. 5. Sobhkhez M, Krasnov A, Chang CJ, Robertsen B. Transcriptome analysis of plasmid-induced genes sheds light on the role of type I IFN as adjuvant in DNA vaccine against infectious salmon anemia virus. PLoS One. 2017; 12(11):e0188456. https://doi.org/10.1371/journal.pone.0188456 PMID: 29161315; PubMed Central PMCID: PMCPMC5697855. 6. Chang CJ, Gu J, Robertsen B. Protective effect and antibody response of DNA vaccine against salmo- nid alphavirus 3 (SAV3) in Atlantic salmon. J Fish Dis. 2017; 40(12):1775–81. https://doi.org/10.1111/ jfd.12644 PMID: 28493514. 7. Chang CJ, Robertsen C, Sun B, Robertsen B. Protection of Atlantic salmon against virus infection by intramuscular injection of IFNc expression plasmid. Vaccine. 2014; 32(36):4695–702. https://doi.org/ 10.1016/j.vaccine.2014.05.059 PMID: 24996122. 8. Purcell MK, Nichols KM, Winton JR, Kurath G, Thorgaard GH, Wheeler P, et al. Comprehensive gene expression profiling following DNA vaccination of rainbow trout against infectious hematopoietic necro- sis virus. Mol Immunol. 2006; 43(13):2089–106. https://doi.org/10.1016/j.molimm.2005.12.005 PMID: 16426680. 9. Pfaffl MW. A new mathematical model for relative quantification in real-time RT–PCR. Nucleic Acids Research. 2001; 29(9):e45–e. PubMed PMID: PMC55695. PMID: 11328886 10. Krasnov A, Timmerhaus G, Afanasyev S, Jorgensen SM. Development and assessment of oligonucleo- tide microarrays for Atlantic salmon (Salmo salar L.). Comp Biochem Physiol Part D Genomics Proteo- mics. 2011; 6(1):31–8. https://doi.org/10.1016/j.cbd.2010.04.006 PMID: 20605756. 11. Vestal DJ, Jeyaratnam JA. The guanylate-binding proteins: emerging insights into the biochemical properties and functions of this family of large interferon-induced guanosine triphosphatase. J Interferon Cytokine Res. 2011; 31(1):89–97. https://doi.org/10.1089/jir.2010.0102 PMID: 21142871; PubMed Central PMCID: PMCPMC3021356. 12. Stirnweiss A, Ksienzyk A, Klages K, Rand U, Grashoff M, Hauser H, et al. IFN regulatory factor-1 bypasses IFN-mediated antiviral effects through viperin gene induction. J Immunol. 2010; 184(9):5179– 85. https://doi.org/10.4049/jimmunol.0902264 PMID: 20308629. 13. Bergan V, Kileng O, Sun B, Robertsen B. Regulation and function of interferon regulatory factors of Atlantic salmon. Mol Immunol. 2010; 47(11–12):2005–14. Epub 2010/05/25. https://doi.org/10.1016/j. molimm.2010.04.015 PMID: 20494444. 14. Paun A, Pitha PM. The IRF family, revisited. Biochimie. 2007; 89(6–7):744–53. https://doi.org/10.1016/ j.biochi.2007.01.014 PMID: 17399883; PubMed Central PMCID: PMCPMC2139905. 15. Takaoka A, Yanai H, Kondo S, Duncan G, Negishi H, Mizutani T, et al. Integral role of IRF-5 in the gene induction programme activated by Toll-like receptors. Nature. 2005; 434(7030):243–9. https://doi.org/ 10.1038/nature03308 PMID: 15665823. 16. Barnes BJ, Kellum MJ, Field AE, Pitha PM. References 1. Anderson ED, Mourich DV, Fahrenkrug SC, LaPatra S, Shepherd J, Leong JA. Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus. Mol Mar Biol Biotechnol. 1996; 5(2):114–22. PMID: 8680524. 1. Anderson ED, Mourich DV, Fahrenkrug SC, LaPatra S, Shepherd J, Leong JA. Genetic immunization of rainbow trout (Oncorhynchus mykiss) against infectious hematopoietic necrosis virus. Mol Mar Biol Biotechnol. 1996; 5(2):114–22. PMID: 8680524. 2. Traxler GS, Anderson E, LaPatra SE, Richard J, Shewmaker B, Kurath G. Naked DNA vaccination of Atlantic salmon Salmo salar against IHNV. Dis Aquat Organ. 1999; 38(3):183–90. Epub 2000/02/25. https://doi.org/10.3354/dao038183 PMID: 10686669. 3. Salonius K, Simard N, Harland R, Ulmer JB. The road to licensure of a DNA vaccine. Current opinion in investigational drugs. 2007; 8(8):635–41. PMID: 17668365. 3. Salonius K, Simard N, Harland R, Ulmer JB. The road to licensure of a DNA vaccine. Current opinion in investigational drugs. 2007; 8(8):635–41. PMID: 17668365. 14 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus 4. Chang CJ, Sun B, Robertsen B. Adjuvant activity of fish type I interferon shown in a virus DNA vaccina- tion model. Vaccine. 2015; 33(21):2442–8. https://doi.org/10.1016/j.vaccine.2015.03.093 PMID: 25865470. Multiple regulatory domains of IRF-5 control activation, cel- lular localization, and induction of chemokines that mediate recruitment of T lymphocytes. Mol Cell Biol. 2002; 22(16):5721–40. https://doi.org/10.1128/MCB.22.16.5721-5740.2002 PMID: 12138184; PubMed Central PMCID: PMCPMC133975. 17. Ai K, Luo K, Xia L, Gao W, Hu W, Qi Z, et al. Functional characterization of interferon regulatory factor 5 and its role in the innate antiviral immune response. Fish Shellfish Immunol. 2018; 72:31–6. https://doi. org/10.1016/j.fsi.2017.10.042 PMID: 29080685. 18. Hallen LC, Burki Y, Ebeling M, Broger C, Siegrist F, Oroszlan-Szovik K, et al. Antiproliferative activity of the human IFN-alpha-inducible protein IFI44. J Interferon Cytokine Res. 2007; 27(8):675–80. https:// doi.org/10.1089/jir.2007.0021 PMID: 17784819. 19. Sun B, Skjaeveland I, Svingerud T, Zou J, Jorgensen J, Robertsen B. Antiviral activity of salmonid gamma interferon against infectious pancreatic necrosis virus and salmonid alphavirus and its depen- dency on type I interferon. J Virol. 2011; 85(17):9188–98. Epub 2011/06/24. https://doi.org/10.1128/ JVI.00319-11 PMID: 21697489; PubMed Central PMCID: PMC3165780. 20. Schroder K, Hertzog PJ, Ravasi T, Hume DA. Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol. 2004; 75(2):163–89. https://doi.org/10.1189/jlb.0603252 PMID: 14525967. 21. McCarthy MK, Weinberg JB. The immunoproteasome and viral infection: a complex regulator of inflam- mation. Front Microbiol. 2015; 6:21. https://doi.org/10.3389/fmicb.2015.00021 PMID: 25688236; PubMed Central PMCID: PMCPMC4310299. 22. Aki M, Shimbara N, Takashina M, Akiyama K, Kagawa S, Tamura T, et al. Interferon-gamma induces different subunit organizations and functional diversity of proteasomes. J Biochem. 1994; 115(2):257– 69. Epub 1994/02/01. PMID: 8206875. 15 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 Transcriptome analyses of salmon muscle injected with a DNA vaccine against salmonid alphavirus 23. Ma W, Lehner PJ, Cresswell P, Pober JS, Johnson DR. Interferon-gamma rapidly increases peptide transporter (TAP) subunit expression and peptide transport capacity in endothelial cells. J Biol Chem. 1997; 272(26):16585–90. Epub 1997/06/27. PMID: 9195970. 24. Abarca-Heidemann K, Friederichs S, Klamp T, Boehm U, Guethlein LA, Ortmann B. Regulation of the expression of mouse TAP-associated glycoprotein (tapasin) by cytokines. Immunology letters. 2002; 83 (3):197–207. Epub 2002/07/04. PMID: 12095710. 25. Kloetzel P-M. Antigen processing by the proteasome. Nature Reviews Molecular Cell Biology. 2001; 2:179. https://doi.org/10.1038/35056572 PMID: 11265247 26. Trombetta ES, Mellman I. Cell biology of antigen processing in vitro and in vivo. Annu Rev Immunol. 2005; 23:975–1028. https://doi.org/10.1146/annurev.immunol.22.012703.104538 PMID: 15771591. 27. van der Poel CE, Spaapen RM, van de Winkel JG, Leusen JH. Functional characteristics of the high affinity IgG receptor, FcgammaRI. J Immunol. 2011; 186(5):2699–704. 4. Chang CJ, Sun B, Robertsen B. Adjuvant activity of fish type I interferon shown in a virus DNA vaccina- tion model. Vaccine. 2015; 33(21):2442–8. https://doi.org/10.1016/j.vaccine.2015.03.093 PMID: 25865470. https://doi.org/10.4049/ jimmunol.1003526 PMID: 21325219. 28. Pearse RN, Feinman R, Ravetch JV. Characterization of the promoter of the human gene encoding the high-affinity IgG receptor: transcriptional induction by gamma-interferon is mediated through common DNA response elements. Proc Natl Acad Sci U S A. 1991; 88(24):11305–9. PMID: 1837149; PubMed Central PMCID: PMCPMC53123. 29. Hultqvist M, Holmdahl R. Ncf1 (p47phox) polymorphism determines oxidative burst and the severity of arthritis in rats and mice. Cellular immunology. 2005; 233(2):97–101. Epub 2005/06/07. https://doi.org/ 10.1016/j.cellimm.2005.04.008 PMID: 15936744. 30. Segal AW, West I, Wientjes F, Nugent JH, Chavan AJ, Haley B, et al. Cytochrome b-245 is a flavocyto- chrome containing FAD and the NADPH-binding site of the microbicidal oxidase of phagocytes. Bio- chem J. 1992; 284 (Pt 3):781–8. Epub 1992/06/15. PMID: 1320378; PubMed Central PMCID: PMCPMC1132607. 31. Morley SC. The Actin-Bundling Protein L-Plastin: A Critical Regulator of Immune Cell Function. Int J Cell Biol. 2012; 2012:10. https://doi.org/10.1155/2012/935173 PMID: 22194750 32. Orange JS, Ramesh N, Remold O, Donnell E, Sasahara Y, Koopman L, et al. Wiskott–Aldrich syn- drome protein is required for NK cell cytotoxicity and colocalizes with actin to NK cell-activating immuno- logic synapses. Proc Natl Acad Sci. 2002; 99(17):11351. https://doi.org/10.1073/pnas.162376099 PMID: 12177428 33. Tsuboi S. A complex of Wiskott-Aldrich syndrome protein with mammalian verprolins plays an important role in monocyte chemotaxis. J Immunol (Baltimore, Md: 1950). 2006; 176(11):6576–85. Epub 2006/ 05/20. PMID: 16709815. 34. Alenghat FJ, Baca QJ, Rubin NT, Pao LI, Matozaki T, Lowell CA, et al. Macrophages require Skap2 and Sirpalpha for integrin-stimulated cytoskeletal rearrangement. J Cell Sci. 2012; 125(Pt 22):5535–45. Epub 2012/09/15. https://doi.org/10.1242/jcs.111260 PMID: 22976304; PubMed Central PMCID: PMCPMC3561861. 35. Turner MD, Nedjai B, Hurst T, Pennington DJ. Cytokines and chemokines: At the crossroads of cell sig- nalling and inflammatory disease. Biochim Biophys Acta. 2014; 1843(11):2563–82. https://doi.org/10. 1016/j.bbamcr.2014.05.014 PMID: 24892271. 36. Kutzler MA, Weiner DB. DNA vaccines: ready for prime time? Nat Rev Genet. 2008; 9(10):776–88. Epub 2008/09/11. https://doi.org/10.1038/nrg2432 PMID: 18781156. 37. Schoenborn JR, Wilson CB. Regulation of interferon-gamma during innate and adaptive immune responses. Advances in immunology. 2007; 96:41–101. https://doi.org/10.1016/S0065-2776(07) 96002-2 PMID: 17981204. 38. Choi NM, Majumder P, Boss JM. Regulation of major histocompatibility complex class II genes. Curr Opin Immunol. 2011; 23(1):81–7. https://doi.org/10.1016/j.coi.2010.09.007 PMID: 20970972; PubMed Central PMCID: PMCPMC3033992. 39. Singer DS, Maguire JE. Regulation of the expression of class I MHC genes. Crit Rev Immunol. 1990; 10 (3):235–57. PMID: 2085409. 40. PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018 4. Chang CJ, Sun B, Robertsen B. Adjuvant activity of fish type I interferon shown in a virus DNA vaccina- tion model. Vaccine. 2015; 33(21):2442–8. https://doi.org/10.1016/j.vaccine.2015.03.093 PMID: 25865470. van den Elsen PJ, Peijnenburg A, van Eggermond MC, Gobin SJ. Shared regulatory elements in the promoters of MHC class I and class II genes. Immunol Today. 1998; 19(7):308–12. PMID: 9666603. 41. Bradley JR. TNF-mediated inflammatory disease. J Pathol. 2008; 214(2):149–60. https://doi.org/10. 1002/path.2287 PMID: 18161752. 16 / 16 PLOS ONE \| https://doi.org/10.1371/journal.pone.0204924 October 1, 2018
https://openalex.org/W2006446338	https://europepmc.org/articles/pmc2247305?pdf=render	English	null	Role of the aclacinomycin A – doxorubicin association in reversal of doxorubicin resistance in K562 tumour cells	British journal of cancer	1,989	cc-by	9,997	J.-M. Millot, T.D.W. Rasoanaivo, H. Morjani & M. Manfait J.-M. Millot, T.D.W. Rasoanaivo, H. Morjani & M. Manfait oratoire de Spectroscopie Biomoleculaire, UFR de Pharmacie, 5I rue Cognacq-Jay, 51096 Reims Cedex, France ctroscopie Biomoleculaire, UFR de Pharmacie, 5I rue Cognacq-Jay, 51096 Reims Cedex, France. Summary Acquired resistance to anthracyclines is characterised by a lower sensitivity to these agents, associated with impaired accumulation of drug. We have examined the ability of aclacinomycin A (ACM) associated with doxorubicin (DOX), to increase intranuclear DOX concentrations and, consequently, to enhance cytotoxic effects against drug resistant cells in vitro. A recently developed microspectrofluorometric technique is used to measure intranuclear DOX concentrations in sensitive and DOX-resistant K562 cells treated with DOX and ACM. Fluorescence emission spectra are collected from a microvolume of single living cell nuclei. From both DOX and ACM model fluorescence spectra (free, DNA-bound and metabolites), the intranuclear spectral profile is analysed according to the amount of each component. This quantitative analysis determines intranuclear DOX concentrations with an error of 10%. Non-cytotoxic doses of ACM, in combination with DOX, increase cytotoxic activity of DOX against K562 resistant cells. When DOX-resistant cells are exposed simultaneously to ACM and DOX, significant increases in intranuclear DOX concentrations are found compared with the case of exposure to DOX alone. The measure of the intranuclear retention of DOX shows that ACM partly blocks the DOX efflux in resistant cell nuclei, resulting in enhanced accumula- tion of DOX. These data lead us to conclude that ACM-DOX association partly reverses the DOX resistance at clinically achievable concentrations. studied with and without simultaneous ACM incubation. We conclude that the increase of intranuclear amoants of DOX, induced by ACM, depends on the DOX efflux inhibition in resistant cell nuclei. A major obstacle to successful use of anthracyclines and other cytotoxic drugs in cancer chemotherapy is the develop- ment of clinical drug resistance. Tumour cell modifications (resulting in multidrug resistance) are characterised by a complex phenotype of cross-resistance to antineoplasic agents (Ling et al., 1983; Bradley et al., 1988). These are accom- panied by gene amplification, a deficient accumulation of drug, an enhanced drug efflux function and a build up of an integral membrane glycoprotein with a molecular weight of 170,000- 180,000 (P-glycoprotein) (Kartner et al., 1983). A study with DNA transfectants (Riordan et al., 1985) has demonstrated that the P-glycoprotein was intimately involved in resistance and might be serving as an active efflux pump to remove the drug from the cell. Correspondence: M. Manfait. Received 20 March 1989; and in revised form 23 May 1989. J.-M. Millot, T.D.W. Rasoanaivo, H. Morjani & M. Manfait One way to overcome in vitro resistance to anti-cancer drugs has been to use simul- taneously calcium channel blockers (Tsuruo et al., 1983) or calmodulin inhibitors (Ganapathi et al., 1984) together with the anti-cancer drug. The mechanism of the resistance rever- sal appears to be related to the inhibition of outward drug transport which subsequently leads to increased intracellular drug levels. Br ~~~~~~~~~ J.Cne 18) 0 7 8 Br. J. Cancer (1989), 60, 678-684 Br ~~~~~~~~~ J.Cne 18) 0 7 8 Br. J. Cancer (1989), 60, 678-684 TeMcilnPesLd,18 '." The Macmillan Press Ltd., 1989 Drugs Stock solutions (10 jAM) of DOX, ACM and ACM metabolites (Laboratoires Roger Bellon, Paris, France) were prepared in Dulbecco's PBS (pH = 7.4, ionic strength I = 0.152 M, with 1 mM EDTA). Drug concentrations in PBS solution were determined by their absorbances at 480 nm for DOX and at 430 nm for ACM. Calf thymus DNA (Sigma Chemical Company, type I) was dissolved in Dulbecco's PBS. Concentration of DNA (phosphate) was estimated on the basis of a molar absorption coefficient e = 6600 M-' cm ' at 260 nm. For in vitro studies, the appropriate amounts of drugs were added to DMEM (Gibco), supplemented with 10% fetal calf serum (Seromed) and 2 mM L-glutamine. With the recent development of microspectrofluorometry, we have studied fluorescence signals from microvolumes within a single living cell (Ginot et al., 1984; Manfait et al., 1987). DOX concentrations in nuclei of sensitive and DOX- resistant K562 human leukaemia cell lines may thus be deter- mined (Gigli et al., 1988). We have demonstrated the impor- tance of the intranuclear concentration, since the cytotoxic effect induced by DOX was dependent on the amount of drug actually incorporated into the nucleus (Gigli et al., 1989). Cells K562 is a human leukaemia cell line, established from a patient with chronic myelogeneous leukaemia in blast trans- formation (Lozzio & Lozzio, 1975). K562 cells were kept in exponential growth at 5-8 x 105 in DMEM supplemented as described above. K562 cells resistant to DOX (K562-DOX) were obtained by continuous exposure to increasing DOX concentrations, and were maintained in DMEM containing DOX (100nM). This subline expresses a unique membrane glycoprotein with a molecular weight of 180,000 and pos- sesses double minute chromosomes and homogeneously staining regions which are supposed to contain amplified DNA sequences responsible for multidrug resistance (Tsuruo et al., 1986; Sugimoto & Tsuruo, 1987). Cell growth and viability were determined by phase contrast microscopy with 0.1% Trypan Blue. A previous report showed that high ACM levels (10pgrml'), in combination with DOX or daunomycin, in- creased the intracellular amount of the latter compounds with concomitant increased cytotoxicity against resistant cells (Tapiero et al., 1988). In this paper, we have extended these inhibition growth studies with non-toxic doses of ACM against DOX-resistant K562 cells in order to determine which kind of exposure with ACM and DOX, simultaneously or sequentially, produced the most cytotoxicity. To explain the reversal of resistance by ACM, we have measured the DOX uptake directly into nuclei from living cells, using microspectrofluorometry. Kinetics of DOX uptake and DOX retention from sensitive and resistant cell nuclei have been yp In drug uptake studies, cells in exponential growth phase were incubated at 5 x 104ml1' density in DMEM containing the appropriate drug concentration, using 24-well multidishes in a moist air/CO2 incubator at 37°C. Cells were then washed free of drug and seeded on a Petri dish containing PBS for the microspectrofluorometric analysis. Routine measurements of the fluorescence of 20 nuclei in each sample were made. DOXORUBICIN RESISTANCE REVERSAL BY ACLACINOMYCIN A 679 quantum yield of each diode and for the transmission of the interference filter. Inhibition of cell growth Growth inhibition (GI) was determined as follows. Cells in exponential growth phase at 5 x I04ml-' density were incubated for 4 h in the medium containing the appropriate amount of drugs. Cells were then washed and resuspended in drug-free medium. After 3 days, triplicate cell counting was performed by phase contrast microscopy. Determination of the DOX concentration in living cell nuclei The fluorescence signal at a given wavelength, arising from the nucleus of a cell treated with DOX F(A), can be expressed as a sum of spectral contributions of free DOX, DNA-bound DOX and an intrinsic nuclear signal, i.e. N2= N,-2Nd(T2 -T1) (1) (1) (2) F(1) = Cf.FKQ) + Cb.Fb(Q) + cx.Fn(Q) F(1) = Cf.FKQ) + Cb.Fb(Q) + cx.Fn(Q) (2) Where Ff and Fb are the fluorescence spectra of free and bound drug referred to unitary concentration. Taking into account the unitary concentration, Cf and Cb represent int- ranuclear concentrations of free and bound drug respectively. ai is the contribution of a fictitious intranuclear component responsible for the intrinsic nuclear spectrum F,. In a recent paper (Gigli et al., 1988), we have shown that each of these contributions has a characteristic spectral shape determined independently by studies of aqueous solutions. Reference spectra, corrected for buffer contribution, for free DOX (0.12 tM) (Ff) in PBS solution and DOX (2pM) bound to DNA (concentration in phosphate: 1 mM) in PBS (Fb) are presented in Figure 1, after normalisation of the integrated surfaces. The fluorescence yield of DOX in the free form is higher than that of the bound form by a factor 48 ± 2. For simultaneous incubation with ACM and DOX, the intranuclear spectral analysis must take into account two additional contributions derived from the following com- pounds: (i) ACM; and (ii) 7-deoxyaklavinone, termed C,. The fluorescence yield of free ACM (5 x 10- M) is 200 times higher compared to the ACM (5 x 10- M) bound to DNA (1 mM) (Manfait et al., 1988). C,, one intracellular metabolite from ACM with an altered chromophore, results from an enzymatic cleavage of the trisaccharide and has been isolated Inhibition of cell growth was calculated as a ratio between doubling number of treated cells and doubling number of untreated cells. G120 for example, is defined as the drug concentration which reduces to 20% the doubling number (Nd) of treated cells in comparison to control. Inhibition of cell growth was calculated as a ratio between doubling number of treated cells and doubling number of untreated cells. G120 for example, is defined as the drug concentration which reduces to 20% the doubling number (Nd) of treated cells in comparison to control. Where Ff and Fb are the fluorescence spectra of free and bound drug referred to unitary concentration. The microspectrofluorometer Fluorescence emission spectra from a microvolume within a living cell were recorded with a microspectrofluorometer (modified Raman spectrometer OMARS 89, DILOR, Lille, France) as described previously (Ginot et al., 1984; Gigli et al., 1988; Millot et al., 1989). An optical microscope (Olympus BH2) equipped with a 100 x water immersion objective (Leitz Fluotar) and phase contrast allows us to: (i) observe the sample; (ii) focus a laser beam at 457.9 nm (Spectra Physics Ar+ 2020/03) on a spot of 1 ltm diameter; and (iii) collect the fluorescence emission through the same optics. The fluorescence sampling was restricted to a volume of about 30ym3 with a pinhole dia- phragm of 200 lum diameter on the image plane of the micro- scope objective. An interference filter (MTO J480) is used to decrease the scattered light intensity from the excitation line by 10'2-fold. The emitted light signal, spectrally dispersed by a diffraction grating, was detected with an optical multichan- nel analyser consisting of a cooled 512 element diode array, optically coupled with an image intensifier. Data were trans- ferred to a Goupil G4 computer for analysis with the specifically developed program Spectre. For simultaneous incubation with ACM and DOX, the intranuclear spectral analysis must take into account two additional contributions derived from the following com- pounds: (i) ACM; and (ii) 7-deoxyaklavinone, termed C,. The fluorescence yield of free ACM (5 x 10- M) is 200 times higher compared to the ACM (5 x 10- M) bound to DNA (1 mM) (Manfait et al., 1988). C,, one intracellular metabolite from ACM with an altered chromophore, results from an enzymatic cleavage of the trisaccharide and has been isolated CU C6 4- enI In Co .E G) C.) a) Ca) a)0 i c Wavelength (nm) Figure 1 Fluorescence emission spectra, corrected for the buffer contribution, of anthracyclines in PBS solution. Free DOX (0.12 uM) (A), DOX (2ILM) bound to DNA (1 mM) (0), after normalisation of the integrated surfaces. Ratio between fluorescence yield of DOX bound to DNA and free DOX: 1/48. Free ACM (0.5 IiM) (), C, (0.5 fiM) (-). Laser excitation wavelength: A = 457.9 nm, laser power at the sample: 4 ItW. CU C6 4- enI In Co .E G) C.) a) Ca) a)0 i c Wavelength (nm) Determination of the DOX concentration in living cell nuclei Taking into account the unitary concentration, Cf and Cb represent int- ranuclear concentrations of free and bound drug respectively. ai is the contribution of a fictitious intranuclear component responsible for the intrinsic nuclear spectrum F,. In a recent paper (Gigli et al., 1988), we have shown that each of these contributions has a characteristic spectral shape determined independently by studies of aqueous solutions. Reference spectra, corrected for buffer contribution, for free DOX (0.12 tM) (Ff) in PBS solution and DOX (2pM) bound to DNA (concentration in phosphate: 1 mM) in PBS (Fb) are presented in Figure 1, after normalisation of the integrated surfaces. The fluorescence yield of DOX in the free form is higher than that of the bound form by a factor 48 ± 2. Cells Cell growth was estimated as doubling number per 24 h, Nd, according to equation (1) for cells in exponential phase, were N, and N2 are cell concentrations at times T, and T2, i.e. Laser power and instrumental response were controlled by the daily use of rhodamine B (C= 70 nM in ethyl alcohol solution) as an external standard. Microspectrofluorometric measurements For the microspectrofluorometric analysis, the cells were incubated in the medium containing DOX and ACM. Then they were washed free of drug in cooled PBS at 4°C and seeded in a Petri dish containing PBS. These survival condi- tions without glucose decrease the energy metabolism and the active outward transport of resistant cells. By repeated measurements at 37°C, at regular intervals for more than 1 h, on the same location on a single cell nucleus, a decrease of intranuclear DOX concentration could also be detected for K562 and K562-DOX cell lines. The results of the above measurements are represented by the following equation: C = CO x (1- t x 0.003) C = CO x (1- t x 0.003) The point from Figure 4 corresponding to a simul- taneous exposure with DOX (1,000 nM) and ACM (1,000 nM) (GI50 at 1 h) is reported in Figure Sb and confirms this conclusion. In contrast the point from Figure 4 corres- ponding to a sequential exposure with ACM (3,000 nM) fol- and identified by HPLC (Ogasawara et al., 1981; Egorin et al., 1982) and microspectrofluorometric studies. Thus, equa- tion (2) becomes: represented on an isobologram as shown in Figure 5, corres- ponding to 4 and 1 h exposure time respectively. This isoeffect plot is defined from the ACM and DOX doses that give growth inhibitory values that add up to the level GI50. The addition is performed by taking the increments in DOX doses starting from zero for calculation by mode I, and from the GIso for calculation by mode II (Steel et al., 1979). The datum from Figure 3 corresponding to an exposure of K562- DOX cells to DOX (1,500 nM) and ACM (50 nM) (GI5o at 4 h) has been reported on the isobologram (Figure 5a). This point appears on the left of the envelope of additivity and could be interpreted in terms of a cytotoxic potentiation mechanism between ACM and DOX against K562-DOX cells. The point from Figure 4 corresponding to a simul- taneous exposure with DOX (1,000 nM) and ACM (1,000 nM) (GI50 at 1 h) is reported in Figure Sb and confirms this conclusion. In contrast the point from Figure 4 corres- ponding to a sequential exposure with ACM (3,000 nM) fol- (3) F(A) = Cf.Ff (A) + Cb.Fb(A) + 1x.F,(A) + P.Fa(I) + y.FC(A) (3 Where Fa and F, are fluorescence spectra of ACM and Cl (Figure 1). P and A are their surface contributions in the intranuclear spectrum F(1) of a K562 cell after exposure to ACM and DOX. Using the resolution of the diode array detector, equation (3) corresponds to a system of 512 equa- tions, which are solved by minimisation of term 4 with a least squares algorithm, and leads to DOX concentrations in the living cell nucleus (Cf, Cb). C = CO x (1- t x 0.003) 700 nm Z [F(A) - (Cf.Ff&A) + Cb.Fb(A) + ,.F.(A) + P.Fa(A) + y-.F(A))J2 (4) A= 500 700 nm Z [F(A) - (Cf.Ff&A) + Cb.Fb(A) + ,.F.(A) + P.Fa(A) + y-.F(A))J2 (4) A= 500 Z [F(A) - (Cf.Ff&A) + Cb.Fb(A) + ,.F.(A) + P.Fa(A) + y-.F(A))J2 (4) A= 500 A fluorescence emission spectrum, as determined from a selected microvolume in the nucleus of a K562 cell after a simultaneous exposure with ACM and DOX, is shown in Figure 2. The resolution into five components (free DOX, bound - DNA DOX, ACM, C, and the intrinsic nuclear contribution), leading to the computed DOX concentration, is also shown in Figure 2. Notice that the free DOX con- tributes about one-quarter of the total DOX signal, although this species only constitutes 0.2% of the total drug concent- ration. This result is general for K562 and K562-DOX cell lines and for the investigated range of drug concentrations in the medium (Gigli et al., 1989). . .o :LI a) a) 8 e) 0 CD) Co 0) 0 Wavelength (nm) Figure 2 Analysis of the emission spectrum from a microvolume within the nucleus of a K562 cell exposed 2 h to I LM DOX and I tLM ACM. The experimental spectrum (A) is resolved, using the minimisation of term 4 reported in the Material and methods section, into five contributions derived from these compounds: free DOX (A), DNA-bound DOX (A), free ACM (), C, (-) and an untreated cell nucleus (0). The corrected fluorescence emission spectrum of DOX from a treated cell nucleus is the spectrum (0). Total intranuclear concentration of DOX, evaluated from equation (3), is 82±8 iM. Conditions of laser excitation as in Figure 1. . .o :LI a) a) 8 e) 0 CD) Co 0) 0 Wavelength (nm) . .o :LI 8 . .o :LI a) a) 8 e) 0 CD) Co 0) 0 C = CO x (1- t x 0.003) Where C is the intranuclear DOX concentration depending on time t, which is expressed in minutes. Where C is the intranuclear DOX concentration depending on time t, which is expressed in minutes. All data reported in this work have been collected from a sample of 20-30 different cell nuclei within the first 15 min after transferring cells in PBS. Under these experimental conditions, the intranuclear concentrations remained essen- tially unaffected (of the order of 5%) during the total time interval within these limits. Twenty spectra from the same intracellular location were accumulated in order to increase the signal to noise ratio. Sample heating and photobleaching were found to be negligible under our experimental condi- tions. A light power of 4 yW at the sample and an illumina- tion time of 1 s were used. Cells always remained viable after repeated fluorescence measurements as determined by phase contrast microscopy. Spectra were numerically corrected for the photodiode array response due to small differences in the Figure 1 Fluorescence emission spectra, corrected for the buffer contribution, of anthracyclines in PBS solution. Free DOX (0.12 uM) (A), DOX (2ILM) bound to DNA (1 mM) (0), after normalisation of the integrated surfaces. Ratio between fluorescence yield of DOX bound to DNA and free DOX: 1/48. Free ACM (0.5 IiM) (), C, (0.5 fiM) (-). Laser excitation wavelength: A = 457.9 nm, laser power at the sample: 4 ItW. J.-M. MILLOT et al. 680 represented on an isobologram as shown in Figure 5, corres- ponding to 4 and 1 h exposure time respectively. This isoeffect plot is defined from the ACM and DOX doses that give growth inhibitory values that add up to the level GI50. The addition is performed by taking the increments in DOX doses starting from zero for calculation by mode I, and from the GIso for calculation by mode II (Steel et al., 1979). The datum from Figure 3 corresponding to an exposure of K562- DOX cells to DOX (1,500 nM) and ACM (50 nM) (GI5o at 4 h) has been reported on the isobologram (Figure 5a). This point appears on the left of the envelope of additivity and could be interpreted in terms of a cytotoxic potentiation mechanism between ACM and DOX against K562-DOX cells. Results p p y K562-DOX cells have been exposed to ACM and DOX, according to two different associations in order to compare their cytotoxic activity: (i) simultaneous exposures with ACM and DOX and (ii) ACM pretreatments followed by DOX alone (Figure 4). For each incubation process, the concentra- tion ratio ACM/DOX = 1, and the exposure time to each drug is 1 h. Compared to the ACM pretreatments, simul- taneous exposures to both anthracyclines produced higher cell growth inhibitions. This observation became more evi- dent for ACM and DOX doses superior to I tIM. A synergistic effect between DOX and ACM against K562- DOX cells can be justified on the basis of deviations from the expected additive cytotoxicity determined with each drug used alone and represented by an isobologram (Steel et al., 1979). Since the dose-response curves of ACM and DOX are non-linear, the expected additive GI50 of K562-DOX cells is y K562-DOX cells have been exposed to ACM and DOX, according to two different associations in order to compare their cytotoxic activity: (i) simultaneous exposures with ACM and DOX and (ii) ACM pretreatments followed by DOX alone (Figure 4). For each incubation process, the concentra- tion ratio ACM/DOX = 1, and the exposure time to each drug is 1 h. Compared to the ACM pretreatments, simul- taneous exposures to both anthracyclines produced higher cell growth inhibitions. This observation became more evi- dent for ACM and DOX doses superior to I tIM. Table I Growth inhibitory concentration (GI) of DOX and ACM in K562 and K562-DOX cells Table I Growth inhibitory concentration (GI) of DOX and ACM in K562 and K562-DOX cells DOX ACM K562 K562-DOX K562 K562-DOX G120 (nM) 50 1000 20 180 G150 (nM) 80 2100 65 450 GIoo (nM) 200 5000 800 1400 Growth inhibition was determined on cells exposed for 4 h to a given drug concentration, and resuspended for 3 days in drug-free medium. Growth inhibition was calculated as per cent ratio between doubling number oftreated cells and doubling number of untreated cells. Growth inhibitory concentrations were calculated from three to five independent determinations. Each standard deviation was less than 15% of the average. Results .o a) a) 8 e) 0 CD) Co 0) 0 Wavelength (nm) Effect of ACM on DOX-induced growth inhibition in sensitive and resistant K562 cells Effect of ACM on DOX-induced growth inhibition in sensitive and resistant K562 cells Effect of ACM on DOX-induced growth inhibition in sensitive and resistant K562 cells The growth inhibition following 4 h exposure to ACM or DOX was determined as reported in the Materials and methods section. Table I shows the growth inhibitory activity of ACM and DOX on K562 and K562-DOX cells. K562- DOX cells, selected for 25-fold resistance to DOX, were 7-fold resistant to ACM, when compared with the GI50 of both cell lines. To detect a synergy mechanism between ACM and DOX on cellular toxicity, cells were incubated for 4h in DOX without and with non-toxic ACM concentrations (growth inhibition < 5%): 10nM for K562, 50nM for K562-DOX. Growth of both cell lines was evaluated by their cellular doubling number. Cytotoxic effects of each incubation pro- cess are compared in Figure 3. For the K562 line, simul- taneous incubation with DOX and a non-toxic ACM dose (10 nM) did not produce any enhancement of the DOX cytotoxicity. On the contrary, for the resistant line, incuba- tions with DOX and a non-toxic ACM dose (50nM) pro- moted a partial restoration of the DOX activity. For exam- ple, DOX (2,000 nM) associated with ACM (50 nM) gives a similar toxicity to DOX (3,000 nM) alone. In this case, ACM (50 nM) allows a DOX dose decrease of 33%, which corres- ponds to a partial reversal of anthracycline resistance. Wavelength (nm) Figure 2 Analysis of the emission spectrum from a microvolume within the nucleus of a K562 cell exposed 2 h to I LM DOX and I tLM ACM. The experimental spectrum (A) is resolved, using the minimisation of term 4 reported in the Material and methods section, into five contributions derived from these compounds: free DOX (A), DNA-bound DOX (A), free ACM (), C, (-) and an untreated cell nucleus (0). The corrected fluorescence emission spectrum of DOX from a treated cell nucleus is the spectrum (0). Total intranuclear concentration of DOX, evaluated from equation (3), is 82±8 iM. Conditions of laser excitation as in Figure 1. Results Table I Growth inhibitory concentration (GI) of DOX and ACM in K562 and K562-DOX cells DOX ACM K562 K562-DOX K562 K562-DOX G120 (nM) 50 1000 20 180 G150 (nM) 80 2100 65 450 GIoo (nM) 200 5000 800 1400 Growth inhibition was determined on cells exposed for 4 h to a given drug concentration, and resuspended for 3 days in drug-free medium. Growth inhibition was calculated as per cent ratio between doubling number oftreated cells and doubling number of untreated cells. Growth inhibitory concentrations were calculated from three to five independent determinations. Each standard deviation was less than 15% of the average. A synergistic effect between DOX and ACM against K562- DOX cells can be justified on the basis of deviations from the expected additive cytotoxicity determined with each drug used alone and represented by an isobologram (Steel et al., 1979). Since the dose-response curves of ACM and DOX are non-linear, the expected additive GI50 of K562-DOX cells is DOXORUBICIN RESISTANCE REVERSAL BY ACLACINOMYCIN A 681 a -i± a) 0) a)0 Qx 0a .3 C 0c 0 4) -C 0) .3 o) a1) cJ a at) -a ~0 0 0 LU nu /U UU U Extracellular DOX concentration (nM) Figure 3 Growth inhibitory effect of doxorubicin on K562 and K562-DOX cells, without and with aclacinomycin A. Cells were simultaneously exposed to both drugs for 4 h at 37'C, resuspended for 3 days in drug-free medium and then counted. (0) K562 (OnM ACM); (-) K562 (lOnM ACM); (0) K562- DOX (0 nM ACM); (-) K562-DOX (50 nM ACM). Growth inhibitory effect was calculated as in Table I. Each point represents the mean and each bar the standard deviation of three experiments. Observed differences in growth inhibitions of K562- DOX cells (,O) are significant: P<0.001. ACM doses (>M) ACM doses (>M) Figure 5 Isobolograms (or isoeffect plots) of G150 against K562- DOX cells with ACM-DOX associations. The expected additive G150 are represented with calculations by mode I (-) and by mode II (0) as defined by Steel et al. (1979). a, Cells were exposed for 4 h to each drug; (0) is a projection from Figure 3 of ACM-DXR simultaneous incubation that gives GI50. Results b, Cells were exposed for I h to each drug; (x) and (0) are projections from Figure 4 of simultaneous and sequential incubations respec- tively that give G150- lowed by DOX (3,000 nM) (GI50 at 1 h) gives a point that lies within the envelope of additivity. lowed by DOX (3,000 nM) (GI50 at 1 h) gives a point that lies within the envelope of additivity. -a0 0 a) 0 ACM 250 500 1000 2000 3000 ACM + DOX doses (nM) Figure 4 Comparison of drug associations (ACM and DOX), on the growth inhibition of K562-DOX cells. (O) Cells were exposed simultaneously to both drugs for I h; (0) cells were pre-treated for I h with ACM, washed and resuspended for I h in ACM-free medium containing DOX. Cells were washed and after 3 days in drug-free medium, growth inhibitory effects were calculated. Drug concentration ratio ACM/DOX = 1. Each point represents the mean and each bar the standard deviation of three experiments. \n 1 -a0 0 a) 0 ACM 250 500 1000 2000 3000 ACM + DOX doses (nM) \n 1 Results b, Cells were exposed for I h to each drug; (x) and (0) are projections from Figure 4 of simultaneous and sequential incubations respec- tively that give G150- 1 \n a -i± a) 0) a)0 Qx 0a a at) -a ~0 0 0 ACM doses (>M) ACM doses (>M) Figure 5 Isobolograms (or isoeffect plots) of G150 against K562- DOX cells with ACM-DOX associations. The expected additive G150 are represented with calculations by mode I (-) and by mode II (0) as defined by Steel et al. (1979). a, Cells were exposed for 4 h to each drug; (0) is a projection from Figure 3 of ACM-DXR simultaneous incubation that gives GI50. b, Cells were exposed for I h to each drug; (x) and (0) are projections from Figure 4 of simultaneous and sequential incubations respec- tively that give G150- 1 a -i± a) 0) a)0 Qx 0a ACM doses (>M) 1 .3 C 0c 0 4) -C \n a -i± a) 0) a)0 Qx 0a a at) -a ~0 0 0 ACM doses (>M) ACM doses (>M) 1 0) .3 o) a1) cJ LU nu /U UU U Extracellular DOX concentration (nM) \n ACM doses (>M) Figure 3 Growth inhibitory effect of doxorubicin on K562 and K562-DOX cells, without and with aclacinomycin A. Cells were simultaneously exposed to both drugs for 4 h at 37'C, resuspended for 3 days in drug-free medium and then counted. (0) K562 (OnM ACM); (-) K562 (lOnM ACM); (0) K562- DOX (0 nM ACM); (-) K562-DOX (50 nM ACM). Growth inhibitory effect was calculated as in Table I. Each point represents the mean and each bar the standard deviation of three experiments. Observed differences in growth inhibitions of K562- DOX cells (,O) are significant: P<0.001. \n Figure 5 Isobolograms (or isoeffect plots) of G150 against K562- DOX cells with ACM-DOX associations. The expected additive G150 are represented with calculations by mode I (-) and by mode II (0) as defined by Steel et al. (1979). a, Cells were exposed for 4 h to each drug; (0) is a projection from Figure 3 of ACM-DXR simultaneous incubation that gives GI50. Effect ofACM on the intranuclear DOX uptake A previous study reported that DOX cytotoxicity was depen- dent on the drug concentration in nuclei and that the resis- tant K562 subline exhibited an altered DOX uptake (Gigli et al., 1989). To determine whether co-incubation with ACM produces an enhancement of DOX intranuclear concentra- tion, we have measured intranuclear concentrations of DOX by microspectrofluorometry, as a function of extracellular doses of ACM. This relationship is shown in Figure 6. For K562-DOX cells, after 4 h of incubation with ACM and DOX (1 JAM), DOX intranuclear concentrations were in- creased proportionally to ACM extracellular doses. In con- trast, for the sensitive line, the DOX uptake into the nucleus was found to be independent of the ACM dose in the medium. The time-course of DOX intranuclear uptake and the effect of ACM (1 JAM) on this process is shown in Figure 7. The rate of DOX accumulation in the parent line is not influenced by the presence of ACM. In resistant cells, the intranuclear concentration of DOX is increased 5-fold with the presence of ACM (1 ILM). During the first hour, this uptake is similar to the one of sensitive cells, but afterwards the DOX uptake reaches a maximum, contrary to the sen- sitive cells. Figure 4 Comparison of drug associations (ACM and DOX), on the growth inhibition of K562-DOX cells. (O) Cells were exposed simultaneously to both drugs for I h; (0) cells were pre-treated for I h with ACM, washed and resuspended for I h in ACM-free medium containing DOX. Cells were washed and after 3 days in drug-free medium, growth inhibitory effects were calculated. Drug concentration ratio ACM/DOX = 1. Each point represents the mean and each bar the standard deviation of three experiments. oC04 . L. 04- ao1 .C c cJ 0 C.) Time (h) Figure 8 Effect of ACM on the DOX retention from nuclei of K562 and K562-DOX cells. Cells were exposed to DOX (K562, 0.5 jIM; K562-DOX, 5lM) for I h at 37°C. Initial intranuclear concentrations of DOX were 83 ± 1311M and 49 ± 11 jiM for K562 and K562-DOX respectively. Cells were washed with PBS and resuspended in DOX-free medium, without and with ACM (I1jM). (0) K562 (OjgM ACM); (A) K562 (I jM ACM); (-) K562-DOX (O IM ACM); (A) K562-DOX (I jiM ACM). Each point represents the mean of three replicates with 30 cell nuclei. Effect ofACM on the intranuclear DOX retention DOX concentration values, as determined from 30 measurements. I I I Where t is time and T1/2 is the time corresponding to decrease of the intranuclear DOX concentration. DO) from resistant cells (T1/2= 30 min) occurs faster tha from sensitive cells (T1/2 = 8 h). The presence of ACM during the DOX efflux phase increased the DOX rel from K562-DOX cell nuclei (T1/2 =90 min), but did no' the efflux from the nuclei of sensitive cells. a 50% ( efflux Ln that (1 JM) tention t affect Figure 8 Effect of ACM on the DOX retention from nuclei of K562 and K562-DOX cells. Cells were exposed to DOX (K562, 0.5 jIM; K562-DOX, 5lM) for I h at 37°C. Initial intranuclear concentrations of DOX were 83 ± 1311M and 49 ± 11 jiM for K562 and K562-DOX respectively. Cells were washed with PBS and resuspended in DOX-free medium, without and with ACM (I1jM). (0) K562 (OjgM ACM); (A) K562 (I jM ACM); (-) K562-DOX (O IM ACM); (A) K562-DOX (I jiM ACM). Each point represents the mean of three replicates with 30 cell nuclei. The results are given with an accuracy of 10%. Data were fitted to equation (5) with the following T,/2: (0) 8 h; (A) 8 h; (O) 30 min; (A) 90 min. Figure 8 Effect of ACM on the DOX retention from nuclei of K562 and K562-DOX cells. Cells were exposed to DOX (K562, 0.5 jIM; K562-DOX, 5lM) for I h at 37°C. Initial intranuclear concentrations of DOX were 83 ± 1311M and 49 ± 11 jiM for K562 and K562-DOX respectively. Cells were washed with PBS and resuspended in DOX-free medium, without and with ACM (I1jM). (0) K562 (OjgM ACM); (A) K562 (I jM ACM); (-) K562-DOX (O IM ACM); (A) K562-DOX (I jiM ACM). Each point represents the mean of three replicates with 30 cell nuclei. The results are given with an accuracy of 10%. Data were fitted to equation (5) with the following T,/2: (0) 8 h; (A) 8 h; (O) 30 min; (A) 90 min. Effect ofACM on the intranuclear DOX retention A more detailed study of DOX uptake and retenti4 explain the weaker efficiency of sequential exposures to followed by DOX, compared to simultaneous exposi both drugs, which have been reported in the Effect of on will ACM ares to r ACM i r%3 - 0 C0 C.) C) c 4) 0) C 0) a ACM concentration (nM) Figure 6 DOX uptake in the nucleus of K562 (0) and I DOX (O) cells, as a function of extracellular concentratic ACM. Cells were exposed simultaneously to DOX (I piM ACM. After 4 h, intranuclear concentrations of DOX were mined. Vertical bars denote standard deviations on the nuclear DOX concentration values, as determined froi measurements. 04 K562- )ns of [) and deter- intr 1sss a- m 30 i r%3 - 0 C0 C.) C) c 4) 0) C 0) a ACM concentration (nM) 04 on DOX-induced growth inhibition section. DOX uptakes in resistant cell nuclei are compared in Figure 9, as a function of the time, according to the type of incubation with associated DOX and ACM. DOX was permanently present in the medium, but ACM was added with various delays. The similarity of curves (O) and () shows that a pretreat- ment with ACM does not influence the intranuclear DOX uptake. A comparison of curves (U) and (A) shows that a removal of ACM from the extracellular medium induces a fast DOX efflux out of the nucleus. Within 30 min the intranuclear DOX concentration has decreased by 50%, corresponding to the initial T,/2 value (30min) of resistant cells without ACM. However, the inhibition of DOX efflux by ACM is reversible and requires the permanent presence of ACM in the medium. Figure 6 DOX uptake in the nucleus of K562 (0) and I DOX (O) cells, as a function of extracellular concentratic ACM. Cells were exposed simultaneously to DOX (I piM ACM. After 4 h, intranuclear concentrations of DOX were mined. Vertical bars denote standard deviations on the nuclear DOX concentration values, as determined froi measurements. K562- )ns of [) and deter- intr 1sss a- m 30 -i i 00 4-1C 0) U 0 C x0a co 0) cB 4-. 0) Exposure time to DOX (h) Figure 9 Effects of sequential and simultaneous ACM-DOX incubations, on the intranuclear uptake of DOX in K562-DOX cells, as a function of the time. J.-M. MILLOT et al. 682 oC04 . L. 04- ao1 .C c cJ 0 C.) Time (h) I Effect ofACM on the intranuclear DOX uptake The results are given with an accuracy of 10%. Data were fitted to equation (5) with the following T,/2: (0) 8 h; (A) 8 h; (O) 30 min; (A) 90 min. I J.-M. MILLOT et al. J.-M. MILLOT et al. Effect ofACM on the intranuclear DOX retention Effect ofACM on the intranuclear DOX retention It has been shown repeatedly that calcium antagonists and calmodulin inhibitors stimulate drug accumulation in resis- tant cells by inhibition of the drug efflux. The intranuclear retention of DOX from preloaded cells, determined by mic- rospectrofluorometry, is shown in Figure 8. When CO is the initial intranuclear DOX concentration of preloaded cells, the DOX concentration (C) retained inside the nucleus in the presence of DOX-free medium is described by an exponential decrease: C = Co. 10--log2/Tj/2 (5) (5) C = Co. 10--log2/Tj/2 Where t is time and T1/2 is the time corresponding to decrease of the intranuclear DOX concentration. DO) from resistant cells (T1/2= 30 min) occurs faster tha from sensitive cells (T1/2 = 8 h). The presence of ACM during the DOX efflux phase increased the DOX rel from K562-DOX cell nuclei (T1/2 =90 min), but did no' the efflux from the nuclei of sensitive cells. A more detailed study of DOX uptake and retenti4 explain the weaker efficiency of sequential exposures to followed by DOX, compared to simultaneous exposi both drugs, which have been reported in the Effect of i r%3 - 0 C0 C.) C) c 4) 0) C 0) a ACM concentration (nM) Figure 6 DOX uptake in the nucleus of K562 (0) and I DOX (O) cells, as a function of extracellular concentratic ACM. Cells were exposed simultaneously to DOX (I piM ACM. After 4 h, intranuclear concentrations of DOX were mined. Vertical bars denote standard deviations on the nuclear DOX concentration values, as determined froi measurements. I 4-C 0 C] a) c0 0 0) C3 0) C 103 102 - lo, 1n A (5) a 50% ( efflux Ln that (1 JM) tention t affect on will ACM ares to r ACM C.) Time (h) Figure 8 Effect of ACM on the DOX retention from nuclei of K562 and K562-DOX cells. Cells were exposed to DOX (K562, 0.5 jIM; K562-DOX, 5lM) for I h at 37°C. Initial intranuclear concentrations of DOX were 83 ± 1311M and 49 ± 11 jiM for K562 and K562-DOX respectively. Cells were washed with PBS and resuspended in DOX-free medium, without and with ACM (I1jM). (0) K562 (OjgM ACM); (A) K562 (I jM ACM); (-) K562-DOX (O IM ACM); (A) K562-DOX (I jiM ACM). Each point represents the mean of three replicates with 30 cell nuclei. Effect ofACM on the intranuclear DOX retention The results are given with an accuracy of 10%. Data were fitted to equation (5) with the following T,/2: (0) 8 h; (A) 8 h; (O) 30 min; (A) 90 min. on DOX-induced growth inhibition section. DOX uptakes in resistant cell nuclei are compared in Figure 9, as a function of the time, according to the type of incubation with associated DOX and ACM. DOX was permanently present in the medium, but ACM was added with various delays. The similarity of curves (O) and () shows that a pretreat- ment with ACM does not influence the intranuclear DOX uptake. A comparison of curves (U) and (A) shows that a removal of ACM from the extracellular medium induces a fast DOX efflux out of the nucleus. Within 30 min the 04 intranuclear DOX concentration has decreased by 50%, corresponding to the initial T,/2 value (30min) of resistant cells without ACM. However, the inhibition of DOX efflux K562- by ACM is reversible and requires the permanent presence of )ns of ACM in the medium. [) and deter- intr 1sss a- m 30 -i i 00 4-1C 0) U 0 C x0a co 0) cB 4-. 0) Exposure time to DOX (h) 0 I -u 1 2 3 4 Exposure time (h) Figure 7 Effect of ACM on the DOX up K562-DOX cell nuclei, as a function of th exposed to DOX (1 giM), without and with K562 (O0 IM ACM); (A) K562 (1 jiM ACM (O giM ACM); (A) K562-DOX (I jiM ACM).' standard deviations on the intranuclear E values, as determined from 30 measurement Figure 9 Effects of sequential and simultaneous ACM-DOX 5 6 7 incubations, on the intranuclear uptake of DOX in K562-DOX cells, as a function of the time. From time 0, K562-DOX cells were exposed to DOX (I pM) at 37TC: (O) without ACM; () )take in K562 and cells were pre-treated for I h with ACM (1 piM) alone, washed ie time. Cells were and resuspended at time 0 in DOX (I piM) alone; (A) cells were ACM (1 piM). (0) exposed for the first hour simultaneously to ACM (I jiM) with [); (U) K562-DOX DOX, washed and resuspended in DOX alone; (-) cells were' Vertical bars denote incubated for 4 h simultaneously with ACM (I piM) and DOX. )OX concentration Vertical bars denote standard deviations on the intranuclear s. Effect ofACM on the intranuclear DOX retention From time 0, K562-DOX cells were exposed to DOX (I pM) at 37TC: (O) without ACM; () cells were pre-treated for I h with ACM (1 piM) alone, washed and resuspended at time 0 in DOX (I piM) alone; (A) cells were exposed for the first hour simultaneously to ACM (I jiM) with DOX, washed and resuspended in DOX alone; (-) cells were' incubated for 4 h simultaneously with ACM (I piM) and DOX. Vertical bars denote standard deviations on the intranuclear DOX concentration values, as determined from 30 measurements. -i i 00 4-1C 0) U 0 C x0a co 0) cB 4-. 0) Exposure time to DOX (h) I 4-C 0 C] a) c0 0 0) C3 0) C 103 102 - lo, 1n A 0 I -u 1 2 3 4 Exposure time (h) 5 6 7 I I I I 4-C 0 C] a) c0 0 0) C3 0) C 103 102 - lo, 1n A 0 I -u 1 2 3 4 Exposure time (h) Figure 7 Effect of ACM on the DOX up K562-DOX cell nuclei, as a function of th exposed to DOX (1 giM), without and with K562 (O0 IM ACM); (A) K562 (1 jiM ACM (O giM ACM); (A) K562-DOX (I jiM ACM).' standard deviations on the intranuclear E values, as determined from 30 measurement 5 6 7 )take in K562 and ie time. Cells were ACM (1 piM). (0) [); (U) K562-DOX Vertical bars denote )OX concentration s. I I I Exposure time to DOX (h) Figure 9 Effects of sequential and simultaneous ACM-DOX incubations, on the intranuclear uptake of DOX in K562-DOX cells, as a function of the time. From time 0, K562-DOX cells were exposed to DOX (I pM) at 37TC: (O) without ACM; () cells were pre-treated for I h with ACM (1 piM) alone, washed and resuspended at time 0 in DOX (I piM) alone; (A) cells were exposed for the first hour simultaneously to ACM (I jiM) with DOX, washed and resuspended in DOX alone; (-) cells were' incubated for 4 h simultaneously with ACM (I piM) and DOX. Vertical bars denote standard deviations on the intranuclear DOX concentration values, as determined from 30 measurements. Figure 9 Effects of sequential and simultaneous ACM-DOX incubations, on the intranuclear uptake of DOX in K562-DOX cells, as a function of the time. Discussion To avoid failure of chemotherapeutic treatment due to the emergence of drug resistance, attempts have been made to circumvent this problem either by using new drugs partic- ularly able to overcome anthracycline resistance, or by an association of drugs where one agent increases the phar- macological effects of the other. Several classes of drugs have been shown to reverse acquired resistance to anthracycline, including calcium channel blockers (verapamil) (Tsuruo et al., 1983; Friche et al., 1987), calmodulin antagonists (trifluoperazine, perhexilene maleate) (Ganapathi et al., 1984), triparanol analogues (tamoxifen) (Ramu et al., 1984), cardiac anti-arhythmics (quinidine, amiodarone) (Chauffert et al., 1986; Tsuruo et al., 1984) and cyclosporins (Twentyman, 1988). g Three hypotheses could explain the increase of the amount of intranuclear DOX caused by ACM: (i) an alteration of DOX influx, (ii) an alteration of DOX efflux, or (iii) a redistribution of DOX inside the cell, allowing it to reach targets more conducive to cytotoxicity. In the light of data implicating increased activity of a membrane-associated pump called P-glycoprotein in acquired resistance, particular attention was given to drug efflux out of the nucleus in our system. With direct observation inside the nucleus of living cells, using microspectrofluorometry, we show that DOX is bound to DNA by a reversible interaction. The DOX efflux from the nucleus occurs faster in K562-DOX cells compared to sensitive cells, with the result that resistant cells are able to reduce intranuclear DOX concentrations to sublethal values by active efflux. ACM, an antineoplasic agent (Umezawa et al., 1987), has been shown to circumvent anthracycline resistance at the high dose of 10 pgml-l (Tapiero et al., 1988). Moreover, a significant synergistic effect of cytotoxicity against P388 leukaemia has been observed with the associations ACM-cyclophosphamide and ACM-vincristine (Fugimoto et al., 1979). Compared with other drug combinations used to reverse the resistance, the interest in ACM is because of its own significant activity against a number of human tumours and its current use in clinical investigations (Kumai et al., 1984; Majima et al., 1987). Moreover, in vitro studies have shown a slight cross-resistance between ACM and DOX (Umezawa et al., 1987; Tapiero et al., 1988). This study demonstrates that the presence of ACM blocks the DOX efflux in the resistant cell line, resulting in enhanced accumulation of DOX and increased cytotoxicity. Discussion Because of the rapid accumulations of ACM into sensitive and multi- drug resistant lines (Seeber et al., 1980; Tapiero et al., 1988), the ACM intracellular deposition might in turn affect the DOX efflux. In contrast, another agent (forskolin) which reversed resistance to DOX, with increase of DOX cellular amounts, has virtually no effect on the rate of drug efflux (Wadler & Wiernik, 1988). The exact function of P- glycoprotein is not well known; however, a recent study has demonstrated drug binding to this protein (Cornwell et al., 1986), suggesting its involvement in drug transport. So, the altered efflux of DOX by ACM in resistant cells could result from a molecular interaction between ACM and the P- glycoprotein or from an altered expression of this protein. In conclusion, our results indicate that ACM partly reverses DOX-resistance at clinically achievable concentra- tions. We find that the reversal of resistance induced by ACM, associated with the DOX efflux inhibition, is reversible and requires the continuous presence of ACM in the medium. Thus, toxicologic, pharmacokinetic studies and clinical trials as resistance modifiers are now in progress by using the ACM-DOX association in simultaneous injections. p Any association of drugs is unable to restore the full sensitivity of resistant cells in vivo. For example, even if, in vitro, verapamil induces increased drug uptake in resistant cells, in vivo experiments indicate that verapamil, at a tolerated dose, can only partially overcome drug resistance (Friche et al., 1987). This result is probably entirely due to the toxicity limitation of dosage, which allows only a insufficient plasma concentration of verapamil during the DOX uptake phase. Since our principal aim was to reverse in vitro drug resistance in incubation conditions close to in vivo conditions, we have implemented short-term incubations with ACM doses corresponding to plasma concentration from 1 ttM to 200 nM during the first hour after an i.v. bolus (Egorin et al., 1982). These short-term incubations probably reflect, better than continuous incubations, the in vivo plasma conditions following a bolus injection. In these incubation conditions, an ACM dose which alone does not produce any growth inhibition on DOX resistant cells, does induce a partial reversal of DOX resistance. As for other agents that reverse r -sistance to DOX, the possible mechanism of this synergistic effect could depend on increased uptake of DOX induced by ACM in resistant cells. Effect ofACM on the intranuclear DOX retention From time 0, K562-DOX cells were exposed to DOX (I pM) at 37TC: (O) without ACM; () cells were pre-treated for I h with ACM (1 piM) alone, washed and resuspended at time 0 in DOX (I piM) alone; (A) cells were exposed for the first hour simultaneously to ACM (I jiM) with DOX, washed and resuspended in DOX alone; (-) cells were' incubated for 4 h simultaneously with ACM (I piM) and DOX. Vertical bars denote standard deviations on the intranuclear DOX concentration values, as determined from 30 measurements. Figure 9 Effects of sequential and simultaneous ACM-DOX incubations, on the intranuclear uptake of DOX in K562-DOX cells, as a function of the time. From time 0, K562-DOX cells were exposed to DOX (I pM) at 37TC: (O) without ACM; (*) cells were pre-treated for I h with ACM (1 piM) alone, washed and resuspended at time 0 in DOX (I piM) alone; (A) cells were exposed for the first hour simultaneously to ACM (I jiM) with DOX, washed and resuspended in DOX alone; (-) cells were' incubated for 4 h simultaneously with ACM (I piM) and DOX. Vertical bars denote standard deviations on the intranuclear DOX concentration values, as determined from 30 measurements. 3 4 Exposure time (h) Figure 7 Effect of ACM on the DOX up K562-DOX cell nuclei, as a function of th exposed to DOX (1 giM), without and with K562 (O0 IM ACM); (A) K562 (1 jiM ACM (O giM ACM); (A) K562-DOX (I jiM ACM).' standard deviations on the intranuclear E values, as determined from 30 measurement )take in K562 and ie time. Cells were ACM (1 piM). (0) [); (U) K562-DOX Vertical bars denote )OX concentration s. Figure 7 Effect of ACM on the DOX up K562-DOX cell nuclei, as a function of th exposed to DOX (1 giM), without and with K562 (O0 IM ACM); (A) K562 (1 jiM ACM (O giM ACM); (A) K562-DOX (I jiM ACM).' standard deviations on the intranuclear E values, as determined from 30 measurement )take in K562 and ie time. Cells were ACM (1 piM). (0) [); (U) K562-DOX Vertical bars denote )OX concentration s. This work was supported in part by Association contre le cancer and by Laboratoires Roger Bellon, Paris, France. The authors are grateful to Prof. R.E. Hester for reading the manuscript, to Dr N. Austin, Dr M. Gigli and Dr P. Joly for helpful discussions and advice, to Dr J.-F. Angiboust for his help in instrument optimisation and to DILOR (Lille, France) for their contribution in instrumenta- tion. Effect ofACM on the intranuclear DOX retention DOXORUBICIN RESISTANCE REVERSAL BY ACLACINOMYCIN A 683 cytotoxic effect of DOX with total cellular drug content is unclear (Lane et al., 1987), we have shown previously a direct relationship between the growth inhibition effect and the DOX amount actually in the nucleus (Gigli et al., 1989). Our findings suggest that there is an enhancement of DOX cytotoxicity with increased intranuclear concentration of this drug induced by ACM. The combined effect of these independent observations strongly supports the hypothesis that the nucleus is a target for DOX. Discussion The mechanism of DOX action has been attributed to inter- calation with DNA (Manfait et al., 1982) with the result that DNA replication and RNA synthesis is inhibited (Zunino et at.,1980), strand-breaking of DNA by bioreductive alkylation occurs (Moore, 1977) and oxygen-free radicals are generated (Bachur et al., 1979). Although the relationship between the This work was supported in part by Association contre le cancer and by Laboratoires Roger Bellon, Paris, France. The authors are grateful to Prof. R.E. Hester for reading the manuscript, to Dr N. Austin, Dr M. Gigli and Dr P. Joly for helpful discussions and advice, to Dr J.-F. Angiboust for his help in instrument optimisation and to DILOR (Lille, France) for their contribution in instrumenta- tion. EGORIN, M.J., VAN ECHO, D., FOX, B.M., WHITACRE, M. & BACHUR, N.R. (1982). Plasma kinetics of aclacinomycin A and its metabolites in man. Cancer Chemother. Pharmacol., 8, 41. FRICHE, E., SKOVSGAARD, T. & NISSEN, N.I. (1987). Effect of verapamil on daunorubicin accumulation in Ehrlich ascites tumor cells. Cancer Chemother. Pharmacol., 19, 35. FUGIMOTO, S., INAGAKI, I., HORIKOSHI, N. & OGAWA, M. (1979). Combination chemotherapy with a new anthracycline glycoside, aclacinomycin A and active drugs for malignant lymphomas in P388 mouse leukemia system. Gann, 70, 411. BACHUR, N.R., GORDON, S.L., GEE, M.V. & KON, H. (1979). NADPH cytochrome P-450 reductase activation of quinone anticancer agents to free radicals. Proc. Natl Acad. Sci. USA, 76, 954. BRADLEY, G., JURANKA, P.F. & LING, V. (1988). Mechanism of multidrug resistance. Biochim. Biophys. Acta, 948, 87. CHAUFFERT, B., MARTIN, M., HAMMAN, A., MICHEL, M. & MAR- TIN, F. (1986). Amiodarone-induced enhancement of doxorubicin and 4'-deoxydoxorubicin cytotoxicity to rat colon cancer cells in vitro and in vivo. Cancer Res., 46, 825. CORNWELL, M.M., GOTTESMAN, M.M. & PASTAN, I.H. (1986). In- d i i References BACHUR, N.R., GORDON, S.L., GEE, M.V. & KON, H. (1979). NADPH cytochrome P-450 reductase activation of quinone anticancer agents to free radicals. Proc. Natl Acad. Sci. USA, 76, 954. EGORIN, M.J., VAN ECHO, D., FOX, B.M., WHITACRE, M. & BACHUR, N.R. (1982). Plasma kinetics of aclacinomycin A and its metabolites in man. Cancer Chemother. Pharmacol., 8, 41. Pharmacol., 8, 41. FRICHE, E., SKOVSGAARD, T. & NISSEN, N.I. (1987). Effect of verapamil on daunorubicin accumulation in Ehrlich ascites tumor cells. Cancer Chemother. Pharmacol., 19, 35. BRADLEY, G., JURANKA, P.F. & LING, V. (1988). Mechanism of multidrug resistance. Biochim. Biophys. Acta, 948, 87. p y , , CHAUFFERT, B., MARTIN, M., HAMMAN, A., MICHEL, M. & MAR- TIN, F. (1986). Amiodarone-induced enhancement of doxorubicin and 4'-deoxydoxorubicin cytotoxicity to rat colon cancer cells in vitro and in vivo. Cancer Res., 46, 825. , , FUGIMOTO, S., INAGAKI, I., HORIKOSHI, N. & OGAWA, M. (1979). Combination chemotherapy with a new anthracycline glycoside, aclacinomycin A and active drugs for malignant lymphomas in P388 mouse leukemia system. Gann, 70, 411. , CORNWELL, M.M., GOTTESMAN, M.M. & PASTAN, I.H. (1986). In- creased vinblastine binding to membrane vesicles from multidrug- resistant KB cells. J. Biol. Chem., 261, 7921. J.-M. MILLOT et al. 684 MOORE, H.W. (1977). Bioactivation as a model for drug design bioreductive alkkylation. Science, 197, 527. GANAPATHI, R., GRABOWSKI, D., ROUSE, W. & RIEGLER, F. (1984). Differential effect of the calmodulin inhibitor trifluoperazine on cellular accumulation, retention, and cytotoxicity of anthracyc- lines in doxorubicin (Adriamycin)-resistant P388 leukemia cells. Cancer Res., 44, 5056. y OGASAWARA, T., MASUDA, Y., GOTO, S., MORI, S. & OKI, T. (1981). High performance liquid chromatographic determination of aclacinomycin A and its related compounds. II: Reverse phase HPLC determination of Aclacinomycin A and its metabolites in biological fluids using fluorescence detection. J. Antibiot., 38, 52. GIGLI, M., DOGLIA, S.M., MILLOT, J.-M., VALENTINI, L. & MAN- FAIT, M. (1988). Quantitative study of doxorubicin in living cell nuclei by microspectrofluorometry. Biochim. Biophys. Acta, 950, 13. y biological fluids using fluorescence detection. J. Antibiot., 38, 52. (1984) l f i d RAMU, A., GLAUBIGER, D. & FUKS, Z. (1984). Reversal of acquired resistance to doxorubicin in P388 murine leukemia cells by tamoxifen and other triparanol analogues. Cancer Res., 44, 4392. GIGLI, M., RASOANAIVO, T.W.D., MILLOT, J.-M. & 5 others (1989). References Correlation between growth inhibition and intranuclear dox- orubicin and 4'-iododoxorubicin quantitated in living K562 cells by microspectrofluorometry. Cancer Res., 49, 560. RIORDAN, J., DEUCHARS, K., KARTNER, N., ALON, N., TRENT, J. & LING. V. (1985). Amplification of P-glycoprotein gene in multidrug-resistant mammalian cell lines. Nature, 316, 817. SEEBER, S., LOTH, H. & CROOKE, S.T. (1980). Comparative nuclear and cellular incorporation of daunorubicin, doxorubicin, car- minomycin, marcellomycin, aclacinomycin A and AD 32 in daunorubicin-sensitive and -resistant Ehrlich Ascites in vitro. Cancer Res. Clin. Oncol., 98, 109. GINOT, L., JEANNESSON, P., ANGIBOUST, J.-F., JARDILLIER, J.-C. & MANFAIT, M. (1984). Interaction of adriamycin in sensitive and resistant leukemic cells, a comparative study by microspec- trofluorometry. Studia Biophys., 104, 145. y p y , , KARTNER, N., RIORDAN, J.R. & LING, V. (1983). Cell surface P- glycoprotein associated with multidrug resistance in mammalian cell lines. Science, 221, 1285. , , STEEL, G.G. & PECKHAM, M.J. (1979). Exploitable mechanisms in combined radiotherapy-chemotherapy: the concept of additivity. Int. J. Radiat. Oncol. Biol. Phys., 5, 85. , , KUMAI, K., KUBOTA, T., ISHIBIKI, K. & ABE, 0. (1984). Experiment- al and clinical studies on aclarubicin in the treatment of solid tumors. Biomed. Pharamacother., 38, 332. y SUGIMOTO, Y. & TSURUO, T. (1987). DNA-mediated transfer and cloning of a human multidrug-resistant myelogeneous leukemia K562. Cancer Res., 47, 2620. , , LANE, P., VICHI, P., BAIN, D.L. & TRITTON, T.R. (1987). Temperature dependence studies of adriamycin uptake and cytotoxicity. Cancer Res., 47, 4038. TAPIERO, H., BOULE, D., TRINCAL, G., FOURCADE, A. & LAM- PIDIS, T.J. (1988). Potentiation of ADM accumulation, and effectiness in ADM-resistant cell by aclacinomycin A. Leukemia Res., 12, 411. y y , , LING, V., KARTNER, N., SUDO, T., SIMINOVITCH, L. & RIORDAN, J.R. (1983). Multidrug resistance phenotype in Chinese hamster ovary cells. Cancer Treat. Rep., 67, 869. TSURUO, T., IIDA, H., NOJIR, M., TSUKAGOSHI, S. & SAKURAI, Y. (1983). Circumvention of vincristine and adriamycin resistance in vitro and in vivo by calcium influx blockers. Cancer Res., 43, 2905. y p , , LOZZIO, C.B. & LOZZIO, B.B. (1975). Human chronic myelogenous leukemia cell line with positive phyladephia chromosome. Blood, 45, 321. TSURUO, T., IIDA, H., KATANI, Y., YOKOTA, K., TSUKAGOSHI, S. & SAKURAI, Y. (1984). Effects of quinidine and related compounds on cytotoxicity and cellular accumulation of vincristine and adriamycin in drug-resistant tumour cells. Cancer Res., 44, 4303. , MAJIMA, H. & OHTA, K. (1987). References Clinical studies of aclacinomycin A (ACM). Biomed. Pharmacother., 41, 233. MANFAIT, M., ALIX, A.J.P., JEANNESSON, P., JARDILLIER, J.-C. & THEOPHANIDES, T. (1982). Interaction of adriamycin with DNA as studied by resonance Raman spectroscopy. Nucleic Acids Res., 10, 3803. TSURUO, T., IIDA, H., KAWATABA, H., OH-HARA, T., HAMADA, H. & UTAKOJI, T. (1986). Characteristics of resistance to adriamycin in human myelogenous leukemia K562 resistant to adriamycin and in isolated clones. Jpn. J. Cancer Res., 77, 682. MANFAIT, M., GIGLI, M., MILLOT, J.-M. & 6 others. (1987). New perspectives of microspectrofluorometry on living cells. In Spec- troscopy of Biological Molecules-New Advances, Schmid, E.D., Schneider, F.W. & Siebert, F. (ed) p. 445. John Wiley & Sons: Chichester. p TWENTYMAN, P.R. (1988). Modification of cytotoxic drug resistance by non-immuno-suppressive cyclosporins. Br. J. Cancer, 57, 254. UMEZAWA, K., KUNIMOTO, S. & TAKEUCHI, T. (1987). Experiment- al studies of new anthracyclines: aclacinomycin, THP adriamycin and ditrisarubicins. Biomed. Pharmacother., 41, 206. p TWENTYMAN, P.R. (1988). Modification of cytotoxic drug resistance by non-immuno-suppressive cyclosporins. Br. J. Cancer, 57, 254. UMEZAWA, K., KUNIMOTO, S. & TAKEUCHI, T. (1987). Experiment- y pp y p , , UMEZAWA, K., KUNIMOTO, S. & TAKEUCHI, T. (1987). Experiment- al studies of new anthracyclines: aclacinomycin, THP adriamycin and ditrisarubicins. Biomed. Pharmacother., 41, 206. MANFAIT, M., RASOANAIVO, T., MILLOT, J.-M., ANGIBOUST, J.-F. & GIGLI, M. (1988). Pharmacokinetic and resistance to anthracyc- lines in living cells studied by microspectrofluorometry. In Neoad- juvant Chemotherapy II, Jacquillat, C., Veil, M. & Khayat, D. (ed) p. 605. John Libbey: London. WADLER, S. & WIERNIK, P.H. (1988). Partial reversal of doxorubicin resistance by forskolin and I,9-dideoxyforsko.n in murine sar- coma S180 variants. Cancer Res, 48, 539. ZUNINO, F., DIMARCO, A., ZACCARA, A. & GAMBETTA, R.A. (1980). The interaction of daunorubicin and doxorubicin with DNA and chromatin. Biochim. Biophys. Acta, 607, 206. p y MILLOT, J.-M., RASOANAIVO, T., ANGIBOUST, J.-F., GIGLI, M. & MANFAIT, M. (1989). Microspectrofluorimetrie sur cellules vivantes: incorporation nucleaire et pharmacocinetique. Innov. Tech. Biol. Med., 10, 160.
https://openalex.org/W2618390447	https://air.unimi.it/bitstream/2434/954225/2/s41598-017-02386-9.pdf	English	null	Time-order-errors and duration ranges in the Episodic Temporal Generalization task	Scientific reports	2,017	cc-by	9,042	Time-order-errors and duration ranges in the Episodic Temporal Generalization task Ezequiel Mikulan1,2, Manuel Bruzzone1, Manuel Serodio1, Mariano Sigman3, Tristán Bekinschtein4, Adolfo M. García1,2,5, Lucas Sedeño1,2 & Agustín Ibáñez 1,2,6,7,8 Ezequiel Mikulan1,2, Manuel Bruzzone1, Manuel Serodio1, Mariano Sigman3, Tristán Bekinschtein4, Adolfo M. García1,2,5, Lucas Sedeño1,2 & Agustín Ibáñez 1,2,6,7,8 Received: 28 October 2016 Accepted: 11 April 2017 Published: xx xx xxxx Received: 28 October 2016 Accepted: 11 April 2017 Published: xx xx xxxx The current model of the Episodic Temporal Generalization task, where subjects have to judge whether pairs of auditory stimuli are equal in duration, predicts that results are scale-free and unaffected by the presentation order of the stimuli. To test these predictions, we conducted three experiments assessing sub- and supra-second standards and taking presentation order into account. Proportions were spaced linearly in Experiments 1 and 2 and logarithmically in Experiment 3. Critically, we found effects of duration range and presentation order with both spacing schemes. Our results constitute the first report of presentation order effects in the Episodic Temporal Generalization task and demonstrate that future studies should always consider duration range, number of trials and presentation order as crucial factors modulating performance. Time has been a matter of ardent debate across many disciplines1. Within neuroscience, not only does it consti- tute an important topic in its own right2, but it also impinges on the field’s fundamental areas of inquiry, includ- ing consciousness3, motor control4, memory5, artificial intelligence6, and neural dynamics7. Likewise, the study of timing abnormalities is pivotal for research on pathologies such as Parkinson’s disease and schizophrenia8. In brief, understanding timing, time perception, and their neural basis proves fundamental for contemporary neuroscience9. Even though many relevant models and theories have been proposed10, some basic questions are still unsol Here we aimed to address one of them: is timing equal across different scales? f Multiple studies have addressed this issue based on a distinction between sub- and supra-second durations11, with contradictory results. Some studies show that response variability increases linearly as a function of duration, thus following Weber’s Law, in a range that goes from a few hundred milliseconds to a few seconds –see ref. 12 for a systematic investigation. However, other reports indicate that this linear property stops holding at some point between 1 and 2 seconds –see ref. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports Received: 28 October 2016 Accepted: 11 April 2017 Published: xx xx xxxx Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 Method i i Participants. Eighteen subjects participated in Experiment 1 (F = 10; x age = 24.22; s age = 3.12; x years of education = 17.78; s years of education = 3.83), 20 in Experiment 2 (F = 10; x age = 25.47; s age = 3.94; x years of education = 18.87; s years of education = 3.04), and 18 in Experiment 3 (F = 11; x age = 24.39; s age = 3.53; x years of education = 18.94; s years of education = 2.82), after signing informed consent. Subjects participated in only one of the experiments. All of them reported normal hearing, right-handedness, and absence of neurological and psychiatric antecedents, and they were naïve as to the purpose of the study. The experiments were approved by the local ethics committee (INECO Foundation) and were conducted in accordance with the Declaration of Helsinki. Stimuli. Auditory stimuli were 500-Hz tones created and delivered using Matlab (Mathworks Inc.) and Psychtoolbox27 through Sennheiser HD202 headphones at 65db on a MacBook Pro notebook. Before the experi- ment, the software created 7 blocks of 16 trials; half of the stimuli corresponded to the Sub-Second condition and the other half to the Supra-Second one. For each trial, a Standard (S) duration was selected from a uniform distri- bution, ranging from 300 to 500 ms in the Sub-Second condition, and from 1200 to 2000 ms in the Supra-Second one. Then, S was multiplied by one of 7 ratios (linearly spaced in Experiments 1 and 2: 0.25, 0.50, 0.75, 1.0, 1.25, 1,50, 1.75; logarithmically spaced in Experiment 3: 0.25, 0.40, 0.63, 1, 1.59, 2.52, 4) to create a Comparison (C) duration, so that on each block there was one trial of each ratio for each condition (i.e., 7 ratios × 2 condi- tions = 14 trials). An additional 1.0 ratio trial was then added to each condition. Finally, the order of the trials was randomized and each trial was assigned a random counterbalanced presentation order so that on half of the trials the S was presented first, and on the other half the first stimulus presented was C. In Experiment 1 a total of 112 trials were obtained from each subject in an average of 16 minutes (σ = 36 s). www.nature.com/scientificreports/ www.nature.com/scientificreports/ of ratios (e.g., 0.5, 1, 1.5). In the original version, a clear standard is learnt at the beginning and participants judge whether it is equal to each of the following durations. In the Episodic version, presentation order is counterbal- anced so that in half of the trials the standard comes first and in the other half it comes second. of ratios (e.g., 0.5, 1, 1.5). In the original version, a clear standard is learnt at the beginning and participants judge whether it is equal to each of the following durations. In the Episodic version, presentation order is counterbal- anced so that in half of the trials the standard comes first and in the other half it comes second. i Both tasks showed a similar pattern of results with sub-second durations: the obtained psychometric functions were asymmetrical, with a higher proportion of “equal” responses on the right tail, that is, when the ratio was higher than 116, 17. The same pattern was found using the original standard version, with durations ranging from 2 up to 8 seconds18. Importantly, results were superimposed between duration ranges when plotted in a relative scale, that is, as a function of the proportion between standard and comparison durations. This was interpreted as confirmation of the common timing hypothesis. However, no studies have yet used supra-second durations as standards in the Episodic version.h p Traditionally, temporal generalization results have been interpreted within Scalar Expectancy Theory, which, in brief, states that durations are estimated via accumulation of pulses. Within this framework, in the temporal generalization task subjects would compare the two values of the estimated durations and then decide based on their normalized absolute difference19. Therefore, according to this model, results should not be affected by pres- entation order of the stimuli, something known as the “balance condition”20, nor by duration range. Presentation order effects have been shown in a wide variety of time perception tasks, termed “time-order-errors” (TOE) within this context (see ref. 21 for a review), but never in the Episodic Temporal Generalization (ETG) task. In fact, no previous studies using the task tested for any effect of this kind17, 22–24. p g yf In addition, evaluating symmetry with linearly spaced proportions (i.e. 0.25, 0.50, 0.75, 1, 1.25, 1.50, 1.75), as is usually the case with the ETG task17, 19, 22, implies an unbalanced comparison. www.nature.com/scientificreports/ Symmetry within this setting would indicate that a similar amount of “equal” responses were obtained when comparing, for example, ratios 0.25–1 (1:4) and 1.75–1 (1.75:1) or vice versa (4:1 and 1:1.75). A more meaningful comparison would arise from using logarithmically spaced proportions, so that, following the example above, ratios, 0.25–1 (1:4) and 4–1 (4:1) could be contrasted. Moreover, the property of superposition has been tested via visual inspection or ANOVAs17, 19, none of which is robust to such an end. The first one proves inadequate because it does not establish a decisional boundary to accept or reject hypotheses, and the second because it can produce spurious results when used with proportional data25, 26. A more convenient approach would be to compare Weber Fractions (WF) between duration ranges and test whether they remain constant, in which case a scalar relationship could be assumed to exist between them. Against this background, our study pursued three main objectives. First, we tested the prediction, derived from the traditional ETG model, that presentation order had no effect on performance. Second, we examined the symmetry/asymmetry of the temporal generalization gradients taking presentation order into account and using linear and logarithmically spaced proportions, so that symmetry could be properly assessed. Third, we compared WFs of sub- and supra-second ranges to test their compliance with the scalar property of timing. g y g To address these aims, we conducted three experiments. Experiment 1 was designed with a number of trials similar to that of previous studies using the task17, 22 and comprised linearly spaced proportions. As taking pres- entation order into account reduced the number of trials of each ratio to a half, we conducted Experiment 2, in which the same task was administered but with a threefold increase in trials. Experiment 3 had the same number of trials as Experiment 2 but proportions were logarithmically spaced. With this combination of experiments, we aimed to address some critical gaps in the ETG framework. Time-order-errors and duration ranges in the Episodic Temporal Generalization task 13 for a review.h This controversy is epitomized by the antinomy between two major conceptual frameworks used to account for timing mechanisms in the brain: the “common timing hypothesis” and the “distinct timing hypothesis”14. Whereas the former assumes a single timing mechanism irrespective of duration, the latter posits dissociable mechanisms for sub- and supra-second durations. A typical approach to study time perception is to have participants judge whether two durations are equal15. This so-called Temporal Generalization task has two main versions for humans. In the original one16, participants learn a standard duration at the beginning of the experiment and are then presented with several to-be-compared durations. Instead, in the Episodic version17, subjects judge the durations of two successive stimuli on a trial-by-trial basis. Stimuli are constructed in a similar way for both versions: a set of comparison durations is generated multiplying a standard duration (e.g., 400 ms, or values from a range such as 300 to 500 ms) by a series 1Laboratory of Experimental Psychology and Neuroscience (LPEN), Institute of Cognitive and Translational Neuroscience (INCYT), INECO Foundation, Favaloro University, Buenos Aires, Argentina. 2National Scientific and Technical Research Council (CONICET), Buenos Aires, Argentina. 3Di Tella University, Buenos Aires, Argentina. 4Consciousness and Cognition Lab, Department of Psychology, University of Cambridge, Cambridge, UK. 5Faculty of Education, National University of Cuyo (UNCuyo), Mendoza, Argentina. 6Universidad Autónoma del Caribe, Barranquilla, Colombia. 7Center for Social and Cognitive Neuroscience (CSCN), School of Psychology, Universidad Adolfo Ibañez, Santiago de Chile, Chile. 8Australian Research Council Centre of Excellence in Cognition and its Disorders, Sydney, Australia. Correspondence and requests for materials should be addressed to E.M. (email: e.mikulan@gmail.com) or A.I. (email: aibanez@ineco.org.ar) Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 1 Method i i In Experiment 2, the task was repeated three times with 5 minute breaks between them, therefore 336 trials were obtained from each subject, in 62 minutes on average (σ = 9 m). In Experiment 3, 336 trials were obtained from each subject in 68 minutes on average (σ = 5 m). See Supplementary Figures 1, 2, and 3 (S1, S2, and S3) for further details. Procedure. Participants were informed that they would hear sequences of two tones and that their task was to decide whether both sounds had the same duration (Fig. 1). Each trial started with a 5-s inter-trial interval that was followed by the presentation of a tone, a gap (randomly chosen from a uniform distribution from 400 Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 2 www.nature.com/scientificreports/ Figure 1. Diagram of the experimental paradigm. Blue rectangles represent Standard durations and green rectangles represent Comparison durations. Linearly spaced proportions (LIN) were used in Experiments 1 and 2 and logarithmically spaced proportions (LOG) in Experiment 3. Figure 1. Diagram of the experimental paradigm. Blue rectangles represent Standard durations and green rectangles represent Comparison durations. Linearly spaced proportions (LIN) were used in Experiments 1 and 2 and logarithmically spaced proportions (LOG) in Experiment 3. Figure 1. Diagram of the experimental paradigm. Blue rectangles represent Standard durations and green rectangles represent Comparison durations. Linearly spaced proportions (LIN) were used in Experiments 1 and 2 and logarithmically spaced proportions (LOG) in Experiment 3. to 600 ms), and a second tone. Participants had to respond with their right hand on the notebook keyboard. To indicate that the tones were equal, they had to press the down arrow key with their index finger; to indicate that they were not, they had to press the right arrow with their middle finger. Importantly, post-task debriefing showed that although all participants detected two duration ranges, none of them realized that there were standard and comparison distributions. to 600 ms), and a second tone. Participants had to respond with their right hand on the notebook keyboard. To indicate that the tones were equal, they had to press the down arrow key with their index finger; to indicate that they were not, they had to press the right arrow with their middle finger. Importantly, post-task debriefing showed that although all participants detected two duration ranges, none of them realized that there were standard and comparison distributions. Data Analysis. Method i i Statistical analyses were performed on R software28. All subjects were included in them. Following previous reports18, we plotted the temporal generalization gradients as a function of comparison dura- tions and tested their asymmetry by comparing the proportion of “equal” responses on the three Ratios below 1 (C-Shorter), against the proportion of “equal” responses on the three Ratios above it (C-Longer) using Wilcoxon Signed-Rank tests. Statistical results were corrected for multiple comparisons using the Holm-Bonferroni method.i In order to characterize the TOE, we first estimated the point of subjective equality (PSE) using the smoothing spline curve-fitting method from Matlab’s (Mathworks Inc.) Curve Fitting Toolbox (with automatic selection of the smoothing parameter), and finding the maximum of the resulting curve. We then defined the TOE following Fechner’s definition of constant error (CE)21: (1) = − CE PSE st (1) = − CE PSE st where PSE denotes the point of subjective equality and st the standard duration. Within this context, the sign of CE denotes the sign of the TOE when the standard is presented first. When the standard was presented in second place, the sign of the TOE was computed as st – PSE. We report the magnitude of the TOE as a percent- age of the standard (%TOE)29. %TOEs were submitted to a rm-ANOVA with standard duration (Sub-second/ Supra-second) and order (S-C/C-S) as factors. b ( ) l l d Weber Fractions (WF) were calculated as: = WF DL PSE / (2) (2) = WF DL PSE / = WF DL PSE / where DL denotes the difference limen and PSE the point of subjective equality30. Within this analysis, the PSE and DL were calculated as the mean and standard deviation of a fitted Gaussian function, respectively. WFs were submitted to a rm-ANOVA with standard duration (Sub-second/Supra-second) and order (S-C/C-S) as factors. To control that the PSEs did not differ between methods we compared them using a Wilcoxon Signed-Rank test.f f p g g Effect sizes in all cases were calculated via generalized eta squared (η2 G)31; these were considered as small if η2 G = 0.02, medium if η2 G = 0.13, and large if η2 G = 0.26. Holm-Bonferroni corrected post hoc t-tests were used for pairwise comparisons. Results i Experiment 1. The proportion of “equal” responses (PE) for each ratio and duration, when collapsing pres- entation orders, is shown in Fig. 2. Even though visual inspection suggests that both temporal generalization gradients were asymmetrical, Wilcoxon tests proved that this was significant only in the Sub-Second condition (V = 135.5, p < 0.01), where a higher PE was found when C-Ratio > 1. In the Supra-Second condition, the differ- ence was not significant (V = 220, p = 0.12).h gi ( , p ) The PE for each ratio and duration when including presentation order as an additional variable is shown in Fig. 3. Visual inspection again suggests that all temporal generalization gradients were asymmetrical, which, in this case, was true for all comparisons. In the Sub-Second condition, both presentations orders, S-C (V = 30, p < 0.05) and C-S (V = 40, p < 0.05) had a greater PE when C-Ratio > 1. In the Supra-Second condition, the C-S presentation order also had a greater PE when C-Ratio > 1 (V = 139, p < 0.01). Interestingly, in the S-C order of the Supra-Second condition the asymmetry was in the opposite direction, that is, the PE was higher when C-Ratio < 1 (V = 1, p < 0.01). Visual inspection also suggests that the temporal generalization gradients of the Supra-second condition are shifted to the left and right in the S-C and C-S orders, respectively.h ptt g p y This part of the analysis revealed that the asymmetries of temporal generalization gradients were not e between the two duration ranges considered. Moreover, when presentation order of the stimuli was taken Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 3 www.nature.com/scientificreports/ Figure 2. Temporal generalization gradients of Experiment 1 (linearly spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. Figure 2. Temporal generalization gradients of Experiment 1 (linearly spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. Figure 3. Temporal generalization gradients of Experiment 1 (linearly spaced ratios) by presentation order. Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. Figure 3. Temporal generalization gradients of Experiment 1 (linearly spaced ratios) by presentation order. Results i In the Supra-Second condition, the difference was not significant (V = 282, p = 0.17).h p The PE for each ratio and duration when including presentation order as an additional variable is shown in Fig. 5. Visual inspection once more suggests that all temporal generalization gradients were asymmetrical, which, in this case, was true for all comparisons but one. In the Sub-Second condition, the S-C presentation order had a higher PE when C-Ratio > 1 (V = 1, p < 0.001) and in the C-S order the difference was not significant (V = 76, p = 0.29). As in Experiment 1, in the Supra-Second condition, the C-S presentation order had a greater PE when C-Ratio > 1 (V = 12, p < 0.001) and in the S-C order the asymmetry was in the opposite direction, that is, the PE was higher when C-Ratio < 1 (V = 196, p < 0.01). Visual inspection also suggests that temporal generalization gradients are shifted but this time also in the Sub-second range. Interestingly, the directions of the shifts seem to be inverted between duration ranges and orders.h g This part of the analysis revealed that, as in Experiment 1, the asymmetries of the temporal generalization gradients were not equal between the duration ranges considered in the study. Experiment 2 showed that when collecting more trials per subject, the asymmetry of the C-S order of the Sub-second condition was not signifi- cant, in contrast with Experiment 1. Furthermore, Experiment 2 suggests that the shifts of the temporal gener- alization gradients that were observed in the Supra-second condition also appeared in the Sub-second range but with opposite sign. In order to characterize the presentation order effect (TOE) we computed the %TOE of each subject of Experiment 2 (Fig. 6) and conducted a rm-ANOVA. It showed a main effect of Duration (F1,19 = 93.15, p < 0.001, η2 G = 0.48), with a higher %TOE in the Sub-second condition; a main effect of Order (F1,19 = 27.36, p < 0.001, η2 G = 0.14), with a higher %TOE in the S-C order; and a non-significant Duration x Order interaction (F1,19 = 0.58, p = 0.45, η2 G = 0.009). Results i Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. account, results suggested that this occurred because of a shift of the temporal generalization gradients in the Supra-Second condition that depended on the presentation order of the stimuli (Fig. 3). The presentation order effect in combination with the asymmetrical gradients (the latter probably due to the linear spacing of the com- parison proportions) undermined the comparison of WFs, as they rely on the estimation of the spread of the temporal generalization gradients. This analysis was therefore not carried out for this Experiment. account, results suggested that this occurred because of a shift of the temporal generalization gradients in the Supra-Second condition that depended on the presentation order of the stimuli (Fig. 3). The presentation order effect in combination with the asymmetrical gradients (the latter probably due to the linear spacing of the com- parison proportions) undermined the comparison of WFs, as they rely on the estimation of the spread of the temporal generalization gradients. This analysis was therefore not carried out for this Experiment. p g gh y p In Experiment 1, we used a similar amount of trials than previous studies that employed the task17, 22. Partitioning trials by presentation order left approximately 3–4 trials per ratio and order combination for each Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 4 www.nature.com/scientificreports/ Figure 4. Temporal generalization gradients of Experiment 2 (linearly spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. Figure 4. Temporal generalization gradients of Experiment 2 (linearly spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. subject, which might have led to inaccurate results. In order to overcome this limitation we replicated Experiment 1 but this time we collected three times more trials per subject. Experiment 2. The PE for each ratio and duration, when collapsing presentation orders, is shown in Fig. 4. Visual inspection again suggests that both temporal generalization gradients were asymmetrical, and again Wilcoxon tests proved that this was significant only in the Sub-Second condition (V = 89, p < 0.001), where a higher PE was found when C-Ratio > 1. Results i The %TOE was positive in the Sub-second range (x = 5.68) and negative in the Supra-second range (x = −13.79).ff g Again, the combination of the different presentation order effects and the asymmetrical gradients hindered the comparison of WFs, so no such analysis was conducted. The linear spacing of the comparison proportions probably caused the asymmetrical gradients. We therefore conducted a third experiment, this time with loga- rithmically spaced ratios, so that the resulting gradients presumably became more symmetrical and allowed a meaningful comparison of WFs. Experiment 3. The PE when collapsing presentation order is shown in Fig. 7. Here, visual inspection sug- gests that temporal generalization gradients were not asymmetrical, which Wilcoxon tests proved to be correct (Sub-second: V = 417.5, p = 0.28; Supra-second: V = 274, p = 1).h The PE for each ratio and duration when including presentation order as an additional variable is shown in Fig. 8. Visual inspection suggests that all temporal generalization gradients were symmetrical, which was true for all comparisons but one. The only significantly asymmetrical gradient was found in the Supra-second S-C con- dition with a higher PE when C-Ratio < 1 (V = 108, p < 0.05). All remaining comparisons were not significant (Sub-second S-C: V = 66, p = 1; Sub-second C-S: V = 144.5, p = 0.54; Supra-second C-S: V = 32, p = 0.08). Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 5 www.nature.com/scientificreports/ Figure 5. Temporal generalization gradients of Experiment 2 (linearly spaced ratios) by presentation order. Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. Figure 5. Temporal generalization gradients of Experiment 2 (linearly spaced ratios) by presentation order. Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. Figure 6. Time-order-errors. Violin plots of the magnitude and sign of the time-order-errors for each duration range and presentation order, as percentage of standard duration, across participants of Experiment 2 (linearly spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.48), a main effect of presentation order (p < 0.001, η2 G = 0.14) and a non-significant interaction (p = 0.45, η2 G = 0.009). Figure 6. Time-order-errors. Results i Violin plots of the magnitude and sign of the time-order-errors for each duration range and presentation order, as percentage of standard duration, across participants of Experiment 2 (linearly spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.48), a main effect of presentation order (p < 0.001, η2 G = 0.14) and a non-significant interaction (p = 0.45, η2 G = 0.009). The TOE analysis from Experiment 3 (Fig. 9) showed a main effect of Order (F1,17 = 2344.41, p < 0.001, η2 G = 0.90), with a higher %TOE in the C-S order; a main effect of Duration (F1,17 = 16.00, p < 0.001, η2 G = 0.21), with a higher %TOE in the Sub-second condition; and a non-significant Duration x Order interaction (F1,17 = 4.20, p = 0.056, η2 G = 0.01). The %TOE was negative in the C-S order (x = −29.38) and positive in the S-C order (x = 24.42).h The more symmetrical gradients obtained using logarithmically spaced ratios allowed us to compute and compare WFs. We first tested that the PSEs obtained for this analysis did not differ from the PSEs obtained in the TOE analysis. Wilcoxon tests proved that they were not significantly different (V = 1024, p = 0.1). We then proceeded to the WF analysis (Fig. 10), where the rm-ANOVA showed a main effect of Duration (F1,17 = 15.87, p < 0.001, η2 G = 0.12), with higher values in the Sub-second condition; a main effect of Order (F1,17 = 5.34, p < 0.05, η2 G = 0.02), with higher values in the C-S order; and a significant Duration x Order interaction (F1,17 = 7.25, Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 6 www.nature.com/scientificreports/ Figure 7. Temporal generalization gradients of Experiment 3 (logarithmically spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. Figure 7. Temporal generalization gradients of Experiment 3 (logarithmically spaced ratios). Proportion of “equal” responses as a function of the ratio of the comparison duration when collapsing presentation orders Vertical lines represent 95% confidence levels. Figure 8. Temporal generalization gradients of Experiment 3 (logarithmically spaced ratios) by presentation order. Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. Figure 8. www.nature.com/scientificreports/ Figure 9. Time-order-errors. Violin plots of the magnitude and sign of the time-order-errors for each duration range and presentation order, as percentage of standard duration, across participants of Experiment 3 (logarithmically spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.21), a main effect of presentation order (p < 0.001, η2 G = 0.90) and a non-significant interaction (p = 0.056, η2 G = 0.01). Figure 9. Time-order-errors. Violin plots of the magnitude and sign of the time-order-errors for each duration range and presentation order, as percentage of standard duration, across participants of Experiment 3 (logarithmically spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.21), a main effect of presentation order (p < 0.001, η2 G = 0.90) and a non-significant interaction (p = 0.056, η2 G = 0.01). Figure 10. Weber Fractions. Violin plots of the Weber Fractions for each duration range and presentation order, across participants of Experiment 3 (logarithmically spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.12), a main effect of presentation order (p < 0.05, η2 G = 0.02) and a significant interaction (p < 0.05, η2 G = 0.04). Holm-Bonferroni corrected post hoc tests showed that the S-C order of the Supra-second condition differed from all others (vs. Sub-second S-C: p < 0.001; vs. Sub- second C-S: p < 0.01; vs. Supra-second C-S: p < 0.01) and that all remaining comparisons were not statistically significant. Figure 10. Weber Fractions. Violin plots of the Weber Fractions for each duration range and presentation order, across participants of Experiment 3 (logarithmically spaced ratios). The corresponding ANOVA showed a main effect of duration range (p < 0.001, η2 G = 0.12), a main effect of presentation order (p < 0.05, η2 G = 0.02) and a significant interaction (p < 0.05, η2 G = 0.04). Holm-Bonferroni corrected post hoc tests showed that the S-C order of the Supra-second condition differed from all others (vs. Sub-second S-C: p < 0.001; vs. Sub- second C-S: p < 0.01; vs. Supra-second C-S: p < 0.01) and that all remaining comparisons were not statistically significant. Discussion Th fi The first aim of our study was to assess whether performance on the ETG task is sensitive to presentation order effects. We showed that TOEs appeared in the two duration ranges under consideration, and that this effect held for both linear and logarithmic spacing. Interestingly, effect sizes between Experiments 2 and 3 were inverted. In the linearly spaced experiment, the duration range showed a large effect (η2 G = 0.48) and presentation order a medium one (η2 G = 0.14), while the use of logarithmically spaced proportions showed a large effect for pres- entation order (η2 G = 0.90) and a medium one for duration range (η2 G = 0.21). Together with the observed signs of the TOEs, these results reveal different patterns between experiments. In the linearly spaced experiment, sub- jects overestimated the first sound in the Sub-second condition and the second sound in the Supra-second range. Instead, when proportions were logarithmically spaced, they overestimated the Comparison duration, with a smaller influence of its position and duration range.h l p g TOEs have been reported in a wide range of tasks21 but never in the ETG task. They have mainly been reported when the standard duration was fixed across trials, even though they have been also reported in experiments where both stimuli varied32, 33. Having a standard that repeats from trial to trial implies that memory and learning can play a major role in the obtained results and might also introduce other sources of variance. For example, sub- jects might realize the existence of the standard and try to find it on each trial. Besides, incorrectly identifying the comparison duration as the standard could also lead to distortions in the memory representation of the standard. In our case, standard (and comparison) durations varied from trial to trial, so these factors can be assumed to have had a lower impact.i p Our results constitute the first report of TOEs in the ETG task. Presentation order was not taken into account in previous studies17, 22–24 and, as our results clearly show, it must be considered when employing the task.h The second aim of our study was to test the symmetry/asymmetry of the temporal generalization gradients while taking presentation order into account, and assessing linear and logarithmically spaced proportions. Results i Temporal generalization gradients of Experiment 3 (logarithmically spaced ratios) by presentation order. Proportion of “equal” responses for the Sub-Second (left) and Supra-Second (right) conditions by presentation order. Vertical lines represent 95% confidence levels. p < 0.05, η2 G = 0.04). Holm-Bonferroni corrected post hoc tests revealed that the WFs of the S-C order of the Supra-second condition differed from all others (vs. Sub-second S-C: p < 0.001; vs. Sub-second C-S: p < 0.01; vs. Supra-second C-S: p < 0.01). All remaining comparisons were not statistically significant (Sub-second S-C vs. Sub-second C-S: p = 0.63; Sub-second S-C vs. Supra-second C-S: p = 0.24; Sub-second C-S vs. Supra-second C-S: p = 0.52). p < 0.05, η2 G = 0.04). Holm-Bonferroni corrected post hoc tests revealed that the WFs of the S-C order of the Supra-second condition differed from all others (vs. Sub-second S-C: p < 0.001; vs. Sub-second C-S: p < 0.01; vs. Supra-second C-S: p < 0.01). All remaining comparisons were not statistically significant (Sub-second S-C vs. Sub-second C-S: p = 0.63; Sub-second S-C vs. Supra-second C-S: p = 0.24; Sub-second C-S vs. Supra-second C-S: p = 0.52). Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 7 www.nature.com/scientificreports/ Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 Discussion Th fi Experiment 1 showed right asymmetrical gradients in the S-C and C-S orders of the Sub-second condition and 8 Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 www.nature.com/scientificreports/ also in the C-S order of the Supra-second range. In the S-C order of the latter the asymmetry emerged in the opposite direction. Experiment 2 showed inverted asymmetries between ranges, that is, right asymmetrical gra- dients were found in the S-C order of the Sub-second condition and in the C-S order of the Supra-second range. Conversely, left asymmetrical gradients were found in the C-S order of the Sub-second condition and in the S-C order of the Supra-second range. Experiment 3 showed that using logarithmically spaced durations yielded more symmetrical gradients but still, results from the S-C order of the Supra-second condition were significantly left asymmetrical. y Wearden and colleagues19 modelled the Episodic version results by modifying the Church and Gibbon model. The original model was created to account for the results of the same task in rats34, where the resulting gradient was not asymmetrical. Wearden added the mean of the two durations as normalizing factor, to account for the asymmetry in humans. Thus, the formula for a “yes-equal” response became: − < ⁎ ⁎ ⁎ t t m b abs( )/ (3) 1 2 − < ⁎ ⁎ ⁎ t t m b abs( )/ 1 2 (3) where t1* and t2* are the two durations to be compared, m is their mean (the normalizing factor), and b* is a threshold. The higher the values of m, the higher the chances of being below the threshold and giving a “yes-equal” response. This way, it predicts only right asymmetrical gradients, despite the presentation order and duration range of the stimuli. Our results showed left asymmetrical gradients that reflected presentation order effects, therefore contradicting these predictions. Consequently, our findings represent a new empirical constraint calling for a modification of the model. gi Our results also have implications for other models of time perception. Apart from Wearden’s proposal, there are other two mainstream models aimed to account for two-interval forced-choice temporal experiments, namely, the Internal Reference Model35–37 and the Sensation Weighting Model21, 32, 38. Both have been developed for com- parative judgements (where subjects have to establish which of the two stimuli was longer), but they have been recently extended to equality judgements36. Discussion Th fi The former was developed for experiments in which the standard was fixed across trials and therefore does not apply in our case. The basic formulation of the Sensation Weighting Model is: = ⋅ − ⋅ + D w X w X u (4) 1 1 2 2 (4) where w1 and w2 are the weighting coefficients of the internal representations of the first (X1) and second (X2) stimuli, and u is a constant to adjust the mean of D. According to this formulation, subjects would judge that durations were equal if a < D < b, where a and b are thresholds. This account implies that the first and second sounds are weighted differently by the subject and therefore predicts and accounts for the time-order-errors that we observed in our study (see ref. 39). Within this framework, the TOEs arise from the formation of a reference level in the midrange of the stimuli, which is then weighted in the comparisons38. This would explain why the TOE was not observed in the Sub-second range when using a small number of trials (Experiment 1) and appeared when such number was increased (Experiment 2), as the reference level requires time to be established. In this regard, our results raise the question of whether the Supra-second range is more susceptible to this influence. g q p g pl Previous reports of the Temporal Generalization task18 claimed that the fact that temporal generalization gradients superimposed across sub- and supra-second ranges supported Scalar Expectancy Theory, one of the emblematic frameworks assuming the “common timing hypothesis”40. We showed that they are different for these two ranges when using the Episodic version of the task and taking presentation order into account, due to time-order-errors. However, presentation order effects have been observed in a wide range of tasks, including non-temporal tasks (i.e. weight comparison)38, 41. They are considered to be caused by processes beyond the specificity of the temporal domain21 and therefore our results should not be interpreted as being in line with the “distinct timing hypothesis”, but rather as the refutation of predictions made by a model that supports the “com- mon timing hypothesis”. g yp Moreover, the property of superposition has been previously tested via visual inspection or ANOVAs17, 19, nei- ther of which is sufficiently robust to such an end. Discussion Th fi The former is not suitable for hypothesis testing and the latter because of its problems when used on proportional data25, 26, as is the case with temporal generalization gradients. Consequently, the third objective of our study was to compare the WFs of the two ranges. If they were not significantly different, they could be assumed to comply with Weber’s Law. One possible confound in this com- parison is that chronometric counting has been shown to improve performance for durations above ~1.18 s42 and to reduce the coefficients of variation (as the WF), therefore disrupting the scalar property of variance43. Interestingly, our results showed that WFs were smaller only for the S-C presentation order of the Supra-second range while not being significantly different in all the remaining comparisons. In other words, WFs were dif- ferent between duration ranges for one presentation order but not for the other, and they were also different between presentation orders in the Supra-second range. If the decrease in the WFs observed in the S-C order of the Supra-second range was caused by chronometric counting, it could be expected to have influenced the C-S order in a similar way, which was not the case. Whichever the cause may be, the answer to the question of whether sub- and supra-second timing rely on the same or different mechanisms remains elusive and future studies will be required to elucidate it. q In sum, even though our results do not provide clear evidence in favour or against the scalar property of human timing in the sub- and supra-second duration ranges, they do demonstrate the importance of taking stim- ulus duration range and presentation order into account. This new constraint should be factored in future studies employing the task and in the models derived from it. Limitations. Our study has two main limitations. The first one is that the supra-second condition included, as comparison durations, stimuli that were below the 1s range. Thus, it was not a purely supra-second condition but Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 9 www.nature.com/scientificreports/ rather a condition in which the standard duration was supra-second. Future studies could use a pure supra-second condition by choosing a longer standard duration range. The second limitation is that we did not explicitly pre- vent chronometric counting. We did so to make our conditions comparable. Discussion Th fi Including a concurrent numerical task within durations of around half a second would have been methodologically incorrect. Not only would that pose higher cognitive demands than if included in a supra-second duration, but it would also be perceptually dif- ficult. It’s worth noting that when chronometric counting was explicitly encouraged in the original version of the task18, the resulting psychometric functions were symmetrical when collapsing presentation orders, which was not the case in our study. To overcome this limitation, modifications of the experimental design will be required. References h 1. Buccheri, R., Saniga, M., Stuckey, W. M. & North Atlantic Treaty Organization. Scientific Affairs Division. The nature of time–geometry, physics, and perception. (Kluwer Academic Publishers, 2003). 1. Buccheri, R., Saniga, M., Stuckey, W. M. & North Atlantic Treaty Organization. Scientific Affairs Division. The nature of time–geometry, physics, and perception. (Kluwer Academic Publishers, 2003). g y p y p p 2. Eagleman, D. M. Time and the Brain: How Subjective Time Relates to Neural Time. Journal of Neuroscience 25, 10369–10371, doi:10.1523/JNEUROSCI.3487-05.2005 (2005).h g y p y p p 2. Eagleman, D. M. Time and the Brain: How Subjective Time Relates to Neural Time. Journal of Neuroscience 25, 10369–10371, doi:10.1523/JNEUROSCI.3487-05.2005 (2005).h J ( ) 3. Dennett, D. C. & Kinsbourne, M. Time and the observer: The where and when of consciousness in the brain. Behavioral and Brain Sciences15, 183–201 (1992).h 4. Berret, B. & Jean, F. Why Don’t We Move Slower? The Value of Time in the Neural Control of Action. J Neurosci 36, 1056–1070, doi:10.1523/JNEUROSCI.1921-15.2016 (2016). 4. Berret, B. & Jean, F. Why Don’t We Move Slower? The Value of Time in the Neural Control of Action. J Neurosci 36, 1056–1070, doi:10.1523/JNEUROSCI.1921-15.2016 (2016). 5. Buzsaki, G. Cognitive neuroscience: Time, space and memory. Nature 497, 568–569, doi:10.1038/497568a (2013). h T l d fi l ll d g p y 6. Chittaro, L. & Montanari, A. Temporal representation and reasoning in artificial intelligence: Issues and approaches. Annals of Mathematics and Artificial Intelligence 28, 47–106, doi:10.1023/a:1018900105153 (2000).l 6. Chittaro, L. & Montanari, A. Temporal representation and reasoning in artificial intelligence: Issues and approaches. Ann Mathematics and Artificial Intelligence 28, 47–106, doi:10.1023/a:1018900105153 (2000).l fi g 7. Johnson, H. A., Goel, A. & Buonomano, D. V. Neural dynamics of in vitro cortical networks reflects experienced temporal patterns. Nat Neurosci 13, 917–919, doi:10.1038/nn.2579 (2010). 8. Allman, M. J. & Meck, W. H. Pathophysiological distortions in time perception and timed performance. Brain 135, 656–677 doi:10.1093/brain/awr210 (2012). 9. Finnerty, G. T., Shadlen, M. N., Jazayeri, M., Nobre, A. C. & Buonomano, D. V. Time in Cortical Circuits. Journal of Neuroscience 35 13912–13916, doi:10.1523/JNEUROSCI.2654-15.2015 (2015). 10. Ivry, R. B. & Schlerf, J. E. Dedicated and intrinsic models of time perception. Trends in Cognitive Sciences 12, 273–280, doi:10.1016/j. tics.2008.04.002 (2008). 1. Chen, L., Bao, Y. & Wittmann, M. Editorial: Sub- and Supra-Second Timing: Brain, Learning and Development. Front Psychol 7, 747 doi:10.3389/fpsyg.2016.00747 (2016). p yg 2. Merchant, H., Zarco, W. Conclusions Our study constitutes the first report of time-order-errors in the ETG task. We also showed differences that arise from the use of sub- and supra-second standards and from linear and logarithmically spaced propor- tions. In addition, we demonstrated that the current model of the task fails to account for the observed results. Presentation order was not taken into consideration by previous studies and, as our results clearly show, should always be considered. Moreover, we found that the number of trials used influences the observed pattern of results and should therefore also be considered as a crucial factor. Finally, we showed that Weber Fractions also vary as a function of duration range and presentation order. These results afford relevant empirical constraints for future research on the topic. Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 References h & Prado, L. Do we have a common mechanism for measuring time in the hundreds of millisecond range Evidence from multiple-interval timing tasks. Journal of Neurophysiology 99, 939–949, doi:10.1152/jn.01225.2007 (2008). p g f p y gy j 3. Grondin, S. In Neurobiology of Interval Timing Vol. 829 (eds Hugo Merchant & Victor de Lafuente) 17–32 (Springer New York 2014). ammsayer, T. H. & Troche, S. J. In search of the internal structure of the processes underlying interval timing in the sub-second and d fi t f t l i h A t P h l (A t) 147 68 74 d i 10 1016/j t 2013 05 004 (2014) 14. Rammsayer, T. H. & Troche, S. J. In search of the internal structure of the processes underlying interval timing in the sub-second the second range: a confirmatory factor analysis approach. Acta Psychol (Amst) 147, 68–74, doi:10.1016/j.actpsy.2013.05.004 (2 the second range: a confirmatory factor analysis approach. Acta Psychol (Amst) 147, 68–74, doi:10.1016/j.actpsy.2013.05.004 (2014). 15. Grondin, S. Timing and time perception: a review of recent behavioral and neuroscience findings and theoretical directions. Atten Percept Psychophys 72, 561–582, doi:10.3758/APP.72.3.561 (2010). gi y y pp y j p y 5. Grondin, S. Timing and time perception: a review of recent behavioral and neuroscience findings and theoretical directions. Atten Percept Psychophys 72, 561–582, doi:10.3758/APP.72.3.561 (2010). rcept Psychophys 72, 561–582, doi:10.3758/APP.72.3.561 (2010). p y p y 6. Wearden, J. H. Temporal generalization in humans. Journal of Experimental Psychology: Animal Behavior Processes 12, 134–144 doi:10.1037/0097-7403.18.2.134 (1992). doi:10.1037/0097-7403.18.2.134 (1992). 17. Wearden, J. H. & Bray, S. Scalar timing without reference memory? Episodic temporal generalization and bisection in humans. Q J Exp Psychol B 54, 289–309, doi:10.1080/713932763 (2001). 7. Wearden, J. H. & Bray, S. Scalar timing without reference memory? Episodic temporal generalization and bisection in humans. Q Exp Psychol B 54, 289–309, doi:10.1080/713932763 (2001). p y , , ( ) 18. Wearden, J. H., Denovan, L. & Haworth, R. Scalar timing in temporal generalization in humans with longer st Journal of Experimental Psychology: Animal Behavior Processes 23, 502 (1997). p y , , ( ) 18. Wearden, J. H., Denovan, L. & Haworth, R. Scalar timing in temporal generalization in humans with longer stimulus durat y ( ) den, J. H., Denovan, L. & Haworth, R. Scalar timing in temporal generalization in humans with longer stimulus durations. References h al of Experimental Psychology: Animal Behavior Processes 23 502 (1997) p y earden, J. H., Denovan, L. & Haworth, R. Scalar timing in tempor 18. Wearden, J. H., Denovan, L. & Haworth, R. Scalar timing in temporal generalization in humans with longer stimulus durations. Journal of Experimental Psychology: Animal Behavior Processes 23, 502 (1997). g p g g Journal of Experimental Psychology: Animal Behavior Processes 23, 502 (1997). J. H. Decision processes in models of timing. Acta Neurobiol Exp ( p g p 20. Falmagne, J.-C. Elements of psychophysical theory. (Clarendon Press; Oxford University Press, 1985).h 21. Hellström, A. The time-order error and its relatives: Mirrors of cognitive processes in comparing. Psychological Bulletin 97, 3 (1985). ( ) 22. Wearden, J. H. & Towse, J. N. Temporal generalizations in humans: Three further studies. Behav Processes 32, 247–263, doi:10.1016/0376-6357(94)90046-9 (1994). 3. McCormack, T., Wearden, J. H., Smith, M. C. & Brown, G. D. Episodic temporal generalization: a developmental study. Q J Exp Psychol A 58, 693–704 (2005).h y ( ) 24. Wearden, J. H. Slowing down an internal clock: Implications for accounts of performance on four timing tasks. The Quarterly Journal of Experimental Psychology 61, 263–274, doi:10.1080/17470210601154610 (2008). f p y gy 25. Dixon, P. Models of accuracy in repeated-measures designs. Journal of Memory and Language 59, 447–456 (2008). 26. Jaeger, T. F. Categorical data analysis: Away from ANOVAs (transformation or not) and towards logit mixed models. Journal of Memory and Language 59, 434–446 (2008).h y g g 27. Brainard, D. H. The Psychophysics Toolbox. Spat Vis 10, 433–436 (1997). y g g 27. Brainard, D. H. The Psychophysics Toolbox. Spat Vis 10, 433–436 (1997). 28. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna, Austria. URL http:// www.R-project.org/ (2015).f 29. Hellstrom, A. & Rammsayer, T. H. Time-order errors and standard-position effects in duration discrimination: An experimental study and an analysis by the sensation-weighting model. Atten Percept Psychophys 77, 2409–2423, doi:10.3758/s13414-015-0946-x (2015).lhfi 30. Garcia-Perez, M. A. Does time ever fly or slow down? The difficult interpretation of psychophysical data on time perception. Front Hum Neurosci 8, 415, doi:10.3389/fnhum.2014.00415 (2014).f ( ) 31. Bakeman, R. Recommended effect size statistics for repeated measures designs. Behav Res Methods 37, 379–384 (2005). 32. Hellstrom, A. Time errors and differential sensation weighting. J Exp Psychol Hum Percept Perform 5, 460–477 (1979). 31. Bakeman, R. www.nature.com/scientificreports/ Elucidating the internal structure of psychophysical timing performance in the sub-second d b l fi f l d d l d ( h Q y J f p y gy , , ( ) 40. Rammsayer, T. H. & Troche, S. J. Elucidating the internal structure of psychophysical timing performance in the sub-second and second range by utilizing confirmatory factor analysis. Adv Exp Med Biol 829, 33–47, doi:10.1007/978-1-4939-1782-2_3 (2014). ll h d d f h h l f 40. Rammsayer, T. H. & Troche, S. J. Elucidating the internal structure of psychophysical timing performance in the sub-second and second range by utilizing confirmatory factor analysis. Adv Exp Med Biol 829, 33–47, doi:10.1007/978-1-4939-1782-2_3 (2014). 41 H ll t A S ti i hti i p i d di i i ti f h i J E p P h l H P pt P f 26 6 17 second range by utilizing confirmatory factor analysis. Adv Exp Med Biol 829, 33–47, doi:10.1007/978-1-4939-1782-2_3 (2014). 1. Hellstrom, A. Sensation weighting in comparison and discrimination of heaviness. J Exp Psychol Hum Percept Perform 26, 6–17 (2000). 42. Grondin, A. When to start explicit counting in a time-intervals discrimination task: A critical point in the timing process of humans. Journal of Experimental Psychology: Human Perception and Performance25, 993–1004 (1999). 42. Grondin, A. When to start explicit counting in a time-intervals discrimination task: A critical point in the timing process of humans. Journal of Experimental Psychology: Human Perception and Performance25, 993–1004 (1999). f p y gy p f 43. Wearden, J. H. & Lejeune, H. Scalar properties in human timing: conformity and violations. Q J Exp Psychol (Hove) 61, 569–587 (2008). Author Contributions Design: E.M., T.B., A.I. Experiment programming: E.M. Data collection: M.S.e., M.B. Data analysis: E.M. Interpretation: E.M., M.S.i., T.B., A.I. Writing: E.M., T.B., M.S.i., A.M.G., A.I. Agreement and approval of the manuscript: E.M., M.S.e., M.B., M.S.i., T.B., A.M.G., A.I. Acknowledgementsh g This work was supported by grants from CONICET, CONICYT/FONDECYT Regular [1130920, 1140114 and 1170010], FONCyT-PICT [2012-0412 and 2012-1309], FONDAP 15150012 and INECO Foundation. The authors thank Marcelo Arlego, Will Harrison and the reviewers for their invaluable contributions. g This work was supported by grants from CONICET, CONICYT/FONDECYT Regular [1130920, 1140114 and 1170010], FONCyT-PICT [2012-0412 and 2012-1309], FONDAP 15150012 and INECO Foundation. The authors thank Marcelo Arlego, Will Harrison and the reviewers for their invaluable contributions. References h Recommended effect size statistics for repeated measures designs. Behav Res Methods 37, 379 384 (2005 32. Hellstrom, A. Time errors and differential sensation weighting. J Exp Psychol Hum Percept Perform 5, 460–477 (1979) Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 10 www.nature.com/scientificreports/ Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 www.nature.com/scientificreports/ 3. Patching, G. R., Englund, M. P. & Hellstrom, A. Time- and space-order effects in timed discrimination of brightness and size o paired visual stimuli. J Exp Psychol Hum Percept Perform 38, 915–940, doi:10.1037/a0027593 (2012). p p y p f ( ) 34. Church, R. M. & Gibbon, J. Temporal generalization. J Exp Psychol Anim Behav Process 8, 165–186 (1982). , , J p g J p y , ( ) 35. Dyjas, O., Bausenhart, K. M. & Ulrich, R. Trial-by-trial updating of an internal reference in discrimination tasks: evidence effects of stimulus order and trial sequence. Atten Percept Psychophys 74, 1819–1841, doi:10.3758/s13414-012-0362-4 (2012)f effects of stimulus order and trial sequence. Atten Percept Psychophys 74, 1819–1841, doi:10.3758/s13414-012-0362-4 (2012). 36. Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination processes in comparative and equality judgements: data and effects of stimulus order and trial sequence. Atten Percept Psychophys 74, 1819 1841, doi:10.3758/s13414 012 0362 4 (2012). 36. Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination processes in comparative and equality judgements: data and models Q J Exp Psychol (Hove) 67 1121–1150 doi:10 1080/17470218 2013 847968 (2014) f 6. Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination processes in comparative and equality judgements: data and models. Q J Exp Psychol (Hove) 67, 1121–1150, doi:10.1080/17470218.2013.847968 (2014).f p y 7. Bausenhart, K. M., Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination sensitivity for short and long durations Attention, Perception, & Psychophysics 77, 1033–1043, doi:10.3758/s13414-015-0875-8 (2015).ff p y p y 8. Hellstrom, A. Comparison is not just subtraction: effects of time- and space-order on subjective stimulus difference. Percep Psychophys 65, 1161–1177 (2003).f y p y ( ) 9. Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination processes in comparative and equality judgements: Data and d l Th Q t l J l f E p i t l P h l 67 1121 1150 d i 10 1080/17470218 2013 847968 (2014) y p y 39. Dyjas, O. & Ulrich, R. Effects of stimulus order on discrimination processes in comparative and equality judgements: Data and models. The Quarterly Journal of Experimental Psychology 67, 1121–1150, doi:10.1080/17470218.2013.847968 (2014). models. The Quarterly Journal of Experimental Psychology 67, 1121 1150, doi:10.1080/17470218.2013.847968 (2014). 40. Rammsayer, T. H. & Troche, S. J. Additional Information Supplementary information accompanies this paper at doi:10.1038/s41598-017-02386-9 Competing Interests: The authors declare that they have no competing interests. Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps 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/. Scientific Reports \| 7: 2643 \| DOI:10.1038/s41598-017-02386-9 11
https://openalex.org/W2952351439	https://www.nature.com/articles/oncsis201715.pdf	English	null	Inhibition of DNA2 nuclease as a therapeutic strategy targeting replication stress in cancer cells	Oncogenesis	2,017	cc-by	9,908	OPEN OPEN OPEN Citation: Oncogenesis (2017) 6, e319; doi:10.1038/oncsis.2017.15 www.nature.com/oncsis 1Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA; 2Department of Clinical Cancer Prevention, University of Texas MD Anderson Cancer Center, Houston, TX, USA; 3Department of Pharmacy, The Third Xiangya Hospital, Central South University, Changsha, China; 4Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, CT, USA; 5Center for Translational Cancer Research, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, USA and 6Department of Internal Medicine, University of Oklahoma, Oklahoma City, OK, USA. Correspondence: Professor P Sung, Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, CT 06510, USA. E-mail: patrick.sung@yale.edu or Dr G Ira, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA. E-mail: gira@bcm.edu or Dr G Peng, Department of Clinical Cancer Prevention, University of Texas MD Anderson Cancer Center, 6767 Bertner Avenue, BSRB S7.8336B, Unit 1013, Houston, TX 77030, USA. E-mail: gpeng@mdanderson.org 7These authors contributed equally to this work. 8Current address: Department of Medical Oncology, Tongji Hospital, Tongji Medical College, The University of Huazhong Science and Technology, Wuhan 430074, China. 9Current address: Department of Molecular and Cellular Biochemistry, Indiana University, Bloomington, IN, 47405 USA. Received 1 February 2017; accepted 10 February 2017 1Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA; 2Department of Clinical Cancer Prevention, University of Texas MD Anderson Cancer Center, Houston, TX, USA; 3Department of Pharmacy, The Third Xiangya Hospital, Central South University, Changsha, China; 4Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, CT, USA; 5Center for Translational Cancer Research, Institute of Biosciences and Technology, Texas A&M Health Science Center, Houston, TX, USA and 6Department of Internal Medicine, University of Oklahoma, Oklahoma City, OK, USA. Correspondence: Professor P Sung, Department of Molecular Biophysics and Biochemistry, Yale University School of Medicine, New Haven, CT 06510, USA. E il t i k @ l d ORIGINAL ARTICLE Inhibition of DNA2 nuclease as a therapeutic strategy targeting replication stress in cancer cells S Kumar1,7, X Peng2,3,7, J Daley4, L Yang2,8, J Shen2, N Nguyen5, G Bae5, H Niu4,5,9, Y Peng2, H-J Hsieh2, L Wang2, C Rao6, CC Stephan5, P Sung4, G Ira1 and G Peng2 Replication stress is a characteristic feature of cancer cells, which is resulted from sustained proliferative signaling induced by activation of oncogenes or loss of tumor suppressors. In cancer cells, oncogene-induced replication stress manifests as replication-associated lesions, predominantly double-strand DNA breaks (DSBs). An essential mechanism utilized by cells to repair replication-associated DSBs is homologous recombination (HR). In order to overcome replication stress and survive, cancer cells often require enhanced HR repair capacity. Therefore, the key link between HR repair and cellular tolerance to replication- associated DSBs provides us with a mechanistic rationale for exploiting synthetic lethality between HR repair inhibition and replication stress. DNA2 nuclease is an evolutionarily conserved essential enzyme in replication and HR repair. Here we demonstrate that DNA2 is overexpressed in pancreatic cancers, one of the deadliest and more aggressive forms of human cancers, where mutations in the KRAS are present in 90–95% of cases. In addition, depletion of DNA2 signiﬁcantly reduces pancreatic cancer cell survival and xenograft tumor growth, suggesting the therapeutic potential of DNA2 inhibition. Finally, we develop a robust high- throughput biochemistry assay to screen for inhibitors of the DNA2 nuclease activity. The top inhibitors were shown to be efﬁcacious against both yeast Dna2 and human DNA2. Treatment of cancer cells with DNA2 inhibitors recapitulates phenotypes observed upon DNA2 depletion, including decreased DNA double strand break end resection and attenuation of HR repair. Similar to genetic ablation of DNA2, chemical inhibition of DNA2 selectively attenuates the growth of various cancer cells with oncogene- induced replication stress. Taken together, our ﬁndings open a new avenue to develop a new class of anticancer drugs by targeting druggable nuclease DNA2. We propose DNA2 inhibition as new strategy in cancer therapy by targeting replication stress, a molecular property of cancer cells that is acquired as a result of oncogene activation instead of targeting currently undruggable oncoprotein itself such as KRAS. Oncogenesis (2017) 6, e319; doi:10.1038/oncsis.2017.15; published online 17 April 2017 Oncogenesis (2017) 6, e319; doi:10.1038/oncsis.2017.15; published online 17 April 2017 y 8Current address: Department of Medical Oncology, Tongji Hospital, Tongji Medical College, The University of Huazhong Science and Technology, Wuhan 430074, China. 9Current address: Department of Molecular and Cellular Biochemistry, Indiana University, Bloomington, IN, 47405 USA. Received 1 February 2017; accepted 10 February 2017 g y Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA. b d p g y or Dr G Ira, Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX 77030, USA. E-mail: gira@bcm.edu Cancer Prevention, University of Texas MD Anderson Cancer Center, 6767 Bertner Avenue, BSRB S7.8336B, Unit 1013, Houston, TX 77030, US INTRODUCTION Furthermore, our analyses showed increased staining of p-CHK1, a key molecule in DNA damage signaling, and reduced staining of Ki67, a marker of cell proliferation, in DNA2-depleted cells is likely due to endogenous DNA damage (Supplementary Figure S1c). Thus, DNA2 over- expression is a common theme in many cancer types including pancreatic cancer and inhibition of DNA2 reduces pancreatic cancer cell survival both in vitro and in vivo. range of DNA lesions including DSBs. The ability of cancer cells to repair DNA damage lowers therapeutic efﬁcacy, thus the simultaneous use of a DNA repair inhibitor holds great promise in sensitizing cancer cells to conventional chemo/radiotherapy. Inhibitors that target enzymes mediating base excision repair, nucleotide excision repair and other DNA repair pathways have been developed and are being evaluated in clinical trials.8–10 Therefore, identiﬁcation of new HR repair inhibitors can open new avenues to cancer therapies.4,11–15 range of DNA lesions including DSBs. The ability of cancer cells to repair DNA damage lowers therapeutic efﬁcacy, thus the simultaneous use of a DNA repair inhibitor holds great promise in sensitizing cancer cells to conventional chemo/radiotherapy. Inhibitors that target enzymes mediating base excision repair, nucleotide excision repair and other DNA repair pathways have been developed and are being evaluated in clinical trials.8–10 Therefore, identiﬁcation of new HR repair inhibitors can open new avenues to cancer therapies.4,11–15 p An evolutionarily conserved protein DNA2 possesses 5′ ﬂap endonuclease and 3′–5′ helicase activities and it has an important role in DNA damage repair, HR and DNA replication. The nuclease activity of DNA2 has several well-documented cellular functions, whereas the biological role of its helicase activity remains enigmatic. Isolation of yDna2 inhibitors We have presented evidence that DNA2 overexpression alleviates the impaired proliferation of U2OS and MCF10A cells after an activation of either H-Ras or cyclin E,33 suggesting that DNA2 could be an effective target in cancer therapy. This prompted us to develop a simple screen for chemical inhibitors of DNA2. The screen entailed the use of oligonucleotide (dT)30 having a reporter ﬂuorescent moiety, 6-FAM on the 5′ end, and the dark quenching group, Iowa Black FQ, on the 3′ end. In addition, a second dark quencher ZEN was placed between the 9th and 10th base (Figure 2a). The ﬂuorescence emitted by 6-FAM is quenched by FRET, but it increases when DNA is degraded by Dna2. In the initial screen, compounds were tested at a ﬁxed concentration of inhibitors (33.33 μM) with a subsaturating amount (0.09 nM) of Dna2. The inhibitory activity of the compounds was determined as a reduction in ﬂuorescent signal compared with the dimethylsulfoxide (DMSO) only control. The primary screen hits were selected based on a ⩾50% inhibition of the Dna2 activity. We screened ~ 50 000 chemical compounds from six different libraries (Supplementary Table S1) to identify chemical inhibitors of the Dna2 nuclease activity. As we could not purify hDNA2 in amounts sufﬁcient for a high-throughput screen, the initial screen was performed with yDna2, followed by testing top candidate compounds with hDNA2. The DNA2 nuclease domain is well conserved among eukaryotes, and therefore we reasoned that inhibitors of yDna2 will likely also work with the human counterpart. DNA2 nuclease efﬁciently cleaves the single-stranded DNA substrate used here (Figure 2b and Stewart et al.38), and inhibitors are expected to block substrate cleavage. First, we determined the concentration of yDna2 and hDNA2 and of the control T5 nuclease needed to generate maximum ﬂuorescence (Figure 2c), and we optimized the screen reaction (Z′ value of 0.8) in a robotic system with a 384-well plate setup to provide high throughput. The ﬁrst screening round identiﬁed 184 compounds, corresponding to a hit rate of ~ 0.4%. These compounds were subject to re-screening to conﬁrm their efﬁcacy and also to eliminate those compounds that also inhibit T5 nuclease. We were able to narrow yDna2-speciﬁc candidate compounds down to 39 (Supplementary Table S2). INTRODUCTION forks.5,6 A marked increase of replication stress and spontaneous DNA damage in cancer cells, which renders them more dependent on DNA repair for survival.7 Among all DNA repair pathways, homologous recombination (HR)-mediated DNA repair is an essential mechanism utilized by cancer cells to repair replication-associated DSBs and thereby to overcome replication stress and survive. The key link between HR repair and cellular tolerance to replication-associated DSBs provides us with a mechanistic rationale for exploiting synthetic lethality between HR repair inhibition and replication stress in cancer cells. The ﬁnding that poly(ADP-ribose) polymerase inhibitors, a new class of DNA repair inhibitors, speciﬁcally kill cancer cells with BRCA1 or BRCA2 mutations but are less cytotoxic to normal cells highlighted the promise of DNA repair inhibitors for targeted cancer treatment.1,2 This ﬁnding also provides proof of the principle that synthetic lethality interactions in the DNA repair network can be exploited for targeted cancer therapy. Early in the process of tumorigenesis, genetic alterations such as activation of oncogenes and loss of tumor-suppressor genes are implicated in inducing replication stress by providing premalignant cells with excessive growth signals.3,4 In cancer cells, oncogene-induced replication stress manifests as a high level of DNA double-strand breaks (DSBs) because of stalling and collapse of DNA replication p p Moreover, owing to their hyper-proliferative state, cancer cells are more vulnerable to killing by DNA-damaging agents. Common cancer treatment strategies use untargeted radiomimetic chemotherapy or radiotherapy that induce a broad DNA2 nuclease as a target in cancer therapy S Kumar et al 2 shows that DNA2 is signiﬁcantly upregulated.33 Consistently, an elevated DNA2 protein level was observed in pancreatic cancer patient specimens (Figure 1a). Moreover, patients with chronic pancreatitis expressed DNA2 at an elevated level in their pancreatic tissue (Figure 1a). To establish the relationship between the K-Ras oncogene and DNA2 expression, we analyzed a transgenic K-RasG12D pancreatic cancer mouse model. Similar to pancreatic cancer patient specimens, the DNA2 protein level was greatly increased in K-RasG12D-activated early pancreatic cancer lesions, suggesting that DNA2 overexpression is an early event during tumorigenesis (Figure 1b). Importantly, the depletion of DNA2 impaired the growth and survival of cells from two pancreatic cancer lines, and also led to increased senescence and apoptosis of these cells (Figure 1c, Supplementary Figure S1a). Consistent with in vitro cell culture ﬁndings, DNA2 knockdown signiﬁcantly reduced tumor growth in the xenograft model (Supplementary Figure S1b). INTRODUCTION Speciﬁcally, DNA2 mediates the resection of the 5′ strand at DNA DSB ends,16–23 an early step in HR, and also at stalled and regressed replication forks.24,25 Dna2 in fungi processes long 5′ DNA ﬂaps during the maturation of Okazaki fragments in lagging strand DNA synthesis,26,27 whereas human DNA2 (hDNA2) has an essential but as yet undeﬁned role in DNA replication.28 hDNA2 has an additional function in mitochondrial DNA maintenance.29 Finally, the unstructured and regulatory N-terminal domain of yeast Dna2 (yDna2) was recently shown to regulate DNA damage response by activating the Mec1 kinase (Mec1 is orthologus to human DNA damage checkpoint kinase ATR).30 Among these important biological functions of DNA2, the nucleolytic processing of the 5′ strands at spontaneous DSBs or regressed replication forks seems to be the most relevant in alleviating replication stress. Consistent with this notion, Dna2 is overexpressed via gene duplication in yeast upon replication stress caused by the mec1-21 mutation.31 Interestingly, similar to this observation in yeast cells, studies from our group and others have demonstrated that DNA2 is overexpressed in a broad spectrum of malignancies including breast cancer, ovarian cancer, lung cancer, prostate cancer, glioblastoma, pancreatic cancer, leukemia, colon cancer, melanoma, Lynch syndrome and heredi- tary nonpolyposis colorectal cancer.32–34 DNA2 overexpression is needed for the tolerance of increased replication stress induced by oncogene activation such as H-Ras or cyclin E in osteosarcoma and breast cancer cells, and partial depletion of DNA2 reduces the tumorigenicity of breast cancer cells.33 These results have raised the prospect of targeting DNA2 in cancer therapy. Isolation of yDna2 inhibitors f As the increased nuclease activity of DNA2 is required in yeast and cancer cells to overcome replication stress and DNA damage, for the ﬁrst time, we developed a robust biochemistry-based screening method for chemical inhibitors of DNA2. We conducted initial screen with nearly 50 000 compounds and we tested efﬁcacy of several candidate inhibitors in pancreatic and breast cancer cell lines with oncogene-induced replication stress. Treatment of cancer cells with the top DNA2 inhibitor recapitulates many phenotypes observed upon DNA2 depletion, and attenuates the growth of various cancer cells. Taken together, in this study, we developed a feasible high-throughput chemical screen assay to identify novel DNA repair inhibitors and more importantly we provided a new synthetic lethal strategy based on inhibition of HR repair by DNA2 inhibitors in cancer cells with oncogene-induced replication stress. Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al Cell proliferation was determined using 4,6-diamidino- 2-phenylindole staining and microscopy. About 10 compounds showed IC50 values between o1 and 10 μM both in cells and in vitro (Supplementary Table S2), and two of them, NSC-5195242 and NSC-105808 (Figures 3a and b), were selected for further analysis. Inhibitory effect of NSC-5195242 and NSC-105808 on hDNA2 The initial inhibitor screen was done with yDna2, thus it was important to test whether NSC-5195242 and NSC-105808 would affect the activity of hDNA2. hDNA2, puriﬁed as described,39 was ﬁrst tested using the triply labeled substrate as above (Figure 2a). We found a similar inhibitory effect on hDNA2 and comparable NSC-105808 interferes with DNA end resection and HR in cells, and suppresses the cisplatin sensitivity of FANCD2 −/ −cells To test whether NSC-105808 inhibits proliferation of U2OS cells by targeting DNA2, we ﬁrst examined whether the ectopic overexpression of DNA2 could reduce the negative effect of NSC-105808 interferes with DNA end resection and HR in cells, and suppresses the cisplatin sensitivity of FANCD2 −/ −cells To test whether NSC-105808 inhibits proliferation of U2OS cells by targeting DNA2, we ﬁrst examined whether the ectopic overexpression of DNA2 could reduce the negative effect of Figure 2. Assay optimization for high-throughput screen for DNA2 inhibitors. (a) Graphical representation of the screen assay. The 30 nucleotides long ssDNA substrate is labeled with three dyes: a ﬂuorophore (6-FAM) at the 5′end, and two dark quenchers, internal ZEN (between 9th and 10th base) and Iowa Black FQ at the 3′ end. The extent of the reaction was determined by measuring increase of ﬂuorescence at wavelength of 520 nm. (b) hDNA2 (0, 0.5 or 1 nM) was incubated with radiolabeled ssDNA (2.5 nM). The radiolabel is denoted by . (c) Degradation of triply labeled ssDNA (100 nM) by yDna2 (0.002 to 5.0 nM), hDNA2 (0.00016 to 166 nM) and T5 nuclease (0.01–10 units). Figure 1. DNA2 is overexpressed in pancreatic cancer lesions. (a) Immunohistochemistry (IHC) staining of DNA2 in representative human pancreatic lesion specimens. Two cases are presented and the lesions (boxed) are also shown at a higher magniﬁcation. (i, ii) normal pancreatic tissue; (iii, iv) hyperplasia in pancreatic tissue with chronic pancreatitis; (v, vi) pancreatic carcinoma. (b) IHC staining of DNA2 in wild-type mouse pancreatic tissue (top) and in LSL-K-RasG12D mouse pancreatic cancer tissue (bottom). Scale bar, 100 μm. Segments (boxed) of the images are also shown at a high magniﬁcation. DNA2 nuclease as a target in cancer therapy S Kumar et al (c) DNA2 depletion affects the survival of pancreatic cancer cells. Pancreatic cancer cell lines AsPC-1 and PANC-1 were stably transfected with short hairpin RNA (shRNA) vectors against two non-overlapping sequences in DNA2 or Figure 2. Assay optimization for high-throughput screen for DNA2 inhibitors. (a) Graphical representation of the screen assay. The 30 nucleotides long ssDNA substrate is labeled with three dyes: a ﬂuorophore (6-FAM) at the 5′end, and two dark quenchers, internal ZEN (between 9th and 10th base) and Iowa Black FQ at the 3′ end. The extent of the reaction was determined by measuring increase of ﬂuorescence at wavelength of 520 nm. (b) hDNA2 (0, 0.5 or 1 nM) was incubated with radiolabeled ssDNA (2.5 nM). The radiolabel is denoted by . (c) Degradation of triply labeled ssDNA (100 nM) by yDna2 (0.002 to 5.0 nM), hDNA2 (0.00016 to 166 nM) and T5 nuclease (0.01–10 units). Figure 1. DNA2 is overexpressed in pancreatic cancer lesions. (a) Immunohistochemistry (IHC) staining of DNA2 in representative human pancreatic lesion specimens. Two cases are presented and the lesions (boxed) are also shown at a higher magniﬁcation. (i, ii) normal pancreatic tissue; (iii, iv) hyperplasia in pancreatic tissue with chronic pancreatitis; (v, vi) pancreatic carcinoma. (b) IHC staining of DNA2 in wild-type mouse pancreatic tissue (top) and in LSL-K-RasG12D mouse pancreatic cancer tissue (bottom) Scale bar 100 μm Segments (boxed) of Figure 2. Assay optimization for high-throughput screen for DNA2 inhibitors. (a) Graphical representation of the screen assay. The 30 nucleotides long ssDNA substrate is labeled with three dyes: a ﬂuorophore (6-FAM) at the 5′end, and two dark quenchers, internal ZEN (between 9th and 10th base) and Iowa Black FQ at the 3′ end. The extent of the reaction was determined by measuring increase of ﬂuorescence at wavelength of 520 nm. (b) hDNA2 (0, 0.5 or 1 nM) was incubated with radiolabeled ssDNA (2.5 nM). The radiolabel is denoted by . (c) Degradation of triply labeled ssDNA (100 nM) by yDna2 (0.002 to 5.0 nM), hDNA2 (0.00016 to 166 nM) and T5 nuclease (0.01–10 units). Figure 1. DNA2 is overexpressed in pancreatic cancer lesions. (a) Immunohistochemistry (IHC) staining of DNA2 in representative human pancreatic lesion specimens. Two cases are presented and the lesions (boxed) are also shown at a higher magniﬁcation. (i, ii) normal pancreatic tissue; (iii, iv) hyperplasia in pancreatic tissue with chronic pancreatitis; (v, vi) pancreatic carcinoma. DNA2 nuclease as a target in cancer therapy S Kumar et al DNA2 nuclease as a target in cancer therapy S Kumar et al DNA2 nuclease as a target in cancer therapy S Kumar et al 4 Cellular screen for the most potent DNA2 inhibitors IC50 values as determined for yDna2 (Figure 3c). Furthermore, we tested the impact of these two compounds in a reconstituted 5′ DNA end resection assay including hDNA2, Bloom syndrome and Replication protein A (RPA) as described39 and in 5′ ﬂap processing by hDNA2. Both compounds inhibited resection in the reconstituted system (Figure 3d) and also 5′ ﬂap processing (Supplementary Figure S2). Considering that compound NSC-105808 shows a lower IC50 in vitro and biologically, further tests were focused on this compound. We found that NSC-105808 has no effect on ATPase activity of DNA2, nor the ability of Bloom syndrome to unwind DNA, nor does it affect DNA binding by DNA2 (Supplementary Figure S2). Furthermore, NSC-105808 does not inhibit the nuclease activity of EXO1 (Supplementary Figure S2), which can also catalyze DNA ends resection. Together, these data provide further evidence that NSC-105808 speciﬁcally inhibits the nuclease activity of DNA2. Cellular screen for the most potent DNA2 inhibitors The biological effectiveness of the 39 compounds selected by re-screening was determined by measuring their half maximal inhibitory concentration (IC50) biochemically and in a cell-based system using the 384-well platform (Supplementary Table S2). We used the cancer cell line, U2OS, as cells of this lineage overexpress DNA2 signiﬁcantly and are sensitive to partial DNA2 depletion.33 We conducted the viability assay in duplicate on three separate occasions using a concentration range of 10 pM to 10 μM for each compound. Cell proliferation was determined using 4,6-diamidino- 2-phenylindole staining and microscopy. About 10 compounds showed IC50 values between o1 and 10 μM both in cells and in vitro (Supplementary Table S2), and two of them, NSC-5195242 and NSC-105808 (Figures 3a and b), were selected for further analysis. p The biological effectiveness of the 39 compounds selected by re-screening was determined by measuring their half maximal inhibitory concentration (IC50) biochemically and in a cell-based system using the 384-well platform (Supplementary Table S2). We used the cancer cell line, U2OS, as cells of this lineage overexpress DNA2 signiﬁcantly and are sensitive to partial DNA2 depletion.33 We conducted the viability assay in duplicate on three separate occasions using a concentration range of 10 pM to 10 μM for each compound. RESULTS Sensitivity of pancreatic cancer cells to DNA2 depletion Pancreatic cancer is particularly deadly because of the lack of an effective treatment regimen. Here, we investigated the expression levels and therapeutic potential of DNA2 depletion in pancreatic cancers, where 90–95% cases are associated with an activating mutation in the K-Ras oncogene.35–37 Analysis of gene expression patterns in pancreatic cancers in the Oncomine database We note that a number of the compounds listed in Supplementary Table S3 inhibited the activity of both yDna2 and T5 and are therefore nonspeciﬁc. Interestingly, many of these Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S l DNA2 nuclease as a target in cancer therapy S Kumar et al 3 compounds also affect the activity of Dna2. Thus, in addition to inducing DSBs by blocking topoisomerase II, it is likely that these drugs affect the processing and repair of these DSBs by nucleases. nonspeciﬁc inhibitors interact with DNA and are already used in cancer therapy, including multiple drugs known to inhibit type II topoisomerases (for example, daunorubicin, doxorubicin and mitoxantrone). Here, we have provided evidence that these Oncogenesis (2 Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al (b) IHC staining of DNA2 in wild-type mouse pancreatic tissue (top) and in LSL-K-RasG12D mouse pancreatic cancer tissue (bottom). Scale bar, 100 μm. Segments (boxed) of the images are also shown at a high magniﬁcation. (c) DNA2 depletion affects the survival of pancreatic cancer cells. Pancreatic cancer cell lines AsPC-1 and PANC-1 were stably transfected with short hairpin RNA (shRNA) vectors against two non-overlapping sequences in DNA2 or scramble control (Mock) (see also Supplementary Figure S1). (Left) Clonogenic assay, cell survival scored 10–15 days after seeding. (Middle) 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, cells proliferation measured 72 h after seeding. (Right) Senescence assay with β-galactosidase staining and apoptosis assay using annexin V staining performed 72 h after DNA2 knockdown. Each value represents the mean ± s.e.m. from three independent experiments (Po0.05; Po0.01; Po0.005; **Po0.001). Oncogenesis (2017), 1 – 10 Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al Figure 3. Validation of the top two inhibitors identiﬁed from the high-throughput screen. (a) Chemical structure of NSC-5195242 2-nitrophenanthrene-9,10-dione) and NSC-105808 (6-amino-7-bromoquinoline-5,8-dione). (b, c) Dose-response curves for indicated compounds, cell based (U2OS cell line) (b) and in vitro (c). The IC50 value and the protein concentration used are indicated. Triple labeled substrate (0.1 μM) as in Figure 2a was used. The concentration range of inhibitors was 0.003–666.67 μM. (d) Degradation of radiolabeled 2-kb dsDNA (0.5 nM ends) with hDNA2 (5 nM), Bloom syndrome (5 nM) and RPA (100 nM) and the indicated concentrations of NSC-5195242 and NSC-105808. The asterisks denote the radiolabel. Plotted are the average data from three independent experiments and the error bars epresent 1 s.d. S Kumar et al S u a et a 5 Figure 3. Validation of the top two inhibitors identiﬁed from the high-throughput screen. (a) Chemical structure of NSC-5195242 (2-nitrophenanthrene-9,10-dione) and NSC-105808 (6-amino-7-bromoquinoline-5,8-dione). (b, c) Dose-response curves for indicated compounds, cell based (U2OS cell line) (b) and in vitro (c). The IC50 value and the protein concentration used are indicated. Triple labeled substrate (0.1 μM) as in Figure 2a was used. The concentration range of inhibitors was 0.003–666.67 μM. (d) Degradation of radiolabeled 2-kb dsDNA (0.5 nM ends) with hDNA2 (5 nM), Bloom syndrome (5 nM) and RPA (100 nM) and the indicated concentrations of NSC-5195242 and NSC-105808. The asterisks denote the radiolabel. Plotted are the average data from three independent experiments and the error bars represent 1 s.d. DNA2 nuclease as a target in cancer therapy S Kumar et al (d) Analysis of colony formation by FANCD2 (+/+) and FANCD2 ( −/ −) cells treated with cisplatin and/or NSC-105808. Each value is relative to the untreated control group of each cell line. The graph represents the mean ± s.e.m. from three independent experiments. Po0.01. Figure 4. NSC-105808 inhibits HR repair and DSB end resection. (a) Analysis of HR efﬁciency with the DR-green ﬂuorescent protein (GFP) assay. Representative images of ﬂow cytometry analysis and bar graph showing percentage of GFP cells from at least three independent experiments. Each value represents the mean ± s.d. (b, c) Analysis of RPA phosphorylation in U2OS cells treated with NSC-105808 (2 h, 0.3 μM) and/or camptothecin (CPT; 2 h, 1 μM). (b) Representative images of cells subjected to immunoﬂuorescent staining of p-RPA and bar graph showing percentage of cells with positive p-RPA staining from three independent experiments. Each value represents the mean ± s.d. (c) Western blot analysis of chromatin fraction with the indicated antibodies. (d) Analysis of colony formation by FANCD2 (+/+) and FANCD2 ( −/ −) cells treated with cisplatin and/or NSC-105808. Each value is relative to the untreated control group of each cell line. The graph represents the mean ± s.e.m. from three independent experiments. Po0.01. the highest tested concentration (0.6 μM) with a slight decrease of G1 and increase of G2 cells, but with no impact on the cellular level of DNA2 protein (Supplementary Figures S7 and 8). We note that longer (448 h) exposure to NSC-105808 causes cell death (Figure 3b). To test DSB end resection, we followed phosphoryla- tion of the single-stranded DNA (ssDNA)-binding protein RPA (p- RPA) as described previously.40 Cells were treated with the topoisomerase I inhibitor camptothecin, which induces DNA nicks, reversed replication forks and single-ended DSBs. As shown in Figure 4b, NSC-105808 treatment signiﬁcantly attenuated the accumulation of camptothecin-induced p-RPA, suggesting an impairment of DSB end resection. We then conﬁrmed this result by conducting chromatin fractionation and observed that NSC-105808 reduced p-RPA accumulation in the chromatin fraction with a concomitant enhancement in the γ-H2AX level, reﬂective of an accumulation of unrepaired DSBs (Figure 4c). Altogether, the above results provide evidence that NSC-105808 interferes with DSB resection and HR in cells. We wished to further verify the DNA2 speciﬁcity of NSC-105808 in cells. DNA2 nuclease as a target in cancer therapy S Kumar et al biological processes. Similar to DNA2 knockdown, NSC-105808 (24-h treatment) signiﬁcantly reduced DSB repair by gene conversion or single stand annealing as determined by using two green ﬂuorescent protein-based reporter assays (Figure 4a, Supplementary Figures S4-6 and Peng et al.33). The treatment with NSC-105808 had only a mild impact on cell cycle distribution at NSC-105808 on cell growth. An approximately 1.5–2.0-fold increase in DNA2 protein level was able to attenuate the negative impact of NSC-105808 on cell proliferation, supporting the idea that DNA2 is the target of NSC-105808 in cells (Supplementary Figure S3). DNA2 functions in DSB end resection and HR,17,18,33 thus we asked whether NSC-105808 interferes with these Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al Figure 4. NSC-105808 inhibits HR repair and DSB end resection. (a) Analysis of HR efﬁciency with the DR-green ﬂuorescent protein (GFP) Representative images of ﬂow cytometry analysis and bar graph showing percentage of GFP cells from at least three indepe experiments. Each value represents the mean ± s.d. (b, c) Analysis of RPA phosphorylation in U2OS cells treated with NSC-105808 (2 h, 0 and/or camptothecin (CPT; 2 h, 1 μM). (b) Representative images of cells subjected to immunoﬂuorescent staining of p-RPA and bar showing percentage of cells with positive p-RPA staining from three independent experiments. Each value represents the mean (c) Western blot analysis of chromatin fraction with the indicated antibodies. (d) Analysis of colony formation by FANCD2 (+/+ FANCD2 ( −/ −) cells treated with cisplatin and/or NSC-105808. Each value is relative to the untreated control group of each cell line graph represents the mean ± s.e.m. from three independent experiments. Po0.01. 6 Figure 4. NSC-105808 inhibits HR repair and DSB end resection. (a) Analysis of HR efﬁciency with the DR-green ﬂuorescent protein (GFP) assay. Representative images of ﬂow cytometry analysis and bar graph showing percentage of GFP cells from at least three independent experiments. Each value represents the mean ± s.d. (b, c) Analysis of RPA phosphorylation in U2OS cells treated with NSC-105808 (2 h, 0.3 μM) and/or camptothecin (CPT; 2 h, 1 μM). (b) Representative images of cells subjected to immunoﬂuorescent staining of p-RPA and bar graph showing percentage of cells with positive p-RPA staining from three independent experiments. Each value represents the mean ± s.d. (c) Western blot analysis of chromatin fraction with the indicated antibodies. DISCUSSION induced replication stress. Initially, we compared the impact of our DNA2 inhibitor in pancreatic or breast cancer cells versus control cells (Figures 5a and b). Transformed pancreatic cancer cell lines (AsPC-1, PANC-1) were more sensitive to NSC-105808 when compared with non-transformed pancreatic ductal epithelial cells (DT-PD59). Similarly, transformed breast cancer cells (Hs578T and BT549) were more sensitive than MCF10A non-transformed breast epithelial cells to the DNA2 inhibitor. During tumorigenesis, the activation of an oncogene or loss of a tumor-suppressor gene leads to excessive growth signals and replication stress. Notably, about 30% of all cancers have an activating mutation in Ras and the vast majority (up to 95%) of pancreatic cancer patients harbor a mutation in K-Ras.35–37 Unfortunately, thus far, no effective inhibitor of the K-Ras oncoprotein has reached the clinic. In this study, we have strived to test the efﬁcacy of DNA2 nuclease inhibition in cancer cells that experience oncogene-induced replication stress. As such stress is a common denominator among multiple cancer types, DNA2 inhibition could be an efﬁcacious strategy to treat a broad spectrum of cancers. Besides targeting cancer cells with an activated oncogene, chemical inhibition of DNA2 may also be useful for sensitizing cells to DNA damage when combined with radiomimetic chemotherapy or radiation therapy. As we expected, recent studies reported that DNA2 inhibition can sensitize breast cancer cells to chemotherapy inducing DNA damage.44 We developed a simple high-throughput screening assay for DNA2 inhibitors, which can screen compound collections in very small reaction volume and in general to screen for inhibitors of other nucleases. We identiﬁed several compounds with the ability to inhibit DNA2 in DNA incision and also in a reconstituted system of DNA end resection. The NSC-105808 compound inhibits DNA2 but not other tested nucleases, however, additional cellular targets cannot be excluded. Importantly, NSC-105808 treatment recapitulates several phenotypes of cells with DNA2 depletion, such as attenuated DSB end resection, decreased HR and suppression of the sensitivity of FANCD2-deﬁcient cells to cisplatin.41 DNA2 has an important role in telomere replication,45 thus the effect of NSC-105808 on telomere stability cannot be excluded. The inhibitor reduces the growth of various cancer cells, and speciﬁcally those cells in which replication stress is induced via oncogene activation. A similar range of IC50 values for the NSC-105808 in vivo and in in vitro assay may indicate that even minimal inhibition of DNA2 in cells could be lethal to cells. DNA2 nuclease as a target in cancer therapy S Kumar et al Previous work revealed that DNA2 depletion in FANCD2- deﬁcient cells partially suppresses the sensitivity of these cells to cisplatin.41 As shown in Figure 4d, FANCD2 −/ −cells were more sensitive to cisplatin treatment compared with control cells, and, importantly, a very low concentration of NSC-105808 lessened the cisplatin sensitivity of these cells. Together, our results are consistent with the premise that DNA2 is the target of NSC-105808 in cells. Oncogene induction renders cells susceptible to NSC-105808 DNA2 overexpression, often observed in cancer cells, has been proposed to alleviate oncogene-induced replication stress. Increased sensitivity of cancer cells to DNA2 depletion33 can likely be explained by an increased amount of substrates that are processed by DNA2 nuclease/helicase. We therefore asked whether NSC-105808 can sensitize cancer cells to oncogene- Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al 7 Figure 5. Oncogene induction enhances sensitivity of cancer cells to NSC-105808. (a-e) Graphs showing proliferation of indicated cell lines in response to a 48-h treatment with the indicated concentrations of NSC-105808. Proliferation relative to cells treated with vehicle alone was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Each value represents the mean ± s.e.m. from three independent experiments. (a) Proliferation of control pancreatic epithelial cells DT-PD59 and transformed pancreatic cancer cells AsPC-1 and PANC-1. (b) Proliferation of normal breast epithelial cells MCF10A and breast cancer cells Hs578T and BT549. (c) Proliferation of MCF10A cells with or without H-Ras expression. H-Ras was induced with doxycycline (2 μg/ml). (d) Proliferation of control pancreatic DT-PD59 cells and transformed DT-PD77 cells with E6/E7 oncoprotein activation. (e) Proliferation of MCF10A breast cells without and with H-Ras activation (AT1, DCIS, CA1a). Figure 5. Oncogene induction enhances sensitivity of cancer cells to NSC-105808. (a-e) Graphs showing proliferation of indicated cell lines in response to a 48-h treatment with the indicated concentrations of NSC-105808. Proliferation relative to cells treated with vehicle alone was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Each value represents the mean ± s.e.m. from three independent experiments. (a) Proliferation of control pancreatic epithelial cells DT-PD59 and transformed pancreatic cancer cells AsPC-1 and PANC-1. (b) Proliferation of normal breast epithelial cells MCF10A and breast cancer cells Hs578T and BT549. (c) Proliferation of MCF10A cells with or without H-Ras expression. H-Ras was induced with doxycycline (2 μg/ml). (d) Proliferation of control pancreatic DT-PD59 cells and transformed DT-PD77 cells with E6/E7 oncoprotein activation. DNA2 nuclease as a target in cancer therapy S Kumar et al (e) Proliferation of MCF10A breast cells without and with H-Ras activation (AT1, DCIS, CA1a). Flow cytometric analysis of apoptosis and BrdU incorporation Apoptosis was measured by ﬂow cytometry using the Annexin V FITC assay kit based on the protocol from the manufacturer (Invitrogen). For BrdU (bromodeoxyuridine (5-bromo-2′-deoxyuridine)) analysis, after treatments, cells were labeled with BrdU (Sigma) for 30 min before ﬁxation with 70% cold ethanol (−20 °C). The next day, cells were washed with cold phosphate-buffered saline and incubated with 2 M HCl for 5 min at room temperature. After adjusting the pH with 0.1 M sodium borate (pH 8.5), cells were incubated in 1% bovine serum albumin with 0.1% Triton X-100 for 30 min and mouse anti-BrdU antibody (1:400) for 1 h to detect BrdU labeling. After washing, cells were incubated with secondary antibody (ﬂuorescein isothiocyanate 1:400) for 30 min. Cells were resuspended in staining solution (10 μg/ml propidium iodide, 20 μg/ml RNase A and 0.05% Triton X-100) and cell cycle analysis was performed at the MD Anderson Cancer Center Flow Cytometry and Cellular Imaging Facility. For ﬂuorescence-activated cell sorting analysis, the percentage of cells with positive BrdU staining in each phase of the cell cycle was quantitated with the CellQuest software and ModFit software (Verity Software House Inc., Topsham, ME, USA). Plasmids, short hairpin RNA and transfection The pLenti6.3/TO/V5-H-Ras vector was obtained from Dr Shiaw-Yih Lin (Department of Systems Biology, MD Anderson Cancer Center). Brieﬂy, PCMV-H-RAS (V12) was purchased from Invitrogen (Waltham, MA, USA) and subcloned into the pLenti6.3/TO/V5-DEST vector (Invitrogen) based on the commercial protocol of ViraPower TM HiPerform TM T-Rex TM Gateway Expression system (Thermo Fisher Scientiﬁc Waltham, MA, USA). pLenti3.3/ TR and pLenti6.3/TO/V5-H-Ras were stably integrated into MCF10A cells. Expression of H-Ras was induced by the addition of doxycycline (1 μg/ml) for 48 h. The vector that expresses FLAG-tagged DNA2 has been described.33 DNA2 short hairpin RNA vectors and control short hairpin RNA vectors against luciferase sequences were obtained from Sigma (St Louis, MO, USA; shDNA2#1: 5′-CCGGACCTGGTGTTGGCAGTCAATACTCGAGT ATTGACTGCCAACACCAGGTTTTTTTG-3′; shDNA2#2: 5′-CCGGAGTTTGTGATG GGCAATATTTCTCGAGAAATATTGCCCATCACAAACTTTTTTTG-3′). On-target smart pool small interfering RNAs against DNA2 were purchased from Dharmacon Research (Lafayette, CO, USA; #9: 5′-AGACAAGGUUCCA GCGCCA-3′; #10: 5′-UAACAUUGAAGUCGUGAAA-3′; #11: 5′-AAGCACAGG UGUACCGAAA-3′; #12: 5′-GAGUCACAAUCGAAGGAUA-3′). Transfection of plasmids was performed with the FuGENE 6 reagent from Roche (Indianapolis, IN, USA). Transfection of small interfering RNAs was performed with Oligofectamine (Invitrogen). Gene conversion and single stand annealing repair assays The DR-green ﬂuorescent protein, pCAGGS and pCBASce plasmids were kindly provided by Dr Maria Jasin (Memorial Sloan-Kettering Cancer Center, New York, NY, USA). Colony-forming assay Colony-forming assay Transfection and colony-forming assay were performed as previously described33 (Supplementary Materials and Methods). DISCUSSION Slides were washed three times with phosphate-buffered saline and incubated in blocking buffer contain- ing 10% goat serum for 30 min. The slides were incubated with DNA2 antibody (1:200) overnight at 4 °C in a wet box, and immunohistochemical staining was done by using the Vectastain ABC avidin biotin-peroxidase enzyme complex kit (Vector Laboratories, Burlingame, CA, USA). Slides were counterstained with hematoxylin and mounted with Permount aqueous medium. obtained from US Biomax (Rockville, MD, USA). Formalin-ﬁxed, parafﬁn- embedded slides of a conditional LSL-K-RasG12D/transgenic pancreatic cancer mouse model was obtained from Dr Chinthalapally Rao (Depart- ment of Internal Medicine, University of Oklahoma, Oklahoma City, OK, USA). After slides were deparafﬁnized with xylene and rehydrated with ethanol, antigen retrieval was carried out by placing the slides in 0.1 M sodium citrate (pH 6.0) in a water bath for 30 min at 95 °C. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in phosphate-buffered saline for 10 min. Slides were washed three times with phosphate-buffered saline and incubated in blocking buffer contain- ing 10% goat serum for 30 min. The slides were incubated with DNA2 antibody (1:200) overnight at 4 °C in a wet box, and immunohistochemical staining was done by using the Vectastain ABC avidin biotin-peroxidase enzyme complex kit (Vector Laboratories, Burlingame, CA, USA). Slides were counterstained with hematoxylin and mounted with Permount aqueous medium. DISCUSSION For Three separate cell models were then used to test whether NSC-105808 can target cancer cells with oncogene activation (Figures 5c-e). First, we generated an MCF10A cell line, an immor- talized non-transformed breast epithelial cell line, with inducible H-Ras expression. Induction of H-Ras expression is shown in Supplementary Figure S9a. Upon activation of H-Ras, cells exhibited increased sensitivity to the DNA2 inhibitor (Figure 5c). Second, we used the E6/E7 oncoprotein to induce replication stress in non-transformed pancreatic ductal cells (DT-PD59) as previously described.42 Again, sensitivity to NSC-105808 was higher in cells with oncoprotein-induced replication stress (Figure 5d). In addition, we also established an inducible K-Ras expression cell line using non-transformed pancreatic ductal epithelial cells (Supplementary Figure S9a). Both 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide and colony formation assays showed that activation of K-Ras expression sensitized pancreatic ductal epithelial cells to NSC-105808 treatment (Supplementary Figures S9b and 9c). Finally, we tested a series of epithelial cell lines that represent different stages of breast cancer progression:43 control non-tumorigenic MCF10A cells; premalignant/tumorigenic MCF10AT1 cells, derived from MCF10A cells by the expression of oncogenic H-Ras; tumorigenic/ locally invasive MCF10-DCIS cells, and tumorigenic/metastatic MCF10CA1a cells, cloned from xenograft lesions induced by premalignant MCF10AT1 cells. Consistent with other observations, the three cell lines with oncogene activation were more sensitive to DNA2 inhibitor treatment compared with the MCF10A control (Figure 5e). Thus, in different cell models, oncogene activation sensitizes cells to NSC-105808. Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al 8 example, a single unrepaired DSB is lethal for cells. Alternatively DNA2 sensitivity to the drug can be altered by cell-speciﬁc features, such as DNA2 posttranslational modiﬁcations. Together our work suggests that temporary inhibition of DNA2 nuclease represents a promising strategy in cancer therapy. obtained from US Biomax (Rockville, MD, USA). Formalin-ﬁxed, parafﬁn- embedded slides of a conditional LSL-K-RasG12D/transgenic pancreatic cancer mouse model was obtained from Dr Chinthalapally Rao (Depart- ment of Internal Medicine, University of Oklahoma, Oklahoma City, OK, USA). After slides were deparafﬁnized with xylene and rehydrated with ethanol, antigen retrieval was carried out by placing the slides in 0.1 M sodium citrate (pH 6.0) in a water bath for 30 min at 95 °C. Endogenous peroxidase activity was quenched with 3% hydrogen peroxide in phosphate-buffered saline for 10 min. Senescence assay The assay was performed with a senescence β-galactosidase staining kit (#9860; Cell Signaling, Danvers, MA, USA) according to the manufacturer’s protocol (Supplementary Materials and Methods). Flow cytometric analysis of apoptosis and BrdU incorporation U2OS cells containing a single copy of the HR repair reporter substrate DR-green ﬂuorescent protein were generated as previously described.48 Single stand annealing reporter cells were kindly provided by Dr Jeremy Stark (Beckman Research Institute of the City of Hope, Duarte, CA, USA).49 Details in Supplementary Materials and Methods. Immunoﬂuorescent staining and chromatin fractionation For the detection of DNA damage-induced p-RPA34 (S4/S8), immuno- ﬂuorescent staining, and the preparation of chromatin fractions and western blot analyses were performed as described previously.33,48 High-throughput screen for inhibitors of DNA2 The screen entailed the use of oligonucleotide (dT)30 having a reporter ﬂuorescent moiety, 6-FAM (excitation max. = 494 nm, emission max. = 520 nm) on the 5′ end, and the dark quenching group, Iowa Black FQ (absorption max. = 531 nm), on the 3′ end. In addition, a second dark High-throughput screen for inhibitors of DNA2 The screen entailed the use of oligonucleotide (dT)30 having a reporter ﬂuorescent moiety, 6-FAM (excitation max. = 494 nm, emission max. = 520 nm) on the 5′ end, and the dark quenching group, Iowa Black FQ (absorption max. = 531 nm), on the 3′ end. In addition, a second dark Human cell cultures Human cell cultures U2OS and MCF10A cells were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA) and were maintained in McCoy’s 5A medium supplemented with 10% fetal bovine serum. MCF10A cells were cultured in mammary epithelial growth medium containing insulin, hydrocortisone, epidermal growth factor and bovine pituitary extract purchased from Clonetics (Lonza Group Ltd, Basel, Switzerland). MCF10A cells with stable H-Ras expression were generated by infection with lentiviral inducible-V5 construct (Life Technologies, Grand Island, NY, USA) and maintained in the presence of G418 (200 μg/ml) and blasticidin (4 μg/ ml). AsPC-1 and PANC-1 cells were cultured in Dulbecco's modiﬁed Eagle's medium (Corning Cellgro, Manassas, VA, USA) with 10% fetal bovine serum, vitamin (100 × ), non-essential amino acids (100 × ) and Penicillin/ Streptomycin (100 × ). BT549 cells were cultured in RPMI-1640 medium with 0.023 IU/ml insulin and Hs578T cells in Dulbecco's modiﬁed Eagle's medium medium with 0.01 mg/ml bovine insulin. The pancreatic cell lines (DT-PD59, DT-PD77-E6/E7) were kindly provided by Dr Michel J Quellette (University of Nebraska Medical Center, Omaha, NE, USA) and cultured as described.46 Pancreatic ductal epithelial cells with inducible K-Ras expression were generated and maintained in the Keratinocyte-SFM medium (Life Technologies) as previously described.47 A series of breast epithelial cell lines transformed by H-Ras (MCF10A, MCF10AT1, MCF10- DCIS and MCF10CA1a) were obtained from Dr Isabelle Bedrosian (Department of Surgical Oncology, MD Anderson Cancer Center, Houston, TX, USA) and maintained as described.43 In the experiments described in Figures 5b and c, MCF10A cells were obtained from ATCC. FANCD2 −/ − and control cell lines were obtained from Dr Randy Legerski (Department of Genetics, MD Anderson Cancer Center). Cells were maintained in Dulbecco's modiﬁed Eagle's medium containing 10% heat-inactivated fetal bovine serum and 1% penicillin/streptomycin at 37 °C in 5% CO2. All cell lines were validated by short tandem repeats DNA ﬁngerprinting using the AmpF short tandem repeats Identiﬁler kit according to the manufacturer instructions (Applied Biosystems, Foster City, CA, USA; cat. # 4322288). The short tandem repeats proﬁles were compared with known ATCC ﬁngerprints, to the Cell Line Integrated Molecular Authentication database (CLIMA) and to the MD Anderson ﬁngerprint database. The short tandem repeats proﬁles matched known DNA ﬁngerprints or were unique. Cell proliferation assay Cell proliferation was measured by 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (Sigma) reduction (Supplementary Materials and Methods). Human cell cultures Cell proliferation assay Cell proliferation was measured by 3-(4,5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (Sigma) reduction (Supplementary Materials and Methods). Nuclease assays Nuclease assays were performed in buffer R (20 mM Na-HEPES pH 7.5, 1 mM ATP, 0.1 mM DTT, 100 μg/ml bovine serum albumin, 0.05% Triton X-100, 2 mM MgCl2, 100 mM KCl) and contained 0.5 nM ends (2-kb substrate) or 5 nM ends (the ssDNA oligonucleotide PSOL6929) of DNA. Reactions containing Bloom syndrome also included an ATP regenerating system consisting of 10 mM creatine phosphate and 50 μg/ml creatine kinase. In control reactions, DMSO was used in place of the inhibitor. Reactions with DNA2 alone were incubated at 30 °C, and those with DNA2-Bloom syndrome-Replication protein A and EXO1 were incubated at 37 °C for the indicated times. After the addition of sodium dodecyl sulfate to 0.02%, proteinase K to 0.25 μg/μl, and 0.08% Orange G dye with a ﬁnal glycerol concentration of 10%, the reaction mixtures were incubated for 5 min at 37 °C. Electrophoresis was done in native 10% polyacrylamide gels in Tris-borate-EDTA buffer (ssDNA substrate) or in 1% agarose gels in Tris-acetate-EDTA buffer (40 mM Tris-acetate, pH 7.4, 0.5 mM EDTA) (2 kb substrate). Gels were dried onto DEAE paper on top of Whatman ﬁlter paper (GE, Issaquah, WA, USA) and then analyzed in a BioRad Personal Molecular Imager FX phosphorimager. Band intensity was quantiﬁed using Image Lab software (Bio-Rad, Hercules, CA, USA). The fraction of DNA substrate digested was determined by measuring the amount of substrate remaining and normalizing to the total substrate in the negative control lane containing no protein. 15 Jackson SP, Bartek J. The DNA-damage response in human biology and disease. Nature 2009; 461: 1071–1078. 16 Cejka P, Cannavo E, Polaczek P, Masuda-Sasa T, Pokharel S, Campbell JL et al. DNA end resection by Dna2-Sgs1-RPA and its stimulation by Top3-Rmi1 and Mre11-Rad50-Xrs2. Nature 2010; 467: 112–116. 17 Hoa NN, Akagawa R, Yamasaki T, Hirota K, Sasa K, Natsume T et al. Relative contribution of four nucleases, CtIP, Dna2, Exo1 and Mre11, to the initial step of DNA double-strand break repair by homologous recombination in both the chicken DT40 and human TK6 cell lines. Genes Cells 2015; 20: 1059–1076. 18 Hoa NN, Kobayashi J, Omura M, Hirakawa M, Yang SH, Komatsu K et al. BRCA1 and CtIP are both required to recruit Dna2 at double-strand breaks in homologous recombination. PLoS ONE 2015; 10: e0124495. 19 Howard SM, Yanez DA, Stark JM. Nuclease assays DNA damage response factors from diverse pathways, including DNA crosslink repair, mediate alternative end joining. PLoS Genet 2015; 11: e1004943. 20 Liao S, Toczylowski T, Yan H. Identiﬁcation of the Xenopus DNA2 protein as a major nuclease for the 5'-43' strand-speciﬁc processing of DNA ends. Nucleic Acids Res 2008; 36: 6091–6100. 21 Nimonkar AV, Genschel J, Kinoshita E, Polaczek P, Campbell JL, Wyman C et al. BLM-DNA2-RPA-MRN and EXO1-BLM-RPA-MRN constitute two DNA end resection machineries for human DNA break repair. Genes Dev 2011; 25: 350–362. machineries for human DNA break repair. Genes Dev 2011; 25: 350– 22 Niu H, Chung WH, Zhu Z, Kwon Y, Zhao W, Chi P et al. Mechanism of the ATP-dependent DNA end-resection machinery from Saccharomyces cerevisiae. Nature 2010; 467: 108–111. The authors declare no conﬂict of interest. The authors declare no conﬂict of interest. 23 Zhu Z, Chung WH, Shim EY, Lee SE, Ira G. Sgs1 helicase and two nucleases dna2 and exo1 resect DNA double-strand break ends. Cell 2008; 134: 981–994. 24 Hu J, Sun L, Shen F, Chen Y, Hua Y, Liu Y et al. The intra-S phase checkpoint targets Dna2 to prevent stalled replication forks from reversing. Cell 2012; 149: 1221–1232. Hu J, Sun L, Shen F, Chen Y, Hua Y, Liu Y et al. The intra-S phase check Dr Kenneth L Scott (Baylor College of Medicine) for providing K-Ras-pancreatic ductal epithelial cells. quencher ZEN (N,N-diethyl-4-(4-nitronaphthalen-1-ylazo)-phenylamine, absorption max. = 532 nm) is placed between 9th and 10th base (Figure 2a). Initial screen, it was performed with yDna2.22 Then, the top candidate compounds were tested on hDNA2. Identical reaction conditions were used for assaying the nuclease activity of yDna2 and hDNA2. We note that yDna2 is more active on the ssDNA substrate than hDNA2, thus the experiment with ssDNA substrates were done with 10 nM of hDNA2. The reaction buffer was 20 mM HEPES/NaOH, pH 7.5, 60 mM KCl, 1 mM DTT, 100 μg/ml bovine serum albumin and 0.05% Triton X-100. The nuclease assay was initiated by adding MgCl2 (3 mM ﬁnal concentration). T5 nuclease was assayed in the buffer (50 mM potassium acetate, pH 7.9, 20 mM Tris-acetate, 10 mM magnesium acetate and 1 mM DTT) provided by the vendor (NEB, Ipswich, MA, USA). The concentration of the ﬂuorogenic substrate was 0.1 μM. The optimal amount of yDna2, hDNA2 and T5 nuclease was determined by utilizing the Kaleidagraph software (Reading, PA, USA) to ﬁt the data according to Michaelis Menton Kinetics (V = V0+Vmax*x/(Km+x)). For curve ﬁtting, the maximum increase in ﬂuorescence (upon adding 5 nM of yDna2 or 166 nM of hDNA2 or 10 units of T5 nuclease) corresponds to Vmax, the background ﬂuorescence (without enzyme) corresponds to V0 and ‘x’ is the enzyme concentration (nM). All the compounds were dissolved in DMSO and DMSO alone was used as control; the ﬁnal concentration of DMSO was at 3.33% for protein-based assay and 0.1% for cell-based assay. The Z’ value was calculated as previously described.50 Cell proliferation was determined using 4,6- diamidino-2-phenylindole staining followed by high-throughput micro- scopy in a 384-well plate setup. In brief, cells were ﬁxed with formaldehyde, washed and stained with 4,6-diamidino-2-phenylindole and the number of nuclei in each well counted by automated microscopy. The data were analyzed as described.51 Nuclease substrates The oligonucleotide PSOL6929 (5′-TTCATGGCTTAGAGCTTAATTGCTGAA TCT-3′) was 5′ end-labeled with γ32P-ATP (Perkin-Elmer, Waltham, MA, USA) and T4 polynucleotide kinase (NEB). The 2-kb dsDNA substrate is a fragment of the hDNA2 gene created by PCR with oligonucleotides PSOL4642 (5′-GATCCTCTAGTACTTCTC-3′) and PSOL6134 (5′-TCTACCT CAAGACTGGTCAG-3′) using the DNA2 expression vector pJD7239 as template; 50 μCi of α32P-dCTP (Perkin-Elmer) was included in the reaction to randomly radiolabel the PCR fragment. 11 Bartek J, Mistrik M, Bartkova J. Thresholds of replication stress signaling in cancer development and treatment. Nat Struct Mol Biol 2012; 19: 5–7. 12 Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K et al. DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 2005; 434: 864–870. 13 Campbell PJ, Yachida S, Mudie LJ, Stephens PJ, Pleasance ED, Stebbings LA et al. The patterns and dynamics of genomic instability in metastatic pancreatic cancer. Nature 2010; 467: 1109–1113. 14 Gorgoulis VG, Vassiliou LV, Karakaidos P, Zacharatos P, Kotsinas A, Liloglou T et al. Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 2005; 434: 907–913. REFERENCES 1 Bryant HE, Schultz N, Thomas HD, Parker KM, Flower D, Lopez E et al. Speciﬁc killing of BRCA2-deﬁcient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature 2005; 434: 913–917. 2 Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 2005; 434: 917–921. 3 Bartkova J, Rezaei N, Liontos M, Karakaidos P, Kletsas D, Issaeva N et al. Oncogene- induced senescence is part of the tumorigenesis barrier imposed by DNA damage checkpoints. Nature 2006; 444: 633–637. 4 Di Micco R, Fumagalli M, Cicalese A, Piccinin S, Gasparini P, Luise C et al. Oncogene-induced senescence is a DNA damage response triggered by DNA hyper-replication. Nature 2006; 444: 638–642. 5 Gaillard H, Garcia-Muse T, Aguilera A. Replication stress and cancer. Nat Rev Cancer 2015; 15: 276–289. 6 Macheret M, Halazonetis TD. DNA replication stress as a hallmark of cancer. Annu Rev Pathol 2015; 10: 425–448. 7 Dobbelstein M, Sorensen CS. Exploiting replicative stress to treat cancer. Nat Rev Drug Disc 2015; 14: 405–423. 8 Carvalho JF, Kanaar R. Targeting homologous recombination-mediated DNA repair in cancer. Expert Opin Ther Targets 2014; 18: 427–458. 9 Dietlein F, Thelen L, Reinhardt HC. Cancer-speciﬁc defects in DNA repair pathways as targets for personalized therapeutic approaches. Trends Genet 2014; 30: 326–339. 10 Kelley MR. Future directions with DNA repair inhibitors: a roadmap for disruptive approaches to cancer therapy In: Kelley MR (ed). DNA Repair in Cancer Therapy. Molecular Targets and Clinical Applications 1st edn. Elsevier: Amsterdam, The Netherlands, 2012, pp 301–310. DNA2 nuclease as a target in cancer therapy S Kumar et al 9 Dr Kenneth L Scott (Baylor College of Medicine) for providing K-Ras-pancreatic ductal epithelial cells. Immunohistochemical staining Tissue microarrays of formalin-ﬁxed, parafﬁn-embedded pancreatitis, pancreatic tumors with their corresponding nonneoplastic tissues were Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy DNA2 nuclease as a target in cancer therapy S Kumar et al DNA2 nuclease as a target in cancer therapy S Kumar et al PLoS ONE 2013; 8: e71755. 33 Peng G, Dai H, Zhang W, Hsieh HJ, Pan MR, Park YY et al. Human nuclease/helicase DNA2 alleviates replication stress by promoting DNA end resection. Cancer Res 2012; 72: 2802–2813. 48 Peng G, Yim EK, Dai H, Jackson AP, Burgt I, Pan MR et al. BRIT1/MCPH1 links chromatin remodelling to DNA damage response. Nat Cell Biol 2009; 11: 865–872. 49 Bennardo N, Cheng A, Huang N, Stark JM. Alternative-NHEJ is a mechanistically distinct pathway of mammalian chromosome break repair. PLoS Genet 2008; 4: e1000110. 34 Strauss C, Kornowski M, Benvenisty A, Shahar A, Masury H, Ben-Porath I et al. The DNA2 nuclease/helicase is an estrogen-dependent gene mutated in breast and ovarian cancers. Oncotarget 2014; 5: 9396–9409. 35 Costello E, Greenhalf W, Neoptolemos JP. New biomarkers and targets in pancreatic cancer and their application to treatment. Nat Rev Gastroenterol Hepatol 2012; 9: 435–444. 50 Zhang JH, Chung TD, Oldenburg KR. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J Biomol Screen 1999; 4: 67–73. p 36 Hidalgo M. Pancreatic cancer. N Engl J Med 2010; 362: 1605–1617. 51 Holbeck SL, Collins JM, Doroshow JH. Analysis of food and drug administration- approved anticancer agents in the NCI60 panel of human tumor cell lines. Mol Cancer Ther 2010; 9: 1451–1460. 51 Holbeck SL, Collins JM, Doroshow JH. Analysis of food and drug administration- approved anticancer agents in the NCI60 panel of human tumor cell lines. Mol Cancer Ther 2010; 9: 1451–1460. 37 Morris JPt, Wang SC, Hebrok M. KRAS, Hedgehog, Wnt and the twisted developmental biology of pancreatic ductal adenocarcinoma. Nat Rev Cancer 2010; 10: 683–695. 38 Stewart JA, Campbell JL, Bambara RA. Dna2 is a structure-speciﬁc nuclease, with afﬁnity for 5'-ﬂap intermediates. Nucleic Acids Res 2010; 38: 920–930. Oncogenesis is an open-access journal published by Nature Publishing Group. This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ 39 Daley JM, Chiba T, Xue X, Niu H, Sung P. DNA2 nuclease as a target in cancer therapy S Kumar et al DNA2 nuclease as a target in cancer therapy S Kumar et al 10 26 Bae SH, Bae KH, Kim JA, Seo YS. RPA governs endonuclease switching during processing of Okazaki fragments in eukaryotes. Nature 2001; 412: 456–461. 42 Bester AC, Roniger M, Oren YS, Im MM, Sarni D, Chaoat M et al. Nucleotide deﬁciency promotes genomic instability in early stages of cancer development. Cell 2011; 145: 435–446. 27 Budd ME, Campbell JL. A yeast replicative helicase, Dna2 helicase, interacts with yeast FEN-1 nuclease in carrying out its essential function. Mol Cell Biol 1997; 17: 2136–2142. 43 Dawson PJ, Wolman SR, Tait L, Heppner GH, Miller FR. MCF10AT: a model for the evolution of cancer from proliferative breast disease. Am J Pathol 1996; 148: 313–319. 28 Duxin JP, Moore HR, Sidorova J, Karanja K, Honaker Y, Dao B et al. Okazaki fragment processing-independent role for human Dna2 enzyme during DNA replication. J Biol Chem 2012; 287: 21980–21991. 44 Liu W, Zhou M, Li Z, Li H, Polaczek P, Dai H et al. A selective small molecule DNA2 inhibitor for sensitization of human cancer cells to chemotherapy. EBioMedicine 2016; 6: 73–86. 29 Zheng L, Zhou M, Guo Z, Lu H, Qian L, Dai H et al. Human DNA2 is a mitochondrial nuclease/helicase for efﬁcient processing of DNA replication and repair inter- mediates. Mol Cell 2008; 32: 325–336. 45 Lin W, Sampathi S, Dai H, Liu C, Zhou M, Hu J et al. Mammalian DNA2 helicase/nuclease cleaves G-quadruplex DNA and is required for telomere integrity. EMBO J 2013; 32: 1425–1439. 30 Kumar S, Burgers PM. Lagging strand maturation factor Dna2 is a component of the replication checkpoint initiation machinery. Genes Dev 2013; 27: 313–321. 46 Campbell PM, Groehler AL, Lee KM, Ouellette MM, Khazak V, Der CJ. K-Ras promotes growth transformation and invasion of immortalized human pancreatic cells by Raf and phosphatidylinositol 3-kinase signaling. Cancer Res 2007; 67: 2098–2106. 31 Vernon M, Lobachev K, Petes TD. High rates of ‘unselected’ aneuploidy and chromosome rearrangements in tel1 mec1 haploid yeast strains. Genetics 2008; 179: 237–247. 47 Tsang YH, Dogruluk T, Tedeschi PM, Wardwell-Ozgo J, Lu H, Espitia M et al. Functional annotation of rare gene aberration drivers of pancreatic cancer. Nat Commun 2016; 7: 10500. 32 Dominguez-Valentin M, Therkildsen C, Veerla S, Jonsson M, Bernstein I, Borg A et al. Distinct gene expression signatures in lynch syndrome and familial colorectal cancer type x. ACKNOWLEDGEMENTS The research presented here was supported by CPRIT grant RP140456 to GI and GP, R01 GM080600 to GI, the University of Texas MD Anderson Cancer Center Duncan Family Institute for Cancer Prevention and Risk Assessment grant to GP, CPRIT grant RP110532 to CCS, NIH grants ES015632 and ES015252 to PS. The authors thank 25 Thangavel S, Berti M, Levikova M, Pinto C, Gomathinayagam S, Vujanovic M et al. DNA2 drives processing and restart of reversed replication forks in human cells. J Cell Biol 2015; 208: 545–562. Oncogenesis (2017), 1 – 10 DNA2 nuclease as a target in cancer therapy S Kumar et al Multifaceted role of the Topo IIIalpha-RMI1-RMI2 complex and DNA2 in the BLM-dependent pathway of DNA break end resection. Nucleic Acids Res 2014; 42: 11083–11091. 40 Polo SE, Blackford AN, Chapman JR, Baskcomb L, Gravel S, Rusch A et al. Regulation of DNA-end resection by hnRNPU-like proteins promotes DNA double-strand break signaling and repair. Mol Cell 2012; 45: 505–516. 41 Karanja KK, Lee EH, Hendrickson EA, Campbell JL. Preventing over-resection by DNA2 helicase/nuclease suppresses repair defects in Fanconi anemia cells. Cell Cycle 2014; 13: 1540–1550. © The Author(s) 2017 Oncogenesis (2017), 1 – 10 Supplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis) © The Author(s) 2017 Supplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis) Supplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis) Supplementary Information accompanies this paper on the Oncogenesis website (http://www.nature.com/oncsis) he Oncogenesis website (http://www.nature.com/onc Oncogenesis (2017), 1 – 10 Oncogenesis (2017), 1 – 10
https://openalex.org/W4361240905	https://figshare.com/articles/journal_contribution/Supplementary_Figure_1_from_Elevated_CRAF_as_a_Potential_Mechanism_of_Acquired_Resistance_to_BRAF_Inhibition_in_Melanoma/22371197/1/files/39816293.pdf	unk	null	Supplementary Figure 1 from Elevated CRAF as a Potential Mechanism of Acquired Resistance to BRAF Inhibition in Melanoma	null	2,023	cc-by	16	A A M14 parental cells M14 parental cells B B M14BRR2 M14BRR2 Figure S1 Figure S1
https://openalex.org/W2529325721	http://journals.iucr.org/e/issues/2016/11/00/su5328/su5328.pdf	English	null	Crystal structure and Hirshfeld surface analysis of 1-carboxy-2-(3,4-dihydroxyphenyl)ethan-1-aminium bromide 2-ammonio-3-(3,4-dihydroxyphenyl)propanoate	Acta crystallographica. Section E, Crystallographic communications	2,016	cc-by	5,190	ISSN 2056-9890 ISSN 2056-9890 Perumal Kathiravan,a Thangavelu Balakrishnan,a* Perumal Venkatesan,b Kandasamy Ramamurthi,c Marı´a Judith Percinob and Subbiah Thamotharand* Perumal Kathiravan,a Thangavelu Balakrishnan,a* Perumal Venkatesan,b Kandasamy Ramamurthi,c Marı´a Judith Percinob and Subbiah Thamotharand* Received 19 September 2016 Accepted 2 October 2016 Received 19 September 2016 Accepted 2 October 2016 aCrystal Growth Laboratory, PG and Research Department of Physics, Periyar EVR Government College (Autonomous), Tiruchirappalli 620 023, India, bLaboratorio de Polı´meros, Centro de Quı´mica Instituto de Ciencias, Beneme´rita Universidad Auto´noma de Puebla (BUAP), Complejo de Ciencias, ICUAP, Edif. 103H, 22 Sur y San Claudio, C.P. 72570 Puebla, Puebla, Mexico, cCrystal Growth and Thin Film Laboratory, Department of Physics and Nanotechnology, SRM University, Kattankulathur 603 203, India, and dBiomolecular Crystallography Laboratory, Department of Bioinformatics, School of Chemical and Biotechnology, SASTRA University, Thanjavur 613 401, India. Correspondence e-mail: balacrystalgrowth@gmail.com, thamu@scbt.sastra.edu Edited by H. Stoeckli-Evans, University of Neuchaˆtel, Switzerland Edited by H. Stoeckli-Evans, University of Neuchaˆtel, Switzerland Keywords: crystal structure; dopa; cyclic N— H Br hydrogen bonds; hydrogen bonding; Hirshfeld surfaces. Keywords: crystal structure; dopa; cyclic N— H Br hydrogen bonds; hydrogen bonding; Hirshfeld surfaces. In the title molecular salt, C9H12NO4 +BrC9H11NO4, one of the dopa molecules is in the cationic form in which the -amino group is protonated and the -carboxylic acid group is uncharged, while the second dopa molecule is in the zwitterion form. The Br anion occupies a special position and is located on a twofold rotation axis. The two dopa molecules are interconnected by short O—H O hydrogen bonds. In the crystal, the various units are linked by O— H O, N—H Br and N—H O hydrogen bonds, forming a three-dimen- sional framework. The title compound was reﬁned as an inversion twin with an absolute structure parameter of 0.023 (8). CCDC reference: 1507715 CCDC reference: 1507715 Supporting information: this article has supporting information at journals.iucr.org/e Supporting information: this article has supporting information at journals.iucr.org/e 1. Chemical context research communications research communications Crystal structure and Hirshfeld surface analysis of 1-carboxy-2-(3,4-dihydroxyphenyl)ethan-1- aminium bromide 2-ammonio-3-(3,4-dihydroxy- phenyl)propanoate 2. Structural commentary hydrogen bonds (Table 1), forming a three-dimensional framework. The cationic dopa molecules form dimers in which the carboxylic acid groups (O4) of the dopa molecules are interconnected via a short O—H O hydrogen bond and the dimers are arranged as ribbons propagating along the b axis (Fig. 3). The protonated amino group forms three hydrogen bonds; two of them with the Br anions and one with the carbonyl oxygen atom, O3, of the carboxylic acid group. The dopa molecules aggregate in a head-to-tail sequence of the type NH3 +—CHR—COO NH3 +—CHR—COO , in which the -amino atom, N1, and the -carboxylate atom O3 form a hydrogen-bonded peptide-like arrangement (layers), as observed in many amino acid–carboxylic acid complexes (Sharma et al., 2006; Selvaraj et al., 2007). Adjacent layers are interconnected by short O—H O hydrogen bonds. These two interactions combine to form an R4 4(18) ring motif (Fig. 4). Similar interactions are observed in dopa and its HCl form (Mostad et al., 1971; Jandacek & Earle, 1971; Mostad & Rømming, 1974). The asymmetric unit of the title salt, Fig. 1, is composed of a Br anion located on a twofold rotation axis, a dopa molecule in the zwitterionic form and a cationic dopa molecule. In the latter, the -amino group is protonated and carries a positive charge and the hydrogen atom (H4O) of the -carboxylic acid group is located on a general position and was reﬁned with 50% occupancy. The crystal structures of l-dopa (Mostad et al., 1971) and its hydrochloride form (Jandacek & Earle, 1971; Mostad & Rømming, 1974) have been reported. Both of these compounds crystallized in the monoclinic space group P21. In the crystal structure of l-dopa HCl, the -amino group is protonated and the -carboxylic acid is neutral. The stoi- chiometry between the cation and the Cl anion is 1:1. The authors of these structures concluded that l-dopa exists as the S enantiomer, based on the R factor and the effects of anomalous scattering. However, the deposited coordinates for these structures belong to the R conﬁguration. Therefore, the l-dopa HCl structure was inverted and used for superposition with one of the dopa molecules of the title compound. These structures superimpose well, with an r.m.s. deviation of 0.045 A˚ (Fig. 2). research communications Figure 1 The molecular structure of the title molecular salt, showing the atom labelling [symmetry code: (#) x + 3, y, z + 1]. Displacement ellipsoids are drawn at the 50% probability level. Table 1 Hydrogen-bond geometry (A˚ , ). D—H A D—H H A D A D—H A O1—H1O O3i 0.82 1.98 2.782 (2) 166 O2—H2O O1ii 0.82 2.32 3.004 (2) 142 O2—H2O O2ii 0.82 2.26 2.9557 (8) 144 O4—H4O O4iii 0.85 (4) 1.61 (4) 2.449 (2) 169 (6) N1—H1A Br1iv 0.95 (3) 2.41 (3) 3.359 (3) 179 (3) N1—H1B Br1 0.91 (3) 2.41 (3) 3.295 (3) 164 (2) N1—H1C O3v 0.89 (3) 1.95 (3) 2.821 (2) 164 (3) Symmetry codes: (i) x 1; y þ 1; z; (ii) x þ 1 2; y 1 2; z þ 1 2; (iii) x þ 3; y; z þ 1; (iv) x; y þ 1; z; (v) x 1; y; z. Table 1 Hydrogen-bond geometry (A˚ , ). Figure 1 Figure 1 The molecular structure of the title molecular salt, showing the atom labelling [symmetry code: (#) x + 3, y, z + 1]. Displacement ellipsoids are drawn at the 50% probability level. Symmetry codes: (i) x 1; y þ 1; z; (ii) x þ 1 2; y 1 2; z þ 1 2; (iii) x þ 3; y; z þ 1; (iv) x; y þ 1; z; (v) x 1; y; z. 2. Structural commentary The amino group (via H1A and H1B) of the cationic dopa molecule participates in intermolecular N—H Br inter- actions with two different Br anions (Table 1). These inter- actions interconnect the cations and anions into a cyclic motif 3. Supramolecular features 1. Chemical context 1. Chemical context An aromatic amino acid enzyme hydroxylase converts l-tyrosine into l-dopa (l-3,4-dihydroxyphenylalanine). After conversion, l-dopa acts as a precursor for the neuro- transmitters dopamine, norepinephrine and epinephrine. The l-dopa molecule is also effectively used in the symptomatic treatment of Parkinson’s disease (Chan et al., 2012). In view of this interest, we have crystallized the title salt and report herein on its crystal structure. The hydrogen-bonding pattern and the relative contributions of various intermolecular interactions present are compared with the closely related chloride counterpart reported on earlier (Jandacek & Earle, 1971; Mostad & Rømming, 1974). An aromatic amino acid enzyme hydroxylase converts l-tyrosine into l-dopa (l-3,4-dihydroxyphenylalanine). After conversion, l-dopa acts as a precursor for the neuro- transmitters dopamine, norepinephrine and epinephrine. The l-dopa molecule is also effectively used in the symptomatic treatment of Parkinson’s disease (Chan et al., 2012). In view of this interest, we have crystallized the title salt and report herein on its crystal structure. The hydrogen-bonding pattern and the relative contributions of various intermolecular interactions present are compared with the closely related chloride counterpart reported on earlier (Jandacek & Earle, 1971; Mostad & Rømming, 1974). y y g g p and the relative contributions of various intermolecular interactions present are compared with the closely related chloride counterpart reported on earlier (Jandacek & Earle, 1971; Mostad & Rømming, 1974). Acta Cryst. (2016). E72, 1544–1548 1544 http://dx.doi.org/10.1107/S2056989016015425 3. Supramolecular features Figure 3 The crystal packing of the title molecular salt, viewed along the b axis. H atoms have been omitted for clarity. Figure 3 The crystal packing of the title molecular salt, viewed along the b axis. H atoms have been omitted for clarity. 3. Supramolecular features The structure of the title compound features a network of intermolecular N—H Br, N—H O and O—H O Figure 2 Superposition of the cationic dopa molecule in the title compound (red) and in l-dopaHCl (blue). Figure 3 The crystal packing of the title molecular salt, viewed along the b axis. H atoms have been omitted for clarity. The structure of the title compound features a network of intermolecular N—H Br, N—H O and O—H O Figure 2 Superposition of the cationic dopa molecule in the title compound (red) and in l-dopaHCl (blue). Figure 3 The crystal packing of the title molecular salt, viewed along the b axis. H atoms have been omitted for clarity. Figure 2 Superposition of the cationic dopa molecule in the title compound (red) and in l-dopaHCl (blue). Figure 2 Superposition of the cationic dopa molecule in the title compound (red) and in l-dopaHCl (blue). Kathiravan et al. C9H12NO4 +BrC9H11NO4 1545 1545 Acta Cryst. (2016). E72, 1544–1548 research communications Figure 4 Part of the crystal structure of the title molecular salt, showing the R4 4(18) ring motifs formed by N—H O and O—H O hydrogen bonds. One of the hydroxy groups (O1 H1O) is involved in an gure 4 art of the crystal structure of the title molecular salt, showing the R4 4(18) ng motifs formed by N—H O and O—H O hydrogen bonds. Figure 6 The side chain side chain interactions of the dopa molecules in the title molecular salt, through intermolecular O—H O hydrogen bonds. Figure 6 The side chain side chain interactions of the dopa molecules in the title molecular salt, through intermolecular O—H O hydrogen bonds. Figure 6 The side chain side chain interactions of the dopa molecules in the title molecular salt, through intermolecular O—H O hydrogen bonds. Figure 4 Part of the crystal structure of the title molecular salt, showing the R4 4(18) ring motifs formed by N—H O and O—H O hydrogen bonds. 3. Supramolecular features One of the hydroxy groups (O1—H1O) is involved in an intermolecular O—H O hydrogen bond with the carbonyl oxygen (O3) of the dopa molecule. This interaction links the dopa molecules into a C(9) chain. The other hydroxy (O2— H2O) group participates in bifurcated hydrogen bonds with two different hydroxy O atoms (O1 and O2) of adjacent dopa layers. The side chain of the dopa molecules in one layer is interconnected by the side chain of the dopa molecules in the adjacent layer through these interactions (Fig. 6). These interactions are also observed in the dopa hydrochloride structure. that can be described as an R4 2(8) ring and it runs parallel to the b axis (Fig. 5). This pattern is also observed in the crystal structure of l-dopaHCl, where two intermolecular N— H Cl hydrogen bonds link the cations and anions into a chain. There, adjacent chains are interconnected through O— H Cl hydrogen bonds (carboxylic acid Cl). Figure 5 Part of the crystal structure of the title molecular salt, showing the R4 2(8) ring motifs formed by N—H Br hydrogen bonds. 4. Hirshfeld surface analysis 4. Hirshfeld surface analysis 4. Hirshfeld surface analysis The Hirshfeld surfaces (HS) mapped with dnorm and 2D ﬁngerprint plots were generated using the program Crystal- Explorer (Wolff et al., 2012). The two different orientations of the HS diagram for complete dopa molecules along with Br Figure 5 Figure 5 Figure 5 Part of the crystal structure of the title molecular salt, showing the R4 2(8) ring motifs formed by N—H Br hydrogen bonds. Figure 5 Part of the crystal structure of the title molecular salt, showing the R4 2(8) ring motifs formed by N—H Br hydrogen bonds. 1546 Kathiravan et al. C9H12NO4 +BrC9H11NO4 Acta Cryst. (2016). E72, 1544–1548 research communications research communications of restraints 1 H-atom treatment H atoms treated by a mixture of independent and constrained reﬁnement max, min (e A˚ 3) 0.37, 0.31 Absolute structure Reﬁned as an inversion twin Absolute structure parameter 0.023 (8) Computer programs: APEX2, SAINT and XPREP (Bruker, 2004), SIR92 (Altomare et al., 1994), Mercury (Macrae et al., 2006), SHELXL2014 (Sheldrick, 2015) and publCIF (Westrip, 2010). C9H12NO4 +BrC9H11NO4 475.29 Monoclinic, I2 293 6.1456 (3), 5.6385 (2), 28.2561 (10) 94.147 (2) 976.57 (7) 2 Mo K 2.16 0.30 0.25 0.25 Figure 7 Figure 7 Figure 7 Two different views of the Hirshfeld surfaces of the dimeric dopa molecules along with a Br anion. Two different views of the Hirshfeld surfaces of the dimeric dopa molecules along with a Br anion. anion are shown in Fig. 7. The two-dimensional ﬁngerprint plots are illustrated in Fig. 8. The HS analysis suggests that the intermolecular O H contacts contribute most (41.4%) to the Computer programs: APEX2, SAINT and XPREP (Bruker, 2004), SIR92 (Altomare et al., 1994), Mercury (Macrae et al., 2006), SHELXL2014 (Sheldrick, 2015) and publCIF (Westrip, 2010). Figure 8 Two-dimensional ﬁngerprint plots: (a) complete unit of dopa and (b) anionic Br in the title salt, and (c) cationic dopa and (d) anionic Cl in l- dopa hydrochloride. The various types of contacts are indicated. crystal packing compared to other contacts. For example, the relative contributions of H H, C H and H Br contacts are 29, 18.6 and 6.1%, respectively, with regard to the complete unit of the dopa molecule. Concerning the Br anion, the relative contributions of H Br and O Br contacts are 64.1 and 10.2%, respectively. In the dopa HCl structure, the relative contributions of O H, H H, C H and H Cl contacts are 40.5, 25.2, 17.1 and 14.1%, respectively, with respect to the cationic dopa molecule. It is of interest to note that O H and H H contacts are reduced by 1.1 and 3.8%, respectively, when compared to the title salt. Concerning the Cl anion, the relative contribution of H Cl contacts is 90.4%. This is approximately 26% higher compared to the relative contri- butions of H Br contacts in the title salt. 5. Synthesis and crystallization research communications anion are shown in Fig. 7. The two-dimensional ﬁngerprint plots are illustrated in Fig. 8. The HS analysis suggests that the intermolecular O H contacts contribute most (41.4%) to the Figure 7 Two different views of the Hirshfeld surfaces of the dimeric dopa molecules along with a Br anion. Table 2 Experimental details. Crystal data Chemical formula C9H12NO4 +BrC9H11NO4 Mr 475.29 Crystal system, space group Monoclinic, I2 Temperature (K) 293 a, b, c (A˚ ) 6.1456 (3), 5.6385 (2), 28.2561 (10) () 94.147 (2) V (A˚ 3) 976.57 (7) Z 2 Radiation type Mo K (mm1) 2.16 Crystal size (mm) 0.30 0.25 0.25 Data collection Diffractometer Bruker Kappa APEXII CCD Absorption correction Multi-scan (SADABS; Bruker, 2004) Tmin, Tmax 0.562, 0.619 No. of measured, independent and observed [I > 2(I)] reﬂections 8138, 2827, 2421 Rint 0.024 (sin /)max (A˚ 1) 0.833 Reﬁnement R[F 2 > 2(F 2)], wR(F 2), S 0.026, 0.056, 0.97 No. of reﬂections 2827 No. of parameters 151 No. of restraints 1 H-atom treatment H atoms treated by a mixture of independent and constrained reﬁnement max, min (e A˚ 3) 0.37, 0.31 Absolute structure Reﬁned as an inversion twin Absolute structure parameter 0.023 (8) Computer programs: APEX2, SAINT and XPREP (Bruker, 2004), SIR92 (Altomare et al., 1994), Mercury (Macrae et al., 2006), SHELXL2014 (Sheldrick, 2015) and publCIF (Westrip, 2010). Figure 7 Two different views of the Hirshfeld surfaces of the dimeric dopa molecules along with a Br anion. Table 2 Experimental details. Crystal data Chemical formula C9H12NO4 +BrC9H11NO4 Mr 475.29 Crystal system, space group Monoclinic, I2 Temperature (K) 293 a, b, c (A˚ ) 6.1456 (3), 5.6385 (2), 28.2561 (10) () 94.147 (2) V (A˚ 3) 976.57 (7) Z 2 Radiation type Mo K (mm1) 2.16 Crystal size (mm) 0.30 0.25 0.25 Data collection Diffractometer Bruker Kappa APEXII CCD Absorption correction Multi-scan (SADABS; Bruker, 2004) Tmin, Tmax 0.562, 0.619 No. of measured, independent and observed [I > 2(I)] reﬂections 8138, 2827, 2421 Rint 0.024 (sin /)max (A˚ 1) 0.833 Reﬁnement R[F 2 > 2(F 2)], wR(F 2), S 0.026, 0.056, 0.97 No. of reﬂections 2827 No. of parameters 151 No. 5. Synthesis and crystallization l-dopa and HBr (1:1 molar ratio) were dissolved in double- distilled water and stirred well for 4 h. The homogeneous solution was ﬁltered and the ﬁltrate allowed to evaporate slowly. Colourless block-like crystals were harvested after a growth period of two weeks. Figure 8 g Two-dimensional ﬁngerprint plots: (a) complete unit of dopa and (b) anionic Br in the title salt, and (c) cationic dopa and (d) anionic Cl in l- dopa hydrochloride. The various types of contacts are indicated. g Two-dimensional ﬁngerprint plots: (a) complete unit of dopa and (b) anionic Br in the title salt, and (c) cationic dopa and (d) anionic Cl in l- dopa hydrochloride. The various types of contacts are indicated. Kathiravan et al. C9H12NO4 +BrC9H11NO4 1547 1547 Acta Cryst. (2016). E72, 1544–1548 6. Refinement Crystal data, data collection and structure reﬁnement details are summarized in Table 2. The amino and carboxylic acid H atoms were located in a difference Fourier map and freely reﬁned. The OH and C-bound H atoms were included in calculated positions and treated as riding atoms: C—H = 0.93– 0.98 A˚ , O—H = 0.82 A˚ with Uiso(H) = 1.2Ueq(C) and Uiso(H) = 1.5Ueq(O). The title compound was reﬁned as an inversion twin; absolute structure parameter = 0.023 (8). References research communications 6. Refinement References Altomare, A., Cascarano, G., Giacovazzo, C., Guagliardi, A., Burla, M. C., Polidori, G. & Camalli, M. (1994). J. Appl. Cryst. 27, 435. Bruker (2004). APEX2, SAINT, XPREP and SADABS. Bruker AXS Inc., Madison, Wisconsin, USA. Chan, S. W., Dunlop, R. A., Rowe, A., Double, K. L. & Rodgers, K. J. (2012). Exp. Neurol. 238, 29–37. ( ) p Jandacek, R. J. & Earle, K. M. (1971). Acta Cryst. B27, 841–845. Macrae, C. F., Edgington, P. R., McCabe, P., Pidcock, E., Shields, G. P., Taylor, R., Towler, M. & van de Streek, J. (2006). J. Appl. Cryst. 39, 453–457. Mostad, A., Otternsen, T. & Rømming, C. (1971). Acta Chem. Scand. 25, 3549–3560. Mostad, A. & Rømming, C. (1974). Acta Chem. Scand. Ser. B, 28, 1161–1168. Acknowledgements Selvaraj, M., Thamotharan, S., Roy, S. & Vijayan, M. (2007). Acta Cryst. B63, 459–468. TB acknowledges the Council of Scientiﬁc and Industrial Research (CSIR), India for providing ﬁnancial support [project ref. No. 03 (1314)/14/EMR-II dt.16–04-14]. ST is extremely grateful to the management of SASTRA University for their encouragement and ﬁnancial support (Professor TRR fund), and also thanks the DST–SERB (SB/YS/LS-19/ 2014) for research funding. Sharma, A., Thamotharan, S., Roy, S. & Vijayan, M. (2006). Acta Cryst. C62, o148–o152. y , Sheldrick, G. M. (2015). Acta Cryst. C71, 3–8. Westrip, S. P. (2010). J. Appl. Cryst. 43, 920–925. Wolff, S. K., Grimwood, D. J., McKinnon, J. J., Turner, M. J., Jayatilaka, D. & Spackman, M. A. (2012). Crystal Explorer. The University of Western Australia. 1548 Kathiravan et al. C9H12NO4 +BrC9H11NO4 Acta Cryst. (2016). E72, 1544–1548 supporting information supporting information Acta Cryst. (2016). E72, 1544-1548 [https://doi.org/10.1107/S2056989016015425] Acta Cryst. (2016). E72, 1544-1548 [https://doi.org/10.1107/S2056989016015425] Crystal structure and Hirshfeld surface analysis of 1-carboxy-2-(3,4-dihydroxy- phenyl)ethan-1-aminium bromide 2-ammonio-3-(3,4-dihydroxyphenyl)propano- ate Perumal Kathiravan, Thangavelu Balakrishnan, Perumal Venkatesan, Kandasamy Ramamurthi, María Judith Percino and Subbiah Thamotharan Perumal Kathiravan, Thangavelu Balakrishnan, Perumal Venkatesan, Kandasamy Ramamurthi, María Judith Percino and Subbiah Thamotharan Computing details Data collection: APEX2 (Bruker, 2004); cell refinement: APEX2 (Bruker, 2004) and SAINT (Bruker, 2004); data reduction: SAINT (Bruker, 2004) and XPREP (Bruker, 2004); program(s) used to solve structure: SIR92 (Altomare et al., 1994); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015); molecular graphics: Mercury (Macrae et al., 2006); software used to prepare material for publication: SHELXL2014 (Sheldrick, 2015) and publCIF (Westrip, 2010). 1-Carboxy-2-(3,4-dihydroxyphenyl)ethan-1-aminium bromide 2-ammonio-3-(3,4-dihydroxyphenyl)propanoate l d sup-1 Acta Cryst. (2016). E72, 1544-1548 Crystal data C9H12NO4+·Br−·C9H11NO4 Mr = 475.29 Monoclinic, I2 a = 6.1456 (3) Å b = 5.6385 (2) Å c = 28.2561 (10) Å β = 94.147 (2)° V = 976.57 (7) Å3 Z = 2 F(000) = 488 Dx = 1.616 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 4553 reflections θ = 2.4–32.1° µ = 2.16 mm−1 T = 293 K Block, colourless 0.30 × 0.25 × 0.25 mm Data collection Bruker Kappa APEXII CCD diffractometer Radiation source: fine-focus sealed tube ω and φ scan Absorption correction: multi-scan (SADABS; Bruker, 2004) Tmin = 0.562, Tmax = 0.619 8138 measured reflections 2827 independent reflections 2421 reflections with I > 2σ(I) Rint = 0.024 θmax = 36.3°, θmin = 2.9° h = −8→8 k = −7→9 l = −37→37 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.026 wR(F2) = 0.056 S = 0.97 2827 reflections 151 parameters 1 restraint Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map Hydrogen site location: mixed sup-1 Acta Cryst. (2016). H atoms treated by a mixture of independent Δρmax = 0.37 e Å−3 Δρmin = −0.31 e Å−3 H atoms treated by a mixture of independent and constrained refinement and constrained refinement Perumal Kathiravan, Thangavelu Balakrishnan, Perumal Venkatesan, Kandasamy Ramamurthi, María Judith Percino and Subbiah Thamotharan E72, 1544-1548 Crystal data C9H12NO4+·Br−·C9H11NO4 Mr = 475.29 Monoclinic, I2 a = 6.1456 (3) Å b = 5.6385 (2) Å c = 28.2561 (10) Å β = 94.147 (2)° V = 976.57 (7) Å3 Z = 2 F(000) = 488 Dx = 1.616 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 4553 reflections θ = 2.4–32.1° µ = 2.16 mm−1 T = 293 K Block, colourless 0.30 × 0.25 × 0.25 mm Data collection Bruker Kappa APEXII CCD diffractometer Radiation source: fine-focus sealed tube ω and φ scan Absorption correction: multi-scan (SADABS; Bruker, 2004) Tmin = 0.562, Tmax = 0.619 8138 measured reflections 2827 independent reflections 2421 reflections with I > 2σ(I) Rint = 0.024 θmax = 36.3°, θmin = 2.9° h = −8→8 k = −7→9 l = −37→37 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.026 wR(F2) = 0.056 S = 0.97 2827 reflections 151 parameters 1 restraint Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map Hydrogen site location: mixed y C9H12NO4+·Br−·C9H11NO4 Mr = 475.29 Monoclinic, I2 a = 6.1456 (3) Å b = 5.6385 (2) Å c = 28.2561 (10) Å β = 94.147 (2)° V = 976.57 (7) Å3 Z = 2 F(000) = 488 Dx = 1.616 Mg m−3 Mo Kα radiation, λ = 0.71073 Å Cell parameters from 4553 reflections θ = 2.4–32.1° µ = 2.16 mm−1 T = 293 K Block, colourless 0.30 × 0.25 × 0.25 mm 1 restraint Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map Hydrogen site location: mixed sup-1 Acta Cryst. (2016). E72, 1544-1548 Acta Cryst. (2016). E72, 1544-1548 supporting information supporting information Δρmax = 0.37 e Å−3 Δρmin = −0.31 e Å−3 Absolute structure: Refined as an inversion twin Absolute structure parameter: 0.023 (8) Δρmax = 0.37 e Å−3 Δρmin = −0.31 e Å−3 Absolute structure: Refined as an inversion twin Absolute structure parameter: 0.023 (8) H atoms treated by a mixture of independent and constrained refinement w = 1/[σ2(Fo2) + (0.0178P)2] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max < 0.001 Δρmax = 0.37 e Å−3 Δρmin = −0.31 e Å−3 Absolute structure: Refined as an inversion twin Absolute structure parameter: 0.023 (8) H atoms treated by a mixture of independent 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. Refined as a 2-component inversion twin. Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso/Ueq Occ. Special details (<1) O1 0.4002 (2) 1.2639 (3) 0.32975 (6) 0.0314 (3) H1O 0.4444 1.3490 0.3519 0.047* O2 0.3013 (2) 0.9618 (3) 0.26190 (5) 0.0296 (3) H2O 0.2781 0.8520 0.2432 0.044* O3 1.4783 (2) 0.5473 (3) 0.40977 (5) 0.0317 (4) O4 1.3241 (2) 0.6326 (4) 0.47670 (4) 0.0279 (3) H4O 1.446 (6) 0.614 (10) 0.4922 (15) 0.021 (11)* 0.5 N1 0.9218 (2) 0.6239 (5) 0.43861 (5) 0.0198 (3) H1A 0.943 (4) 0.765 (6) 0.4564 (11) 0.033 (8)* H1B 0.927 (4) 0.503 (5) 0.4600 (9) 0.020 (6)* H1C 0.788 (4) 0.613 (7) 0.4245 (8) 0.042 (6)* C1 0.8776 (3) 0.8691 (4) 0.34526 (6) 0.0209 (4) C2 0.7366 (3) 1.0550 (4) 0.35295 (7) 0.0227 (4) H2 0.7713 1.1616 0.3775 0.027* C3 0.5451 (3) 1.0840 (3) 0.32468 (6) 0.0198 (4) C4 0.4911 (3) 0.9215 (4) 0.28859 (6) 0.0205 (4) C5 0.6293 (3) 0.7354 (4) 0.28093 (7) 0.0248 (4) H5 0.5935 0.6270 0.2568 0.030* C6 0.8220 (3) 0.7097 (4) 0.30924 (7) 0.0245 (4) H6 0.9148 0.5838 0.3039 0.029* C7 1.0906 (3) 0.8400 (4) 0.37490 (7) 0.0230 (4) H7A 1.1131 0.9771 0.3954 0.028* H7B 1.2091 0.8349 0.3540 0.028* C8 1.0982 (2) 0.6168 (5) 0.40531 (6) 0.0183 (3) H8 1.0753 0.4786 0.3845 0.022* C9 1.3203 (3) 0.5942 (4) 0.43256 (6) 0.0195 (4) Br1 1.0000 0.13069 (5) 0.5000 0.05492 (14) onal atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) Atomic displacement parameters (Å2) U11 U22 U33 U12 U13 U23 O1 0.0287 (8) 0.0279 (8) 0.0363 (9) 0.0055 (7) −0.0067 (7) −0.0073 (7) O2 0.0219 (7) 0.0347 (9) 0.0306 (8) −0.0001 (6) −0.0101 (6) −0.0027 (7) Atomic displacement parameters (Å2) sup-2 supporting information O3 0.0150 (6) 0.0536 (11) 0.0260 (7) 0.0049 (6) −0.0012 (5) −0.0099 (7) O4 0.0154 (6) 0.0493 (8) 0.0180 (6) 0.0013 (9) −0.0051 (5) −0.0025 (10) N1 0.0134 (7) 0.0274 (7) 0.0183 (7) −0.0010 (9) −0.0006 (5) 0.0023 (10) C1 0.0184 (9) 0.0280 (10) 0.0159 (9) −0.0020 (8) −0.0010 (7) 0.0059 (7) C2 0.0241 (10) 0.0240 (9) 0.0193 (9) −0.0038 (8) −0.0029 (7) −0.0005 (7) C3 0.0199 (9) 0.0199 (13) 0.0195 (8) −0.0005 (7) 0.0011 (7) 0.0028 (7) C4 0.0172 (9) 0.0260 (10) 0.0180 (9) −0.0024 (8) −0.0017 (7) 0.0046 (8) C5 0.0265 (10) 0.0280 (10) 0.0193 (9) −0.0016 (9) −0.0018 (8) −0.0040 (8) C6 0.0217 (10) 0.0292 (10) 0.0225 (10) 0.0054 (8) 0.0001 (8) 0.0005 (8) C7 0.0169 (9) 0.0307 (11) 0.0207 (9) −0.0045 (8) −0.0035 (7) 0.0072 (8) C8 0.0128 (7) 0.0256 (9) 0.0163 (7) 0.0001 (9) −0.0011 (6) 0.0009 (10) C9 0.0149 (8) 0.0232 (13) 0.0197 (8) 0.0002 (8) −0.0032 (6) −0.0022 (8) Br1 0.1055 (3) 0.01895 (14) 0.03896 (18) 0.000 −0.00415 (18) 0.000 Geometric parameters (Å, º) O1—C3 1.364 (2) C1—C7 1.511 (3) O1—H1O 0.8200 C2—C3 1.384 (3) O2—C4 1.362 (2) C2—H2 0.9300 O2—H2O 0.8200 C3—C4 1.393 (3) O3—C9 1.232 (2) C4—C5 1.377 (3) O4—C9 1.265 (2) C5—C6 1.388 (3) O4—H4O 0.85 (4) C5—H5 0.9300 N1—C8 1.486 (2) C6—H6 0.9300 N1—H1A 0.95 (3) C7—C8 1.523 (3) N1—H1B 0.91 (3) C7—H7A 0.9700 N1—H1C 0.89 (3) C7—H7B 0.9700 C1—C6 1.382 (3) C8—C9 1.523 (2) C1—C2 1.387 (3) C8—H8 0.9800 C3—O1—H1O 109.5 C4—C5—C6 119.94 (19) C4—O2—H2O 109.5 C4—C5—H5 120.0 C9—O4—H4O 116 (3) C6—C5—H5 120.0 C8—N1—H1A 106.1 (17) C1—C6—C5 120.78 (19) C8—N1—H1B 114.1 (15) C1—C6—H6 119.6 H1A—N1—H1B 106.3 (17) C5—C6—H6 119.6 C8—N1—H1C 114.1 (14) C1—C7—C8 113.12 (16) H1A—N1—H1C 113 (3) C1—C7—H7A 109.0 H1B—N1—H1C 103 (3) C8—C7—H7A 109.0 C6—C1—C2 118.87 (18) C1—C7—H7B 109.0 C6—C1—C7 119.72 (18) C8—C7—H7B 109.0 C2—C1—C7 121.41 (18) H7A—C7—H7B 107.8 C3—C2—C1 120.92 (18) N1—C8—C7 109.9 (2) C3—C2—H2 119.5 N1—C8—C9 110.50 (14) C1—C2—H2 119.5 C7—C8—C9 110.12 (18) O1—C3—C2 124.17 (17) N1—C8—H8 108.8 O1—C3—C4 116.33 (17) C7—C8—H8 108.8 C2—C3—C4 119.50 (17) C9—C8—H8 108.8 O3 0.0150 (6) 0.0536 (11) 0.0260 (7) 0.0049 (6) −0.0012 (5) −0.0099 (7) O4 0.0154 (6) 0.0493 (8) 0.0180 (6) 0.0013 (9) −0.0051 (5) −0.0025 (10) N1 0.0134 (7) 0.0274 (7) 0.0183 (7) −0.0010 (9) −0.0006 (5) 0.0023 (10) C1 0.0184 (9) 0.0280 (10) 0.0159 (9) −0.0020 (8) −0.0010 (7) 0.0059 (7) C2 0.0241 (10) 0.0240 (9) 0.0193 (9) −0.0038 (8) −0.0029 (7) −0.0005 (7) C3 0.0199 (9) 0.0199 (13) 0.0195 (8) −0.0005 (7) 0.0011 (7) 0.0028 (7) C4 0.0172 (9) 0.0260 (10) 0.0180 (9) −0.0024 (8) −0.0017 (7) 0.0046 (8) C5 0.0265 (10) 0.0280 (10) 0.0193 (9) −0.0016 (9) −0.0018 (8) −0.0040 (8) C6 0.0217 (10) 0.0292 (10) 0.0225 (10) 0.0054 (8) 0.0001 (8) 0.0005 (8) C7 0.0169 (9) 0.0307 (11) 0.0207 (9) −0.0045 (8) −0.0035 (7) 0.0072 (8) C8 0.0128 (7) 0.0256 (9) 0.0163 (7) 0.0001 (9) −0.0011 (6) 0.0009 (10) C9 0.0149 (8) 0.0232 (13) 0.0197 (8) 0.0002 (8) −0.0032 (6) −0.0022 (8) Br1 0.1055 (3) 0.01895 (14) 0.03896 (18) 0.000 −0.00415 (18) 0.000 Geometric parameters (Å, º) sup-3 Acta Cryst. Acta Cryst. (2016). E72, 1544-1548 Symmetry codes: (i) x−1, y+1, z; (ii) −x+1/2, y−1/2, −z+1/2; (iii) −x+3, y, −z+1; (iv) x, y+1, z; (v) x−1, y, z. Special details (2016). E72, 1544-1548 supporting information supporting information pp g O2—C4—C5 123.61 (18) O3—C9—O4 126.40 (17) O2—C4—C3 116.40 (17) O3—C9—C8 117.71 (15) C5—C4—C3 119.98 (18) O4—C9—C8 115.85 (15) C6—C1—C2—C3 −1.1 (3) C7—C1—C6—C5 −178.73 (18) C7—C1—C2—C3 178.05 (17) C4—C5—C6—C1 0.1 (3) C1—C2—C3—O1 −179.13 (18) C6—C1—C7—C8 −66.3 (2) C1—C2—C3—C4 1.3 (3) C2—C1—C7—C8 114.5 (2) O1—C3—C4—O2 0.9 (2) C1—C7—C8—N1 −60.3 (2) C2—C3—C4—O2 −179.49 (16) C1—C7—C8—C9 177.70 (16) O1—C3—C4—C5 179.62 (17) N1—C8—C9—O3 168.4 (2) C2—C3—C4—C5 −0.7 (3) C7—C8—C9—O3 −70.0 (3) O2—C4—C5—C6 178.75 (18) N1—C8—C9—O4 −13.5 (3) C3—C4—C5—C6 0.1 (3) C7—C8—C9—O4 108.1 (2) C2—C1—C6—C5 0.4 (3) Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A O1—H1O···O3i 0.82 1.98 2.782 (2) 166 O2—H2O···O1ii 0.82 2.32 3.004 (2) 142 O2—H2O···O2ii 0.82 2.26 2.9557 (8) 144 O4—H4O···O4iii 0.85 (4) 1.61 (4) 2.449 (2) 169 (6) N1—H1A···Br1iv 0.95 (3) 2.41 (3) 3.359 (3) 179 (3) N1—H1B···Br1 0.91 (3) 2.41 (3) 3.295 (3) 164 (2) N1—H1C···O3v 0.89 (3) 1.95 (3) 2.821 (2) 164 (3) Symmetry codes: (i) x−1, y+1, z; (ii) −x+1/2, y−1/2, −z+1/2; (iii) −x+3, y, −z+1; (iv) x, y+1, z; (v) x−1, y, z. O2—C4—C5 123.61 (18) O3—C9—O4 126.40 (17) O2—C4—C3 116.40 (17) O3—C9—C8 117.71 (15) C5—C4—C3 119.98 (18) O4—C9—C8 115.85 (15) C6—C1—C2—C3 −1.1 (3) C7—C1—C6—C5 −178.73 (18) C7—C1—C2—C3 178.05 (17) C4—C5—C6—C1 0.1 (3) C1—C2—C3—O1 −179.13 (18) C6—C1—C7—C8 −66.3 (2) C1—C2—C3—C4 1.3 (3) C2—C1—C7—C8 114.5 (2) O1—C3—C4—O2 0.9 (2) C1—C7—C8—N1 −60.3 (2) C2—C3—C4—O2 −179.49 (16) C1—C7—C8—C9 177.70 (16) O1—C3—C4—C5 179.62 (17) N1—C8—C9—O3 168.4 (2) C2—C3—C4—C5 −0.7 (3) C7—C8—C9—O3 −70.0 (3) O2—C4—C5—C6 178.75 (18) N1—C8—C9—O4 −13.5 (3) C3—C4—C5—C6 0.1 (3) C7—C8—C9—O4 108.1 (2) C2—C1—C6—C5 0.4 (3) Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A O1—H1O···O3i 0.82 1.98 2.782 (2) 166 O2—H2O···O1ii 0.82 2.32 3.004 (2) 142 O2—H2O···O2ii 0.82 2.26 2.9557 (8) 144 O4—H4O···O4iii 0.85 (4) 1.61 (4) 2.449 (2) 169 (6) N1—H1A···Br1iv 0.95 (3) 2.41 (3) 3.359 (3) 179 (3) N1—H1B···Br1 0.91 (3) 2.41 (3) 3.295 (3) 164 (2) N1—H1C···O3v 0.89 (3) 1.95 (3) 2.821 (2) 164 (3) Symmetry codes: (i) x−1 y+1 z; (ii) −x+1/2 y−1/2 −z+1/2; (iii) −x+3 y −z+1; (iv) x y+1 z; (v) x−1 y z Symmetry codes: (i) x−1, y+1, z; (ii) −x+1/2, y−1/2, −z+1/2; (iii) −x+3, y, −z+1; (iv) x, y+1, z; (v) x−1, y, z. Acta Cryst. (2016). E72, 1544-1548 sup-4 sup-4
https://openalex.org/W4226151367	https://cris.unibo.it/bitstream/11585/883905/1/62-GRL_2022%20_NikkhooRivalta.pdf	English	null	Analytical solutions for gravity changes caused by triaxial volumetric sources	null	2,022	cc-by	9,316	© 2022. The Authors. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Citation: Nikkhoo, M., & Rivalta, E. (2022). Analytical solutions for gravity changes caused by triaxial volumetric sources. Geophysical Research Letters, 49, e2021GL095442. https://doi. org/10.1029/2021GL095442 Author Contributions: Conceptualization: Mehdi Nikkhoo, Eleonora Rivalta Formal analysis: Mehdi Nikkhoo Funding acquisition: Eleonora Rivalta Investigation: Mehdi Nikkhoo, Eleonora Rivalta Methodology: Mehdi Nikkhoo, Eleonora Rivalta Project Administration: Mehdi Nikkhoo, Eleonora Rivalta Resources: Mehdi Nikkhoo, Eleonora Rivalta Software: Mehdi Nikkhoo Supervision: Eleonora Rivalta Validation: Mehdi Nikkhoo Visualization: Mehdi Nikkhoo Writing – original draft: Mehdi Nikkhoo, Eleonora Rivalta Writing – review & editing: Mehdi Nikkhoo, Eleonora Rivalta Key Points: Mehdi Nikkhoo1 and Eleonora Rivalta1,2 • We develop analytical solutions for gravity changes due to the point Compound Dislocation Model simulating triaxial expansions 1GFZ German Research Centre for Geosciences, Potsdam, Germany, 2Department of Physics and Astronomy, Alma Mater Studiorum University of Bologna, Bologna, Italy • Rapid coupled inversions of deformation and gravity changes, accounting for deformation-induced gravity changes are now possible Abstract Volcanic crises are often associated with magmatic intrusions or the pressurization of magma chambers of various shapes. These volumetric sources deform the country rocks, changing their density, and cause surface uplift. Both the net mass of intruding magmatic fluids and these deformation effects contribute to surface gravity changes. Thus, to estimate the intrusion mass from gravity changes, the deformation effects must be accounted for. We develop analytical solutions and computer codes for the gravity changes caused by triaxial sources of expansion. This establishes coupled solutions for joint inversions of deformation and gravity changes. Such inversions can constrain both the intrusion mass and the deformation source parameters more accurately. • For shallow sources, estimation errors in the chamber volume change may lead to large biases in the simulated gravity changes Correspondence to: M. Nikkhoo and E. Rivalta, mehdi.nikkhoo@gfz-potsdam.de; eleonora.rivalta@unibo.it M. Nikkhoo and E. Rivalta, mehdi.nikkhoo@gfz-potsdam.de; eleonora.rivalta@unibo.it Plain Language Summary Volcanic crises are usually associated with magmatic fluids that intrude and deform the host rocks before potentially breaching the Earth's surface. It is important to estimate how much fluid (mass and volume) is on the move. Volume can be determined from the measured surface uplift. Mass can be determined from surface gravity changes. The fluid intrusion increases the mass below the volcano, thereby increasing the gravity and pressurizing the rocks. This dilates parts of the host rock and compresses other parts, changing the rock density and redistributing the rock mass. This causes secondary gravity changes, called deformation-induced gravity changes. The measured gravity change is always the sum of the mass and deformation-induced contributions. Here, we develop mathematical equations for the rapid estimation of these deformation-induced gravity changes caused by arbitrary intrusion shapes. This way we can take the mass contribution apart from the deformation contribution. We show that by using this solution not only the intrusion mass, but also other intrusion parameters, including the volume, depth, and shape can be calculated more accurately. Analytical Solutions for Gravity Changes Caused by Triaxial Volumetric Sources Analytical Solutions for Gravity Changes Caused by Triaxial Volumetric Sources 10.1029/2021GL095442 RESEARCH LETTER Key Points: 1. Introduction Intrusion of magma through the host rock or into an existing magma chamber deforms the Earth's crust and also changes the surface gravity field. The intrusion mass is a key parameter for characterizing the nature of the activity and its future evolution. Joint analyses of the measured surface displacements and gravity changes can constrain the intrusion mass, beside the other parameters of the deformation source, that is, its location, shape, spatial orientation, and some strength parameter (pressure or volume change; Battaglia et al., 1999, 2003; Okubo et al., 1991). Both the mass transport and the ensuing country-rock deformations contribute to the gravity changes (Bonafede & Mazzanti, 1998; Hagiwara, 1977; Lisowski, 2007; Walsh & Rice, 1979). Such deformation-induced effects may be substantial for nonspherical sources, as shown through numerical models based on the finite element method (FEM; see Currenti, 2014; Currenti et al., 2007, 2008; Trasatti & Bonafede, 2008). The deformation effects caused by tabular sources such as dikes and sills can be estimated through the Okubo (1992) analyt- ical solutions. There are no analytical solutions for other source geometries, such as ellipsoids, yet rigorous joint inversions of surface displacements and gravity changes demand models accounting for the source shape (Amoruso et al., 2008). © 2022. The Authors. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. A source model composed of three orthogonal tensile dislocations can simulate the deformation field associated with triaxial sources (Amoruso & Crescentini, 2013; Bonafede & Ferrari, 2009; Lisowski et al., 2008). Based on this concept, Nikkhoo et al. (2017) developed the point Compound Dislocation Model (point CDM), which represents the far-field deformation of generic triaxial sources. This source model spans a wider parameter space than ellipsoids (Ferrari et al., 2015) while retaining the simplicity of the Mogi (1958) model. 1 of 12 1 of 12 NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA Geophysical Research Letters 10.1029/2021GL095442 Figure 1. Schematic mass redistribution and surface uplift caused by chamber pressurization. Compressed magma (red) is surrounded by the interface cavity. The dashed ellipse depicts chamber walls prior to pressurization and separates the δVm and δVc portions of the interface cavity (see Equation 1). The country rocks are subjected to positive dilatation/density decrease (light gray and white contours) and negative dilatation/density increase (dark gray and black contours). Thick black contour marks zero dilatation. The gravity station (black triangle) has been subjected to gravity change δg and vertical displacement uv Fi 1 S h ti di t ib ti d f lift d b h b i ti C d ( d) i Figure 1. Schematic mass redistribution and surface uplift caused by chamber pressurization. Compressed magma (red) is surrounded by the interface cavity. The dashed ellipse depicts chamber walls prior to pressurization and separates the δVm and δVc portions of the interface cavity (see Equation 1). The country rocks are subjected to positive dilatation/density decrease (light gray and white contours) and negative dilatation/density increase (dark gray and black contours). Thick black contour marks zero dilatation. The gravity station (black triangle) has been subjected to gravity change δg and vertical displacement uv. In this study, we use the Okubo (1991) expressions to derive analytical solutions for the gravity changes asso- ciated with the point CDM. We show how gravity changes due to point and finite ellipsoidal sources can be calculated by using the point CDM. We compare the point CDM gravity changes with the Hagiwara (1977) and Trasatti and Bonafede (2008) solutions. Finally, we elaborate on the potential of the model for coupled inversions of surface displacements and gravity changes. 2. Methods Deformation-induced gravity changes are usually expressed as the sum of contributions due to deformation in the source region and country rocks and the surface uplift. Here, we adopt a decomposition scheme compatible with the point CDM formulation. We assume a homogeneous, isotropic elastic half-space as a model for the Earth's crust. We denote the Poisson's ratio, shear modulus, and bulk modulus in the medium by ν, μ, and K, respectively. We adopt a right-handed xyz Cartesian coordinate system with the origin at the free surface and the z axis pointing upward. By “gravity change” we refer to the change in the absolute value of the gravity vector's z component. Thus, a positive mass change (mass increase) below a gravimeter leads to a positive gravity change (gravity increase). 2.1. Gravity Changes Caused by Magma Chamber Pressurization As an example, suppose that magma degassing pressurizes a magma chamber (Figure 1). We assume that the exsolved gases all gather at the interface between the chamber walls and the degassed magma, forming a 2 of 12 NIKKHOO AND RIVALTA Geophysical Research Letters 10.1029/2021GL095442 shell-shaped cavity. The outward expansion of the chamber walls and inward compression of the magma lead to the oppositely signed chamber volume change, δVc, and magma volume change, δVm, respectively. The total volume created by the expansion-compression process—namely the interface volume change, ΔVint—is given by shell-shaped cavity. The outward expansion of the chamber walls and inward compression of the magma lead to the oppositely signed chamber volume change, δVc, and magma volume change, δVm, respectively. The total volume created by the expansion-compression process—namely the interface volume change, ΔVint—is given by (1) Δ𝑉𝑉int = 𝛿𝛿𝛿𝛿c −𝛿𝛿𝛿𝛿m, (1) or equivalently by or equivalently by Δ𝑉𝑉int = 𝑉𝑉c −𝑉𝑉m, (2) (2) Δ𝑉𝑉int = 𝑉𝑉c −𝑉𝑉m, where Vc = V + δVc and Vm = V + δVm are the chamber volume and magma volume in the deformed state, respectively, and V represents both the chamber volume and magma volume in the undeformed state. The cham- ber expansion also uplifts the surface and generates a strain field, ϵij, in the surrounding rocks. This changes the density of the rocks by δρr = −ρrϵkk, where ρr is the rock density in the undeformed state and ϵkk = ϵ11 + ϵ22 + ϵ33 is the volumetric strain or dilatation—a positive dilatation reduces the density (see Figure 1). Similarly, the magma density change, δρm, due to the compression is related to the magma compressibility, βm, through δρm = ρmβmδp, where ρm is the magma density in the undeformed state and δp is the pressure change in the chamber (Rivalta & Segall, 2008). Provided that βm and δp are known, we have 𝛿𝛿𝛿𝛿m = 𝑉𝑉𝑉𝑉m𝛿𝛿𝛿𝛿𝛿 (3) 𝛿𝛿𝛿𝛿m = 𝑉𝑉𝑉𝑉m𝛿𝛿𝛿𝛿𝛿 (3) Since we consider the created volume ΔVint as void, the density change in the δVc and δVm portions is −ρr and −ρm, respectively. Similarly, uplift, or subsidence, at the Earth's surface will either fill void space or create a void space. So, the other zone of substantial density change is the Earth's surface, where areas of uplift and subsidence are subjected to density changes +ρr and −ρr, respectively. 2.1. Gravity Changes Caused by Magma Chamber Pressurization The same deformation-induced density changes exist if instead of exsolved gases, the interface cavity is formed by, and filled with, the intrusion of some external fluids. In such case, the interface cavity is filled with a net mass Δ𝑀𝑀= 𝜌𝜌intΔ𝑉𝑉int, (4) Δ𝑀𝑀= 𝜌𝜌intΔ𝑉𝑉int, (4) where ρint is the intrusion density. 2.1.1. The Far Field Approximations The far field gravity changes caused by the intruded fluid mass can be calculated through a point-mass approx- imation as Δ𝑔𝑔Δ𝑀𝑀= 𝐺𝐺Δ𝑀𝑀𝑑𝑑 𝑟𝑟3 , (8) (8) where d is the depth to the center of the chamber and r is the distance between the center of the chamber and the surface observation point. This approximation can also be applied to the far field gravity changes caused by the other density changes in the chamber as Δ푔훽= 퐺훿휌m푉m 푑 푟3 , Δ푔훿푉m = 퐺휌m훿푉m 푑 푟3 , Δ푔훿푉c = −퐺휌r훿푉c 푑 푟3 , Δ푔Δ푉int = −퐺휌rΔ푉int 푑 푟3 . (9) (9) The conservation of the initial magma mass in the chamber implies δρmVm = −ρmδVm, which together with Equa- tion 9 yields The conservation of the initial magma mass in the chamber implies δρmVm = −ρmδVm, which together with Equa- tion 9 yields Δ𝑔𝑔𝛽𝛽+ Δ𝑔𝑔𝛿𝛿𝛿𝛿m = 0. (10) (10) Note that for shallow finite sources Equation 10 does not necessarily hold, as mass redistribution within the chamber may lead to measurable gravity changes. The far field form of Equation 5 can now be written as Note that for shallow finite sources Equation 10 does not necessarily hold, as mass redistribution within the chamber may lead to measurable gravity changes. The far field form of Equation 5 can now be written as 𝛿𝛿𝛿𝛿= Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘+ Δ𝑔𝑔SM + Δ𝑔𝑔FA + Δ𝑔𝑔Δ𝑀𝑀, (11) (11) 𝛿𝛿𝛿𝛿= Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘+ Δ𝑔𝑔SM + Δ𝑔𝑔FA + Δ𝑔𝑔Δ𝑀𝑀, which expresses the surface gravity changes associated with a deep pressurized chamber as the sum of contribu- tions due to displaced mass at the chamber walls 𝐴𝐴 ( Δ𝑔𝑔𝛿𝛿𝛿𝛿c ) , volumetric strain in the host rocks 𝐴𝐴 ( Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 ) , displaced mass at the Earth's surface (ΔgSM), and the vertical displacement of gravity stations (ΔgFA), superimposed on the mass change contribution (ΔgΔM). which expresses the surface gravity changes associated with a deep pressurized chamber as the sum of contribu- tions due to displaced mass at the chamber walls 𝐴𝐴 ( Δ𝑔𝑔𝛿𝛿𝛿𝛿c ) , volumetric strain in the host rocks 𝐴𝐴 ( Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 ) , displaced mass at the Earth's surface (ΔgSM), and the vertical displacement of gravity stations (ΔgFA), superimposed on the mass change contribution (ΔgΔM). Note that Equations 1–11 hold for any chamber shape and boundary conditions on the chamber walls. where ρint is the intrusion density. The other terms in Equation 5 can be estimated only by using a deformation model for the NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA Geophysical Research Letters 10.1029/2021GL095442 chamber pressurization. Note that Equation 5 is valid for sources both in the near field and the far field. In the following, we first introduce an analytical point-source model, which can be applied to sources in the far field, and show that in this case Equation 5 can be simplified. Next, we present a semi-analytical finite-source solution and elaborate on the issues that may limit its applicability to near-field problems. chamber pressurization. Note that Equation 5 is valid for sources both in the near field and the far field. In the following, we first introduce an analytical point-source model, which can be applied to sources in the far field, and show that in this case Equation 5 can be simplified. Next, we present a semi-analytical finite-source solution and elaborate on the issues that may limit its applicability to near-field problems. 1.1. The Far Field Approximations where ρint is the intrusion density. The magma chamber expansion leads to a vertical displacement, uv, and the following gravity change contribu- tions for each observation point at the surface: 1. Δgβ, due to density change δρm in the magma volume in the deformed state, Vm, 1. Δgβ, due to density change δρm in the magma volume in the deformed state, Vm, 2. Δ𝑔𝑔𝛿𝛿𝛿𝛿m , due to density change −ρm within the δVm volume, 2. Δ𝑔𝑔𝛿𝛿𝛿𝛿m , due to density change −ρm within the δVm volume, 3. Δ𝑔𝑔𝛿𝛿𝛿𝛿c , due to density change −ρr within the δVc volume, 3. Δ𝑔𝑔𝛿𝛿𝛿𝛿c , due to density change −ρr within the δVc volume, 4. Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 , due to density changes δρr throughout the country rocks, 4. Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 , due to density changes δρr throughout the country rocks, 5. ΔgSM, due to the presence of the displaced surface mass layer with density +ρr, 6. ΔgFA, due to the free air change in gravity associated with uv, 6. ΔgFA, due to the free air change in gravity associated with uv, 6. ΔgFA, due to the free air change in gravity associated with uv 7. ΔgΔM, due to the added intrusion mass ΔM that leads to density change ρint within the interface cavity for a total surface gravity change of (5) 𝛿𝛿𝛿𝛿= Δ𝑔𝑔𝛽𝛽+ Δ𝑔𝑔𝛿𝛿𝛿𝛿m + Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘+ Δ𝑔𝑔SM + Δ𝑔𝑔FA + Δ𝑔𝑔Δ𝑀𝑀. (5) ΔgΔM, also known as residual gravity, can be used to constrain ΔM (see Battaglia et al., 2008). However, this requires all the other terms in Equation 5 to be quantified first. At each station, δg and uv can be determined through repeated gravity and deformation measurements, respectively. Then we have Δ𝑔𝑔FA = 𝛾𝛾𝛾𝛾𝑣𝑣, (6) Δ𝑔𝑔FA = 𝛾𝛾𝛾𝛾𝑣𝑣, (6) where γ ≃ −0.3086 mGal/m is the free air gradient, and where γ ≃ −0.3086 mGal/m is the free air gradient, and where γ ≃ −0.3086 mGal/m is the free air gradient, and Δ𝑔𝑔SM = 2𝜋𝜋𝜋𝜋𝜋𝜋r𝑢𝑢𝑣𝑣, (7) (7) where G is the gravitational constant. Note that Equation 7 uses the Bouguer plate approximation and is valid for flat topographies. The other terms in Equation 5 can be estimated only by using a deformation model for the where G is the gravitational constant. Note that Equation 7 uses the Bouguer plate approximation and is valid for flat topographies. 2.2. Gravity Changes Caused by the Point CDM The point CDM represents the far field of triaxial sources of expansion with arbitrary spatial orientations (Nikkhoo et al., 2017). The point CDM is composed of three mutually orthogonal point tensile dislocations (see Figure 2a) constrained to either expand or contract together. The strength of each point tensile dislocation is determined by its potency, defined as the product of dislocation surface area and opening (Aki & Richards, 2002; Nikkhoo et al., 2017, see also Appendix A). The point CDM has 10 parameters: three location coordinates, three rotation angles, three potencies specifying the expansion intensity along the three principal axes of the source, and Poisson's ratio, ν. The total potency of the point CDM, denoted by ΔV, is the sum of the three potencies. ΔV has the units of volume but it is not a physical quantity. Rather, it is a measure of the source strength and it holds ΔV = ΔVint, provided that Km = K, where Km = 1/βm is the bulk modulus of magma. Triaxial sources of differing shapes, but identical far field deformation, have the same point CDM representation and thus, the same ΔV. However, in order to have the same δVc, these sources must also have identical shapes (except for ν = 0.5, which leads to ΔV = δVc). For example, the uniformly pressurized cuboidal and ellipsoidal chambers in Figure 2 have the same potencies but their volume changes are different. Analytical expressions NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA 4 of 12 Geophysical Research Letters 10.1029/2021GL095442 Figure 2. Triaxial volumetric sources. (a) A point Compound Dislocation Model (point CDM) with potencies ΔVx (yellow ΔVy (green), and ΔVz (blue), where ΔVx = ΔVy > ΔVz. Inset shows the equivalent CDM (see Nikkhoo et al., 2017). (b) A uniformly pressurized cuboidal source with Km = K. The two interface cavity portions 𝐴𝐴 𝐴𝐴𝐴𝐴Cub c and 𝐴𝐴 𝐴𝐴𝐴𝐴Cub m are indicated, where Δ𝑉𝑉Cub = 𝛿𝛿𝛿𝛿Cub c + 𝛿𝛿𝛿𝛿Cub m . (c) Same as (b), but for a uniformly pressurized ellipsoidal source. The interface cavity portions a 𝐴𝐴𝐴𝐴Ell c and 𝐴𝐴 𝐴𝐴𝐴𝐴Ell m , with 𝐴𝐴 Δ𝑉𝑉Ell = 𝛿𝛿𝛿𝛿Ell c + 𝛿𝛿𝛿𝛿Ell m . Note that 𝐴𝐴 𝐴𝐴𝐴𝐴Cub c ≠𝛿𝛿𝛿𝛿Ell c and 𝐴𝐴 𝐴𝐴𝐴𝐴Cub m ≠𝛿𝛿𝛿𝛿Ell m but ΔV Cub = ΔV Ell. (d) A set of N point CDMs uniformly distributed within the ellipsoidal cavity in (c). 2.2. Gravity Changes Caused by the Point CDM The point CDM in (a) represents the far field of all th finite sources in (b–d). Provided N → ∞, the near fields of (c and d) are equivalent. For the models in (b and c) ν = 0.25. Figure 2. Triaxial volumetric sources. (a) A point Compound Dislocation Model (point CDM) with potencies ΔVx (yellow), ΔVy (green), and ΔVz (blue), where ΔVx = ΔVy > ΔVz. Inset shows the equivalent CDM (see Nikkhoo et al., 2017). (b) A uniformly pressurized cuboidal source with Km = K. The two interface cavity portions 𝐴𝐴 𝐴𝐴𝐴𝐴Cub c and 𝐴𝐴 𝐴𝐴𝐴𝐴Cub m are indicated, where Δ𝑉𝑉Cub = 𝛿𝛿𝛿𝛿Cub c + 𝛿𝛿𝛿𝛿Cub m . (c) Same as (b), but for a uniformly pressurized ellipsoidal source. The interface cavity portions are 𝐴𝐴𝐴𝐴Ell c and 𝐴𝐴 𝐴𝐴𝐴𝐴Ell m , with 𝐴𝐴 Δ𝑉𝑉Ell = 𝛿𝛿𝛿𝛿Ell c + 𝛿𝛿𝛿𝛿Ell m . Note that 𝐴𝐴 𝐴𝐴𝐴𝐴Cub c ≠𝛿𝛿𝛿𝛿Ell c and 𝐴𝐴 𝐴𝐴𝐴𝐴Cub m ≠𝛿𝛿𝛿𝛿Ell m but ΔV Cub = ΔV Ell. (d) A set of N point CDMs uniformly distributed within the ellipsoidal cavity in (c). The point CDM in (a) represents the far field of all the finite sources in (b–d). Provided N → ∞, the near fields of (c and d) are equivalent. For the models in (b and c) ν = 0.25. relating ΔV to δVc are available for ellipsoidal sources from Eshelby (1957). For uniformly pressurized ellipsoids, we have (Nikkhoo et al., 2017): Δ𝑉𝑉Ell = 𝛿𝛿𝛿𝛿Ell c + 𝑉𝑉𝑉𝑉𝑉𝑉 𝐾𝐾. (12) (12) Recalling that 𝐴𝐴 𝐴𝐴= 2𝜇𝜇(1+𝜈𝜈) 3(1−2𝜈𝜈) and that for a spherical source of radius a the total volume and volume change are 𝐴𝐴Sph = 4 3𝜋𝜋𝜋𝜋3 and 𝐴𝐴 𝐴𝐴𝐴𝐴Sph c = 𝜋𝜋 𝜇𝜇𝑎𝑎3𝛿𝛿𝛿𝛿 , respectively, Equation 12 becomes Δ𝑉𝑉Sph = 3 (1 −𝜈𝜈) (1 + 𝜈𝜈) 𝛿𝛿𝛿𝛿Sph c , (13) (13) which for ν = 0.25 leads to 𝐴𝐴 Δ𝑉𝑉Sph = 1.8𝛿𝛿𝛿𝛿Sph c (see also Aki & Richards, 2002; Bonafede & Ferrari, 2009; Ichi- hara et al., 2016). which for ν = 0.25 leads to 𝐴𝐴 Δ𝑉𝑉Sph = 1.8𝛿𝛿𝛿𝛿Sph c (see also Aki & Richards, 2002; Bonafede & Ferrari, 2009; Ichi- hara et al., 2016). NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA 5 of 12 Geophysical Research Letters 10.1029/2021GL095442 Gravity changes caused by point tensile dislocations can be calculated through the Okubo (1991) analytical expressions (Appendix A). By superimposing the gravity changes associated with three mutually orthogonal point dislocations (Equation A1), we derive the analytical gravity changes associated with the point CDM as 𝛿𝛿𝛿𝛿= Δ𝑔𝑔Δ𝑉𝑉+ Δ𝑔𝑔MD + Δ𝑔𝑔SM + Δ𝑔𝑔FA + Δ𝑔𝑔Δ𝑀𝑀, (14) (14) where ΔgΔV is the interface cavity contribution (white space in Figures 2b and 2c) and ΔgMD is the contribution due to the medium dilatation both inside and outside the source (gray space in Figures 2b and 2c). Noting that Δ𝑔𝑔Δ𝑉𝑉= Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝛿𝛿𝛿𝛿m and 𝐴𝐴 Δ𝑔𝑔MD = Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘+ Δ𝑔𝑔𝛽𝛽 and using Equation 10 we have where ΔgΔV is the interface cavity contribution (white space in Figures 2b and 2c) and ΔgMD is the contribution due to the medium dilatation both inside and outside the source (gray space in Figures 2b and 2c). Noting that Δ𝑔𝑔Δ𝑉𝑉= Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝛿𝛿𝛿𝛿m and 𝐴𝐴 Δ𝑔𝑔MD = Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘+ Δ𝑔𝑔𝛽𝛽 and using Equation 10 we have Δ𝑔𝑔Δ𝑉𝑉+ Δ𝑔𝑔MD = Δ𝑔𝑔𝛿𝛿𝛿𝛿c + Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘, (15) (15) from which it follows that the δg from Equation 14 and the δg from Equation 11 are equivalent. Therefore, the point CDM can be used to compute the effects of deformation on gravity change and thus estimate the mass change ΔM. 2.2.1. Gravity Changes Caused by Point and Finite Pressurized Ellipsoidal Cavities For any point ellipsoidal model after Davis (1986), there is an equivalent point CDM related to the elastic param- eters of the medium and the ellipsoid semi-axes and pressure change through the Eshelby (1957) tensor (see Nikkhoo et al., 2017). Thus, Equation 14 also holds for point ellipsoidal sources. By calculating δVc for ellipsoi- dal cavities 𝐴𝐴 Δ𝑔𝑔𝛿𝛿𝛿𝛿c (Equation 9) and thus, 𝐴𝐴 Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 (Equation 15) can be determined for ellipsoidal sources. Assume that a point CDM with potencies 𝐴𝐴 (Δ𝑉𝑉𝑎𝑎, Δ𝑉𝑉𝑏𝑏, Δ𝑉𝑉𝑐𝑐) represents the far field of a pressurized ellipsoidal cavity with semi-axes 𝐴𝐴 (𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎) . Then, a set of N point CDMs with potencies 𝐴𝐴 (Δ𝑉𝑉𝑎𝑎∕𝑁𝑁𝑁Δ𝑉𝑉𝑏𝑏∕𝑁𝑁𝑁Δ𝑉𝑉𝑐𝑐∕𝑁𝑁) , uniformly distributed within the ellipsoid (see Figure 2d), approximates the near field deformations of the pressurized cavity (Amoruso & Crescentini, 2011; Amoruso et al., 2008; Davis, 1986; Eshelby, 1957; Segall, 2010; Yang et al., 1988). Provided that N → ∞, this procedure leads to an accurate solution, unless the cavity is immediately below the free surface (Amoruso & Crescentini, 2011; Segall, 2010; Yang et al., 1988). Similar accuracies can be achieved by using the finite Ellipsoidal Cavity Model (finite ECM) after Nikkhoo and Rivalta (2022), which uses a smaller number of point sources with depth-dependent spacing and strengths. By incorporating the expres- sions for the point CDM gravity changes in these configurations, we derive new solutions for the gravity changes caused by a finite pressurized ellipsoidal cavity. While the finite ECM is more accurate than the point CDM in modeling shallow pressurized ellipsoidal cavities, it is still an approximate solution for both deformation and gravity change calculations. Similar to the Yang et al. (1988) solution, the finite ECM provides excellent accu- racies in the limit that the source dimensions are small compared to its depth (see Nikkhoo & Rivalta, 2022, for further details). 3.1. Comparisons With Other Gravity Change Solutions 3.1. Comparisons With Other Gravity Change Solutions Hagiwara (1977) derived closed-form expressions for the gravity change contributions caused by the Mogi (1958) source, later used to validate analytical (Okubo, 1991) and numerical solutions (Currenti et al., 2007, 2008; Tras- atti & Bonafede, 2008). An isotropic point CDM is equivalent to the Mogi (1958) model (Bonafede & Ferrari, 2009). Assuming potency ΔV Sph and depth d for such a point CDM, Equation A1 yields: Δ푔Sph MD = 1 3퐺휌r (1 −2휈) Δ푉Sph 푑 푟3 , Δ푔Sph SM = 2 3퐺휌r (1 + 휈) Δ푉Sph 푑 푟3 , Δ푔Sph Δ푉= −퐺휌rΔ푉Sph 푑 푟3 . (16) (16) By using Equations 9, 13, and 15, we rewrite Equation 16 in terms of 𝐴𝐴 𝐴𝐴𝐴𝐴Sph c : By using Equations 9, 13, and 15, we rewrite Equation 16 in terms of 𝐴𝐴 𝐴𝐴𝐴𝐴Sph c : By using Equations 9, 13, and 15, we rewrite Equation 16 in terms of 𝐴𝐴 𝐴𝐴𝐴𝐴Sph c : NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA 6 of 12 Geophysical Research Letters 10.1029/2021GL095442 Figure 3. Comparing the finite Ellipsoidal Cavity Model (finite ECM) with the Trasatti and Bonafede (2008) finite element method solution for a vertical prolate spheroidal cavity with semi-major axes 1.842 km, aspect ratio 0.4, and depth to the center 5 km. (a) Radial (ur) and vertical (uv) displacements, normalized by the maximum vertical displacement of the finite ECM solution. (b) Gravity change contributions, normalized by the maximum ΔgSM of the finite ECM solution. Δ푔Sph 휖푘푘= −퐺휌r (1 −2휈) 훿푉Sph c 푑 푟3 , Δ푔Sph SM = 2퐺휌r (1 −휈) 훿푉Sph c 푑 푟3 , Δ푔Sph 훿푉c = −퐺휌r훿푉Sph c 푑 푟3 , (17) (17) which are equivalent to the Hagiwara (1977) expressions (see also Okubo, 1991; Rundle, 1978; Savage, 1984; Walsh & Rice, 1979). This validates the gravity change solution for the point CDM in the case of point spherical cavities. As proved by Walsh and Rice (1979), the sum of the three terms in each set of Equations 16 and 17 vanishes. Note also that, for any point CDM, if ν = 0.5 then 𝐴𝐴 Δ𝑔𝑔MD = Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘= 0 . We now show that the gravity change solutions for the point CDM also provide a basis for rigorous benchmarking of numerical solutions. 3.1. Comparisons With Other Gravity Change Solutions We use the point CDM and the finite ECM to calculate the surface displacements (Figure 3a) and gravity changes (Figure 3b) associated with the Trasatti and Bonafede (2008) FEM solution for a pressurized vertical prolate spheroidal cavity. In the far field, the point CDM and the finite ECM displacements are indistinguishable. The FEM solution shows a small deviation that can be attributed to the finite domain of the model. In the near field, the finite ECM and FEM displacements show a very good agreement. The maximum ∼9% difference between the finite ECM and point CDM reflects the difference between a point-source and a finite-source solution. There is also a good agreement between the gravity changes from all approaches (Figure 3b). The maximum differences between 𝐴𝐴 Δ𝑔𝑔𝛿𝛿𝛿𝛿c , 𝐴𝐴 Δ𝑔𝑔𝜖𝜖𝑘𝑘𝑘𝑘 , ΔgSM, and Δg from the finite ECM and point CDM are ∼6%, ∼9%, ∼9%, and ∼6%, respectively. Since the cavity in this example is relatively deep, the finite ECM calculations are very accurate. Thus, in this particular case, the subtle differences between the finite ECM and the FEM gravity change contributions mostly reflect the errors in the FEM vertical displacements and cavity volume change. The largest difference between the Trasatti and Bonafede (2008) and the other solutions is slightly above 1 μGal, which is more than double the error that Trasatti and Bonafede (2008) estimated by comparison with Hagiwara (1977). This suggests that comparing numerical models with the solution for spherical cavities alone may lead to underesti- mated errors for the numerical models. Figure 3. Comparing the finite Ellipsoidal Cavity Model (finite ECM) with the Trasatti and Bonafede (2008) finite element method solution for a vertical prolate spheroidal cavity with semi-major axes 1.842 km, aspect ratio 0.4, and depth to the center 5 km. (a) Radial (ur) and vertical (uv) displacements, normalized by the maximum vertical displacement of the finite ECM solution. (b) Gravity change contributions, normalized by the maximum ΔgSM of the finite ECM solution. 3.2. Implications for the Retrieval of Deformation Source Parameters 3.2. Implications for the Retrieval of Deformation Source Parameters Dieterich and Decker (1975) showed that different source shapes produce almost indistinguishable uplift patterns if the source depths are appropriately adjusted. However, the associated horizontal displacements will be completely different. The implication is that in order to constrain all source parameters reliably, horizontal and vertical displacement data must be inverted together. Similar to horizontal and vertical surface displacements, the deformation-induced gravity changes depend on the deformation source parameters. Thus, gravity changes can potentially help better constrain them (Trasatti & Bonafede, 2008). We use the point CDM to simulate the radial and vertical displacements and the gravity changes associated with three different axially symmetric deformation sources: a horizontal sill, an isotropic source, and a prolate source (see Figure 4). For all sources, ΔM = 0. The source depths in Figure 4a lead to similar vertical displacements (Figure 4c) but distinct horizontal displacements (Figure 4d) and distinct gravity changes (free air contribution removed; Figure 4b). Adjusting the source depths differently (Figure 4e) such that the horizontal displacements match (Figure 4h), leads to distinct vertical displacements (Figure 4f) and distinct gravity changes (Figure 4g). This implies that, from a theoretical perspective, gravity changes may also help to better constrain the deformation NIKKHOO AND RIVALTA 7 of 12 Geophysical Research Letters 10.1029/2021GL095442 gure 4. Gravity changes (Δg = δg − ΔgFA), vertical displacements (uv), and radial displacements (ur) for point sou different aspect ratios and depths. Top block: The sources illustrated in (a) give rise to different Δg (b), similar uv ( nd different ur (d). Bottom block: The sources in (e) cause different Δg (f), similar ur (h), and different uv (g). The po ctors of the point spherical source, point prolate source, and point sill in both (a and e) may be any positive multipl , 1, 1) , 𝐴𝐴 (1, 1, 0.44) , and 𝐴𝐴 (0, 0, 1) , respectively. The gravity changes are normalized by the maximum ΔgSM (b and f). T splacements are normalized by the maximum vertical displacement (c and d) and the maximum radial displacemen . All distances are normalized by the depth of the point spherical source, D. Figure 4. Gravity changes (Δg = δg − ΔgFA), vertical displacements (uv), and radial displacements (ur) for point sources of different aspect ratios and depths. 4. Discussion Volcanic gravity changes caused by the net mass of intruding magmatic fluids and the induced host rock defor- mations may have comparable magnitudes to those of hydrological origin, such as changes in the water table. Such hydrogravimetric disturbances can be estimated by employing hydrological monitoring and modeling tech- niques (Battaglia et al., 2003, 2006; Creutzfeldt et al., 2010; Kazama et al., 2015; Lien et al., 2014; Van Camp et al., 2010) or by analyzing time-lapse gravity data (Güntner et al., 2017). Thus, the mass of intruding fluids at volcanoes can be inferred reliably once such effects are removed. New generation, low-cost, and accurate gravimeters might soon provide gravity measurements at an unprece- dented spatiotemporal resolution (Carbone et al., 2017, 2020). Permanent networks provide opportunities for new insight on magmatic plumbing systems (Battaglia et al., 2008; Carbone et al., 2019). One main challenge associated with these developments is to perform both detailed Bayesian inferences for in-depth understanding of the volcano, and rapid inversions for hazard assessment and early warning. The available FEM gravity change models can incorporate various chamber shapes (Currenti, 2014; Currenti et al., 2007, 2008; Trasatti & Bonafede, 2008), the Earth's surface topography (Charco et al., 2009; Currenti et al., 2007), crustal density and material heterogeneities (Currenti et al., 2007, 2008; Trasatti & Bonafede, 2008; Wang et al., 2006), viscoelasticity of the Earth's crust (Currenti, 2018), self-gravitation effects (Charco et al., 2005, 2006; Fernández et al., 2001, 2005), and magma compressibility (Currenti, 2014). Beside difficulties in imple- menting the FEM, such as meshing issues, this powerful method is computationally too demanding to be used for detailed inverse modeling. In contrast, the point CDM is a half-space model, but has already proven to be suitable for exploring the parameter space in both detailed Bayesian inferences (see Lundgren et al., 2017) and rapid and unsupervised inversions of deformation data (see Beauducel et al., 2020). The gravity change solutions for the point CDM, which we provide here, extend this potential to joint inversions of surface displacements and gravity changes. Volcanic deformation sources are often deep or far enough from the observation point to be treated as far field sources. The point CDM can provide a first order solution, which can be later improved by more sophis- ticated numerical models. 3.2. Implications for the Retrieval of Deformation Source Parameters Top block: The sources illustrated in (a) give rise to different Δg (b), similar uv (c), and different ur (d). Bottom block: The sources in (e) cause different Δg (f), similar ur (h), and different uv (g). The potency vectors of the point spherical source, point prolate source, and point sill in both (a and e) may be any positive multiple of (1, 1, 1) , 𝐴𝐴 (1, 1, 0.44) , and 𝐴𝐴 (0, 0, 1) , respectively. The gravity changes are normalized by the maximum ΔgSM (b and f). The displacements are normalized by the maximum vertical displacement (c and d) and the maximum radial displacements (g and h). All distances are normalized by the depth of the point spherical source, D. 8 of 12 NIKKHOO AND RIVALTA 8 of 12 Geophysical Research Letters 10.1029/2021GL095442 source parameters, beside the mass changes. In practice, however, if ΔM ≠ 0, gravity changes may be dominated by ΔgΔM and thus, depending on the signal-to-noise ratio of the data, the Δg curves (Figures 4b and 4f) may become indistinguishable. source parameters, beside the mass changes. In practice, however, if ΔM ≠ 0, gravity changes may be dominated by ΔgΔM and thus, depending on the signal-to-noise ratio of the data, the Δg curves (Figures 4b and 4f) may become indistinguishable. 4. Discussion This shows that unlike mass change estimates, the estimates of the intrusion density are prone to large uncertainties. density can be estimated from the inferred mass only if the interface volume change, ΔVint, is known (ΔVint should not be mistaken for the chamber volume change δVc). It can be shown from Equations 2 and 3 that the determina- tion of ΔVint requires knowledge of the fluid compressibility. This shows that unlike mass change estimates, the estimates of the intrusion density are prone to large uncertainties. Appendix A: Gravity Changes Caused by Point Tensile Dislocations Following the conventions in Section 2 and Okubo (1991), a point tensile dislocation below the origin with depth d, azimuth 0, dip angle θ, potency ΔV, and filled with an intrusion mass ΔM, causes the following gravity change contributions at 𝐴𝐴 (𝑥𝑥𝑥𝑥𝑥𝑥0) Δ푔Δ푉= −퐺휌rΔ푉푑 푟3 , Δ푔MD = 퐺휌rΔ푉(1 −2휈) [ 푑 푟3 − 1 푟(푟+ 푑) + 푥2(2푟+ 푑) 푟3(푟+ 푑)2 ] sin2휃, Δ푔MD = 퐺휌rΔ푉(1 −2휈) [ 푑 푟3 − 1 푟(푟+ 푑) + 푥2(2푟+ 푑) 푟3(푟+ 푑)2 ] sin2휃, Δ푔SM = 2휋퐺휌r푢푣, (A1) [ 푟3 푟(푟+ 푑) 푟3(푟+ 푑)2 ] Δ푔SM = 2휋퐺휌r푢푣, (A1) ( ) Δ푔SM = 2휋퐺휌r푢푣, (A1) (A1) Δ푔SM = 2휋퐺휌r푢푣, Δ푔FA = 훾푢푣, 푔 훾 Δ푔Δ푀= 퐺Δ푀푑 푟3 , Δ푔Δ푀= 퐺Δ푀푑 푟3 , where ΔgΔV, ΔgMD, ΔgSM, ΔgFA, and ΔgΔM are the contributions due to dislocation cavity, medium dilatation, displaced surface mass, free air effect, and intruded mass, respectively, 𝐴𝐴 𝐴𝐴= (𝑥𝑥2 + 𝑦𝑦2 + 𝑑𝑑2)1∕2 and uv is the surface uplift (see Okada, 1985; Okubo, 1991). Note that for ν = 0.5 and also, for horizontal tensile cracks (θ = 0) we have ΔgMD = 0. where ΔgΔV, ΔgMD, ΔgSM, ΔgFA, and ΔgΔM are the contributions due to dislocation cavity, medium dilatation, displaced surface mass, free air effect, and intruded mass, respectively, 𝐴𝐴 𝐴𝐴= (𝑥𝑥2 + 𝑦𝑦2 + 𝑑𝑑2)1∕2 and uv is the surface uplift (see Okada, 1985; Okubo, 1991). Note that for ν = 0.5 and also, for horizontal tensile cracks (θ = 0) we have ΔgMD = 0. Acknowledgments This research was funded by the EU Horizon 2020 programme NEWTON-g project, under the FETOPEN-2016/2017 call (Grant Agreement No 801221) and by the German Research Foundation (DFG), Grant 634756, RI 2782/2. The manuscript greatly benefited from constructive reviews by Paul Segall and Maurizio Battaglia. The authors are thankful to Daniele Carbone, Flavio Cannavò, and María Charco for fruitful discussions. The authors thank Elisa Trasatti for sharing and discussing the FEM results used in Figure 3. Open access funding enabled and organized by Projekt DEAL. Data Availability Statement Data were not used, nor created for this research. The MATLAB codes for computations are available to down- load at https://volcanodeformation.com/onewebmedia/pCDMgravity.zip. Further details about the dislocation models can be found under https://www.volcanodeformation.com/. 4. Discussion Some complexities such as layering or viscoelasticity can be accounted for (Amoruso et al., 2008) by using appropriate Green's functions for point dislocations (Okubo, 1993; Sun & Okubo, 1993; Wang et al., 2006). Besides, theory errors, arising from ignoring real Earth complexities, can be estimated in terms of noise covariance matrices within a Bayesian framework (see Duputel et al., 2014; Minson et al., 2013; Vasyura-Bathke et al., 2021). Finite pressurized ellipsoidal cavities can be approximated by a set of point CDMs uniformly distributed in the cavity volumes. With a high number of point CDMs, this approach can be used for benchmarking numerical models. An alternative solution is the finite ECM after Nikkhoo and Rivalta (2022), which provides comparable accuracies for a lesser number of point CDMs. The finite ECM is very fast, and thus, provides a practical way for performing coupled inversions of surface displacements and gravity changes. It is important to recall that for ellipsoidal deformation models in the half-space, including the finite ECM and the Yang et al. (1988) spheroid, the full-space expressions are used to calculate δVc (Amoruso & Crescentini, 2009). While this approximation may often be acceptable for deformation studies, it may lead to large errors in gravity change calculations involving shallow finite sources. This warrants future systematic comparisons with numeri- cal models in order to quantify the associated error. Deformation-induced gravity changes may be substantial (see Figure 3b) and should be accounted for in joint inversions of surface displacements and gravity changes. Provided that coupled models are employed for such inversions, the gravity changes may be exploited to better constrain the deformation source parameters besides the mass change. How practical this may be, depends on the observation uncertainties and the signal-to-noise ratio. We will explore this feature in future studies. Coupled inversions of surface displacements and gravity changes constrain the deformation source parameters and the intrusion mass without making any assumption on the properties of the intruding fluid. The intrusion NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA 9 of 12 Geophysical Research Letters 10.1029/2021GL095442 density can be estimated from the inferred mass only if the interface volume change, ΔVint, is known (ΔVint should not be mistaken for the chamber volume change δVc). It can be shown from Equations 2 and 3 that the determina- tion of ΔVint requires knowledge of the fluid compressibility. 5. Conclusions 1. Surface gravity changes are sensitive to both the intruding fluid mass and the deformation-induced surface uplift (subsidence) and country rock dilatation. Due to this coupling between the gravity changes and host rock deformations, gravity changes can also be used to constrain deformation source parameters, namely the location, spatial orientation, and potency of triaxial source models for expanding reservoirs. 1. Surface gravity changes are sensitive to both the intruding fluid mass and the deformation-induced surface uplift (subsidence) and country rock dilatation. Due to this coupling between the gravity changes and host rock deformations, gravity changes can also be used to constrain deformation source parameters, namely the location, spatial orientation, and potency of triaxial source models for expanding reservoirs. 2. We provide analytical solutions and MATLAB codes for surface displacements and gravity changes caused by both the point CDMs, a model for triaxial sources of expansion and the finite ECM, a model for ellipsoidal sources of uniform pressurization. 2. We provide analytical solutions and MATLAB codes for surface displacements and gravity changes caused by both the point CDMs, a model for triaxial sources of expansion and the finite ECM, a model for ellipsoidal sources of uniform pressurization. 3. While modeling gravity changes caused by shallow sources, it may be necessary to account for the mass redis- tribution within the source. This issue and also the inherent error in δVc for half-space solutions may limit the applicability of the finite ECM. 3. While modeling gravity changes caused by shallow sources, it may be necessary to account for the mass redis- tribution within the source. This issue and also the inherent error in δVc for half-space solutions may limit the applicability of the finite ECM. 4. The analytical solutions presented here can be used to validate new numerical gravity change models. Such validations should ideally consider various source depths and aspect ratios. 4. The analytical solutions presented here can be used to validate new numerical gravity change models. Such validations should ideally consider various source depths and aspect ratios. 5. By using the point CDM and the finite ECM, coupled inversions of surface displacements and gravity changes can now be performed. 5. By using the point CDM and the finite ECM, coupled inversions of surface displacements and gravity changes can now be performed. Geophysical Research Letters 10.1029/2021GL095442 Amoruso, A., & Crescentini, L. (2011). Modelling deformation due to a pressurized ellipsoidal cavity, with reference to the Campi Flegrei caldera, Italy. Geophysical Research Letters, 38(1), L01303. https://doi.org/10.1029/2010gl046030 moruso, A., & Crescentini, L. (2011). Modelling deformation due to a pressurized ellipsoidal cavity, with reference to the Campi Flegrei caldera, Italy. Geophysical Research Letters, 38(1), L01303. https://doi.org/10.1029/2010gl046030 moruso, A., & Crescentini, L. (2013). Analytical models of volcanic ellipsoidal expansion sources. Annals of Geophysics, 56(4), 0435. https:// doi.org/10.4401/ag-6441 Amoruso, A., Crescentini, L., & Berrino, G. (2008). Simultaneous inversion of deformation and gravity changes in a horizontally layered half- space: Evidences for magma intrusion during the 1982–1984 unrest at Campi Flegrei caldera (Italy). Earth and Planetary Science Letters, 272(1–2), 181–188. https://doi.org/10.1016/j.epsl.2008.04.040 ttaglia, M., Gottsmann, J., Carbone, D., & Fernández, J. (2008). 4D volcano gravimetry. Geophysics, 73(6), WA3–WA18. https://doi. org/10.1190/1.2977792 Battaglia, M., Roberts, C., & Segall, P. (1999). Magma intrusion beneath Long Valley caldera confirmed by temporal changes in gravity. Science, 285(5436), 2119–2122. https://doi.org/10.1126/science.285.5436.2119 Battaglia, M., Segall, P., & Roberts, C. (2003). The mechanics of unrest at Long Valley caldera, California. 2. Constraining the nature of the source using geodetic and micro-gravity data. Journal of Volcanology and Geothermal Research, 127(3–4), 219–245. https://doi.org/10.1016/ s0377-0273(03)00171-9 Battaglia, M., Troise, C., Obrizzo, F., Pingue, F., & De Natale, G. (2006). Evidence for fluid migration as the source of deformation at Ca Flegrei caldera (Italy). Geophysical Research Letters, 33(1), L01307. https://doi.org/10.1029/2005gl024904 Beauducel, F., Peltier, A., Villié, A., & Suryanto, W. (2020). Mechanical imaging of a volcano plumbing system from GNSS unsupervised m ling. Geophysical Research Letters, 47(17), e2020GL089419. https://doi.org/10.1029/2020GL089419 Bonafede, M., & Ferrari, C. (2009). Analytical models of deformation and residual gra medium. Tectonophysics, 471(1–2), 4–13. https://doi.org/10.1016/j.tecto.2008.10.006 Bonafede, M., & Ferrari, C. (2009). Analytical models of deformation and residual gravity changes due to a Mogi source in a viscoel medium. Tectonophysics, 471(1–2), 4–13. https://doi.org/10.1016/j.tecto.2008.10.006 Bonafede, M., & Mazzanti, M. (1998). Modelling gravity variations consistent with ground deformation in the Campi Flegrei caldera (It Journal of Volcanology and Geothermal Research, 81(1–2), 137–157. https://doi.org/10.1016/s0377-0273(97)00071-1 Carbone, D., Antoni-Micollier, L., Hammond, G., Zeeuw-van Dalfsen, D., Rivalta, E., Bonadonna, C., et al. (2020). The NEWTON-g gr imager: Towards new paradigms for terrain gravimetry. Frontiers in Earth Science, 8, 452. https://doi.org/10.3389/feart.2020.573396 Carbone, D., Cannavó, F., Greco, F., Reineman, R., & Warburton, R. J. (2019). The benefits of using a network of superconducting gravime monitor and study active volcanoes. Geophysical Research Letters Surface deformation due to inflation of an arbitrarily oriented triaxial ellipsoidal cavity in an elastic half-space, with refer- Kil l H ii J l f G h i l R h 91(B7) 7429 7438 h //d i /10 1029/jb091ib07 07429 Davis, P. M. (1986). Surface deformation due to inflation of an arbitrarily oriented triaxial ellipsoidal cavity in an elastic half-space, with refer- ence to Kilauea volcano, Hawaii. Journal of Geophysical Research, 91(B7), 7429–7438. https://doi.org/10.1029/jb091ib07p07429 Dieterich, J. H., & Decker, R. W. (1975). Finite element modeling of surface deformation associated with volcanism. Journal of Geoph Research, 80(29), 4094–4102. https://doi.org/10.1029/JB080i029p04094 Duputel, Z., Agram, P. S., Simons, M., Minson, S. E., & Beck, J. L. (2014). Accounting for prediction uncertainty when inferring subsurface slip. Geophysical Journal International, 197(1), 464–482. https://doi.org/10.1093/gji/ggt517 Duputel, Z., Agram, P. S., Simons, M., Minson, S. E., & Beck, J. L. (2014). Accounting for predicti slip. Geophysical Journal International, 197(1), 464–482. https://doi.org/10.1093/gji/ggt517 p p y , ( ), p g gj gg Eshelby, J. D. (1957). The determination of the elastic field of an ellipsoidal inclusion, and related problems. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences, 241(1226), 376–396. https://doi.org/10.1098/rspa.1957.0133 Eshelby, J. D. (1957). The determination of the elastic field of an ellipsoidal inclusion, and related problems. Proceedings of the Royal Socie Fernández, J., Tiampo, K. F., & Rundle, J. B. (2001). Viscoelastic displacement and gravity changes due to point magmatic intrusions in a g tational layered solid Earth. Geophysical Journal International, 146(1), 155–170. https://doi.org/10.1046/j.0956-540x.2001.01450.x rnández, J., Tiampo, K. F., & Rundle, J. B. (2001). Viscoelastic displacement and gravity changes due to point magmatic intrusions in a gravi tational layered solid Earth. Geophysical Journal International, 146(1), 155–170. https://doi.org/10.1046/j.0956-540x.2001.01450.x Fernández, J., Tiampo, K. F., Rundle, J. B., & Jentzsch, G. (2005). On the interpretation of vertical gravity gradients produced by magm intrusions. Journal of Geodynamics, 39(5), 475–492. https://doi.org/10.1016/j.jog.2005.04.005 Fernández, J., Tiampo, K. F., Rundle, J. B., & Jentzsch, G. (2005). On the interpretation of verti intrusions. Journal of Geodynamics, 39(5), 475–492. https://doi.org/10.1016/j.jog.2005.04.005 Ferrari, C., Bonafede, M., & Trasatti, E. (2015). Relations between pressurized triaxial cavities and moment tensor distributions. Anna Geophysics, 58(4), 0438. https://doi.org/10.4401/ag-6737 Güntner, A., Reich, M., Mikolaj, M., Creutzfeldt, B., Schroeder, S., & Wziontek, H. (2017). Landscape-scale water balance monitoring with an iGrav superconducting gravimeter in a field enclosure. Hydrology and Earth System Sciences, 21(6), 3167–3182. https://doi.org/10.5194/ hess-21-3167-2017 Hagiwara, Y. (1977). References Aki, K., & Richards, P. G. (2002). Quantitative seismology (2nd ed.). University Science Books. Amoruso, A., & Crescentini, L. (2009). Shape and volume change of pressurized ellipsoidal cavities from deformation and seismic data. Journal of Geophysical Research, 114(B2), B02210. https://doi.org/10.1029/2008JB005946 10 of 12 NIKKHOO AND RIVALTA Geophysical Research Letters Journal of Geophysical Research: Solid Earth, 124(4), 4035–4050. https://doi.org/10.1029/2018JB0 Carbone, D., Poland, M. P., Diament, M., & Greco, F. (2017). The added value of time-variable microgravimetry to the understanding of volcanoes work. Earth-Science Reviews, 169, 146–179. https://doi.org/10.1016/j.earscirev.2017.04.014 Charco, M., Camacho, A. G., Tiampo, K. F., & Fernández, J. (2009). Spatiotemporal gravity changes on volcanoes: Assessing the importance of topography. Geophysical Research Letters, 36(8), L08306. https://doi.org/10.1029/2009gl037160 Research Letters, 36(8), L08306. https://doi.org/10.1029/2009gl03716 Charco, M., Fernández, J., Luzón, F., & Rundle, J. (2006). On the relative importance of self-gravitation and elasticity in mod ground deformation and gravity changes. Journal of Geophysical Research, 111(B3). https://doi.org/10.1029/2005jb003754 Charco, M., Fernández, J., Luzón, F., & Rundle, J. (2006). On the relative importance of self-gravitation and elasticity in modeling vol ground deformation and gravity changes Journal of Geophysical Research 111(B3) https://doi org/10 1029/2005jb003754 p g y g ground deformation and gravity changes. Journal of Geophysical Research, 111(B3). https://doi.org/10.1029/2005jb003754 Charco, M., Tiampo, K. F., Luzón, F., & Fernández Torres, J. (2005). Modelling gravity changes and crustal deformation in active volcanic a Física de la Tierra, 17, 129–146. Creutzfeldt, B., Güntner, A., Wziontek, H., & Merz, B. (2010). Reducing local hydrology from high-precision gravity measurements: A lysime- ter-based approach. Geophysical Journal International, 183(1), 178–187. https://doi.org/10.1111/j.1365-246x.2010.04742.x Currenti, G. (2014). Numerical evidence enabling reconciliation gravity and height changes in volcanic areas. Geophysical Journal Internati 197(1), 164–173. https://doi.org/10.1093/gji/ggt507 197(1), 164–173. https://doi.org/10.1093/gji/ggt507 Currenti, G. (2018). Viscoelastic modeling of deformation and gravity changes induced by pressurized magmatic sources. Journal of Volcanology and Geothermal Research, 356, 264–277. https://doi.org/10.1016/j.jvolgeores.2018.03.020 Currenti, G., Del Negro, C., & Ganci, G. (2007). Modelling of ground deformation and gravity fields using finite element method: An applica to Etna volcano Geophysical Journal International 169(2) 775 786 https://doi org/10 1111/j 1365 246x 2007 03380 x Currenti, G., Del Negro, C., & Ganci, G. (2007). Modelling of ground deformation and gravity fields using finite element method: An application to Etna volcano. Geophysical Journal International, 169(2), 775–786. https://doi.org/10.1111/j.1365-246x.2007.03380.x to Etna volcano. Geophysical Journal International, 169(2), 775 786. https://doi.org/10.1111/j.1365 246x.2007.03380.x Currenti, G., Del Negro, C., & Ganci, G. (2008). Finite element modeling of ground deformation and gravity field at Mt. Etna. Annals of Geop ics 51(1) 105 119 https://doi org/10 4401/ag 3037 Currenti, G., Del Negro, C., & Ganci, G. (2008). Finite element modeling of ground deformation and gravity field at Mt. Etna. Annals of Geop ics, 51(1), 105–119. https://doi.org/10.4401/ag-3037 Davis, P. M. (1986). Geophysical Research Letters The Mogi model as a possible cause of the crustal uplift in the eastern part of Izu Peninsula and the related gravity change. Bulletin of the Earthquake Research Institute, 52, 301–309. Ichihara, M., Kusakabe, T., Kame, N., & Kumagai, H. (2016). On volume-source representations based on the representation theorem. Earth, Planets and Space, 68(1), 1–10. https://doi.org/10.1186/s40623-016-0387-3 Kazama, T., Okubo, S., Sugano, T., Matsumoto, S., Sun, W., Tanaka, Y., & Koyama, E. (2015). Absolute gravity change associated with magma mass movement in the conduit of Asama Volcano (Central Japan), revealed by physical modeling of hydrological gravity disturbances. Journal of Geophysical Research: Solid Earth, 120(2), 1263–1287. https://doi.org/10.1002/2014jb011563 Lien, T., Cheng, C.-C., Hwang, C., & Crossley, D. (2014). Assessing active faulting by hydrogeological mod case study for Hsinchu Fault, Taiwan. Journal of Geophysical Research: Solid Earth, 119(9), 7319–7335 Lien, T., Cheng, C.-C., Hwang, C., & Crossley, D. (2014). Assessing active faulting by hydrogeological modeling and superconducting gravimetry: A case study for Hsinchu Fault, Taiwan. Journal of Geophysical Research: Solid Earth, 119(9), 7319–7335. https://doi.org/10.1002/2014JB011285 Lisowski, M. (2007). Analytical volcano deformation source models. In Volcano deformation: Geodetic monitoring techniques (pp. 279–304). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-49302-0_8 case study for Hsinchu Fault, Taiwan. Journal of Geophysical Research: Solid Earth, 119(9), 7319–7335. https://doi.org/10.1002/2014JB011285 Lisowski, M. (2007). Analytical volcano deformation source models. In Volcano deformation: Geodetic monitoring techniques (pp. 279–304). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-49302-0_8 NIKKHOO AND RIVALTA 11 of 12 Geophysical Research Letters https://doi.org/10.1111/j.1365-246x.1993.tb01501.x nal, 115(3), 921–928. https://doi.org/10.1111/j.1365-246x.1993.tb0150 Okubo, S., Hirata, Y., Sawada, M., & Nagasawa, K. (1991). Gravity change caused by the 1989 earthquake swarm and submarine eruption off Ito, Japan test on the magma intrusion hypothesis. Journal of Physics of the Earth, 39(1), 219–230. https://doi.org/10.4294/jpe1952.39.219 Rivalta, E., & Segall, P. (2008). Magma compressibility and the missing source for some dike intrusions. Geophysical Research Letters, 3 L04306. https://doi.org/10.1029/2007gl032521 Rundle, J. B. (1978). Gravity changes and the Palmdale uplift. Geophysical Research Letters, 5(1), 41–44. https://doi.org/10.1029/gl005i001p00041 Rundle, J. B. (1978). Gravity changes and the Palmdale uplift. Geophysical Research Letters, 5(1), 41–44. https://doi.org/10.1029/gl005i001p00041 Savage, J. (1984). Local gravity anomalies produced by dislocation sources. Journal of Geophysical Research, 89(B3), 1945–1952. https://doi. org/10 1029/jb089ib03p01945 Rundle, J. B. (1978). Gravity changes and the Palmdale uplift. Geophysical Research Letters, 5(1), 41–44. https://doi.org/10.1029/gl005i001p00041 Savage, J. (1984). Local gravity anomalies produced by dislocation sources. Journal of Geophysical Research, 89(B3), 1945–1952. https://doi. ndle, J. B. (1978). Gravity changes and the Palmdale uplift. Geophysical Research Letters, 5(1), 41 44. https://doi.org/10.1029/gl005i001p00041 vage, J. (1984). Local gravity anomalies produced by dislocation sources. Journal of Geophysical Research, 89(B3), 1945–1952. https://doi. org/10.1029/jb089ib03p01945 g j p Segall, P. (2010). Earthquake and volcano deformation. Princeton University Press. Segall, P. (2010). Earthquake and volcano deformation. Princeton University Press. all, P. (2010). Earthquake and volcano deformation. Princeton Universi Segall, P. (2010). Earthquake and volcano deformation. Princeton University Press. Sun, W., & Okubo, S. (1993). Surface potential and gravity changes due to internal dislocations in a spherical Earth—I. Theory for a point d cation. Geophysical Journal International, 114(3), 569–592. https://doi.org/10.1111/j.1365-246x.1993.tb06988.x Trasatti, E., & Bonafede, M. (2008). Gravity changes due to overpressure sources in 3D heterogeneous media: Application to Campi Flegrei caldera, Italy. Annals of Geophysics, 51, 121–135. https://doi.org/10.4401/ag-4442 Trasatti, E., & Bonafede, M. (2008). Gravity changes due to overpressure sources in caldera, Italy. Annals of Geophysics, 51, 121–135. https://doi.org/10.4401/ag-4442 Van Camp, M., Métivier, L., De Viron, O., Meurers, B., & Williams, S. (2010). Characterizing long-time scale hydrological effects on gravit improved distinction of tectonic signals. Journal of Geophysical Research, 115(B7), B07407. https://doi.org/10.1029/2009jb006615 Vasyura-Bathke, H., Dettmer, J., Dutta, R., Mai, P. M., & Jonsson, S. (2021). Accounting for theory errors with empirical Bayesian noise mo in nonlinear centroid moment tensor estimation. Geophysical Journal International, 225(2), 1412–1431. https://doi.org/10.1093/gji/ggab in nonlinear centroid moment tensor estimation. Geophysical Journal International, 225(2), 1412–1431. https://doi.org/10.1093/gji/ggab034 Walsh, J., & Rice, J. (1979). Local changes in gravity resulting from deformation. Geophysical Research Letters 10.1029/2021GL095442 Lisowski, M., Dzurisin, D., Denlinger, R. P., & Iwatsubo, E. Y. (2008). Analysis of GPS-measured deformation associated with the 2004–2006 dome-building eruption of Mount St. Helens, Washington. In A volcano rekindled: The renewed eruption of Mount St. Helens, 2004–2006 (Vol. 1750, pp. 301–333). US Geological Survey Reston. https://doi.org/10.3133/pp175015 Lundgren, P., Nikkhoo, M., Samsonov, S. V., Milillo, P., Gil-Cruz, F., & Lazo, J. (2017). Source model for the Copahue volcano magma plumbing system constrained by InSAR surface deformation observations. Journal of Geophysical Research: Solid Earth, 122(7), 5729–5747. https:// doi.org/10.1002/2017jb014368 Minson, S., Simons, M., & Beck, J. (2013). Bayesian inversion for finite fault earthquake source models I—Theory and algorithm. Geophysical Journal International, 194(3), 1701–1726. https://doi.org/10.1093/gji/ggt180 Mogi, K. (1958). Relations between the eruptions of various volcanoes and the deformations of the ground surfaces around them. Bulletin of the Earthquake Research Institute, 36, 99–134. kkhoo, M., & Rivalta, E. (2022). Surface deformations caused by pressurized finite ellipsoidal cavities. Earth and Space Science Open Archive, 19. https://doi.org/10.1002/essoar.10510624.1 Nikkhoo, M., Walter, T. R., Lundgren, P. R., & Prats-Iraola, P. (2017). Compound dislocation models (CDMs) for volcano deformation analyses. Geophysical Journal International, 208(2), 877–894. https://doi.org/10.1093/gji/ggw427 Nikkhoo, M., Walter, T. R., Lundgren, P. R., & Prats-Iraola, P. (2017). Compound dislocation models (CDMs) for volcano deformation analy Geophysical Journal International, 208(2), 877–894. https://doi.org/1 Okada, Y. (1985). Surface deformation due to shear and tensile faults in a half-space. Bulletin of th Okada, Y. (1985). Surface deformation due to shear and tensile faults in a half-space. Bulletin of the Seismological Society of America, 7 1135–1154. https://doi.org/10.1785/bssa0750041135 1135–1154. https://doi.org/10.1785/bssa0750041135 Okubo, S. (1991). Potential and gravity changes raised by point dislocations. Geophysical Journal International, 105(3), 573–586. https:/ org/10.1111/j.1365-246x.1991.tb00797.x Okubo, S. (1992). Gravity and potential changes due to shear and tensile faults in a half-space. Journal of Geophysical Research, 97( 7137–7144. https://doi.org/10.1029/92jb00178 p g j Okubo, S. (1993). Reciprocity theorem to compute the static deformation due to a point dislocation buried in a spherically symmetric Earth. Geophysical Journal International, 115(3), 921–928. https://doi.org/10.1111/j.1365-246x.1993.tb01501.x S. (1993). Reciprocity theorem to compute the static deformation due to a point dislocation buried in a spherically symmetric Earth. h l l l 115(3) 921 928 h //d i /10 1111/j 1365 246 1993 b01501 Okubo, S. (1993). Reciprocity theorem to compute the static deformation due to a point dislocation buried in a Geophysical Journal International, 115(3), 921–928. https://doi.org/10.1111/j.1365-246x.1993.tb01501.x ( ) p y p p Geophysical Journal International, 115(3), 921–928. Geophysical Research Letters Journal of Geophysical Research, 84(B1), 165–170. https:// doi.org/10.1029/jb084ib01p00165 J., & Rice, J. (1979). Local changes in gravity resulting from deformation. Journal of Geophysical Research, 84(B1), 165–170. https:// rg/10.1029/jb084ib01p00165 Wang, R., Lorenzo-Martin, F., & Roth, F. (2006). PSGRN/PSCMP—A new code for calculating co-and post-seismic deformation, geoid and grav- ity changes based on the viscoelastic-gravitational dislocation theory. Computers & Geosciences, 32(4), 527–541. https://doi.org/10.1016/j. cageo.2005.08.006 Yang, X.-M., Davis, P. M., & Dieterich, J. H. (1988). Deformation from inflation of a dipping finite prolate spheroid in an elastic half-space as a model for volcanic stressing. Journal of Geophysical Research, 93(B5), 4249–4257. https://doi.org/10.1029/JB093iB05p04249 NIKKHOO AND RIVALTA NIKKHOO AND RIVALTA 12 of 12
https://openalex.org/W4321994226	https://zenodo.org/records/7675123/files/B-114.pdf	English	null	INNOVATIVE REFORMS IN THE MODERN PRESCHOOL EDUCATION SYSTEM	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	3,005	INNOVATIVE REFORMS IN THE MODERN PRESCHOOL EDUCATION SYSTEM O’rinova Feruza Uljayevna FarSU, Associate Professor of Preschool Education Department, Candidate of Pedagogical Sciences https://doi.org/10.5281/zenodo.7675123 https://doi.org/10.5281/zenodo.7675123 Abstract. The article talks about the achievements, opportunites gained as a result of the reforms carried out in recent years in the field of preshool education and at the same time the shortcomings allowed in the system, the solutions and proposals aimed at eliminating them. The experiences of the Korean country in the development of the sphere were discussed. The place of higher education in the training of specialists-educators is analytical and critical. Practical proposals were made on the direction of future educators to innovative education. Keywords: educational program “First step”, Organization of preschool educations, professor-teacher, teacher, Korean model, teacher-educator, reform, Third Renaissance. Today, as science progresses, many scientific research works are being carried out on the basis of exact sciences. However, there is no clear and concise suggestion about education and improvement of the spiritual image of a person. Therefore, education does not always lose its relevance. In all the conducted experiments, it was confirmed that the worldview formed in a person from the youngest age is preserved throughout his life. That is, these are formed as initial competencies and shape the child as a person. When the pedagogue can determine the personal orientation of the child, he can determine the task before him. President Sh. Mirziyoyev said, "If you ask me what is bothering you, I will answer that it is the education of our children." No matter how many centuries and eras pass, child education is always considered an important duty to parents and teachers. It is important that they not only aim to become a "great" person in the future, but also to mature as a real person, a person who has formed real human qualities in himself. For this, it is necessary to pay great attention to them from a young age and create the most suitable conditions for education. The introduction of the "Improved First Step" state program into the educational process in preschool educational organizations introduced a new direction in educational activities. The program is structured in a non-traditional way compared to the previous basic programs, and it serves as a method of developing individual abilities of children. Now, the child moves based on his own internal motivation, not on the basis of the teacher's instructions. SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ SCIENCE AND INNOVATION changing and improving. This situation is clearly visible in the outlook of the generations of the 21st century. Because today's children are growing up side by side with information and technical resources from a young age. They cannot imagine their daily activities without the virtual world. changing and improving. This situation is clearly visible in the outlook of the generations of the 21st century. Because today's children are growing up side by side with information and technical resources from a young age. They cannot imagine their daily activities without the virtual world. Therefore, the issue of education requires deep responsibility from parents and teachers. Therefore, the issue of education requires deep responsibility from parents and teachers. The laying of the foundation for the Third Renaissance by President Sh. Mirziyoyev motivated everyone to set a new noble goal. Everyone who knows that he is in harmony with the country's destiny strives to contribute to the country's development. Preschool education organizations are the most important base for educating the future generations as well-rounded people who love the country, respect the elders, and feel that they are involved in the development of the country. Taking this into account, we are looking for solutions to the following go - creation of mechanisms for evaluating and improving the knowledge of pedagogues working in the field; - creation of mechanisms for evaluating and improving the knowledge of pedagogues working in the field; - to determine ways to improve the knowledge of future specialists studying in higher educational institutions; - to determine ways to improve the knowledge of future specialists studying in higher educational institutions; - development of proposals and recommendations that provide solutions to the problems faced by pedagogues in the implementation of innovative technologies in the educational activities of children of a preschool educational organization. The reforms applied by many developed foreign countries in the field of preschool education and their effectiveness are studied, and the most appropriate ones are being put into practice. In particular, the Ministry of Preschool Education emphasized the elements of the Korean model. This decision was made after the meetings held during the visits of the head of state to this country. Because within the framework of this visit, during more than 20 meetings and studies, a number of achievements were noticed. INNOVATIVE REFORMS IN THE MODERN PRESCHOOL EDUCATION SYSTEM The child feels that his personality is respected, that his interests and needs are satisfied, that there are conditions for free creative movement. Forming the essence, goals and tasks of the "Improved First Step" state program in the minds of pedagogues is a time-consuming process. Since the implementation of the program, the working mechanism within it has not yet been completely created. That is why pedagogues- educators have many problems to be solved in order to improve their spiritual, scientific and worldly knowledge. There is a question of providing optimal solutions for pedagogues who are stuck in the old mold and cannot adapt to the new direction. But time is progressing in constant change and development. A specialist who does not constantly work on himself and sharpen his knowledge cannot continue his work in his field. Because the educational system is constantly 557 SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ pedagogue spends a lot of time on preparing documents and on educating children's creative potential. A pedagogue who knows that the supervisor who comes to evaluate the organization will pay attention to the students, always works to expand the worldview of himself and the children. A teacher-pedagogue should start his work incorporating this content. Children of preschool age have a high level of imitability and attention to the behavior of adults in the environment. Children's psychology is structured in such a way that they want to be encouraged by adults through their every action. Naturally, in order to reach this level, they strive to find in themselves any quality they have. That is, as P.F. Lestgaft said: "The period of kindergarten age is a period in which character traits that will be formed are determined and moral qualities emerge" [6]. Therefore, the spiritual growth of children depends to a large extent on the work of educators. At the solemn ceremony dedicated to the "Teachers and Mentors" Day, the head of our state said, "We will create the foundations of a new renaissance, that is, the third renaissance, in Uzbekistan through large-scale democratic changes, including educational reforms, we set as our main goal" was the president’s speech [9] It can be seen that the prosperity of the new renaissance period will be ensured by the initiative and selfless work of today's teachers-coaches. Special attention should be paid to the activities of higher education institutions in training future knowledgeable and advanced specialists. The weight of the potential and scientific power of the growing staff depends on the quality of the education they receive today. In the areas of higher education, the network of subjects far from specialization is reduced, and the scope of subjects related to specialization is being expanded. This leads to enrichment and improvement of students' knowledge in their field. At the same time, there are a number of other tasks that await solutions aimed at eliminating them. We all know that today every parent is in favor of their children learning foreign languages from a young age and actively participating in various circles. Heads of pre-school education organizations naturally want to work together with staff who know one of the foreign languages perfectly. SCIENCE AND INNOVATION The Ministry of Preschool Education has studied the activities of a number of pre-school education organizations in Korea, the created conditions, peculiarities, methods of education, means, the state of financial and material security, and the work being done to protect children's health. The cases of providing quality food were thoroughly studied. As a result, many strategies that can be used in the preschool education system of our country have been developed. Creation of maximum favorable conditions for the work of pedagogues, creative freedom of the pedagogue, absence of large-scale reports, high social status of pedagogues of state kindergartens, creative environment created for children to express themselves, including comfortable design and equipment of the rooms even down to the smallest details, For them, the availability of rich literature and educational games, social behavior and personal hygiene skills from early childhood are the basis of the success of the Korean preschool education system. [7] Over the past 2-3 years, the red tape in preschool education organizations has been somewhat optimized. Perfectly standardized criteria were created and applied to the process, from the material and educational base to the quality of food. Educational programs have been created aimed at ensuring more cooperative work of educators-pedagogues with children. However, as a result of the lack of knowledge and lack of work on the part of some officials, they cause excessive confusion during the inspection of educational activities. Therefore, it is necessary to review the requirements for the organization of inspection and control work. It would be expedient if the analysis of activities of pre-school educational organizations is focused on checking the knowledge indicators of more educated students. That is, if it is focused on conducting interviews and evaluation activities with children within the framework of state requirements, the educator- 558 SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ SCIENCE AND INNOVATION of higher education. For example, while teaching the subject "Innovations in preschool education" to students, the teacher can present innovative educational activities in the environment of preschool education based on his creative potential. This increases the opportunity for the student to gain experience. It also serves as a "Skills lesson" for the employees of the educational organization. That's why it is extremely effective to conduct practice classes together on the basis of each specialty. Also, practicing students can conduct only one open educational activity without preparing various practical documents. It is necessary to have the head of practice and other volunteers participate in such an open training, and at the same time, it is necessary to give the opportunity to conduct another open training based on the student's desire. Then there will be an opportunity to fill in the shortcomings. For the pedagogues of the preschool education organization, there was a need to reconstruct the entire system of preschool education, to change the methods of communication and interaction not only with the child, but also with all the subjects of preschool education. Therefore, the main task of pedagogues of the preschool organization is to choose the most suitable innovative pedagogical technologies in order to reveal the potential of the individual, the methods and forms of organizing work with children. in order to create an innovative process, the help of leaders- experts will definitely be necessary. Therefore, conducting trainings, consultation hours, and experimental trainings by professors among the employees of pre-school educational organizations gives a very useful result. Therefore, it is necessary to establish cooperation with teachers of higher educational institutions and pedagogues of preschool educational organizations. In such cooperation, pedagogues of preschool education organizations can be involved in higher education seminars. This is useful in sharing experience with students and identifying some shortcomings. As a result of such approaches, it is possible to achieve many achievements in the field. The goal of the state policy in the field of preschool education is to realize the right of every child to receive quality and free education. The variety of modern system forms of preschool education in Uzbekistan, flexible attitude to the needs of society and the individual, the emergence of new types of preschool education for children being characterized by various pedagogical services. [2. SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ 8 pages] Development indicators in our country can be ensured by quality education and competent individuals. For this, it is necessary that all participants of educational activities do not miss the activity of solidarity in this way. SCIENCE AND INNOVATION INTERNATIONAL SCIENTIFIC JOURNAL VOLUME 2 ISSUE 2 FEBRUARY 2023 UIF-2022: 8.2 \| ISSN: 2181-3337 \| SCIENTISTS.UZ Today, the reason why graduates of higher education institutions have difficulty finding their place in society is their lack of knowledge of foreign languages. Therefore, it is necessary to fundamentally revise the curriculum of preschool education. It takes time to introduce a new mechanism of personnel training in the field. There is a need to specialize preschool education in English or Russian. Because a future specialist who does not know one of these two languages will not find his place in the field. As time passes, this becomes more evident. Therefore, in higher education, if one of the two languages is mastered in depth on the basis of specialized subjects, the personnel potential will increase even more. Having a creative description of pedagogical activity requires the educator to study advanced work practices, master innovations and put them into practice. The rapid introduction of innovations creates the need to involve educators in scientific research. Involvement of educators in such research work will allow them to successfully use the results obtained later in their work. does not hesitate and achieves high results in creative activities. In the course of higher education, the professional practice program also needs to be revised. Experience in professional and pedagogical practices is not effective enough. Therefore, it is necessary to enrich the internship program with a new approach. For example, it is recommended to add practice hours to the basis of the specialized subjects of the leading professors 559 SCIENCE AND INNOVATION 5. Uljayevna U. F. THE ACTIVITY OF DEVELOPMENTAL CENTERS IN THE FORMATION OF CHILDREN'S ABILITIES //Science and innovation. – 2022. – Т. 1. – №. B2. – С. 107-110. 5. Uljayevna U. F. THE ACTIVITY OF DEVELOPMENTAL CENTERS IN THE FORMATION OF CHILDREN'S ABILITIES //Science and innovation. – 2022. – Т. 1. – №. B2. – С. 107-110. 6. Uljayevna O. R. F., Qizi A. Z. N. TARBIYACHINING INNOVATSION FAOLIYATGA TAYYORLASHNING DIDAKTIK KOMPONENTLARI //Science and innovation. – 2022. – Т. 1. – №. B3. – С. 340-344. 6. Uljayevna O. R. F., Qizi A. Z. N. TARBIYACHINING INNOVATSION FAOLIYATGA TAYYORLASHNING DIDAKTIK KOMPONENTLARI //Science and innovation. – 2022. – Т. 1. – №. B3. – С. 340-344. 7. Уринова Ф. У. ОСОБЕННОСТИ РАЗВИТИЯ ПСИХОЛОГИИ РЕБЁНКА ДОШКОЛЬНОГО И МЛАДШЕГО ШКОЛЬНОГО ВОЗРАСТА //Science and innovation. – 2022. – Т. 1. – №. B2. – С. 68-71. 7. Уринова Ф. У. ОСОБЕННОСТИ РАЗВИТИЯ ПСИХОЛОГИИ РЕБЁНКА ДОШКОЛЬНОГО И МЛАДШЕГО ШКОЛЬНОГО ВОЗРАСТА //Science and innovation. – 2022. – Т. 1. – №. B2. – С. 68-71. 8. Uljayevna O. R. F. MAKTABGACHA TALIM TASHKILOTI TARBIYACHISINING KASBIY PEDAGOGIK KOMPETENSIYANI TAKOMILLASHTIRISH XUSUSIYATLARI //Science and innovation. – 2022. – Т. 1. – №. B2. – С. 88-91. 9. Uljayevna O. R. F., Saydamatovna S. Z. MAKTABGACHA YOSHDAGI BOLALARNING IJODIY FAOLIYATINI SHAKLLANTIRISHNING O’QUV-USLUBIY TA’MINOTI //Science and innovation. – 2022. – Т. 1. – №. B3. – С. 349-355. 9. Uljayevna O. R. F., Saydamatovna S. Z. MAKTABGACHA YOSHDAGI BOLALARNING IJODIY FAOLIYATINI SHAKLLANTIRISHNING O’QUV-USLUBIY TA’MINOTI //Science and innovation. – 2022. – Т. 1. – №. B3. – С. 349-355. 10. Uljaevna U. F. Didactic games in preschool educational system //Проблемы современной науки и образования. – 2020. – №. 4-2 (149). – С. 27-29. 10. Uljaevna U. F. Didactic games in preschool educational system //Проблемы современной науки и образования. – 2020. – №. 4-2 (149). – С. 27-29. 11. Uljayevna O. F. et al. BO’LAJAK TARBIYACHILARNING KASBIY SALOHIYATINI TAKOMILLASHTIRISHDA ILMIY FAOLIYATGA TAYYORLASHNING METODIK TIZIMI //BARQARORLIK VA YETAKCHI TADQIQOTLAR ONLAYN ILMIY JURNALI. – 2022. – С. 660-667. 12. Uljayevna U. F. et al. CREOLIZED TEXTS, THEIR FEATURES AND POSSIBILITIES //BARQARORLIK VA YETAKCHI TADQIQOTLAR ONLAYN ILMIY JURNALI. – 2022. – С. 668-673. 13. Uljayevna O. F. et al. MAKTABGACHA TA'LIM TASHKILOTLARIDAGI TAYYORLOV GURUHI TARBIYALANUVCHILARINI MAKTABGA TAYYORLASH TAMOYILLARI //BARQARORLIK VA YETAKCHI TADQIQOTLAR ONLAYN ILMIY JURNALI. – 2022. – С. 654-659. 14. Uljaevna U. F. et al. REFERENCES 1. “O’zbekiston Respublikasi maktabgacha ta’lim tizimini 2019-2030 yillarda rivojlantirish konsepsiyasi” 2019 y 1. “O’zbekiston Respublikasi maktabgacha ta’lim tizimini 2019-2030 yillarda rivojlantirish konsepsiyasi” 2019 y 2. Grosheva I.B., Mirziyoyeva SH.SH., Tashmuhammedova F.B., Rustamova M.E., Madjitova H.A., Mikailova U.T., Ismailova M.A “Maktabgacha ta’lim tashkilotlarida ta’lim dasturlarini ishlab chiqish” Metodik qo’llanma, Toshkent, 2020 yil 8 bet 3. Grosheva I.B., Olimov K.T., Nazarova V.A., Djanpeisova G.E., Mikailova U.T., Kenjabayeva D.A., Gulyamova H.B., Miftayeva H.A “Kuzatish va baholash” Metodik qo’llanma, Toshkent, 2020 yil 8 bet 4. Shin A.B., Mirziyoyeva Sh.SH., Grosheva I.B., Mikailova U.T., Suleymanova Yu.T., Daukaeva A.G., Vlasova E.H., Galimova E.F “Shaxsga yo’naltirilgan yondashuv asosida ta’lim jarayonini rejalashtirish” Metodik qo’llanma, Toshkent, 2020 yil 12 bet 560 SCIENCE AND INNOVATION DEVELOPMENT OF CREATIVE ACTIVITY OF PRESCHOOL CHILDREN IN ART GAME AS A PSYCHOLOGICAL AND PEDAGOGICAL PROBLEM //BARQARORLIK VA YETAKCHI TADQIQOTLAR ONLAYN ILMIY JURNALI. – 2022. – С. 47-51. 15. Uljaevna U. F. et al. NECESSARY CONDITIONS FOR THE DEVELOPMENT OF CREATIVE THINKING IN FUTURE TEACHERS //Modern Journal of Social Sciences and Humanities. – 2022. – Т. 4. – С. 444-448. 16. Uljayevna O. R. F., Saydamatovna S. Z. MAKTABGACHA YOSHDAGI BOLALARNING IJODIY FAOLIYATINI SHAKLLANTIRISHNING O’QUV-USLUBIY TA’MINOTI //Science and innovation. – 2022. – Т. 1. – №. B3. – С. 349-355. 561
https://openalex.org/W1969650498	http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4084993/pdf/pone.0101700.pdf	English	null	Do the Historical Biogeography and Evolutionary History of the Digenean Margotrema spp. across Central Mexico Mirror Those of Their Freshwater Fish Hosts (Goodeinae)?	PloS one	2,014	cc-by	10,278	Andre´s Martı´nez-Aquino1,2¤, Fadia Sara Ceccarelli3, Luis E. Eguiarte4, Ella Va´zquez-Domı´nguez5, Gerardo Pe´rez-Ponce de Leo´ n2 1 Posgrado en Ciencias Biolo´gicas, Universidad Nacional Auto´noma de Me´xico, Me´xico, D.F., Me´xico, 2 Departamento de Zoologı´a, Instituto de Biologı´a, Universidad Nacional Auto´noma de Me´xico, Me´xico, D.F., Me´xico, 3 Divisio´n de Aracnologı´a, Museo Argentino de Ciencias Naturales ‘‘Bernardino Rivadavia’’, Buenos Aires, Argentina 4 Departamento de Ecologı´a Evolutiva, Instituto de Ecologı´a, Universidad Nacional Auto´noma de Me´xico, Me´xico, D.F., Me´xico, 5 Departamento de Ecologı´a de la Biodiversidad, Instituto de Ecologı´a, Universidad Nacional Auto´noma de Me´xico, Me´xico, D.F., Me´xico Abstract doi:10.1371/ journal.pone.0101700 Editor: Diego Fontaneto, Consiglio Nazionale delle Ricerche (CNR), Italy Editor: Diego Fontaneto, Consiglio Nazionale delle Ricerche (CNR), Italy Received November 28, 2013; Accepted June 11, 2014; Published July 7, 2014 Received November 28, 2013; Accepted June 11, 2014; Published July 7, 2014 Received November 28, 2013; Accepted June 11, 2014; Published July 7, 2014 Copyright: 2014 Martı´nez-Aquino et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: AMA was supported by a scholarship from Consejo Nacional de Ciencia y Tecnologı´a (CONACyT) to obtain his PhD degree within Posgrado en Ciencias Biolo´gicas, Universidad Nacional Auto´noma de Me´xico, and actually his research is supported by a postdoctoral grant from CONACYT (# 207983). FSC was supported by a postdoctoral fellowships provided by Universidad Nacional Auto´noma de Me´xico (UNAM-DGAPA). GPPL was supported by grants from the Consejo Nacional de Ciencia y Tecnologı´a (CONACyT 83043) and Programa de Apoyo a Proyectos de Investigacio´n e Innovacio´n Tecnologı´ca (PAPIIT-UNAM IN 202111 and IN204514). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. Email: maandres_@hotmail.com * Email: maandres_@hotmail.com ¤ Current address: Divisio´n Zoologı´a Invertebrados, Museo de La Plata, FCNyM, UNLP, La Plata, Argentina Citation: Martı´nez-Aquino A, Ceccarelli FS, Eguiarte LE, Va´zquez-Domı´nguez E, Pe´rez-Ponce de Leo´n G (2014) Do the Historical Biogeography and Evolutionary History of the Digenean Margotrema spp. across Central Mexico Mirror Those of Their Freshwater Fish Hosts (Goodeinae)? PLoS ONE 9(7): e101700. doi:10.1371/ journal.pone.0101700 Do the Historical Biogeography and Evolutionary History of the Digenean Margotrema spp. across Central Mexico Mirror Those of Their Freshwater Fish Hosts (Goodeinae)? Andre´s Martı´nez-Aquino1,2¤, Fadia Sara Ceccarelli3, Luis E. Eguiarte4, Ella Va´zquez-Domı´nguez5, Gerardo Pe´rez-Ponce de Leo´ n2 July 2014 \| Volume 9 \| Issue 7 \| e101700 Editor: Diego Fontaneto, Consiglio Nazionale delle Ricerche (CNR), Italy Abstract Host-parasite systems provide an ideal platform to study evolution at different levels, including codivergence in a historical biogeography context. In this study we aim to describe biogeographic and codivergent patterns and associated processes of the Goodeinae freshwater fish and their digenean parasite (Margotrema spp.) over the last 6.5 Ma (million years), identifying the main factors (host and/or hydrogeomorphology) that influenced the evolution of Margotrema. We obtained a species tree for Margotrema spp. using DNA sequence data from mitochondrial and nuclear molecular markers (COI and ITS1, respectively) and performed molecular dating to discern divergence events within the genus. The dispersal-extinction- cladogenesis (DEC) model was used to describe the historical biogeography of digeneans and applied to cophylogenetic analyses of Margotrema and their goodeine hosts. Our results showed that the evolutionary history of Margotrema has been shaped in close association with its geographic context, especially with the geological history of central Mexico during the Pleistocene. Host-specificity has been established at three levels of historical association: a) Species-Species, represented by Xenotaenia resolanae-M. resolanae exclusively found in the Cuzalapa River Basin; b) Species-Lineage, represented by Characodon audax-M. bravoae Lineage II, exclusive to the Upper and Middle Mezquital River Basin, and c) Tribe-Lineage, including two instances of historical associations among parasites and hosts at the taxonomical level of tribe, one represented by Ilyodontini-M. bravoae Lineage I (distributed across the Ayuquila and Balsas River Basins), and another comprised of Girardinichthyini/Chapalichthyini-M. bravoae Lineage III, found only in the Lerma River Basin. We show that the evolutionary history of the parasites is, on several occasions, in agreement with the phylogenetic and biogeographic history of their hosts. A series of biogeographic and host-parasite events explain the codivergence patterns observed, in which cospeciation and colonisation via host-switching and vicariant plus dispersal events are appreciated, at different times during the diversification history of both associates, particularly during the Pleistocene. Citation: Martı´nez-Aquino A, Ceccarelli FS, Eguiarte LE, Va´zquez-Domı´nguez E, Pe´rez-Ponce de Leo´n G (2014) Do the Historical Biogeography and Evolutionary History of the Digenean Margotrema spp. across Central Mexico Mirror Those of Their Freshwater Fish Hosts (Goodeinae)? PLoS ONE 9(7): e101700. 1. Taxa, molecular dataset and phylogenetic analyses 1. Taxa, molecular dataset and phylogenetic analyses The present study is a follow-up to a previous work in which Bayesian phylogenetic inference was used to reconstruct phylog- enies of 127 individuals belonging to the genus Margotrema allowing us to establish topological congruence between various programs and algorithms (for more details see [21]). For the present phylogenetic analyses the same taxa and gene sequences as in the previous study were used: samples of Margotrema spp. and published fragments of Cytochrome c Oxidase subunit I (COI; mitochondrial DNA) and Internal Transcribed Spacer 1 (ITS1; nuclear DNA), encompassing 750 and 831 base pairs including gaps, respectively [21]. We constructed a combined dataset to perform a multispecies coalescent analysis as implemented in BEAST v1.7.4 [24] to obtain a species tree to infer the genealogical relationship between M. resolanae and the three M. bravoae lineages [21]. The aligned data file in nexus format was deposited in the Dryad Digital Repository (DRYAD) [25], DOI: 10.5061/dryad.bq7q0. Central Mexico, and particularly the Trans-Mexican Volcanic Belt (TMVB), has been used to explain diversification because it represents a transition zone between the Neotropical and Nearctic biogeographic regions [18]. The TMVB is considered an area of endemism for different taxa, mainly as a result of its complex hydro-geomorphological history. It is a remarkable area of endemism for the freshwater fish fauna [19]. Goodeines, endemic elements of the central Mexican freshwater fish fauna, are a monophyletic group of cyprinodontiforms that experienced a remarkable diversification in this area ([20] and references therein). The digenean genus Margotrema Lamothe-Argumedo, 1972 encompasses species that are relatively common intestinal parasites of goodeines across Mexico [21]. Their members are part of the biogeographic ‘‘core’’ helminth fauna of goodeines [17]. Based on the fact that the evolutionary and biogeographic history of goodeines was deeply influenced by tectovolcanic activity in central Mexico, Martı´nez-Aquino et al. [21] recently deciphered the genealogical structure of Margotrema. The genus contains M. resolanae Pe´rez-Ponce de Leo´n, Martı´nez-Aquino and Mendoza- Garfias, 2013 and three independent genetic lineages within M. bravoae Lamothe-Argumedo, 1972. These authors also demon- strated that the parasite is tightly associated with the genus’ geographic distribution across different hydrological systems in this region. In addition, the three M. bravoae lineages apparently show certain specific association with their hosts, at the goodeine taxonomic level of tribe, representing monophyletic groups. Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution usually focuses on evaluating the geographic distribution and the phylogenetic relationships of the hosts. These two elements can be analysed by probabilistic methods in biogeography [6], as well as through evolutionary tangled trees of hosts and parasites [7]. In turn, hypotheses of evolutionary history and biogeography can be explicitly tested in time and space using parametric biogeography methods [8,9,10] coupled with dated species-tree estimation [11] and by contrasting the phylogenetic relationships between two groups that have a narrow biological association in a particular geographic area [10,12,13]. Previous studies have shown that geographic features may significantly shape genealogical relation- ships of hosts and parasites by causing co-differentiation between parasitic organisms and the evolutionary history of their hosts [14]. However, it is difficult to find a biological model of parasitism capable of explaining the process of diversification between hosts and parasites, as outlined by Caira and Jensen [15] and Althoff et al. [16]. The biogeographic ‘‘core’’ parasite fauna, i.e. widely distributed species characteristically associated with - and restrict- ed to - a monophyletic group of host species (see [17]), offer a unique opportunity to test diversification processes between hosts and their parasites. scenario of central Mexico, by their close association with their goodeine hosts, or by a combination of the two. Therefore, we test the general null hypothesis that the parasite phylogeny is independent of the host phylogeny [23]. Consequently, the main alternative hypothesis is that the biogeographic congruence between the genealogical history of the Margotrema lineages and the hydro-geomorphological history of central Mexico is similar to the historical biogeography of their goodein hosts. If this main hypothesis is supported, the following two specific hypotheses can be tested: 1) the biogeographic congruence is further reflected in the evolutionary histories of the Goodeinae tribes and the associated lineages of Margotrema and 2) the divergence times of the main clades of Goodeinae and those of Margotrema are relatively similar. 1. Taxa, molecular dataset and phylogenetic analyses These results support the hypothesis that the distribution patterns and the host associations of the four Margotrema lineages were concordant with the hydro-geomorphologic events occurring in central Mexico. Also, that the vicariant and dispersal events associated with the goodeine diversification promoted at the same time the diversification of the Margotrema lineages. In this context, two patterns of evolution of Margotrema were uncovered: restricted geographic distribution in hydrological systems and host-specificity at the host-species and host-tribe levels [21]. Introduction hosts plays a greater role in its diversification process than coevolution with the host, resulting in highly incongruent phylogenies, as shown in some host-parasite systems such as pinnipeds and their cestode parasites [1,2]. On the other hand, strict cospeciation was found between pocket gophers and their chewing lice [3]. As host-parasite systems are wide and varied, a broad spectrum of factors influences the associates’ evolution (for reviews on this topic see [4,5]). The study of a host-parasite association, in terms of the evolution of parasitic organisms, Host-parasite associations represent exceptional systems for linking evolution and ecology to obtain a broader view of how biotic interactions shape life on earth. Evolutionary studies of host- parasite systems vary in their approaches and outcomes, ranging from very close evolutionary associations between hosts and their parasites, to the apparent lack of such evolutionary associations. In some cases for example, the ability of the parasite to colonise new July 2014 \| Volume 9 \| Issue 7 \| e101700 1 PLOS ONE \| www.plosone.org 4. Cophylogenetic analyses y y To test evolutionary associations such as codivergence between goodeine fishes and their Margotrema parasites, we implemented a DEC model using Lagrange. Trees involving the association of two taxa were used to infer their common evolutionary history, both in a geographic (area and organism trees) and in a cophylogenetic (host and parasite) scenario. Considering that biogeographic investigations can be paralleled to cophylogenetic ones, the following analogies were used, as study units, in accordance with Page and Charleston [32]: area/host like organism/parasite, dispersal/host-switch, vicariance/cospeciation, sympatric speciation/parasite speciation in one host and extinc- tion/parasite extinction (lineage loss). In this context, we built two matrices of 15 ‘‘areas ( = hosts)’’ (see Table 1 for codes), in which the maximum ‘‘range’’ size for ancestral areas ( = host species) was set to two. The dispersal probabilities were constrained using similar premises to those used for the geographic DEC analyses (input files in DRYAD; DOI: 10.5061/dryad.bq7q0) taking into account the genealogical relationships between Margotrema spp. and their geographic areas. Also, the following three terms were used in accordance with Charleston [12]. Codivergence, also referred to as cospeciation, implies an event where a parasite lineage (species) infecting a host lineage diverges into two new lineages following the divergence of their host (approximately at the same time). Duplication, an event where the parasite lineage diverges into two new lineages, independently of its host, and both new lineages remain on that host lineage. Host-switching, defined as the event where the parasite diverges by switching from one host to establish in another host lineage. Following Choudhury et al. [33], parasite speciation may either be concomitant with and resulting from host speciation (cospeciation) or follow the colonisation of a new host from an existing one (host-switching). Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution nodes. Each terminal of the species tree was set to incorporate the individuals from one lineage found in one of the 12 areas chosen (see below for more details). Two separate analyses were run for 50 million generations each with a sampling frequency of one in every 1000 generations and outputs were combined using LogCombiner v. 1.7.4 [24]. Tracer v. 1.5 (available online from http://beast.bio. ed.ac.uk/) was used to assess convergence of the model parameters, where an effective sample size (ESS) value .200 was considered adequate. Branch support for the different tree topologies was evaluated by Posterior Probability (PP) values for nodal support (PP.0.95). The tree with the highest clade probability was chosen from the BEAST output files using the program TreeAnnotator v.1.7 [24]. The BEAST phylogenetic reconstructions were run through the CIPRES Science Gateway V. 3.3 [30]. The BEAST.xml input file was deposited in DRYAD DOI: 10.5061/dryad.bq7q0. to consider the presence of the central Mexican Palaeolakes and thus increased dispersal probabilities between the areas included during the Pleistocene. The input files for the analyses were constructed using the web-based Lagrange configurator (http://www. reelab.net/lagrange/configurator/index) and are available from DRYAD; DOI: 10.5061/dryad.bq7q0. 3. Historical biogeography To uncover the events that influenced the historical biogeog- raphy and diversification processes of Margotrema spp., we applied dispersal–extinction–cladogenesis models (DEC) using the pro- gram Lagrange v. 20130526 [8,9] with the dated ultrametric area tree obtained from BEAST. We coded the geographic distribu- tion of the terminal taxa into 12 different areas, which were used to recover the main clades within Margotrema spp. (see [21]), following the area codes used by Domı´nguez-Domı´nguez et al. [20,22] (Table 1; Fig. 1; Fig. S1). The 12 areas were delimited based on the Mexican hydrological basins and sub-basins map produced by CONABIO (www.conabio.gob.mx). All maps shown in this study were modified in DIVA-GIS 7.5 from sources ([31] freely available through www.diva-gis.org). The DEC algorithm permits uncovering dispersal and extinction events along branches of a phylogeny, while estimating the ranges of the MRCA and descendent species at each node, returning maximum likelihood (ML) and relative probability values for each area/event as well as providing a global ML value for the total analysis. This value allows for comparisons of different runs incorporating varying degrees of spatio-temporal constraints. Accordingly, we compared three models varying in maximum number of ancestral areas and dispersal probability constraints, keeping the same adjacency matrix for all three runs. Dispersal probability constraints were set We also compared the phylogenies of the subfamily Goodeinae and Margotrema to establish whether a significant match existed Table 1. Parasite-host-area associations of Margotrema species and lineages used in this study. 2. Divergence dating To determine an accurate time frame for phylogenetic divergence processes for each Margotrema lineage (see Table 1), we estimated mean node ages and their 95% highest posterior densities (HPDs) using a Bayesian relaxed molecular clock method [26] implemented in BEAST. In this framework, tests of evolutionary hypotheses are not conditioned to a single tree topology, allowing for the simultaneous evaluation of topology and divergence times, while incorporating their uncertainty. Hetero- zygote sites in the nuclear gene fragment were identified using ‘‘Find Heterozygotes Plugin’’ in Geneiuos Pro v5.1.7 [27], using a threshold of 90% peak heights. We applied the same optimal model as the one obtained by Martı´nez-Aquino et al. [21]: COI with HKY+I+G and ITS1 with HKY+G invariant sites with BEAST package (BEAUti v1.7.4; [24]). An uncorrelated relaxed log-normal molecular clock was applied to model rate variation across branches and the uniform Yule tree prior was chosen, appropriate for hierarchical rather than reticulate relationships. In absence of specific information on substitution rates of the gene fragments for our species, we applied lognormal distributions with a mean of 1 and standard deviation of 0.33 to both markers, allowing for auto-optimizations as the runs progressed. Based on the geographic distribution of Margotrema (Martı´nez-Aquino et al. unpublished data), we applied a geological calibration based on the uplifting of the western zone of the TMVB, which started around 11 million years ago (Ma) [28,29]. This age was set as a maximum for the Most Recent Common Ancestor (MRCA) of Margotrema spp. Monophyly was not enforced for any of the other In this study, we explored the evolutionary processes that drove the diversification of a host-parasite association, based on biogeographic and phylogenetic hypotheses of goodeine fishes [20,22] and the phylogenetic relationships of the parasite Margotrema spp. We also examined the potential role that their respective geographic distributions had on each other’s evolution- ary history. In this context, we established whether the evolution of Margotrema spp. was influenced by the complex geographic PLOS ONE \| www.plosone.org July 2014 \| Volume 9 \| Issue 7 \| e101700 2 3. Historical biogeography Parasite species (lineages) Host species (code) Area (sub-basin) (code) Hydrological systems Margotrema resolanae Xenotaenia resolanae (m) Cuzalapa River (I) Cuzalapa River Margotrema bravoae Lineage I Codoma ornata (h) Lower Conchos River (A) Conchos River Basin Margotrema bravoae Lineage I Allodontichthys zonistius (g) Ilyodon furcidens (k) Armerı´a-Ayuquila River (H) Ayuquila River Margotrema bravoae Lineage I Chapalichthys pardalis (j) Ilyodon whitei (l) Lower Balsas River (K) Balsas River Basin Margotrema bravoae Lineage I Ilyodon whitei (l) Upper Balsas River (L) Balsas River Basin Margotrema bravoae Lineage II Characodon audax (i) Upper and Middle Mezquital River (B) Mezquital River Basin Margotrema bravoae Lineage III Zoogoneticus purhepechus (o) Lower Lerma River (C) Lerma River Basin Margotrema bravoae Lineage III Allotoca zacapuensis (f) Zoogoneticus quitzeoensis (n) Zacapu Lake (D) Lerma River Basin Margotrema bravoae Lineage III Alloophorus robustus (e) Zoogoneticus quitzeoensis (n) Cuitzeo Lake (E) Lerma River Basin Margotrema bravoae Lineage III Allotoca diazi (a) Allotoca duguesi (b) Pa´tzcuaro Lake (F) Lerma River Basin Margotrema bravoae Lineage III Allotoca meeki (c) Zirahue´n Lake (G) Lerma River Basin Margotrema bravoae Lineage III Neoophorus regalis (d) Cotija (J) Lerma River Basin Letters in brackets next to host and area names correspond to the codes used in the biogeographic and cophylogenetic analyses. doi:10.1371/journal.pone.0101700.t001 Table 1. Parasite-host-area associations of Margotrema species and lineages used in this study. July 2014 \| Volume 9 \| Issue 7 \| e101700 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org 3 Margotrema Biogeography: Cophylogeny and Evolution Figure 1. Pattern of geographic distribution of Margotrema spp. and their associated goodeins from central Mexico. Hydrological systems in colours correspond to distributions of each lineage of Margotrema spp. collected in this study: green = Margotrema resolanae; orange = Margotrema bravoae Linaege I; blue = Margotrema bravoae Linaege II; red = Margotrema bravoae Linaege III. Coloured fish outlines correspond to the family/tribe(s)/species associated with each Margotrema lineage: grey fish = Cyprinidae (in Codoma ornata); red fish = Characodontini (in Characodon audax); yellow fish = Ilyodontini; blue fish = Chapalichthyini and Girardinichthyini; green fish = Xenotaenia resolanae (of the tribe Ilyodontini). doi:10.1371/journal.pone.0101700.g001 Figure 1. Pattern of geographic distribution of Margotrema spp. and their associated goodeins from central Mexico. Hydrological systems in colours correspond to distributions of each lineage of Margotrema spp. collected in this study: green = Margotrema resolanae; orange = Margotrema bravoae Linaege I; blue = Margotrema bravoae Linaege II; red = Margotrema bravoae Linaege III. 2. Historical Biogeography between host and parasite trees. We used a statistic test of p values with a 95% of confidence intervals with TreeMap 3b [12], employing the tree topologies obtained from cytochrome b gene sequences from the hosts and COI+ITS1 sequences from four lineages of Margotrema spp. The dated tree from the probabilistic analyses of Goodeinae [20] was obtained from a co-author of the cited study and edited with MESQUITE 2.72 [34], to select terminal taxa that are found in a host-parasite association with Margotrema, removing the remaining terminal taxa from the tree. Similarly, the dated molecular phylogeny of Margotrema spp. was edited in MESQUITE 2.72, where samples of each locality were trimmed to a single terminal taxon. This was done because TreeMap 3b only reconciles strictly dichotomous trees. The input file for TreeMap is available from DRYAD (DOI: 10.5061/ dryad.bq7q0). The ancestral area for Margotrema bravoae and M. resolanae was recovered with extremely low relative probabilities, making its unequivocal identification almost impossible. This was the case whether or not dispersal probability constraints were enforced in the DEC analyses. Enforcing dispersal constraints increased the global ML of the run, but generally decreased the ML and relative probability of the individual events (see File S1A). For this reason we considered areas and events from all runs as long as they showed combined relative probabilities .0.7 and were in agreement between runs. Overall, low probabilities were also recovered for the ancestral area of M. bravoae. On the other hand, the ancestral areas of M. bravoae lineages II and III were recovered with relatively high probabilities that support the ancestral area containing the Upper and Middle Mezquital River [B] and a combination of the Cotija area [J] plus one of the areas of the Paleolakes (either the Lower Lerma River [C], Zacapu Lake [D], Cuitzeo Lake [E], Pa´tzcuaro Lake [F] or Zirahue´n Lake [G]). A vicariant event was found separating the MRCA of M. bravoae lineages II and III (populations found in area B from the ones found in the remaining ancestral areas) during the Pleistocene, approximately 1.04 Ma (see Figs. 2, 3A and File S1A). The recovered ancestral area of the MRCA of M. bravoae Lineage III included a combination of the Cotija area [J] plus two of the areas of the Paleolakes (either the Lower Lerma River [C], Zacapu Lake 3. Historical biogeography Coloured fish outlines correspond to the family/tribe(s)/species associated with each Margotrema lineage: grey fish = Cyprinidae (in Codoma ornata); red fish = Characodontini (in Characodon audax); yellow fish = Ilyodontini; blue fish = Chapalichthyini and Girardinichthyini; green fish = Xenotaenia resolanae (of the tribe Ilyodontini). doi 10 1371/jo rnal pone 0101700 g001 y doi:10.1371/journal.pone.0101700.g001 doi:10.1371/journal.pone.0101700.g001 3. Codivergence patterns [D], Cuitzeo Lake [E], Pa´tzcuaro Lake [F] or Zirahue´n Lake [G]; Fig. 3B, only one area shown), with a vicariant event separating the populations in Cotija from the populations of the remaining areas, approximately 170,000 years ago. Subsequent vicariant events, which progressively fragmented the ancestral populations from the areas of present-day central Mexico that once formed part of the greater Paleolakes during the Pleistocene (Figs. 2 and 3B), were uncovered. The ancestral area of M. bravoae Lineage I was found to span the Upper and Middle Mezquital River [B], the Armerı´a-Ayuquila River [H] and the Upper Balsas River [L]. Maximum likelihood values lend support to a vicariant event separating the ancestral populations of the Upper Balsas River from those found in areas B and H, with interrupted gene flow causing diversification approximately 900,000 years ago (Figs 2 and 3C). [D], Cuitzeo Lake [E], Pa´tzcuaro Lake [F] or Zirahue´n Lake [G]; Fig. 3B, only one area shown), with a vicariant event separating the populations in Cotija from the populations of the remaining areas, approximately 170,000 years ago. Subsequent vicariant events, which progressively fragmented the ancestral populations from the areas of present-day central Mexico that once formed part of the greater Paleolakes during the Pleistocene (Figs. 2 and 3B), were uncovered. The ancestral area of M. bravoae Lineage I was found to span the Upper and Middle Mezquital River [B], the Armerı´a-Ayuquila River [H] and the Upper Balsas River [L]. Maximum likelihood values lend support to a vicariant event separating the ancestral populations of the Upper Balsas River from those found in areas B and H, with interrupted gene flow causing diversification approximately 900,000 years ago (Figs 2 and 3C). The results of our second round of DEC analyses (where we equated parasite-host to species-range) reflect the difficulty of inferring ancestral hosts and events that pre-date the Pleistocene. For the MRCA of M. bravoae moderate support (combined relative probabilities .0.5; File S1B) is given to a model of lineage duplication in an ancestral host during the Pliocene, postulated to have been a member of the tribe Characodontini (i), while also being present in a host of the Ilyodontini tribe (g). During the Plio- Pleistocene boundary, a host-switch from the Characodontini (i) to Chapalichthyini (n) occurred in the MRCA of M. bravoae Lineages II and III, followed by the MRCA’s divergence (approx. 1. Divergence times The divergence time estimates for the MRCA of the M. resolanae + M. bravoae clade was around 6.53 Ma (Fig. 2). The divergence time between the ancestor of Lineage I and ancestor of Lineages II and III was dated at 3.20 Ma, while Lineages II and III diverged approximately 1.04 Ma. PLOS ONE \| www.plosone.org July 2014 \| Volume 9 \| Issue 7 \| e101700 4 Margotrema Biogeography: Cophylogeny and Evolution Figure 2. Molecular dating of cladogenetic events. Ultrametric tree derived from BEAST and the combined dataset showing divergence time estimates of selected clades including 95% confidence intervals (blue bars). Black circles above branches represent Bayesian posterior probability (BPP) values $0.95. The terminals (areas/lineages) correspond to codes described in Table 1. doi:10.1371/journal.pone.0101700.g002 Figure 2. Molecular dating of cladogenetic events. Ultrametric tree derived from BEAST and the combined dataset showing divergence time estimates of selected clades including 95% confidence intervals (blue bars). Black circles above branches represent Bayesian posterior probability (BPP) values $0.95. The terminals (areas/lineages) correspond to codes described in Table 1. Figure 2. Molecular dating of cladogenetic events. Ultrametric tree derived from *BEAST and the combined dataset showing divergence time estimates of selected clades including 95% confidence intervals (blue bars). Black circles above branches represent Bayesian posterior probability (BPP) values $0.95. The terminals (areas/lineages) correspond to codes described in Table 1. doi:10.1371/journal.pone.0101700.g002 July 2014 \| Volume 9 \| Issue 7 \| e101700 3. Codivergence patterns 1.04 Ma), as a result of a vicariant event separating the ancestral populations of C. audax (i) in the Mezquital River from the other ancestor that colonised Chapalychthyini (n) in the Lerma River (Fig. 3A). Particularly, in M. bravoae Lineage III, the ancestor underwent PLOS ONE \| www.plosone.org 5 July 2014 \| Volume 9 \| Issue 7 \| e101700 July 2014 \| Volume 9 \| Issue 7 \| e101700 Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution July 2014 \| Volume 9 \| Issue 7 \| e101700 PLOS ONE \| www.plosone.org Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution Figure 3. Biogeographic and host-parasite events of the evolution of Margotrema spp. from central Mexico associated to Goodeinae. A) Evolution of M. resolanae and of each M. bravoae lineage. B) Specific events that influenced the evolution of M. bravoae Lineage I. C) Specific events that influenced the evolution of M. bravoae Lineage III. Squares on branches and at terminal nodes with uppercase letters represent areas and coloured fish shapes with lowercase letters represent current (at terminals) or ancestral (along branches) hosts. The area and host codes correspond to the letters found in Table 1. Additionally, fish shapes are coloured according to their tribes (as in the in-figure Legend). Postulated historical biogeographic and host-parasite events are represented by the different shapes explained in the figure itself. doi:10.1371/journal.pone.0101700.g003 Margotrema in our analyses (see below). This is further complicated by our limited understanding of geological events that took place in central Mexico during the past 7 million years, following a pattern of complex reticulation, i.e. biogeographic scenarios in which areas and biota underwent several events of fragmentation, fusion, and re-fragmentation in iterative cycles of dispersal and vicariance [35]. Similar problems are observed with inferences made regarding the ancestral host at the time of the diversification of the MRCA of Margotrema. When discussing ancestral areas in a freshwater fishes-helminth parasite system it is important to keep in mind that the hydrological configuration most likely was very different during past geological epochs (akin to Sanmartı´n’s [35] ‘‘range evolution’’), even though the ancestral areas may have spanned similar geographic spaces as today. The same applies to the host species, where the present-day species are not the same as the ‘‘ancestral’’ hosts. 3. Codivergence patterns For these reasons, when talking about ancestral areas and hosts, events and areas/hosts receive increasingly higher support values as the reconstructions approach the present. We will therefore discuss the history of each species and lineage of Margotrema in turn, starting from the most derived lineage and moving as far into the past as the statistically supported reconstructions allow. colonisation via host-switching from Zoogoneticus quitzeoensis (n) to Neoophorus regalis (d) in the Lerma River Basin (Fig. 3B; between 900,000 and 200,000 years ago). Subsequently, other host- switching events occurred once the ancestor was in that particular geographic area. For instance, a host-switching event allowed the colonisation of Alloophorus robustus (e), while another five host- switching events permitted the colonisation of A. diazi (a), A. duguesi (b), Allotoca meeki (c), A. zacapuensis (f) and Z. purhepechus (o) (Fig. 3B; from 0.20 Ma to the present). colonisation via host-switching from Zoogoneticus quitzeoensis (n) to Neoophorus regalis (d) in the Lerma River Basin (Fig. 3B; between 900,000 and 200,000 years ago). Subsequently, other host- switching events occurred once the ancestor was in that particular geographic area. For instance, a host-switching event allowed the colonisation of Alloophorus robustus (e), while another five host- switching events permitted the colonisation of A. diazi (a), A. duguesi (b), Allotoca meeki (c), A. zacapuensis (f) and Z. purhepechus (o) (Fig. 3B; from 0.20 Ma to the present). In the case of the MRCA of M. bravoae Lineage I, we found an association with postulated ancestors of A. zonistius (g) + C. audax (i) (since 3.20 Ma, Fig. 3A). Initial lineage duplication (in g) allowed the diversification of M. bravoae Lineage I, with subsequent colonisations of several host species belonging to the Ilyodontini, i.e. in A. zonistius (g), I. furcidens (k) and I. withei (l) (since 1.03 Ma, Fig. 3C). Of particular interest is the history of the present-day M. bravoae Lineage I found Codoma ornata (h) (host species of the family Cyprinidae), where the ancestral population originated in an ancestor of C. audax (i). However, this lineage was lost in this host following a host-switch to C. ornata in the Lower Conchos River (area A) (Fig. 3C). Finally, codivergent events between Goodeinae and Margotrema, as found when comparing of the phylogenetic trees of both associates in TreeMap, showed the following three hierarchical levels of codivergence (Fig. 4; also see Fig. 1). 1. Patterns of regional codivergence of Goodeinae- Margotrema bravoae across central Mexico g 1.1. Girardinichthyini/Chapalichthyini-Margotrema bravoae Lineage III: a Pleistocenic model. Our results indicate that the MCRA of M. bravoae Lineages II and III was associated with the ancestor of the Characodontini and Chapa- lychthyini in the Upper and Middle Mezquital and the Cotija and Lower Lerma rivers between 3.2 and 1.03 Ma. The divergence of Lineages II and III was recovered as a vicariant geological event with lineage codivergence leaving Margotrema Lineage III associ- ated with an ancestral Chapalychthyini. The subsequent diversi- fication of Lineage III took place between 1 million and 167,000 years ago (Fig. 2), which corresponds with mid- to late Pleistocenic events. The results of the DEC model support the idea of the dispersal of Lineage III during the Pleistocene across hydrological systems of the Lower Lerma River and the Cuitzeo and Zirahue´n lakes; apparently, the dispersal and vicariant events occurred in concordance with the diversification processes of their hosts, in turn driven by postulated Pleistocenic events such as the fragmentation of the central Mexican Paleolakes due to a combination of volcanic activity and Paleoclimatic changes [20,36]. Two area relationships are recovered from this model, the first one between the Zacapu, Pa´tzcuaro and Zirahue´n lakes, which further support the idea of ancestral connections between these water bodies [20,22] (Fig. 5A). The second, the Cuitzeo Lake and the Lower Lerma River, supports the ancestral connection of this river basin. These area relationships are also supported by the current distribution patterns of several freshwater fish taxa, an idea originally proposed by A´ lvarez del Villar [37] and more recently by particular phylogenetic and biogeographic analyses [20,36,38]. 1 2 Ch d d M b Li 3. Codivergence patterns Level 1 Species-Species: this level represents strict cospeciation between X. resolanae and M. resolanae in the Cuzalapa River; Level 2 Species-Lineage: a host species exhibits a close association with a parasite lineage, in this case represented by M. bravoae Lineage II parasitising C. audax in the Upper and Middle Mezquital River. This level is herein referred to as Type I Codivergence, in which the divergence of a parasite lineage occurs in response to the speciation event of its host. Level 3 Tribe-Lineage: a monophyletic group of hosts (tribe in this case) exhibits a close association with a parasite lineage. This is represented by M. bravoae Lineage I parasitising members of the tribe Ilyodontyni, distributed in several water bodies of the Balsas River Basin, the Ayuquila River and the Lower Conchos River; and by M. bravoae Lineage III parasitising tribes Chapalichthyni and Girardinichthyni, in several water bodies exclusive to the Lerma River Basin. This level is herein referred to as Type II Codivergence. This codivergence occurs at deeper levels of the phylogenetic history of the hosts (e.g. tribe). In other words, the divergence of the parasite lineage is a result of the diversification process of its host, and subsequent colonisation (host-switching events) of parasite lineages to new hosts strictly belonging to the same tribe (i.e. with close phylogenetic affinities). This may be viewed as vertical transmission sensu lato when considering the host tribe as a phylogenetic unit. Discussion The results of the biogeographic analyses reflect the difficulty of making pre-Pleistocene inferences of ancestral areas and events shaping the present distribution of Margotrema and their hosts since the divergence of the MRCA of the genus Margotrema, ca. 6.53 Ma. Rather than a methodological constraint, this difficulty relates to the fact that we were unable to include the sister group of 1.2. Characodon audax-Margotrema bravoae Lineage II. As mentioned earlier, lineage codivergence of M. bravoae Lineages II and III- Characodontini and Chapalychthyini around July 2014 \| Volume 9 \| Issue 7 \| e101700 PLOS ONE \| www.plosone.org 7 Margotrema Biogeography: Cophylogeny and Evolution Figure 4. Tanglegram depicting the three levels of codivergent associations between Goodeinae and Margotrema. Level 1: Species- Species, representing the association Xenotaenia resolanae-Margotrema resolanae (green line). Level 2: Species-Lineage, representing the association Characodon audax-Margotrema bravoae Lineage II (red line). Level 3: Tribe-Lineage, representing the associations between Ilyodontini-M. bravoae Lineage I (blue lines) and Girardinichthyini/Chapalichthyini-M. bravoae Lineage III (blue lines). Red circles at nodes represent statistically supported codivergence events established by cophylogenetic analysis in TreeMap. doi:10.1371/journal.pone.0101700.g004 Figure 4. Tanglegram depicting the three levels of codivergent associations between Goodeinae and Margotrema. Level 1: Species- Species, representing the association Xenotaenia resolanae-Margotrema resolanae (green line). Level 2: Species-Lineage, representing the association Characodon audax-Margotrema bravoae Lineage II (red line). Level 3: Tribe-Lineage, representing the associations between Ilyodontini-M. bravoae Lineage I (blue lines) and Girardinichthyini/Chapalichthyini-M. bravoae Lineage III (blue lines). Red circles at nodes represent statistically supported codivergence events established by cophylogenetic analysis in TreeMap. doi:10.1371/journal.pone.0101700.g004 Ilyodontini tribe diversified in Ameca, Armerı´a-Ayuquila, and Coahuayana-Tamazula river basins [20], and M. bravoae Lineage I codiversified through a vertical transmission pattern [39] into derived illiodontins. Margotrema bravoae Lineage I is the only lineage found in Ilyodontini, yet Ilyodontini is not the only host group of this lineage. For example, there is a record of M. bravoae Lineage I in a member of the Chapalichthyini (Chapalicthys pardalis -j-), inhabiting the same river drainages (Martı´nez-Aquino et al. unpublished data), recovered as a host-switching and sharing event, congruent with a horizontal transmission pattern [39]. Additionally, M. bravoae Lineage I has also been found in Cyprinidae (Codoma ornata), following a postulated host-switch from and lineage loss in an ancestor of Characodon audax in the Lower Conchos River in the last 500,000 years. Discussion Although goodeids no longer inhabit the Lower Conchos River, this host- switch-extinction event is highly plausible, as a goodeid fossil (Empetrichthys erdisi) was found in the Yaqui River in Sonora, North- western Mexico, a locality close to the area where goodeids are currently distributed (Minckley et al. [40], and citations therein). In addition, recent discoveries of Margotrema sp. associated with other cyprinid fish species from hydrological systems in northern Mexico can be seen as a shadow of past events, possibly reflecting this host-switch-extinction event (Table S1 and Fig. 5B). Ilyodontini tribe diversified in Ameca, Armerı´a-Ayuquila, and Coahuayana-Tamazula river basins [20], and M. bravoae Lineage I codiversified through a vertical transmission pattern [39] into derived illiodontins. Margotrema bravoae Lineage I is the only lineage found in Ilyodontini, yet Ilyodontini is not the only host group of this lineage. For example, there is a record of M. bravoae Lineage I in a member of the Chapalichthyini (Chapalicthys pardalis -j-), inhabiting the same river drainages (Martı´nez-Aquino et al. unpublished data), recovered as a host-switching and sharing event, congruent with a horizontal transmission pattern [39]. Additionally, M. bravoae Lineage I has also been found in Cyprinidae (Codoma ornata), following a postulated host-switch from and lineage loss in an ancestor of Characodon audax in the Lower Conchos River in the last 500,000 years. Although goodeids no longer inhabit the Lower Conchos River, this host- switch-extinction event is highly plausible, as a goodeid fossil (Empetrichthys erdisi) was found in the Yaqui River in Sonora, North- western Mexico, a locality close to the area where goodeids are currently distributed (Minckley et al. [40], and citations therein). In addition, recent discoveries of Margotrema sp. associated with other cyprinid fish species from hydrological systems in northern Mexico can be seen as a shadow of past events, possibly reflecting this host-switch-extinction event (Table S1 and Fig. 5B). 1 Ma left Margotrema Lineage II associated with an ancestor of C. audax. The presence of the ancestor of the Characodontini in Cotija and the Lower Lerma River is incongruent with the current distribution patterns established for the genus Characodon. Never- theless, the phylogenetic history of goodeines shows that the formation of the Salto waterfall dated 1.8 Ma represents a vicariant event that caused the allopatric speciation of C. audax and C. lateralis [20,22] (Fig. 5B). However, no records of Margotrema spp. have been established for C. Discussion lateralis, even though several surveys have been carried out in the area (Martı´nez-Aquino et al. unpublished data). Apparently, the codiverence pattern uncovered in this study, related with the association of C. audax and M. bravoae Lineage II (exclusive to the Upper and Middle Mezquital river), can be explained by an extinction event of ancestral populations of the Lineage II in the Lower Mezquital River, following the event of cladogenesis in the host caused by the Salto waterfall as a geographic barrier. 1 Ma left Margotrema Lineage II associated with an ancestor of C. audax. The presence of the ancestor of the Characodontini in Cotija and the Lower Lerma River is incongruent with the current distribution patterns established for the genus Characodon. Never- theless, the phylogenetic history of goodeines shows that the formation of the Salto waterfall dated 1.8 Ma represents a vicariant event that caused the allopatric speciation of C. audax and C. lateralis [20,22] (Fig. 5B). However, no records of Margotrema spp. have been established for C. lateralis, even though several surveys have been carried out in the area (Martı´nez-Aquino et al. unpublished data). Apparently, the codiverence pattern uncovered in this study, related with the association of C. audax and M. bravoae Lineage II (exclusive to the Upper and Middle Mezquital river), can be explained by an extinction event of ancestral populations of the Lineage II in the Lower Mezquital River, following the event of cladogenesis in the host caused by the Salto waterfall as a geographic barrier. 1.3. Ilyodontini-Margotrema bravoae Lineage I. The differentiation process of the MRCA of M. bravoae Lineage I from Lineages II and III was dated at 3.2 Ma. The MRCA of Lineage I was found to have diversified approximately 900,000 years ago in the areas that today correspond to the Balsas, Armerı´a-Ayuquila and Mezquital Rivers, associated with ancestors of the tribes Characodontini and mainly Ilyodontini (Fig. 3C). The entire July 2014 \| Volume 9 \| Issue 7 \| e101700 PLOS ONE \| www.plosone.org 8 Margotrema Biogeography: Cophylogeny and Evolution Figure 5. Biogeographic history, codivergence patterns and evolution of Margotrema spp. in central Mexico. Hydrological systems i green and orange in Figure 5A correspond to areas where Margotrema were collected in this study and previous records, respectively. 3. Early beginnings of Margotrema: future investigations into the past The evolutionary history and divergence date of M. bravoae Lineage I (3.2 Ma) are congruent with the hypothesis of the origin of the genus Allodontichthys (3.6–2.9 Ma) in the Ameca, Armerı´a- Ayuquila, and Coahuayana-Tamazula river basins, configured by vicariant events [20] (Fig. 5C). These patterns were associated with geological events that shaped the biogeographic history of several freshwater fish taxa, such as the uprising of the Sierras de Manantla´n and Cacoma, the volcanic activity of the Talpa- Mascota graben (geological depression; 3.6 Ma) and the reactiva- tion of the Colima and Tamazula graben in the Pliocene [38,41]. According to Domı´nguez-Domı´nguez et al. [20], the MRCA of Allodontichthys and Ilyodon began diversifying during the early Pleistocene (ca. 2 Ma), dispersing from the Armerı´a-Ayuquila River into several hydrological systems in central Mexico such as Ameca, Balsas, Coahuayana-Tamazula and Purificacio´n-Mascota river basins. Previous records of Margotrema (Martı´nez-Aquino et al. unpublished data) in illyodontins in the Coahuayana river basin (Allodonthichthys hubbsi in El Tule, Jalisco, and A. tamazulae in Tamazula River, Jalisco), are congruent with the hypotheses of area expansion of M. bravoae Lineage I, following the diversifica- tion process of the Ilyodontini tribe. Similarly, these records also support the ancestral connection between the Armerı´a-Ayuquila and Coahuayana hydrological systems where the MRCA of the Ilyodontini was distributed [20]. The evolutionary history and divergence date of M. bravoae Lineage I (3.2 Ma) are congruent with the hypothesis of the origin of the genus Allodontichthys (3.6–2.9 Ma) in the Ameca, Armerı´a- Ayuquila, and Coahuayana-Tamazula river basins, configured by vicariant events [20] (Fig. 5C). These patterns were associated with geological events that shaped the biogeographic history of several freshwater fish taxa, such as the uprising of the Sierras de Manantla´n and Cacoma, the volcanic activity of the Talpa- Mascota graben (geological depression; 3.6 Ma) and the reactiva- tion of the Colima and Tamazula graben in the Pliocene [38,41]. In light of the current available data, any inferences made regarding the geographic range and host association of the MRCA of Margotrema are speculative. The ancestral area of Margotrema may be similar to that of the MRCA for Goodeinae (divergence dated at 15.5 to 8 Ma; see [20] Fig. 5D). Our results neither support nor reject the Nearctic affinity of this digenean, as postulated by Pe´rez- Ponce de Leo´n et al. [42] and later supported by Curran et al. 4. Codivergence patterns This study provides empirical evidence that demonstrates that the historical biogeography and evolutionary history of the digenean Margotrema spp. across central Mexico in part mirrors that of their goodeine freshwater fish hosts at varying levels. The diversification process of the genus Margotrema found in this study is driven by a combination of geography and host-specificity at three hierarchical levels of codivergence: a) Species-Species (Cospeciation), b) Species-Lineage (Type I Codivergence) and c) Tribe-Lineage (Type II Codivergence). A series of biogeographic and host-parasite events explain these three codivergence patterns. These results clearly show codivergent patterns for the Goodeinae-Margotrema association, even though these taxa display intermittent periods of independent evolutionary histories. In this context, the allopatric cospeciation process proposed between X. resolanae-M. resolanae agrees with a one by one model of coevolution [44]. Combining the approaches of historical association (sensu Page and Charleston [32]) and parametric methods in biogeography [35], we can integrate events, processes and patterns on a temporal scale to discriminate between several hypotheses of codivergence. We have shown in this study that the DEC algorithm can successfully be applied to reconstruct host-parasite evolutionary histories in addition to biogeographic processes (for which it was originally intended) and it can complement other methods in detecting fine- scale events. For example, the multiple host-switching events detected by DEC support the vertical transmission patterns of M. bravoae (Lineage I, II and III) throughout the evolutionary history of Goodeinae, allowing for the distinction between ‘‘speciation’’ and ‘‘colonisation’’ events (see [23,45,46]). 3. Early beginnings of Margotrema: future investigations into the past [43], who found a close sister group relationship between Margotrema and Crepidostomum, which is a common intestinal parasite of centrarchiids in other parts of North America. To contribute to a more precise understanding of potential ancestral areas and hosts in this host-parasite system, future studies will need to include the sister groups of Margotrema. According to Domı´nguez-Domı´nguez et al. [20], the MRCA of Allodontichthys and Ilyodon began diversifying during the early Pleistocene (ca. 2 Ma), dispersing from the Armerı´a-Ayuquila River into several hydrological systems in central Mexico such as Ameca, Balsas, Coahuayana-Tamazula and Purificacio´n-Mascota river basins. Previous records of Margotrema (Martı´nez-Aquino et al. unpublished data) in illyodontins in the Coahuayana river basin (Allodonthichthys hubbsi in El Tule, Jalisco, and A. tamazulae in Tamazula River, Jalisco), are congruent with the hypotheses of area expansion of M. bravoae Lineage I, following the diversifica- tion process of the Ilyodontini tribe. Similarly, these records also support the ancestral connection between the Armerı´a-Ayuquila and Coahuayana hydrological systems where the MRCA of the Ilyodontini was distributed [20]. Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution this study (Figure 5B; for more details see Table S1). The evolutionary history of Margotrema is interpreted as follows. A) Hypothetical area of the Pleistocenic palaeolakes where M. bravoae Lineage III is distributed; the orange, blue and green shadows correspond to the phylogenetic relationships of areas. B) Geographic scenario of the diversification of M. bravoae Lineage II associated to a vicariant event (El Salto waterfall) that caused the diversification of Characodon audax and range expansion of MRCA of M. bravoae Lineage III (L3) associated to a complex area of Pleistocenic palaeolakes. The area shaded in red represents a hypothetical ancestral area of Margotrema potentially associated to records of Margotrema from northern Mexico and the distribution area of fossil records of species of Goodeidae. C) Range expansion of M. bravoae Lineage I (L1) along the Balsas River Basin, associated to the biogeographic history of the Ilyodontini [I] tribe; and range expansion of the MRCA of M. bravoae Lineage III (L3) associated to vicariant-dispersal events that caused the diversification of the Chapalichthyini and Girardinichthyini tribes. The yellow arrow refers to the dispersal of L1 influenced by the dispersal of its ilyodontini hosts; dark grey arrows refer to dispersal of L3 influenced by the historical patterns of distribution of the chapalichthyines and girardinichthyines. D) Fragmentation of the ancestral area of the most recent common ancestor (MRCA) of Margotrema spp. (area shaded in red with red arrows). Diversification process of the MRCA of the tribe Ilyodontini and Margotrema, respectively, influenced by vicariant-dispersal events in Cuzalapa River (I) and Armerı´a-Ayuquila River (H). doi:10.1371/journal.pone.0101700.g005 Discussion Full gree circles with a capital letter correspond to localities analysed in this study; full red circles with a number indicate Margotrema records not collected i PLOS ONE \| l J l \| V l \| I \| Figure 5. Biogeographic history, codivergence patterns and evolution of Margotrema spp. in central Mexico. Hydrological systems in green and orange in Figure 5A correspond to areas where Margotrema were collected in this study and previous records, respectively. Full green circles with a capital letter correspond to localities analysed in this study; full red circles with a number indicate Margotrema records not collected in PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org July 2014 \| Volume 9 \| Issue 7 \| e101700 9 References 16. Althoff DM, Segraves KA, Johnson MTJ (2014) Testing for coevolutionary diversification: linking pattern with process. Trends Ecol Evol 29: 82–89. 1. Hoberg E, Adams AM (1992) Phylogeny, historical biogeography, and ecology of Anophryocephalus spp. (Eucestoda: Tetrabothriidae) among pinnipeds of the Holarctic during the late Tertiary and Pleistocene. Can J Zoo 70: 703–719. 17. Pe´rez-Ponce de Leo´n G, Choudhury A (2005) Biogeography of helminth parasites of freshwater fishes in Mexico: the research for patterns and processes. J Biogeogr 32: 645–649. 2. Hoberg E, Adams AM (2000). Phylogeny, history and biodiversity: understand- ing faunal structure and biogeography in the marine realm. Bull Scand Soc Para 10: 19–37. 18. Morrone JJ (2006) Biogeographic areas and transition zones of Latin America and the Caribbean islands based on panbiogeographic and cladistic analyses of the entomofauna. Annu Rev Entomol 51: 467–494. 3. Hafner MS, Nadler SA (1990) Cospeciation in host–parasite assemblages: comparative analysis of rates of evolution and timing of cospeciation. Syst Zool 39: 192–204. 19. Miller RR, Smith ML (1986) Origin and geography of the fishes of Central Mexico. In Hocutt CH, Wiley EO, editors. The Zoogeography of North American Freshwater Fishes. John Wiley and Sons, Inc. New York. pp. 487– 517. 4. Morand S, Krasnov BR (2010) The Biogeography of Host-Parasite Interactions. Oxford, University Press Inc, New York. 277 p. 4. Morand S, Krasnov BR (2010) The Biogeograp Oxford, University Press Inc, New York. 277 p Oxford, University Press Inc, New York. 277 p 5. Froeschke G, von der Heyden S. 2014. A review of molecular approaches for investigating patterns of coevolution in marine host-parasite relationships. Adv. Parasitol 84: 209–252. 20. Domı´nguez-Domı´nguez O, Pedraza-Lara C, Gurrola-Sa´nchez N, Pe´rez- Rodrı´guez R, Israde-Alca´ntara I, et al. (2010) Historical biogeography of the Goodeinae (Cyprinodontiforms). In Uribe-Aranzabal MC, Grier H, editors. Viviparous Fishes II. New Life Publications, Homestead, Florida. pp. 33–74. 6. Ronquist F, Sanmartı´n I (2011) Phylogenetic methods in biogeography. Annu Rev Ecol Evol Syst 42: 441–464. y 7. Page RDM (2003) Tangled Trees: Phylogeny, Cospeciation, and Coevolution. Chicago: University of Chicago, Press. 350 p. 7. Page RDM (2003) Tangled Trees: Phylogeny, C Chicago: University of Chicago, Press. 350 p. 21. Martı´nez-Aquino A, Ceccarelli FS, Pe´rez-Ponce de Leo´n G (2013) Molecular phylogeny of the genus Margotrema (Digenea: Allocreadiidae), parasitic flatworms of goodeid freshwater fishes across central Mexico: species boundaries, host- specificity and geographical congruence. Zool J Linn Soc-Lond 168: 1–16. http://dx.doi.org/10.1111/zoj.12027 8. Acknowledgments We thank C. Pedraza-Lara for the information used for the phylogenetic analysis and molecular evolution of Goodeinae. Professor Julio Rozas, A. Sa´nchez-Gracia and R. Pe´rez-Rodrı´guez for earlier discussions of this manuscript. AMA and FSC thank the Swiss Systematics Society for the course Model-based methods in Biogeography given by Professor Isabel Sanmartı´n during the 5th. Scientific meeting of the Swiss Systematics Society. We thank Dan Proud for help with language editing and two anonymous reviewers for their constructive criticisms. Margotrema Biogeography: Cophylogeny and Evolution the parasites have shorter generation times compared to their hosts [47,48]. However, this study shows a higher diversity of host species compared to Margotrema on a similar timescale. In addition, individuals of M. bravoae Lineage I were found in hydrological systems that are more than 1600 km apart and have been separated for at least 5 Ma [20], suggesting that minimal diversification has taken place in the parasite during the last 900,000 years. These different rates of diversification can be explained by diversifying selection acting more strongly on the fishes, due to their more variable external environment relative to the one of the endohelminth parasites. Divergence times for each host and parasite lineage - keeping in mind that we are talking about different taxonomic levels for the two - are relatively congruent and demonstrate the concordance of the evolutionary and biogeographic history between the two taxa, both in time and space, i.e. the divergence date of C. audax is 1.5 Ma, while for M. bravoae lineage II it was estimated at 1.04 Ma. This is in agreement with Page [23] mentioning that the relative ages of speciation events in the host and parasite lineages are key to reconstructing the history of the host-parasite assemblages [49,50]. The coevolutionary patterns herein described follow a geographic mosaic in which the populations differ in their characteristics and specialisations with respect to the species with which they interact [51]. Therefore, the patterns of codivergence uncovered for the Goodeinae-M. bravoae associations occur at the three aforemen- tioned levels, and are thus congruent with the geographic mosaic theory. This might also be related with the macroevolutionary mosaic as described by Hoberg and Brooks [52], where environmental changes may drive both the persistence and diversification of host-parasite systems, generating opportunities for host-switching during geographic expansion, but also permit- ting cospeciation during episodes of geographic isolation. Finally, the geographic scenario produced by the complex geologic and climatic history of the region seems to be the most important determinant that drives the evolution of Margotrema spp. and its codivergent association with their goodeine hosts. Author Contributions Conceived and designed the experiments: AMA GPPL. Performed the experiments: AMA FSC. Analyzed the data: AMA FSC. Contributed reagents/materials/analysis tools: GPPL LEE EVD. Wrote the paper: AMA FSC LEE EVD GPPL. Figure S1 Hydrological systems of this study. (PDF) Figure S1 Hydrological systems of this study. (PDF) Table S1 Additional records of Margotrema in fresh- water fish species from Mexico. (DOCX) File S1 Detailed output of DEC analyses in Lagrange for ancestral area (S1A) and host (S1B) reconstructions. (DOC) 2. Xenotaenia resolanae-Margotrema resolanae: an allopatric cospeciation model p p The separation of M. resolanae from its MRCA with M. bravoae was dated at approximately 6.53 Ma, meaning that the biogeo- graphic history of M. resolanae in particular is difficult to reconstruct, especially given the many postulated reticulate geological changes that have taken place in the area during the last 6 to 7 Ma. Nevertheless, the strict host-specificity shown by M. resolanae alludes to cospeciation, further supported by the reconstruction carried out in TreeMap. The restricted distribution range of both species may be the result of allopatric speciation by peripheral isolates, where populations of both hosts and parasites were isolated at the edge of the distributional range of the MRCAs, resulting in a strong host-specificity pattern that illustrates a classical model of cospeciation reflecting reciprocal selection, i.e. coevolution. The divergence date of the MRCA of Illydontini (6.9 Ma) and the posterior diversification process of X. resolanae, restricted to the Purificacio´n-Mascota River Basin (between 5.6 to 5.1 Ma) support this pattern. Unfortunately, geological information of this region is scarce, where only the volcanic activity of the Talpa-Mascota graben dated ca. 4.6 Ma has been documented [20]. The codivergent patterns we uncovered between Goodeinae and M. bravoae (Lineages I, II and III) support Vernon Kellog’s parasitological rule in that ‘‘the parasites evolve more slowly than their hosts’’ [33]. The intuitive expectation would be that digeneans should evolve more rapidly than their freshwater fish hosts, since PLOS ONE \| www.plosone.org July 2014 \| Volume 9 \| Issue 7 \| e101700 July 2014 \| Volume 9 \| Issue 7 \| e101700 10 Margotrema Biogeography: Cophylogeny and Evolution 1. Hoberg E, Adams AM (1992) Phylogeny, historical biogeography, and ecology of Anophryocephalus spp. (Eucestoda: Tetrabothriidae) among pinnipeds of the Holarctic during the late Tertiary and Pleistocene. Can J Zoo 70: 703–719. Margotrema Biogeography: Cophylogeny and Evolution Margotrema Biogeography: Cophylogeny and Evolution tectonism. In Hocutt CH, Wiley EO, editors. The Zoogeography of North American Freshwater Fishes. John Wiley and Sons, Inc. New York. pp. 519– 613. tectonism. In Hocutt CH, Wiley EO, editors. The Zoogeography of North American Freshwater Fishes. John Wiley and Sons, Inc. New York. pp. 519– 613. 28. Ferrari L, Lo´pez-Martı´nez M, Aguirre-Dı´az G, Carrasco-Nu´n˜ez G (1999) Space-time patterns of Cenozoic arc volcanism in central Mexico: From the Sierra Madre Occidental to the Mexican Volcanic Belt. Geology 27: 303–306. gy 29. Ferrari L (2004) Slab detachment control on volcanic pulse and mantle heterogeneity in Central Mexico. Geology 32: 77–80. 41. Mateos M, Sanjur OI, Vrijenhoek RC (2002) Historical biogeography of the livebearing fish genus Poeciliopsis (Poeciliidae: Cyprinodontiformes). Evolution 56: 972–984. 30. Miller MA, Pfeiffer W, Schwartz T (2010) ‘‘Creating the CIPRES Science Gateway for inference of large phylogenetic trees’’ in Proceedings of the Gateway Computing Environments Workshop (GCE), 14 Nov. 2010, New Orleans, LA. 1–8 p. 42. Pe´rez-Ponce de Leo´n G, Choudhury A, Rosas-Valdez R, Mejı´a-Madrid H (2007) Systematic position of Wallinia spp. and Margotrema spp. parasites of Middle-American and Neotropical freshwater fishes based on 28S ribosomal RNA gene. Syst Parasitol 68: 49–55. 31. Hijmans RJ, Guarino L, Mathur P (2012) DIVA-GIS version 7.5 manual. California, USA: University of California Davis. Manual available at http:// www.diva-gis.org/ 43. Curran SS, Tkach VV, Overstreet RM (2011) Phylogenetic affinities of Auriculostoma (Digenea: Allocreadiidae), with descriptions of two new species from Peru. J Parasitol 97: 661–670. 32. Page RDM, Charleston MA (1998). Trees within trees: phylogeny and historical associations. TREE 13: 356–359. J 44. Ehrlich PR, Raven PH (1964) Butterflies and Plants: a study in Coevolution. Evolution 18: 586–608. 33. Choudhury A, Moore BR, Marques FL (2002) Vernon Kellog, host-switching, and cospeciation: rescuing straggled ideas. J Parasitol 88: 1045–1048. 45. Charleston MA, Perkins SL (2006) Traversing the tangle: Algorithms and applications for cophylogenetic studies. J Biomed Inform 39: 62–71. 34. Maddison WP, Maddison DR (2011) MESQUITE: A Modular System for Evolutionary Analysis. Version 2.72. Available: http://mesquiteproject.org. Accessed 2012 Nov. pp p y g J 46. Keller-Schmidt S, Wieseke N, Klemm K, Middendorf M (2011) Evaluation of host parasite reconciliation methods using a new approach for cophylogeny generation. Working paper from Bioinformatics Leipzig. URL http://www. bioinf.uni-leipzig.de/working/11-013 35. Sanmartı´n I (2012) Historical Biogeography: Evolution in time and space. Evo Edu Outreach 5: 555–568. doi: 10.1007/s12052-012-0421-2 47. References Ree RH, Moore BR, Webb CO, Donoghue MJ (2005) A likelihood framework for inferring the evolution of geographic range on phylogenetic trees. Evolution 59: 2299–2311. 9. Ree RH, Smith SA (2008) Maximum likelihood inference of geographic range evolution by dispersal, local extinction, and cladogenesis. Sys Biol 57: 4–14. 22. Domı´nguez-Domı´nguez O, Doadrio I, Pe´rez-Ponce de Leo´n G (2006) Historical biogeography of some river basins in central Mexico evidenced by their goodeine freshwater fishes: a preliminary hypothesis using secondary Brooks parsimony analysis. J Biogeogr 33: 1437–1447. 10. Sanmartı´n I, Van der Mark P, Ronquist F (2008) Inferring dispersal: a Bayesian approach to phylogeny based island biogeography, with special reference to the Canary Islands. J Biogeogr 35: 428–449. 23. Page RDM (1994) Parallel phylogenies: reconstructing the history of host- parasite assemblages. Cladistics 10:155–73. 11. Heled J, Drummond AJ (2010) Bayesian inference of species trees from multilocus data. Mol Biol Evol 27: 570–80. 24. Drummond AJ, Suchard MA, Xie D, Rambaut A (2012) Bayesian phylogenetics with BEAUti and the BEAST 1.7. Mol Biol Evol 29: 1969–1973. 12. Charleston MA (2011) TreeMap 3b. Available: http://sites.google.com/site/ cophylogeny. Accessed 2012 Nov. 25. Heneghan C, Thompson M, Billingsley M, Cohen D (2011) Data from: Medical-device recalls in the UK and the device-regulation process: retrospec- tive review of safety notices and alerts. Dryad Digital Repository. http://dx.doi. org/10.5061/dryad.585t4 13. Ceccarelli FS, Zaldı´var-Rivero´n A (2013) Broad polyphyly and historical biogeography of the Neotropical wasp genus Notiospathius (Braconidae, Doryctinae). Mol Phylogenet Evol 69: 142–152. 14. Nieberding CM, Durette-Desset MC, Vanderpoorten A, Casanova JC, Ribas A, et al. (2008) Geography and host biogeography matter for understanding the phylogeography of a parasite. Mol Phylogenet Evol 47: 538–554. 26. Drummond AJ, Ho SYW, Phillips MJ, Rambaut A (2006) Relaxed phylogenetics and dating with confidence. PLoS Biol 4(5): e88. 15. Caira JN, Jensen K (2001) An investigation of the co-evolutionary relationships between onchobothriid tapeworms and their elasmobranch hosts. Int J Parasitol 31: 960–975. 27. Drummond AJ, Ashton B, Buxton S, Cheung M, Cooper A, et al. (2010) Geneious v5.0.4 Available at:http://www.geneious.com/ July 2014 \| Volume 9 \| Issue 7 \| e101700 11 PLOS ONE \| www.plosone.org Margotrema Biogeography: Cophylogeny and Evolution Huyse T, Audenaert V, Volckaert FAM (2003) Speciation and host-parasite relationships in parasite genus Gyrodactylus (Monogenea, Platyhelminthes) infecting gobies of the genus Pomatoschistus (Gobiidae, Teleostei). Int J Parasitol 33: 1679–1689. 36. Domı´nguez-Domı´nguez O, Alda F, Pe´rez-Ponce de Leo´n G, Garcı´a-Garitagoi- tia JL, Doadrio I (2008) Evolutionary history of the endangered fish Zoogoneticus quitzeoensis (Bean, 1898) (Cyprinodontiformes: Goodeidae) using a sequential approach to phylogeography based on mitochondrial and nuclear DNA data. BMC Evol Biol 8: 161. ´ 48. Price PW (1980) Evolutionary biology of parasites. Princeton, NJ, Princeton University 237 p. 37. A´ lvarez del Villar J (1972) Ictiologı´a Michoacana V. Origen y distribucio´n de la ictiofauna dulceacuı´cola michoacana. An Esc Nac Cienc Biol 19: 155–161. 49. Hafner MS, Sudman PD, Villablanca FX, Spradling TA, Demastes JW et al. (1994) Disparate rates of molecular evolution in cospeciating hosts and parasites. Science 265: 1087–1090. 38. Pe´rez-Rodrı´guez R, Domı´nguez-Domı´nguez O, Pe´rez-Ponce de Leo´n G, Doadrio I (2009) Phylogenetic relationships and biogeography of the genus Algansea Girard (Cypriniformes: Cyprinidae) of central Mexico inferred from molecular data. BMC Evol Biol 9: 223. 50. Page RDM (1996) Temporal congruence revisited: comparison of mitocondrial DNA sequence divergence in cospeciating pocket gophers and their chewing lice. Syst Biol 45:151–67 39. Rannala B, Michalakis Y (2003) Populations genetics and cospeciation: from process to pattern. In Page RDM, editor. Tangled Trees: Phylogeny, Cospeciation, and Coevolution. The University of Chicago Press: Chicago and London, U.S.A. pp. 120–143. y 51. Thompson JN (2005) The Geographic Mosaic of Coevolution. University of Chicago Press, Chicago. p. 443. 52. Hoberg EP, Brooks DR (2008) A macroevolutionary mosaic: episodic host- switching, geographical colonization and diversification in complex host-parasite systems. J Biogeogr 35: 1533–1550. pp 40. Minckley WL, Hendrickson DA, Bond CE (1986) Geography of western North American freshwater fishes: description and relationships to intracontinental PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org July 2014 \| Volume 9 \| Issue 7 \| e101700 12
https://openalex.org/W1970640173	https://molecularneurodegeneration.biomedcentral.com/track/pdf/10.1186/1750-1326-8-S1-P32	English	null	Peripheral inflammation increases PKR activation, Tau phosphorylation and amyloid β production in wild-type mice	Molecular neurodegeneration	2,013	cc-by	469	Authors’ details 1 1Inserm UMR-S839, Paris, France. 2Hopital Lariboisière APHP, Paris, France. 3University of Southampton, Southampton, UK. Learning objective Published: 13 September 2013 Published: 13 September 2013 To investigate whether PKR-mediated neuroinflammation could play a role in AD neurodegenerative process. doi:10.1186/1750-1326-8-S1-P32 Cite this article as: Mouton-Liger et al.: Peripheral inflammation increases PKR activation, Tau phosphorylation and amyloid b production in wild-type mice. Molecular Neurodegeneration 2013 8(Suppl 1):P32. Background Systemic inflammation is correlated with dementia pro- gression. Pro-inflammatory molecules can communicate from the periphery to the central nervous system to induce neuroinflammation and neurodegeneration. Our protein of interest is the pro-apoptotic kinase PKR (the double strand-RNA dependent protein kinase). Increased activated PKR levels were found in AD patients brain and cerebrospinal fluid. PKR activation can be triggered by inflammatory stresses and induces neurotoxicity in vitro. Is in vivo PKR-mediated inflammation involved in AD neurodegenerative process? 1Inserm UMR-S839, Paris, France Full list of author information is available at the end of the article Mouton-Liger et al. Molecular Neurodegeneration 2013, 8(Suppl 1):P32 http://www.molecularneurodegeneration.com/content/8/S1/P32 Mouton-Liger et al. Molecular Neurodegeneration 2013, 8(Suppl 1):P32 http://www.molecularneurodegeneration.com/content/8/S1/P32 Mouton-Liger et al. Molecular Neurodegeneration 2013, 8(Suppl 1):P32 http://www.molecularneurodegeneration.com/content/8/S1/P32 Open Access Open Access Peripheral inflammation increases PKR activation, Tau phosphorylation and amyloid b production in wild-type mice François Mouton-Liger1,2*, Anne-Sophie Rebillat1, Clarisse Pace1, Sarah Gourmaud1, Mariko Taga1,3, Claire Paquet1,2, Jacques Hugon1,2 From Molecular Neurodegeneration: Basic biology and disease pathways Cannes, France. 10-12 September 2013 Microglial activation, neuronal apoptosis, increase of PKR, GSK3b and Tau phosphorylation and amyloid b production were found in hippocampus and cortex of LPS-treated mice. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review Conclusions PKR could be involved in the signalling of neurofibri- llary tangles formation after a systemic inflammatory challenge. Methods C57BL/6 wild type mice were injected intraperitoneally with LPS (1mk/kg) versus saline once a day for 3 days to induce PKR activation and neuroinflammation (LPS is the bacilli gram negative endotoxin lipopolysaccharide). Brains were collected and dissected; immunohisto- chemistry and western blotting were performed for neu- roinflammation, PKR activation and AD pathological hallmarks (as Tau hyperphosphorylation). © 2013 Mouton-Liger 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. Results Mice showed endotoxin-induced sickness behaviour includingbody weight loss and elevated serum cytokine levels. 1Inserm UMR-S839, Paris, France Full list of author information is available at the end of the article © 2013 Mouton-Liger 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.
https://openalex.org/W2964017340	https://link.springer.com/content/pdf/10.1140/epjc/s10052-018-6045-0.pdf	English	null	Khuri–Treiman equations for $$\pi \pi $$ scattering	European physical journal. C, Particles and fields	2,018	cc-by	15,193	Received: 28 March 2018 / Accepted: 2 July 2018 / Published online: 13 July 2018 © The Author(s) 2018 Abstract The Khuri–Treiman formalism models the partial-wave expansion of a scattering amplitude as a sum of three individual truncated series, capturing the low-energy dynamics of the direct and cross channels. We cast this for- malism into dispersive equations to study ππ scattering, and compare their expressions and numerical output to the Roy and GKPY equations. We prove that the Khuri–Treiman equations and Roy equations coincide when both are trun- cated to include only S- and P-waves. When higher partial waves are included, we ﬁnd an excellent agreement between the Khuri–Treiman and the GKPY results. This lends cre- dence to the notion that the Khuri–Treiman formalism is a reliable low-energy tool for studying hadronic reaction amplitudes. S-matrix theory, such as analyticity, unitarity, and crossing symmetry for the analysis of three-particle ﬁnal states are mandatory. One of the main issues posed by the presence of hadrons in any reaction is their ﬁnal-state interactions, which are for- mally expressed in terms of the unitarity of the T -matrix. In the case of 2 →2 scattering, this effect can often be incorpo- rated by neglecting unitarity in the t- and u-channels while preservings-channelunitarity.Forexample,inthecaseofππ scattering, the results obtained for the σ meson with the Roy equations (or other dispersive approaches that account for the left-hand cut singularities in a nonperturbative way) are very similar to those obtained with approaches that neglect alto- gether the left-hand cut, or take it into account perturbatively (see for example Refs. [7–9] and references therein). Despite crossing symmetry, this is certainly not the case for a 1 →3 decay, where ideally one wants to take into account unitarity in the three possible two-hadron channels in the ﬁnal state. a e-mail: miguelalbaladejo@gmail.com b e-mail: nlsherri@iu.edu Regular Article - Theoretical Physics Regular Article - Theoretical Physics Khuri–Treiman equations for ππ scattering M. Albaladejo1,a , N. Sherrill2,3,b, C. Fernández-Ramírez4, A. Jackura2,3, V. Mathieu5, M. Mikhasenko6, J. Nys2,3,5,7, A. Pilloni5, A. P. Szczepaniak2,3,5, Joint Physics Analysis Center 1 Departamento de Física, Universidad de Murcia, 30071 Murcia, Spain 2 Center for Exploration of Energy and Matter, Indiana University, Bloomington, IN 47403, USA 3 Physics Department, Indiana University, Bloomington, IN 47405, USA 4 Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, 04510 Mexico City, Mexico 5 Theory Center, Thomas Jefferson National Accelerator Facility, Newport News, VA 23606, USA 6 Helmholtz-Institut für Strahlen- und Kernphysik, Universität Bonn, 53115 Bonn, Germany 7 Department of Physics and Astronomy, Universiteit Gent, Gent, Belgium M. Albaladejo1,a , N. Sherrill2,3,b, C. Fernández-Ramírez4, A. Jackura2,3, V. Mathieu5, M. Mikhasenko6, J. Nys2,3,5,7, A. Pilloni5, A. P. Szczepaniak2,3,5, Joint Physics Analysis Center 1 Departamento de Física, Universidad de Murcia, 30071 Murcia, Spain 2 Center for Exploration of Energy and Matter, Indiana University, Bloomington, IN 47403, USA 3 Physics Department, Indiana University, Bloomington, IN 47405, USA 4 Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, 04510 Mexico City, Mexico 5 Theory Center, Thomas Jefferson National Accelerator Facility, Newport News, VA 23606, USA 6 Helmholtz-Institut für Strahlen- und Kernphysik, Universität Bonn, 53115 Bonn, Germany 7 Department of Physics and Astronomy, Universiteit Gent, Gent, Belgium M. Albaladejo1,a , N. Sherrill2,3,b, C. Fernández-Ramírez4, A. Jackura2,3, V. Mathieu5, M. Mikhasenko6, J. Nys2,3,5,7, A. Pilloni5, A. P. Szczepaniak2,3,5, Joint Physics Analysis Center 1 Departamento de Física, Universidad de Murcia, 30071 Murcia, Spain 2 Center for Exploration of Energy and Matter, Indiana University, Bloomington, IN 47403, USA 3 Physics Department, Indiana University, Bloomington, IN 47405, USA 4 Instituto de Ciencias Nucleares, Universidad Nacional Autónoma de México, 04510 Mexico City, Mexico 5 Theory Center, Thomas Jefferson National Accelerator Facility, Newport News, VA 23606, USA 6 Helmholtz-Institut für Strahlen- und Kernphysik, Universität Bonn, 53115 Bonn, Germany 7 Department of Physics and Astronomy, Universiteit Gent, Gent, Belgium Received: 28 March 2018 / Accepted: 2 July 2018 / Published online: 13 July 2018 © The Author(s) 2018 Eur. Phys. J. C (2018) 78:574 https://doi.org/10.1140/epjc/s10052-018-6045-0 1 Introduction Three-body decays offer a unique window into hadron dynamics. They are an especially useful tool for exploring the hadron spectrum in the exotic sectors, where resonances appear that cannot be accurately described by constituent quark models. Some notable examples are the mysterious XYZ peaks observed in three-body decays of heavy quarko- nia [1–3]. Moreover, new methods have recently been devel- opedthatenabledirectmappingfromthree-particlespectrain a ﬁnite volume to three-particle scattering amplitudes in the inﬁnite volume, opening the door to lattice QCD calculations [4–6]. Generally speaking, robust methods for constructing reaction amplitudes that fulﬁll well-known properties from In the 1960s, Khuri and Treiman proposed a simple ampli- tude model to study K →3π decays [10]. This model is based on the factorization of the scattering amplitude A(s, t, u) into a sum of three functions, each of which depends on a single Mandelstam variable only. Several more studies later appeared that expanded upon this representation of the amplitude [11–16] (see also the recent lectures in Ref. [17]). For the lowest waves, this approach, which we refer to as the Khuri–Treiman (KT) formalism, can be justiﬁed in chiral perturbation theory (χPT) at lowest order via the so- called reconstruction theorem [18–20]. From a broader point of view, as we will discuss in detail below, the KT formalism is a rather simple approach for modeling A(s, t, u). Gener- ally speaking, an inﬁnite sum of s-channel partial waves that contain both right-hand cut (RHC) and left-hand cut (LHC) 12 3 3 Eur. Phys. J. C (2018) 78 :574 574 Page 2 of 14 574 Page 2 of 14 terms of a single function A(s, t, u) and permutations of its arguments by virtue of crossing symmetry. Bose symmetry also requires this function to be symmetric in the second and third variable, A(s, t, u) = A(s, u, t). Our primary interest in the structure Ai jkl(s, t, u) is its decomposition into ampli- tudes of well-deﬁned isospin in the s-channel, A(I)(s, t, u), which is accomplished by means of projectors: discontinuities is substituted by a ﬁnite sum of s-, t-, and u-channel isobar amplitudes, each of which exhibit only a RHC structure emanating from unitarity in the respective channel. Although it is known that this model fails to prop- erly account for asymptotic behavior at high energies, it is expected to accurately describe the physics at low energies. 1 For a detailed derivation of this decomposition we refer the reader to Refs. [41,42]. 1 Introduction (1) and (2): ⎡ ⎣ A(0)(s, t, u) A(1)(s, t, u) A(2)(s, t, u) ⎤ ⎦= ⎡ ⎣ 3 1 1 0 1 −1 0 1 1 ⎤ ⎦ ⎡ ⎣ A(s, t, u) A(t, s, u) A(u, t, s) ⎤ ⎦ ≡K ⎡ ⎣ A(s, t, u) A(t, s, u) A(u, t, s) ⎤ ⎦. (4) ≡K ⎡ ⎣ A(s, t, u) A(t, s, u) A(u, t, s) ⎤ ⎦. (4) (4) The left-hand side of Eq. (4) can be decomposed into an inﬁnite series of s-channel partial-wave amplitudes t(I) l (s): A(I)(s, t, u) = ∞ ℓ=0 (2ℓ+ 1)Pℓ(zs)t(I) ℓ(s), (5) t(I) ℓ(s) = 1 2 +1 −1 dzs Pℓ(zs)A(I)(s, t(s, zs), u(s, zs)), (6) (5) (6) 1 Introduction For this reason, it has often been used to study meson decays, where the energy range is limited by the decay kinematics. This formalism has recently been reviewed and applied to various three-body decay channels of light and heavy mesons [21–32]. Ai jkl(s, t, u) = 2 I=0 P(I) i jkl A(I)(s, t, u), (2) P(0) i jkl = 1 3δi jδkl, P(1) i jkl = 1 2 δikδ jl −δilδ jk , P(2) i jkl = 1 2 δikδ jl + δilδ jk −1 3δi jδkl. (3) (2) Because the KT formalism has been applied to such a wide range of processes, it is important to establish and validate its range of applicability. This is what we propose to do in this paper. Speciﬁcally, we compare the KT model amplitude for ππ scattering with the results of other, arguably more sophis- ticated dispersive approaches. In the same spirit, we will not strictly enforce elastic unitarity, unlike in Refs. [10,21–32]. The manuscript is organized as follows. In Sect. 2 we present our notation for the general description of ππ scattering, as well as the form of the scattering amplitude in the KT formalism. In Sect. 3 we analytically compare the KT and Roy equations [33] for the lowest partial waves. We explic- itly demonstrate that both formalisms exactly coincide when truncated to S- and P-waves, complementing former results discussed in Refs. [34–39]. In Sect. 4 we numerically study the results obtained with the KT equations considering also D- and F-waves, using as an input the parameterization of the ππ scattering shift in Ref. [40]. We compare the results obtained with the latter parameterization and with the results of the GKPY equations discussed in the same work. The con- clusions are given in Sect. 5. In the Appendix we speciﬁcally discuss the contributions of the LHC to the partial waves. (3) The amplitudes A(I)(s, t, u) can be written, in turn, as com- binations of the amplitude A(s, t, u) and permutations of the arguments by comparing Eqs. (1) and (2): The amplitudes A(I)(s, t, u) can be written, in turn, as com- binations of the amplitude A(s, t, u) and permutations of the arguments by comparing Eqs. 2 Amplitudes where Pℓ(zs) are Legendre polynomials in the variable zs, the s-channel scattering angle, which has the form We begin by brieﬂy reviewing the structure of ππ scattering as it relates to our analysis of the KT equations. The gen- eral form of the isopin invariant ππ scattering amplitude is determined from the matrix elements of the transition-matrix operator ˆT : zs = zs(s, t, u) = t −u 4p2(s). (7) (7) Additionally, the cross-channel variables under this projec- tion are given by 1 32π ⟨πk(p3)πl(p4)\| ˆT \|πi(p1)π j(p2)⟩≡Ai jkl(s, t, u) = δi jδkl A(s, t, u) + δikδ jl A(t, s, u) + δilδ jk A(u, t, s). (1) t(s, zs) = −2p2(s)(1 −zs), (8) u(s, zs) = −2p2(s)(1 + zs), (9) (8) (8) (8) (9) (9) (1) where p2(s) = (s −4m2)/4 is the momentum squared in the center-of-mass frame, and m is the charge-averaged pion mass. The normalization of the partial-wave amplitudes is chosen such that TheLatinindicesinEq.(1)denoteCartesianisospinindices.1 The variables s, t, and u refer to the usual Mandelstam vari- ables. In addition, the invariant amplitude can be expressed in TheLatinindicesinEq.(1)denoteCartesianisospinindices.1 The variables s, t, and u refer to the usual Mandelstam vari- ables. In addition, the invariant amplitude can be expressed in t(I) ℓ(s) = η(I) ℓ(s)e2iδ(I) ℓ(s) −1 2iσ(s) , (10) 1 For a detailed derivation of this decomposition we refer the reader to Refs. [41,42]. (10) 3 12 Eur. Phys. J. C (2018) 78 :574 Page 3 of 14 574 A(s, t, u) = ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zs)p2ℓ(s) ¯a(0) ℓ(s) −¯a(2) ℓ(s) 3 + ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zt)p2ℓ(t) ¯a(1) ℓ(t) + ¯a(2) ℓ(t) 2 + ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zu)p2ℓ(u) ¯a(1) ℓ(u) + ¯a(2) ℓ(u) 2 (−1)ℓ, (14) where σ(s) = 1 −4m2 s = 2p(s)/√s. The threshold parameters (which are dimensionless with our deﬁnitions) are deﬁned through e threshold deﬁnitions) (11) ) i hi A(s, t, u) = ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zs)p2ℓ(s) ¯a(0) ℓ(s) −¯a(2) ℓ(s) 3 + ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zt)p2ℓ(t) ¯a(1) ℓ(t) + ¯a(2) ℓ(t) 2 + ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zu)p2ℓ(u) ¯a(1) ℓ(u) + ¯a(2) ℓ(u) 2 (−1)ℓ, m2ℓ p2ℓ(s)Re t(I) ℓ(s) = a(I) ℓ + b(I) ℓ p2(s) m2 + · · · . (11) (11) ℓ=0 (14) (14) We now wish to describe the amplitude A(s, t, u) within the KT formalism. 2 Amplitudes This involves truncating the inﬁnite series of s-channel partial-wave amplitudes t(I) ℓ(s), which contain both RHC and LHC structure. Truncating this expansion at some maximum orbital angular momentum ℓmax deﬁnes an amplitude that is regular in the variables t and u. The singu- larities generated by t- and u-channel physics can be partially recovered by adding truncated t- and u-channel series expan- sions of partial-wave-like functions [24,28,31,43]: where zt and zu are the t- and u-channel center-of-mass scat- tering angles, zt = zt(s, t, u) = s −u 4p2(t), (15) zu = zu(s, t, u) = t −s 4p2(u). (16) (15) (16) A(s, t, u) = ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zs)p2ℓ(s)as ℓ(s) + (s →t) + (s →u) , (12) For the rest of the paper we will use the condensed notation zt(s, z) = zt(s, t(s, z), u(s, z)) and the analogous notation for zu. In Eq. (14) it should be understood that ¯a(I) ℓ = 0 unless ℓ+ I is an even integer so that Bose symmetry is respected. We remind the reader that the speciﬁc choice of the nomenclature in Eq. (14) is such that the well-deﬁned isospin amplitudes A(I)(s, t, u) have isobars ¯a(I) ℓ(s) in the s- channel projection. More speciﬁcally, inserting Eq. (14) into Eq. (4), one obtains (12) where the functions as ℓ(s), at ℓ(t), and au ℓ(u) are isobars in the indicated channel. By assumption, the isobars contain only RHC singularities in their respective channel variable and thus exclusively account for the singularity structure of A(s, t, u). It is important to recognize that the isobars are not independent functions: the symmetry A(s, t, u) = A(s, u, t) implies at ℓ(t) = (−1)ℓau ℓ(t). Note that crossing symmetry is respected in Eq. (12) since we take ℓmax to be the same integer for each truncated series and the inclusion of the p2ℓfactors enforces the proper behavior of A(s, t, u) near threshold [44, 45]. As we mentioned in the introduction, this representation oftheamplitudeisconsistentwiththereconstructiontheorem [18], but only if ℓmax ⩽1. In this regard, the representation given by Eq. (12) is more general, however, its effectiveness is restricted to low energies and partial waves, as it lacks Mandelstam analyticity in the angular momentum plane. We accept the limitations of this representation and aim to test the limits of its effectiveness. 2 Strictly speaking, it is more correct to write n = nℓ,I since the number of subtractions performed can vary for each wave. While this possibility has been explored, the results presented in Sect.4 consider only the same number of subtractions for all waves. 2 Amplitudes Inserting this dispersive representation into Eq. (17), and the resulting amplitudes A(I)(s, t, u) into the partial- wave projection, Eq. (6), one obtains the KT representation of the partial-wave amplitudes of well-deﬁned isospin in the s-channel, P(I) ℓ (s) = ℓ′,I ′ n−1 j=0 (s −s1) j j! ¯a(I ′)( j) ℓ′ (s1)δℓℓ′δI I ′ + (2ℓ′ + 1)CI I ′ R( j) ℓℓ′ (s) p2ℓ(s) ¯a(I ′)( j) ℓ′ (s1) j! , (24) QI I ′ ℓℓ′ (s, t′) = 1 π(t′ −s1)n (s −s1)n t′ −s δℓℓ′δI I ′ + CI I ′(2ℓ′ + 1) S(n) ℓℓ′ (s, t′) p2ℓ(s) . (25) + (2ℓ′ + 1)CI I ′ R( j) ℓℓ′ (s) p2ℓ(s) ¯a(I ′)( j) ℓ′ (s1) j! , (24) QI I ′ ℓℓ′ (s, t′) = 1 π(t′ −s1)n (s −s1)n t′ −s δℓℓ′δI I ′ (24) + CI I ′(2ℓ′ + 1) S(n) ℓℓ′ (s, t′) p2ℓ(s) . (25) (25) This is our ﬁnal form of the KT equations for ππ scattering. This is our ﬁnal form of the KT equations for ππ scattering. (tKT)(I) ℓ(s) = p2ℓ(s)¯a(I) ℓ(s) + ℓ′,I ′ n−1 j=0 (2ℓ′ + 1)CI I ′ R( j) ℓℓ′ (s) j! ¯a(I ′)( j) ℓ′ (s1) + ℓ′,I ′ (2ℓ′ + 1)CI I ′ 1 π ∞ sth dt′S(n) ℓℓ′ (s, t′)Im¯a(I ′) ℓ′ (t′) (t′ −s1)n . (20) 3 The explicit form of the polynomial contributions P(I) ℓ (s) is shown in Appendix B for the cases n = 1 and n = 2. 2 Amplitudes (s −s1) j + (s −s1)n π ∞ sth ds′ Im¯a(I) ℓ(s′) (s′ −s1)n(s′ −s), (19) (tKT)(I) ℓ(s) = p2ℓ(s) × ⎛ ⎝P(I) ℓ (s) + ℓ′,I ′ ∞ sth dt′ QI I ′ ℓℓ′ (s, t′) p2ℓ′(t′) Im(tKT)(I ′) ℓ′ (t′) ⎞ ⎠, (23) where we used Im(tKT)(I) ℓ(s) = p2ℓ(s)Im¯a(I) ℓ(s) for s > relation with an arbitrary number of subtractions n,2 ¯a(I) ℓ(s) = n−1 j=0 ¯a(I)( j) ℓ (s1) j! (s −s1) j + (s −s1)n π ∞ sth ds′ Im¯a(I) ℓ(s′) (s′ −s1)n(s′ −s), (19) (23) (19) where we used Im(tKT)(I) ℓ(s) = p2ℓ(s)Im¯a(I) ℓ(s) for s > sth. The polynomial term3 P(I) ℓ (s) and the integral kernels QI I ′ ℓℓ′ (s, t′) are given by where sth = 4m2. For simplicity, all of the subtractions are taken at the same point s1 < sth, and the subtraction constants are denoted as ¯a(I)( j) ℓ (s1). As it is written, Eq. (19) might lead one to think that the subtraction constants have no physical meaning. This, however, depends on the subtraction point chosen and the contribution of the crossed channels. In any case, with appropriate (linear) transformations, the subtrac- tions constants at any subthreshold point can be related to the effective range parameters [introduced in Eq. (11)]. We will see this in an speciﬁc case in Sect. 3. Dispersion relations similar to Eq. (19) are implied for the t- and u-channel iso- bars. Inserting this dispersive representation into Eq. (17), and the resulting amplitudes A(I)(s, t, u) into the partial- wave projection, Eq. (6), one obtains the KT representation of the partial-wave amplitudes of well-deﬁned isospin in the s-channel, where sth = 4m2. For simplicity, all of the subtractions are taken at the same point s1 < sth, and the subtraction constants are denoted as ¯a(I)( j) ℓ (s1). As it is written, Eq. (19) might lead one to think that the subtraction constants have no physical meaning. This, however, depends on the subtraction point chosen and the contribution of the crossed channels. In any case, with appropriate (linear) transformations, the subtrac- tions constants at any subthreshold point can be related to the effective range parameters [introduced in Eq. (11)]. We will see this in an speciﬁc case in Sect. 3. Dispersion relations similar to Eq. (19) are implied for the t- and u-channel iso- bars. 2 Amplitudes A(I)(s, t, u) = ℓmax ℓ=0 (2ℓ+ 1)Pℓ(zs)p2ℓ(s)¯a(I) ℓ(s) + ℓmax ℓ=0 I ′ (2ℓ+ 1)Pℓ(zt)p2ℓ(t)1 2CI I ′ ¯a(I ′) ℓ (t) + ℓmax ℓ=0 I ′ (2ℓ+ 1)Pℓ(zu)p2ℓ(u) × 1 2CI I ′ ¯a(I ′) ℓ (u)(−1)I+I ′, (17) (17) where the coefﬁcients CI I ′ are the matrix elements of where the coefﬁcients CI I ′ are the matrix elements of where the coefﬁcients CI I ′ are the matrix elements of From here, we construct isobars of deﬁnite isospin in the s-channel and denote them as ¯a(I) ℓ(s). These can be deﬁned by comparing the KT amplitude in Eq. (12) with Eq. (4). Speciﬁcally, we deﬁne: C = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎣ 2 3 2 10 3 2 3 1 −5 3 2 3 −1 1 3 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎦ . (18) (18) ⎡ ⎢⎣ ¯a(0) ℓ(x) ¯a(1) ℓ(x) ¯a(2) ℓ(x) ⎤ ⎥⎦= K ⎡ ⎣ as ℓ(x) at ℓ(x) au ℓ(x) ⎤ ⎦, (13) (13) We again remind the reader that the functions ¯a(I) ℓ(s) have only a RHC, and for each of them we write a dispersion where x = s, t, u and the matrix K has been introduced in Eq. (4). Using this relation in Eq. (12) we obtain 12 Eur. Phys. J. C (2018) 78 :574 574 Page 4 of 14 574 Page 4 of 14 574 relation with an arbitrary number of subtractions n,2 ¯a(I) ℓ(s) = n−1 ¯a(I)( j) ℓ (s1) j! (s −s1) j (tKT)(I) ℓ(s) = p2ℓ(s) × ⎛ ⎝P(I) ℓ (s) + ∞ dt′ QI I ′ ℓℓ′ (s, t′) 2ℓ′(t′) Im(tKT)(I ′) ℓ′ (t′) ⎞ ⎠, relation with an arbitrary number of subtractions n,2 n−1 ¯(I)( j)( ) (tKT)(I) ℓ(s) = p2ℓ(s) ⎛ ⎞ relation with an arbitrary number of subtractions n,2 (tKT)(I) ℓ(s) = p2ℓ(s) (tKT)(I) ℓ(s) = p2ℓ(s) × ⎛ ⎝P(I) ℓ (s) + ℓ′,I ′ ∞ sth dt′ QI I ′ ℓℓ′ (s, t′) p2ℓ′(t′) Im(tKT)(I ′) ℓ′ (t′) ⎞ ⎠, (23) relation with an arbitrary number of subtractions n,2 (tKT)(I) ℓ(s) = p2ℓ(s) relation with an arbitrary number of subtractions n,2 ¯a(I) ℓ(s) = n−1 j=0 ¯a(I)( j) ℓ (s1) j! 3 Analytical comparison of KT and Roy equations for S- and P-waves We begin in this section with a brief introduction to the Roy equations before comparing them with the KT equa- tions developed in Sect. 2. The Roy equations have been used extensively to study ππ scattering, with initial stud- ies following Roy’s original paper [46–50], but also more recently in the context of newer data [39,40]. The Roy equa- tions impose, within a given kinematical region, rigorous conditions on the determination of partial-wave amplitudes of deﬁnite isospin, based on analyticity and crossing sym- metry. In their exact form, the Roy equations couple the inﬁ- nite set of partial waves. However, any practical application requires a ﬁnite truncation, which makes them resemble the KT equations as we will demonstrate. The starting point for deriving these equations is to write down a twice-subtracted ﬁxed-t dispersion relation for these amplitudes. Roy realized these could be written as a matrix equation [33]: (20) The following functions have been introduced above: R( j) ℓℓ′ (s) = 1 2 +1 −1 dz (t(s, z) −s1) j Pℓ(z) ×Pℓ′(zt(s, z))p2ℓ′ (t(s, z)) , (21) S(n) ℓℓ′ (s, t′) = 1 2 +1 −1 dz ×(t(s, z) −s1)n Pℓ(z)Pℓ′(zt(s, z))p2ℓ′ (t(s, z)) t′ −t(s, z) . (22) R( j) ℓℓ′ (s) = 1 2 +1 −1 dz (t(s, z) −s1) j Pℓ(z) ×Pℓ′(zt(s, z))p2ℓ′ (t(s, z)) , (21) S(n) ℓℓ′ (s, t′) = 1 2 +1 −1 dz (21) S(n) ℓℓ′ (s, t′) = 1 2 +1 −1 dz (22) The functions R( j) ℓℓ′ (s) are polynomials in s. Furthermore, both R( j) ℓℓ′ (s) and S(n) ℓℓ′ (s, t′) behave as p2ℓ(s) for s →4m2. In the s-channel physical region, Eq. (20), together with Eq. (19), can be written more compactly as a Roy-like equa- tion, The functions R( j) ℓℓ′ (s) are polynomials in s. Furthermore, both R( j) ℓℓ′ (s) and S(n) ℓℓ′ (s, t′) behave as p2ℓ(s) for s →4m2. In the s-channel physical region, Eq. (20), together with Eq. 3 Analytical comparison of KT and Roy equations for S- and P-waves They contain two parameters; the S0- and S2-wave scattering lengths a(0) 0 and a(2) 0 , respec- tively. The integral kernels K I I ′ ℓℓ′ (s, x) are known as well and are fully documented in Appendix A of Ref. [39]. Including the results of Refs. [51,52] with experimental information on the scattering lengths and imaginary parts Im (tRoy)(I ′) ℓ′ (s) for sth < s < ∞, the real part of Eq. (27) can be fully deter- mined on the interval −4m2 < s < 60m2, and analytically continued to the complex plane in a region limited by the Lehmann ellipse [39]. This implies the Roy equations are particularly useful for studying the resonance properties of low-energy ππ scattering. Indeed, they have proven to be a popular resource in this regard. P(0) 0 (s) = 5 3 ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) + 3 4(3s −4m2)¯a(1) 1 (s1), (29) P(2) 0 (s) = 2 3 ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) −3 8(3s −4m2)¯a(1) 1 (s1), (30) P(1) 1 (s) = 3 2 ¯a(1) 1 (s1). (31) P(0) 0 (s) = 5 3 ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) + 3 4(3s −4m2)¯a(1) 1 (s1), (29) P(2) 0 (s) = 2 3 ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) −3 8(3s −4m2)¯a(1) 1 (s1), (30) (29) P(1) 1 (s) = 3 2 ¯a(1) 1 (s1). (31) (31) Although we have made a total of three subtractions, there are only two effective parameters, the subtraction constant ¯a(1) 1 (s1) and the linear combination ¯a(0) 0 (s1)+2¯a(2) 0 (s1). This is similar to what happens in the Roy equations. Let us now pay attention to the difference between the kernels of the KT and Roy equations, Although we have made a total of three subtractions, there are only two effective parameters, the subtraction constant ¯a(1) 1 (s1) and the linear combination ¯a(0) 0 (s1)+2¯a(2) 0 (s1). This is similar to what happens in the Roy equations. Let us now pay attention to the difference between the kernels of the KT and Roy equations, Now that we have introduced the Roy equations, we pro- ceed by comparing them with our formalism of the KT equa- tions. If restricted to the elastic region, the Roy equations Eq. 4 They are related to the matrix C used in Sect. 2 through (Cst)I I ′ = 1 2CI I ′ and (Cus)I I ′ = 1 2CI I ′(−1)I+I ′. 3 Analytical comparison of KT and Roy equations for S- and P-waves (27) That the Roy equations imply this KT-like structure for A(s, t, u), with the functions ¯a(I) ℓ(s) having only a RHC, has been already discussed in Refs. [34–39]. In what follows, we show in full detail that Roy and KT equations under the truncation speciﬁed give the same partial-wave amplitudes as in Eq. (28). To that end, let us consider a single subtraction n = 1 in each of the KT isobars. Since we are considering only one subtraction in each isobar, we simplify the notation throughout this section by writing ¯a(I)( j=0) ℓ (s1) = ¯a(I) ℓ(s1). In this way, the KT polynomials of Eq. (24) are That the Roy equations imply this KT-like structure for A(s, t, u), with the functions ¯a(I) ℓ(s) having only a RHC, has been already discussed in Refs. [34–39]. In what follows, we show in full detail that Roy and KT equations under the truncation speciﬁed give the same partial-wave amplitudes as in Eq. (28). To that end, let us consider a single subtraction n = 1 in each of the KT isobars. Since we are considering only one subtraction in each isobar, we simplify the notation throughout this section by writing ¯a(I)( j=0) ℓ (s1) = ¯a(I) ℓ(s1). In this way, the KT polynomials of Eq. (24) are (27) In this equation, the kinematical functions k(I) ℓ(s) are known linear polynomials in s resulting from projecting over the subtraction polynomials. They contain two parameters; the S0- and S2-wave scattering lengths a(0) 0 and a(2) 0 , respec- tively. The integral kernels K I I ′ ℓℓ′ (s, x) are known as well and are fully documented in Appendix A of Ref. [39]. Including the results of Refs. [51,52] with experimental information on the scattering lengths and imaginary parts Im (tRoy)(I ′) ℓ′ (s) for sth < s < ∞, the real part of Eq. (27) can be fully deter- mined on the interval −4m2 < s < 60m2, and analytically continued to the complex plane in a region limited by the Lehmann ellipse [39]. This implies the Roy equations are particularly useful for studying the resonance properties of low-energy ππ scattering. Indeed, they have proven to be a popular resource in this regard. In this equation, the kinematical functions k(I) ℓ(s) are known linear polynomials in s resulting from projecting over the subtraction polynomials. 3 Analytical comparison of KT and Roy equations for S- and P-waves (19), can be written more compactly as a Roy-like equa- tion, A(I)(s, t, u) = I ′ (Cst)I I ′ [αI ′(t) + βI ′(t)(s −u)] + 1 π ∞ sth dx x2 s2 x2 −s2 δI I ′ + u2 x2 −u2 (Cus)I I ′ ×ImA(I ′)(x, t, u(x, t)), (26) A(I)(s, t, u) = I ′ (Cst)I I ′ [αI ′(t) + βI ′(t)(s −u)] + 1 π ∞ sth dx x2 s2 x2 −s2 δI I ′ + u2 x2 −u2 (Cus)I I ′ ×ImA(I ′)(x, t, u(x, t)), (26) A(I)(s, t, u) = I ′ (Cst)I I ′ [αI ′(t) + βI ′(t)(s −u)] A(I)(s, t, u) = I ′ (Cst)I I ′ [αI ′(t) + βI ′(t)(s −u)] + 1 π ∞ sth dx x2 s2 x2 −s2 δI I ′ + u2 x2 −u2 (Cus)I I ′ + 1 π ∞ sth dx x2 s2 x2 −s2 δI I ′ + u2 x2 −u2 (Cus)I I ′ ×ImA(I ′)(x, t, u(x, t)), (26) ×ImA(I ′)(x, t, u(x, t)), (26) 3 The explicit form of the polynomial contributions P(I) ℓ (s) is shown in Appendix B for the cases n = 1 and n = 2. 12 3 Eur. Phys. J. C (2018) 78 :574 Page 5 of 14 574 parison to the KT equations under truncation does not exactly coincide with those Roy equation analyses. where we are using the notation in Eq. (4). The column vectors characterizing the t-dependent subtraction constants αI ′(t) and βI ′(t) have isospin 0, 2 and isospin 1 compo- nents, respectively. The matrices (Cst)I I ′ and (Csu)I I ′ are the crossing matrices and are given explicitly in Ref. [33].4 The Roy equations themselves are the partial-wave projec- tion of Eq. (26). Using the notation of Ref. [39], these can be compactly written as From Eq. (14), the KT amplitude A(s, t, u) with ℓmax = 1 has the expansion A(s, t, u) = ¯a(0) 0 (s) −¯a(2) 0 (s) 3 + ¯a(2) 0 (t) 2 + 3 8(s −u)¯a(1) 1 (t) + ¯a(2) 0 (u) 2 + 3 8(s −t)¯a(1) 1 (u). (28) (28) (tRoy)(I) ℓ(s) = k(I) ℓ(s) + ℓ′,I ′ ∞ sth dx K I I ′ ℓℓ′ (s, x) × Im (tRoy)(I ′) ℓ′ (x). 4 Numerical results Since the proof of the result in the previous section only holds when restricting to S- and P-waves, in this section we numerically study to what extent the KT equations are useful when higher partial waves are considered. Speciﬁcally, we now set ℓmax = 3 in Eq. (14) and thus consider up to the F- wave. In this way, partial waves with ℓ⩾4 (G- and higher waves) are set to zero, and do not contribute to the imaginary parts in Eq. (23). As stated above, the KT and Roy equations give the real part of the amplitudes once the imaginary part is given, up to a polynomial contribution. As an input for the KT equations, we shall use the CFD parameterization of Ref. [40], which we now brieﬂy discuss. This work parame- terizes the inelasticities η(I) ℓ(s) and phase shifts δ(I) ℓ(s) of the partial-wave amplitudes, Eq. (10). Two different parameter- izations are given in Ref. [40], called UFD and CFD, which respectively stand for unconstrained and constrained ﬁt to data. They do not differ on the form of the parameterization but in the values the parameters take. In the UFD parameter- ization only the data are ﬁtted, while in the CFD dispersive constraints are imposed on the amplitudes. Among these, the most relevant ones are the Roy and GKPY equations, which are, respectively, twice- and once-subtracted dispersion rela- tions for the ππ amplitude. Hence, the amplitudes computed with the CFD parameterization satisfy, within uncertainties, these dispersive equations. Both parameterizations of Ref. [40]providethephaseshiftsandinelasticitiesalongtheRHC. These quantities can be used as inputs to the Roy or GKPY equations so that the amplitudes can be computed at any point on the complex plane. The real part of the amplitudes obtained with the Roy or GKPY equations along the RHC are very similar to those obtained with the CFD parameterization for the amplitudes, since the latter are constrained to satisfy the former. We can use as well the CFD parameterization of Ref. [40] as an input for our KT equations to obtain the real part of the partial-wave amplitudes (tKT)(I) ℓ(s), and compare our results with the original input. 5 As advanced in Sect. 3, we explicitly see in Eqs. (37) and (38) that the subtraction constants can be linearly related to the threshold parameters. 4 Numerical results (tKT)(I) ℓ(s) −(tRoy)(I) ℓ(s) = p2ℓ(s)P(I) ℓ (s) −k(I) ℓ(s) + ℓ′,I ′ ∞ sth dt′ I I ′ ℓℓ′ (4m2, t′) Im(tRoy)(I ′) ℓ′ (t′) + (s −4m2) ℓ′,I ′ ∞ sth dt′ ∂ ∂s I I ′ ℓℓ′ (4m2, t′) Im(tRoy)(I ′) ℓ′ (t′) ≡p2ℓ(s)P(I) ℓ (s) −k(I) ℓ(s) + x(I) ℓ + (s −4m2)y(I) ℓ , (35) (35) where the imaginary part Im(tKT)(I ′) ℓ′ (t′) = Im(tRoy)(I ′) ℓ′ (t′) are given as an input, and x(I) ℓ and y(I) ℓ are constants not depending on s. Moreover, they are related through: 6y(1) 1 = −2y(2) 0 = y(0) 0 = 2x(0) 0 −5x(2) 0 12m2 . (36) (36) The above difference (tKT)(I) ℓ(s) −(tRoy)(I) ℓ(s) is also a lin- ear polynomial in s. The next question is: can we choose our constants ¯a(1) 1 (s1) and ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) such that (tKT)(I) ℓ(s) −(tRoy)(I) ℓ(s) = 0 for the three S- and P-waves? The P-wave is proportional to (s −4m2), so x(1) 1 = 0. We thus have to satisfy ﬁve conditions with only two parameters. It turns out that the solution 3 ¯a(0) 0 (s1) + 2¯a(2) 0 (s1) = a(0) 0 + 2a(2) 0 − x(0) 0 + 2x(2) 0 , (37) 27m2 ¯a(1) 1 (s1) = 2a(0) 0 −5a(2) 0 − 2x(0) 0 −5x(2) 0 , (38) These quantities can be used as inputs to the Roy or GKPY equations so that the amplitudes can be computed at any point on the complex plane. The real part of the amplitudes obtained with the Roy or GKPY equations along the RHC are very similar to those obtained with the CFD parameterization for the amplitudes, since the latter are constrained to satisfy the former. We can use as well the CFD parameterization of Ref. [40] as an input for our KT equations to obtain the real part of the partial-wave amplitudes (tKT)(I) ℓ(s), and compare our results with the original input. fulﬁlls indeed the ﬁve conditions.5 Hence, we have proved that, under the conditions indicated above, (tKT)(I) ℓ(s) −(tRoy)(I) ℓ(s) = 0. 3 Analytical comparison of KT and Roy equations for S- and P-waves (27) can be seen as a closed set of coupled nonlinear integral equations for amplitudes (tRoy)(I) ℓ(s), since each of the former can be written in terms of a single function, the phase shift δ(I) ℓ(s). The same can be said of the KT equations, but this approach will not be pursued here. We instead con- sider the Roy and KT equations as integral representations of the analytical partial waves in terms of their discontinuities and subtraction constants. As mentioned in Sect. 1, the aim in this section is to analytically compare the KT and Roy equations when truncating both formalisms to include only S- and P-waves (ℓ, ℓ′ = 0, 1). One needs to consider the previous sentence with care: by truncation of the Roy equa- tions to S- and P-waves, we mean letting the values of ℓand ℓ′ in Eq. (27) to be either 0 or 1. This truncation should not be confused with what is usually done in several analyses of the Roy equations [39,40,46–50], where one solves the integral equation for a ﬁnite number of low partial waves. Consequently, what we refer to as the Roy equations in com- π I I ′ ℓℓ′ (s, t′) ≡p2ℓ(s) p2ℓ′(t′) QI I ′ ℓℓ′ (s, t′) −K I I ′ ℓℓ′ (s, t′). (32) (32) For instance, for the case (ℓ, I) = (0, 0) and (ℓ′, I ′) = (1, 1) (recall that we are taking the number of subtractions n = 1), we have: 1 p2(t′) Q01 01(s, t′) = 6 π 2s + t′ −4m2 (t′ −4m2)(s −4m2) log 1 + s −4m2 t′ + 34m2 −3s −2(t′ −s1) π(t′ −s1)(t′ −4m2) , (33) K 01 01(s, t′) = 6 π 2s + t′ −4m2 (t′ −4m2)(s −4m2) log 1 + s −4m2 t′ + 34m2 −3s −2t′ πt′(t′ −4m2) . (34) (33) (34) It can be seen that the logarithmic terms cancel, and thus the difference is a ﬁrst order polynomial in s. This happens for all I I ′ ℓℓ′ (t′, s) as long as ℓ, ℓ′ ⩽1. Recalling that these 12 123 574 Page 6 of 14 574 Page 6 of 14 Eur. Phys. J. C (2018) 78 :574 differences appear under an integral in which the integration variable is t′, we can write 4 Numerical results (39) (39) This is a highly nontrivial result: the kernels of both approaches are originally different, but it turns out that they are such that the differences in the partial-wave amplitudes are only polynomials, and that the three polynomials can be put to zero with only two free parameters. The real parts of the input amplitudes are the black solid lines in Fig. 1. The CFD parameterization reaches up to a center-of-mass energy squared sm = (1.42 GeV)2 ≃ 2 GeV2. Since the dispersive integrals in the KT equa- tions extend to inﬁnity, we shall take as an approximation δ(I) ℓ(s) = δ(I) ℓ(sm) and η(I) ℓ(s) = η(I) ℓ(sm) for s ⩾sm. It is important to mention here that we do not use the full CFD parameterization since above s ≃2 GeV2 we set the phase shift to a constant instead of using the Regge formu- las of Ref. [40]. With the input amplitudes ﬁxed, the only remaining freedom in the KT equations are the subtraction constants appearing in the polynomial terms. The subtrac- tion constants are chosen so as to minimize the difference in the region sth ⩽s ⩽s f between the real part of the input amplitudes and those computed with the KT equations. To p y p For higher waves, ℓ, ℓ′ ⩾2, the kernels K I I ′ ℓℓ′ (s, t′) have more complicated structures. For instance, when ℓ′ = 2, one also ﬁnds the term log 1 + s−4m2 2t′ , in addition to log 1 + s−4m2 t′ . In that case, the differences in the kernels are not purely polynomials, and the above result cannot be proved. For higher waves, ℓ, ℓ′ ⩾2, the kernels K I I ′ ℓℓ′ (s, t′) have more complicated structures. For instance, when ℓ′ = 2, one also ﬁnds the term log 1 + s−4m2 2t′ , in addition to log 1 + s−4m2 t′ . In that case, the differences in the kernels are not purely polynomials, and the above result cannot be proved. 12 3 Page 7 of 14 574 Eur. Phys. J. C (2018) 78 :574 Eur. Phys. J. 4 Numerical results C (2018) 78 :574 Page 7 of 14 574 0.5 1 1.5 2 −0.5 0 0.5 s (GeV2) Re t(0) 0 (s) KT-A KT-B CFD 0.5 1 1.5 2 −0.6 −0.4 −0.2 0 s (GeV2) Re t(2) 0 (s) 0.5 1 1.5 2 −0.5 0 0.5 s (GeV2) Re t(1) 1 (s) 0.5 1 1.5 2 −0.4 −0.2 0 0.2 0.4 s (GeV2) Re t(0) 2 (s) 0.5 1 1.5 2 −0.06 −0.04 −0.02 0 0.02 s (GeV2) Re t(2) 2 (s) 0.5 1 1.5 2 0 0.02 0.04 0.06 0.08 s (GeV2) Re t(1) 3 (s) Fig. 1 Comparison of the real part of the different input partial-wave amplitudes (tCFD)(I) ℓ(s) taken from Ref. [40] (black lines, labeled as CFD) with (tKT)(I) ℓ(s). The two setups A (blue dashed line) and B (red dash-dotted line) represent different choices of the number of subtrac- tions in each wave and the maximum value of s (s f ) taken into account into the ﬁt of the subtraction constants (see text for more details) 0.5 1 1.5 2 −0.6 −0.4 −0.2 0 s (GeV2) Re t(2) 0 (s) 0.5 1 1.5 2 −0.5 0 0.5 s (GeV2) Re t(0) 0 (s) KT-A KT-B CFD s (GeV2) s (GeV2) 0.5 1 1.5 2 −0.4 −0.2 0 0.2 0.4 s (GeV2) Re t(0) 2 (s) 0.5 1 1.5 2 −0.5 0 0.5 s (GeV2) Re t(1) 1 (s) Re t(1) 1 (s) s (GeV2) 0.5 1 1.5 2 0 0.02 0.04 0.06 0.08 s (GeV2) Re t(1) 3 (s) 0.5 1 1.5 2 −0.06 −0.04 −0.02 0 0.02 s (GeV2) Re t(2) 2 (s) Fig. 1 Comparison of the real part of the different input partial-wave amplitudes (tCFD)(I) ℓ(s) taken from Ref. [40] (black lines, labeled as CFD) with (tKT)(I) ℓ(s). The two setups A (blue dashed line) and B (red dash-dotted line) represent different choices of the number of subtrac- tions in each wave and the maximum value of s (s f ) taken into account into the ﬁt of the subtraction constants (see text for more details) be more speciﬁc, we are minimizing the following χ2-like function: approximately equal to those given by the CFD parameter- ization. Two different setups for the ﬁts will be considered. 6 In Setup B we choose s f = 1.9 GeV2 close to but smaller than sm ≃2 GeV2, in which our input for Imt(I) ℓ(t′) changes from the CFD parameterization to a constant phase and inelasticity one. By choosing s f ≲sm, we avoid the point s = sm in which, despite the fact that the amplitude is continuous, some small numerical perturbations could appear. 6 In Setup B we choose s f = 1.9 GeV2 close to but smaller than sm ≃2 GeV2, in which our input for Imt(I) ℓ(t′) changes from the CFD parameterization to a constant phase and inelasticity one. By choosing s f ≲sm, we avoid the point s = sm in which, despite the fact that the amplitude is continuous, some small numerical perturbations could appear. 4 Numerical results [40] (the results labeled as CFD in that work are quoted here). We see that the agreement is also quite good. This is expected since these parameters control the low-energy behavior of the amplitudes, which the KT equations are able to reproduce in the whole energy range (sth ⩽s ⩽sm) considered here. above. In setup A we seek a description of low energies with a small number of subtractions. In contrast, in setup B we have extended the range of the ﬁt up to the maximum allowed by the CFD parameterization and the number of subtractions is increased. In setup A, n = 1 and thus there are a total of six subtrac- tions, but only ﬁve free parameters, since the subtraction con- stants a(0)(0) 0 (s1) and a(2)(0) 0 (s1) always appear in the same linear combination [see also the discussion in Sect. 3 and Eqs. (29)–(31)]. Analogously, in setup B, although n = 2 and there is a total of twelve subtractions, the number of free parameters is only seven and setup B introduces two addi- tional free parameters with respect to setup A. The values of the χ2-like function deﬁned in Eq. (40) are χ2 = 1.5 · 10−3 and 2.7 · 10−3 for setups A and B, respectively. We remark here that this is not a χ2 function, and hence the compari- son between the values for the two setups has no statistical meaning. We further remark that the inclusion of D- and F- waves is found to be necessary for an accurate description of the dispersive output with the KT equations, but only in the sense of the additional polynomial degrees of freedom they introduce. That is, the explicit inclusion of the imaginary contributions Imt(I) ℓ(t′) for ℓ> 1 appearing in the disper- sive projections in Eq. (23) are negligible. Despite the fact that these contributions are actually relevant over the entire interval s ∈(4m2, ∞), their inclusion is typically smooth as we further discuss in the Appendix and their net contribution is found to be small in comparison to the aforementioned polynomials. The partial waves obtained within each setup are shown in Fig. 1. We see that the agreement with the orig- inal input is quite good and clearly better for setup B. In Fig. 4 Numerical results In setup A we choose for the number of subtractions n = 1 and s f = 1 GeV2, while setup B is computed with n = 2 and s f = 1.9 GeV2 ≲sm,6 where sm has been deﬁned χ2 = 1 s f −sth ℓ,I s f sth ds Re (tKT)(I) ℓ(s) −(tCFD)(I) ℓ(s) 2 . (40) χ2 = 1 s f −sth ℓ,I s f sth ds Re (tKT)(I) ℓ(s) −(tCFD)(I) ℓ(s) 2 . ℓ,I (40) (40) We recall that the goal here is not to describe the phase shifts and inelasticities parameterized in Ref. [40], but rather a comparison between the amplitudes used as an input and the output given by the KT equations. For this reason we will not dwell on the calculation of errors, which should be 12 3 Eur. Phys. J. C (2018) 78 :574 574 Page 8 of 14 574 Page 8 of 14 400 600 800 1000 −0.5 0 0.5 √s (MeV) Re t(0) 0 (s) GKPY KT-A KT-B CFD 400 600 800 1000 −0.5 −0.4 −0.3 −0.2 −0.1 0 √s (MeV) Re t(2) 0 (s) 400 600 800 1000 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 √s (MeV) Re t(1) 1 (s) Fig. 2 Same as Fig. 1, but restricted to S- and P-waves, and including the dispersive output (green dotted lines, GKPY) of Ref. [40]. The (green) error band is associated with the GKPY dispersive output, as given in Ref. [40] 400 600 800 1000 −0.5 0 0.5 √s (MeV) Re t(0) 0 (s) GKPY KT-A KT-B CFD (2)( ) 400 600 800 1000 −0.5 −0.4 −0.3 −0.2 −0.1 0 √s (MeV) Re t0 (s) (1) 400 600 800 1000 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 √s (MeV) Re t(1) 1 (s) Fig. 2 Same as Fig. 1, but restricted to S- and P-waves, and including the dispersive output (green dotted lines, GKPY) of Ref. [40]. The (green) error band is associated with the GKPY dispersive output, as given in Ref. [40] Togetherwiththegeneralcomparisonoftherealpartofthe input and KT amplitudes, we can more speciﬁcally compare the threshold parameters of the different waves that we have included in our study, cf. Eq. (11), obtained with the two approaches. In Table 1 we show the values of the a(I) ℓ and b(I) ℓ parameters computed with the KT equations and compared with those of Ref. 4 Numerical results For comparison, we also show in the fourth column the same quantities extracted from the GKPY equations or the CFD parameterization of Ref. [40], as quoted in Refs. [40,53,54] KT-A KT-B GKPY–CFD a(0) 0 0.217 0.213 0.221(9) [40] b(0) 0 0.274 0.275 0.278(7) [40] a(2) 0 −0.044 −0.047 −0.043(8) [40] b(2) 0 −0.078 −0.079 −0.080(9) [40] 103 · a(1) 1 37.5 37.9 38.5(1.2) [40] 103 · b(1) 1 5.6 5.7 5.1(3) [40] 104 · a(0) 2 17.8 17.8 18.8(4) [40] 104 · b(0) 2 −3.4 −3.4 −4.2(1.0) [40] 104 · a(2) 2 1.9 1.8 2.8(1.0) [40] 104 · b(2) 2 −3.2 −3.2 −2.8(8) [40] 105 · a(1) 3 5.7 5.7 5.1(1.3) [40] 105 · b(1) 3 −4.0 −4.0 −4.6(2.5) [40] √sσ (MeV) (448, 270) (448, 269) (457+14 −13, 279+11 −7 ) [53] \|gσ \| GeV 3.36 3.37 3.59+0.11 −0.13 [53] √sρ (MeV) (762.2, 72.4) (763.4, 73.5) (763.7+1.7 −1.5, 73.2+1.0 −1.1) [53] gρ 5.95 6.01 6.01+0.04 −0.07 [53] √s f0 (MeV) (1000, 24) (995, 26) (996 ± 7, 25+10 −6 ) [53] g f0 (GeV) 2.4 2.3 2.3 ± 0.2 [53] √s f2 (MeV) (1275.1, 89.5) (1268.9, 86.4) (1267.3+0.8 −0.9, 87 ± 9) [54] g f2 (GeV−1) 5.6 5.5 5.0 ± 0.3 [54] lytically continued. Denoting as tI I(s) the amplitude7 on the second Riemann sheet, we take its customary deﬁnition in terms of the amplitude on the physical sheet, tI(s): lytically continued. Denoting as tI I(s) the amplitude7 on the second Riemann sheet, we take its customary deﬁnition in terms of the amplitude on the physical sheet, tI(s): withtheKTequationsandthosefromthedispersiveapproach of Refs. [53,54], which use our same input amplitudes but into dispersive equations in principle very different from KT equations. t−1 I I (s) = t−1 I (s) + 2iσ(s). (41) (41) (41) Our ﬁnal discussion about the results for the partial-wave amplitudes obtained with the KT formalism concerns the unitarity of said amplitudes. As explained before, Ref. [40] parameterizes the amplitudes through the phase shifts δ(I) ℓ(s) and inelasticities η(I) ℓ(s) as given in Eq. (10), with the latter chosen such that η(I) ℓ(s) ⩽1. Hence, the input amplitudes are unitary by construction. However, there is nothing in the KT equations (neither in the Roy nor GKPY equations) that force the partial waves to fulﬁll unitarity. For instance, in the derivation of our KT equations in Sect. 7 Here, for simplicity in the notation, we drop the ℓ, I scripts notation. 8 We choose this particular deﬁnition of the coupling to directly com- pare with the results given in Refs. [40,53,54]. 9 As discussed for instance in Ref. [39], the unitarity of the partial wave is an additional requirement to be imposed on the Roy equations. The same can be said about the other dispersive approaches referred to in this 4 Numerical results 2, showing only S- and P-waves up to √s = 1.1 GeV, we include in the comparison the dispersive output obtained with the GKPY equations and its associated error band, as given in Ref. [40]. We see that all curves lie well within or at the edges of the error bands. ( t m) Up to now we have checked the agreement between the real part of the amplitudes along the RHC obtained from KT equations and those from the CFD parameterization and the GKPY dispersive amplitudes of Ref. [40]. Since the subtrac- tion constants have been ﬁxed so as to minimize the differ- ence between the input amplitudes and the output from the KT equations, this agreement could be seen as natural. One may therefore ask: to what extent does the agreement stand away from the real axis? In this context, the CFD parameter- ization of Ref. [40] is used in Refs. [53,54] to compute the amplitudes on the complex plane by means of GKPY and Roy equations, as described above. In particular, the position and coupling of poles associated with several resonances are computed in those works. As done with the Roy and GKPY equations, our KT equations allow us to compute the ampli- tudes at any point on the complex plane. Hence, we now com- pare the results obtained in Refs. [53,54] with those obtained with the KT equations. To that end, let us ﬁrst brieﬂy discuss the relation between resonances and amplitudes, mainly to deﬁne our notation. Resonances manifest in the amplitudes as poles on the unphysical Riemann sheets which are con- tinuously connected with the physical one on the real axis. To reach the unphysical sheets the amplitudes must be ana- 12 123 123 Eur. Phys. J. C (2018) 78 :574 Page 9 of 14 574 Table 1 Threshold parameters and the pole positions and couplings for the σ, ρ, f0(980), and f2(1270) (the latter two are simply denoted above simply as f0 and f2, respectively) obtained with the KT equations (second and third columns for setups A and B, respectively) are shown. Table 1 Threshold parameters and the pole positions and couplings for the σ, ρ, f0(980), and f2(1270) (the latter two are simply denoted above simply as f0 and f2, respectively) obtained with the KT equations (second and third columns for setups A and B, respectively) are shown. 4 Numerical results The real part of the KT or GKPY amplitudes is given by the dispersive approach used (for instance, Eq. (20) for the KT approach), whereas the imaginary part, by construction, is given by the amplitude used to feed the dispersive equations. In both cases, this is the CFD parameterization KT determination of η(0) 0 closely follows the inelasticity of the input amplitude. We also show in Fig. 3 η(0) 0 as computed with the partial wave that results from the GKPY equations (green dotted line), as well as a simple estimation of its error from one of the partial waves (see Fig. 2). We observe that our KT dispersive output lies well within this error band. Furthermore, the value η(0) 0 (s) = 1 for s ⩽4m2 K is well comprised in this error band. All this considered, and given that unitarity is not imposed in KT equations, we can safely say that unitarity is well fulﬁlled for low energies. KT determination of η(0) 0 closely follows the inelasticity of the input amplitude. We also show in Fig. 3 η(0) 0 as computed with the partial wave that results from the GKPY equations (green dotted line), as well as a simple estimation of its error from one of the partial waves (see Fig. 2). We observe that our KT dispersive output lies well within this error band. Furthermore, the value η(0) 0 (s) = 1 for s ⩽4m2 K is well comprised in this error band. All this considered, and given that unitarity is not imposed in KT equations, we can safely say that unitarity is well fulﬁlled for low energies. straint is imposed relating the real and the imaginary parts. A second point to be taken into account is that, as explained in Sect. 3, in this work we are not obtaining t(I) ℓ(s) as the solu- tion of the integral equation in Eq. (23). Rather, our approach in this work is to consider Eq. (23) as an integral representa- tion for the amplitude t(I) ℓ(s), which can be obtained when a given input (namely, a speciﬁc parameterization of the imag- inary part) is used to feed the equations. The fact that the input used fulﬁlls unitary does not guarantee that the ampli- tude obtained as an output will fulﬁll unitarity. Hence, in this approach, unitarity violations10 are to be expected and, indeed, they are present. 4 Numerical results 2, the unitarity requirement is not used, and only dispersion relations for the different functions entering in the full amplitude A(s, t, u) are used. A similar statement can be made about the Roy equations.9 Unitarity is only achieved when a speciﬁc con- Around the pole s ≃sp, Around the pole s ≃sp, tI I(s) ≃ ˜g2 p s −sp , (42) (42) and we deﬁne the coupling gp of the resonance to the ππ channel in terms of the residue ˜gp as8 g2 p = −16π 2ℓ+ 1 (4p2(sp))ℓ˜g2 p. (43) (43) In Table 1 we show the poles and couplings of the different resonances that show up in the amplitudes considered in this work (S-, P-, D-, and F-waves). It can be seen that there is an excellent agreement between the determination obtained 12 3 3 574 Page 10 of 14 Eur. Phys. J. C (2018) 78 :574 400 600 800 1000 0 60 120 180 240 √s (MeV) δ(0) 0 (s) KT-A KT-B CFD GKPY 400 600 800 1000 0 0.25 0.5 0.75 1 √s (MeV) η(0) 0 (s) Fig. 3 Phase shift (left) and inelasticity (right) for the S0 wave. The notation for the different curves and the band is as in Fig. 2. These quan- tities are constructed from the real and imaginary parts of the S0 partial wave. The real part of the KT or GKPY amplitudes is given by the dispersive approach used (for instance, Eq. (20) for the KT approach), whereas the imaginary part, by construction, is given by the amplitude used to feed the dispersive equations. In both cases, this is the CFD parameterization 400 600 800 1000 0 60 120 180 240 √s (MeV) δ(0) 0 (s) KT-A KT-B CFD GKPY 400 600 800 1000 0 0.25 0.5 0.75 1 √s (MeV) η(0) 0 (s) 400 600 800 1000 0 0.25 0.5 0.75 1 √s (MeV) η(0) 0 (s) dispersive approach used (for instance, Eq. (20) for the KT approach), whereas the imaginary part, by construction, is given by the amplitude used to feed the dispersive equations. In both cases, this is the CFD parameterization Fig. 3 Phase shift (left) and inelasticity (right) for the S0 wave. The notation for the different curves and the band is as in Fig. 2. These quan- tities are constructed from the real and imaginary parts of the S0 partial wave. Footnote 9 continued paper. This can be achieved, for instance, by looking only for solutions in a subspace of unitary solutions. 10 These refer to either η(I) ℓ(s) < 1 in the elastic region, or η(I) ℓ(s) > 1 in any region. 4 Numerical results Yet, as we shall see below, we can advance the idea that using a unitary parameterization of the input amplitude makes these unavoidable unitary violations quite small. Forhigherenergies,s →∞,duetothe p2ℓ(s)factorscho- seninourdeﬁnitionsoftheamplitudestosatisfythethreshold behavior, the absolute value of the real part of the partial- wave amplitudes grows as sλ (with some positive integer λ), and then the inelasticity does not satisfy unitarity either. This behavior could be corrected by cutting the KT equation for t(I) ℓ(s) at some value of s and imposing there some appro- priate asymptotic behavior, but this is beyond the scope of our exploratory study. Furthermore, as said, this unwanted behavior occurs for s →∞, whereas the KT equations are meant to be low-energy approximations. Let us take as an example the S0 wave. In Fig. 3 we show the phase shift and inelasticity of this wave as given by the CFD parameterization of Ref. [40] (black solid line), and as obtained from t(0) 0 (s) calculated with KT equations (blue dashed and red dash-dotted lines for setup A and B, respec- tively). Since the real part of t(0) 0 (s) closely follows the input amplitude (see Figs. 1 and 2), and the imaginary parts are equal by construction, we expect δ(0) 0 (s) and η(0) 0 (s) to be similar in both approaches and, indeed, they are. However, for √s ⩽1 GeV (K ¯K threshold), we see that the input inelasticity is exactly one due to the parameterization. This is not the case for the KT equation, which shows a value η(0) 0 ⩾1 in a small region around √s ≃1 GeV. For low energies, η(0) 0 is very close to 1, and for √s ⩾1 GeV the Appendix A: LHC contributions A dispersive representation of the ππ partial waves t(I) ℓ(s) would have, generally speaking, three different contribu- tions: two terms from the integrals of the discontinuities along the RHC and LHC, and a polynomial term stemming from the subtractions performed in the dispersion relation. These contributions are easily identiﬁable in our KT repre- sentation of the partial waves, Eq. (23). The RHC and LHC respectively arise from the singular and nonsingular terms of the kernels, Eq. (25), whereas the polynomial term is given by Eq. (24). In this appendix we discuss the relative impor- tance of the LHC contribution to the partial waves in the KT dispersive representation. In Fig. 4 we show, for the S- and P-waves, the three contributions to the partial waves, together with the total amplitude. For simplicity, we consider the number of subtractions given by setup A discussed above for the KT approach. We immediately mention two features. First, in the S0 and P waves, where prominent resonances show up, the general shape of the amplitude is given by that of the RHC contribution. Second, the LHC contribution is of non-negligible magnitude in the three waves. However, even if the LHC contribution is large, we see that, for the S0 and P waves, it has no particular structure in the region s ⩽1 GeV2 (the maximum range for setup A), whereas it has a more complicated structure in the S2 wave. In Fig. 5 we show the LHC contribution for the three waves compared with order n polynomials, whose coefﬁcients have been ﬁt- ted to reproduce the LHC contribution. For the S2 wave, although the polynomial is able to account for the bulk of the LHC contribution, the particular details of the former cannot be described even with higher-order polynomials. Although Acknowledgements MA and NS contributed equally to this work. MA thanks J. R. Peláez, A. Rodas, and J. Ruiz de Elvira for use- ful discussions and for providing numerical results concerning Ref. [40]. NS thanks E. Passemar for useful discussions. This work was supported by BMBF, by the U.S. Department of Energy under grants No. DE-AC05-06OR23177 and No. DE-FG02-87ER40365, by PAPIIT- DGAPA (UNAM, Mexico) grant No. IA101717, CONACYT (Mex- ico) grant No. 251817, by Research Foundation – Flanders (FWO), by U.S. National Science Foundation under award Nos. PHY-1415459 and PHY-1205019, and Ministerio de Economía y Competitividad through grants Nos. FPA2016-77313-P, FIS2014-51948-C2-1-P and SEV-2014- 0398. 5 Discussion and conclusions While the connection between the Roy equations and the reconstruction theorem under a similar truncation has already been made in previous works, here we demonstrated that a general representation of the amplitude in terms of three distinct expansions in all of the scattering channels also reproduces this result. Numerical results testing the validity of the KT equations including higher waves was explored in Sect. 4. The dispersive output from the KT equations using as an input the CFD parameterization of Ref. [40] up to the F-wave and a center-of-mass energy √s = 1.42 GeV was compared with the GKPY equations and the CFD input itself. The KT equations provide an excellent agreement with both the CFD parameterization and GKPY equations at the level of partial-wave amplitudes and subsequent resonance pole parameters. In addition, since the KT equations (as well as the Roy or GKPY equations) do not imply per se the uni- tarity of the partial waves, we have studied how much they deviate from exact unitarity. We found that the KT equations satisfy unitarity within the CFD parameterization error for low energies. This supports the idea that the KT formalism is a good and simple approach for modeling amplitudes at low energies. The contribution to the partial waves coming from the LHC in the KT approach is also explored in some detail. It is found that, in the scattering region, and for some waves, this contribution can be well accounted for by polynomials. ons.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. Funded by SCOAP3. 5 Discussion and conclusions In this work, we explored the various aspects of the KT for- malism within the context of ππ scattering. Our main goal was to assess the validity of the KT formalism within a kine- matical range characteristic of hadronic processes in general. Since ππ scattering is the most well studied and simplest, purelyhadronicprocess, this makes it anideal systemfor test- ing the KT formalism. To accomplish this, we derived the KT equations for ππ scattering in Sect. 2 and followed this with a proof showing the KT and Roy equations are equivalent when 12 3 Eur. Phys. J. C (2018) 78 :574 Page 11 of 14 574 Page 11 of 14 574 truncating both formalisms to include only S- and P-waves in Sect. 3. While the connection between the Roy equations and the reconstruction theorem under a similar truncation has already been made in previous works, here we demonstrated that a general representation of the amplitude in terms of three distinct expansions in all of the scattering channels also reproduces this result. Numerical results testing the validity of the KT equations including higher waves was explored in Sect. 4. The dispersive output from the KT equations using as an input the CFD parameterization of Ref. [40] up to the F-wave and a center-of-mass energy √s = 1.42 GeV was compared with the GKPY equations and the CFD input itself. The KT equations provide an excellent agreement with both the CFD parameterization and GKPY equations at the level of partial-wave amplitudes and subsequent resonance pole parameters. In addition, since the KT equations (as well as the Roy or GKPY equations) do not imply per se the uni- tarity of the partial waves, we have studied how much they deviate from exact unitarity. We found that the KT equations satisfy unitarity within the CFD parameterization error for low energies. This supports the idea that the KT formalism is a good and simple approach for modeling amplitudes at low energies. The contribution to the partial waves coming from the LHC in the KT approach is also explored in some detail. It is found that, in the scattering region, and for some waves, this contribution can be well accounted for by polynomials. truncating both formalisms to include only S- and P-waves in Sect. 3. Appendix A: LHC contributions Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecomm 0.2 0.4 0.6 0.8 1 −2 0 2 s (GeV2) Re t(0) 0 (s) RHC Polyn. LHC Total 0.2 0.4 0.6 0.8 1 −0.5 0 s (GeV2) Re t(2) 0 (s) 0.2 0.4 0.6 0.8 1 −0.5 0 0.5 s (GeV2) Re t(1) 1 (s) Fig. 4 Contributions to the total partial-wave amplitude using setup A as described above (black dashed line): RHC integral (purple dash-dotted line), LHC integral (blue solid line), and polynomial term (orange dash-double-dotted line). Only the S- and P-waves are shown 0.2 0.4 0.6 0.8 1 −2 0 2 s (GeV2) Re t(0) 0 (s) RHC Polyn. LHC Total 0.2 0.4 0.6 0.8 1 −0.5 0 s (GeV2) (1) 0.2 0.4 0.6 0.8 1 −0.5 0 0.5 s (GeV2) s (GeV2) Fig. 4 Contributions to the total partial-wave amplitude using setup A as described above (black dashed line): RHC integral (purple dash-dotted line), LHC integral (blue solid line), and polynomial term (orange dash-double-dotted line). Only the S- and P-waves are shown 123 12 3 3 Eur. Phys. J. C (2018) 78 :574 574 Page 12 of 14 574 Page 12 of 14 0.2 0.4 0.6 0.8 1 −1.5 −1 −0.5 0 s (GeV2) Re t(0) 0 (s) LHC n = 1 n = 2 0.2 0.4 0.6 0.8 1 0 0.1 0.2 0.3 s (GeV2) Re t(2) 0 (s) LHC n = 2 n = 4 0.2 0.4 0.6 0.8 1 0 0.05 0.1 s (GeV2) Re t(1) 1 (s) LHC n = 1 n = 2 Fig. 5 Comparison of the LHC integral (blue solid line) shown in Fig. 4 with order n polynomials (red dashed and green dash-dotted lines) with coefﬁcients ﬁtted to reproduce the LHC integral 0.2 0.4 0.6 0.8 1 −1.5 −1 −0.5 0 s (GeV2) Re t(0) 0 (s) LHC n = 1 n = 2 0.2 0.4 0.6 0.8 1 0 0.05 0.1 s (GeV2) Re t(1) 1 (s) LHC n = 1 n = 2 0.2 0.4 0.6 0.8 1 0 0.1 0.2 0.3 s (GeV2) Re t(2) 0 (s) LHC n = 2 n = 4 Fig. 5 Comparison of the LHC integral (blue solid line) shown in Fig. Appendix A: LHC contributions 4 with order n polynomials (red dashed and green dash-dotted lines) with coefﬁcients ﬁtted to reproduce the LHC integral + ˆs2 25 36(¯a(0)(0) 2 + 5¯a(2)(0) 2 ) −315 32 ¯a(1)(0) 3 + 245 128 ˆs3 ¯a(1)(0) 3 , (B1) P(2) 0 (s) = 2 3 ¯a(0)(0) 0 + 4 3 ¯a(2)(0) 0 + 3 2 ¯a(1)(0) 1 + 5 9 2¯a(0)(0) 2 + ¯a(2)(0) 2 + 7 4 ¯a(1)(0) 3 + ˆs −9 8 ¯a(1)(0) 1 −10 9 (2¯a(0)(0) 2 + ¯a(2)(0) 2 ) −105 16 ¯a(1)(0) 3 + ˆs2 25 72(2¯a(0)(0) 2 + ¯a(2)(0) 2 ) + 315 64 ¯a(1)(0) 3 −245 256 ˆs3 ¯a(1)(0) 3 , (B2) this agrees with the expectation for the LHC to dominate the S2 lineshape, in absence of any resonance pole, the devia- tions of the polynomial from the LHC are of the same order as the deviations of KT from the Roy result shown in Fig. 1, and we cannot derive any strong conclusions from that. On the other hand, for the S0 and P waves it can be seen that it is possible to accurately reproduce the LHC contribution with low-order polynomials. Therefore, if one writes down independent (i.e., crossing symmetry violating) dispersion relations for the partial waves t(I) ℓ(s), the contributions from the LHC can be reabsorbed into the polynomial coefﬁcients thatarealreadypresentinthedispersionrelationfortheRHC. This is the meaning of the common statement that LHC con- tributions can be neglected (or at least treated perturbatively) in scattering process like ππ →ππ. (B1) P(1) 1 (s) = 4320¯a(1)(0) 1 −800(2¯a(0)(0) 2 −5¯a(2)(0) 2 ) + 3024¯a(1)(0) 3 1 2880 + 1000(2¯a(0)(0) 2 −5¯a(2)(0) 2 ) −9072¯a(1)(0) 3 ˆs 2880 + 441 320 ˆs2 ¯a(1)(0) 3 , (B3) P(0) 2 (s) = 5 9(2¯a(0)(0) 2 + ¯a(2)(0) 2 ) −7 10 ¯a(1)(0) 3 + 49 40 ˆs ¯a(1)(0) 3 , (B4) −5¯a(2)(0) 2 ) + 3024¯a(1)(0) 3 1 2880 Appendix B: Polynomials (B3) In this Appendix we collect the polynomials used in the KT equations, Eq. (24). The variable ˆs stands for ˆs = s/m2, and the analogous deﬁnition is implied for the subtraction point, ˆs1 = s1/m2.Forcompactness,wehavedroppedthes1 depen- dence in the subtraction constants, ¯a(I)( j) ℓ (s1) →¯a(I)( j) ℓ . The ﬁrst six equations, Eqs. (B1)–(B6), represent the polynomial contributions P(I) ℓ (s) (we remind the reader that the appro- priate factors p2ℓ(s) have been explicitly factored out) to the partial-wave amplitudes when one subtraction (n = 1, as in setup A) is performed on each isobar, ¯a(I) ℓ(s). Analogously, the six equations Eqs. (B7)–(B12) represent the additional contributions, P(I) ℓ (s), when two subtractions (n = 2, as in setup B) are performed. P(2) 2 (s) = −49 80 ˆs ¯a(1)(0) 3 P(2) 2 (s) = −49 80 ˆs ¯a(1)(0) 3 + 1 720 80¯a(0)(0) 2 + 760¯a(2)(0) 2 + 252¯a(1)(0) 3 , (B5) P(1) 3 (s) = 21 20 ¯a(1)(0) 3 . (B6) P(0) 0 (s) = 4 3 ¯a(0)(1) 0 −5 3 ˆs1 ¯a(0)(1) 0 + 2 9 30¯a(2)(1) 0 −18¯a(1)(1) 1 + 5(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) −14 5 ¯a(1)(1) 3 −1 9 ˆs1 30¯a(2)(1) 0 −27¯a(1)(1) 1 + 10(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 7 2 ˆs1 ¯a(1)(1) 3 + ˆs 2 3 ¯a(0)(1) 0 + 1 36 −60¯a(2)(1) 0 + 9(20 −9ˆs1)¯a(1)(1) 1 + 10(−11 + 8ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) P(1) 3 (s) = 21 20 ¯a(1)(0) 3 . Appendix B: Polynomials (B12) + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 833 160 ¯a(1)(1) 3 −245 128 ˆs1 ¯a(1)(1) 3 −49 80 ˆs4 ¯a(1)(1) 3 , (B7) P(2) 0 (s) = 4 3 ¯a(0)(1) 0 −2 3 ˆs1 ¯a(0)(1) 0 + 2 3 ¯a(2)(1) 0 −4 3 ˆs1 ¯a(2)(1) 0 + 2¯a(1)(1) 1 −3 2 ˆs1 ¯a(1)(1) 1 −5 9(−1 + ˆs1)(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 7 5 ¯a(1)(1) 3 + 49 160 ˆs4 ¯a(1)(1) 3 −7 4 ˆs1 ¯a(1)(1) 3 + ˆs −1 3 ¯a(0)(1) 0 + 5 6 ¯a(2)(1) 0 −5 2 ¯a(1)(1) 1 + 9 8 ˆs1 ¯a(1)(1) 1 −55 36(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 10 9 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −133 20 ¯a(1)(1) 3 + 105 16 ˆs1 ¯a(1)(1) 3 + ˆs2 1 2 ¯a(1)(1) 1 + 125 144(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −25 72 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 567 80 ¯a(1)(1) 3 −315 64 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 192(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −833 320 ¯a(1)(1) 3 +245 ˆ ¯(1)(1) (B8) −80¯a(2)(1) 2 −760ˆs1 ¯a(2)(1) 2 −36(4 + 7ˆs1)¯a(1)(1) 3 + 1 720 ˆs 360¯a(0)(1) 2 + 900¯a(2)(1) 2 + 828¯a(1)(1) 3 + 441ˆs1 ¯a(1)(1) 3 −4 5 ˆs2 ¯a(1)(1) 3 , (B11) P(1) 3 (s) = 2 21 ¯a(0)(1) 2 −5 21 ¯a(2)(1) 2 −1 5 ¯a(1)(1) 3 −21 20 ˆs1 ¯a(1)(1) 3 + 3 2 ˆs ¯a(1)(1) 3 . (B12) References 1. A. Esposito, A. Pilloni, A.D. Polosa, Phys. Rep. 668, 1 (2016). arXiv:1611.07920 [hep-ph] 2. R.F. Lebed, R.E. Mitchell, E.S. Swanson, Prog. Part. Nucl. Phys. 93, 143 (2017). arXiv:1610.04528 [hep-ph] 3. S.L. Olsen, T. Skwarnicki, D. Zieminska, Rev. Mod. Phys. 90, 015003 (2018). arXiv:1708.04012 [hep-ph] 4. M.T. Hansen, S.R. Sharpe, Phys. Rev. D 92, 114509 (2015). arXiv:1504.04248 [hep-lat] 5. H.W. Hammer, J.Y. Pang, A. Rusetsky, JHEP 10, 115 (2017). arXiv:1707.02176 [hep-lat] 6. R.A. Briceño, M.T. Hansen, S.R. Sharpe, Phys. Rev. D 95, 074510 (2017). arXiv:1701.07465 [hep-lat] 7. I. Caprini, G. Colangelo, H. Leutwyler, Phys. Rev. Lett. 96, 132001 (2006). arXiv:hep-ph/0512364 [hep-ph] 8. M. Albaladejo, J.A. Oller, Phys. Rev. D 86, 034003 (2012). Appendix B: Polynomials arXiv:1205.6606 [hep-ph] 9. J.R. Peláez, Phys. Rep. 658, 1 (2016). arXiv:1510.00653 [hep-ph] 10. N.N. Khuri, S.B. Treiman, Phys. Rev. 119, 1115 (1960) 11. J.B. Bronzan, C. Kacser, Phys. Rev. 132, 2703 (1963) 12. I.J.R. Aitchison, Phys. Rev. 137, B1070 (1965) 13 I J R Aitchison R Pasquier Phys Rev 152 1274 (1966) (B12) j arXiv:1205.6606 [hep-ph] 9. J.R. Peláez, Phys. Rep. 658, 1 (2016). arXiv:1510.00653 [hep-ph] 10. N.N. Khuri, S.B. Treiman, Phys. Rev. 119, 1115 (1960) 11. J.B. Bronzan, C. Kacser, Phys. Rev. 132, 2703 (1963) 12. I.J.R. Aitchison, Phys. Rev. 137, B1070 (1965) (B8) 13. I.J.R. Aitchison, R. Pasquier, Phys. Rev. 152, 1274 (1966) 14. R. Pasquier, J.Y. Pasquier, Phys. Rev. 170, 1294 (1968) 15. R. Pasquier, J.Y. Pasquier, Phys. Rev. 177, 2482 (1969) 16. A. Neveu, J. Scherk, Ann. Phys. 57, 39 (1970) 17. I.J.R. Aitchison, Unitarity, analyticity and crossing symme- try in two- and three-hadron ﬁnal state interactions (2015). arXiv:1507.02697 [hep-ph] Appendix B: Polynomials + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 833 160 ¯a(1)(1) 3 −245 128 ˆs1 ¯a(1)(1) 3 −49 80 ˆs4 ¯a(1)(1) 3 , (B7) P(2) 0 (s) = 4 3 ¯a(0)(1) 0 −2 3 ˆs1 ¯a(0)(1) 0 + 2 3 ¯a(2)(1) 0 −4 3 ˆs1 ¯a(2)(1) 0 + 2¯a(1)(1) 1 −3 2 ˆs1 ¯a(1)(1) 1 −5 9(−1 + ˆs1)(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 7 5 ¯a(1)(1) 3 + 49 160 ˆs4 ¯a(1)(1) 3 −7 4 ˆs1 ¯a(1)(1) 3 + ˆs −1 3 ¯a(0)(1) 0 + 5 6 ¯a(2)(1) 0 −5 2 ¯a(1)(1) 1 + 9 8 ˆs1 ¯a(1)(1) 1 −55 36(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 10 9 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −133 20 ¯a(1)(1) 3 + 105 16 ˆs1 ¯a(1)(1) 3 + ˆs2 1 2 ¯a(1)(1) 1 + 125 144(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −25 72 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 567 80 ¯a(1)(1) 3 −315 64 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 192(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −833 320 ¯a(1)(1) 3 +245 256 ˆs1 ¯a(1)(1) 3 , (B8) P(1) 1 (s) = 4 9 ¯a(0)(1) 0 −10 9 ¯a(2)(1) 0 −3 2 ˆs1 ¯a(1)(1) 1 +1 9(−2 + 5ˆs1)¯a(0)(1) 2 −5 18(−2 + 5ˆs1)¯a(2)(1) 2 −7 60(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs 3 2 ¯a(1)(1) 1 + 1 9 ¯a(0)(1) 2 −25 36 ˆs1 ¯a(0)(1) 2 −5 18 ¯a(2)(1) 2 + 125 72 ˆs1 ¯a(2)(1) 2 + 7 20(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs2 1 8 ¯a(0)(1) 2 −5 16 ¯a(2)(1) 2 −49 320(−4 + 9ˆs1)¯a(1)(1) 3 , (B9) + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 833 160 ¯a(1)(1) 3 −245 128 ˆs1 ¯a(1)(1) 3 −49 80 ˆs4 ¯a(1)(1) 3 , (B P(2) 0 (s) = 4 3 ¯a(0)(1) 0 −2 3 ˆs1 ¯a(0)(1) 0 + 2 3 ¯a(2)(1) 0 −4 3 ˆs1 ¯a(2)(1) 0 + 2¯a(1)(1) 1 −3 2 ˆs1 ¯a(1)(1) 1 −5 9(−1 + ˆs1)(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 7 5 ¯a(1)(1) 3 + 49 160 ˆs4 ¯a(1)(1) 3 −7 4 ˆs1 ¯a(1)(1) 3 + ˆs −1 3 ¯a(0)(1) 0 + 5 6 ¯a(2)(1) 0 −5 2 ¯a(1)(1) 1 + 9 8 ˆs1 ¯a(1)(1) 1 −55 36(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 10 9 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −133 20 ¯a(1)(1) 3 + 105 16 ˆs1 ¯a(1)(1) 3 + ˆs2 1 2 ¯a(1)(1) 1 + 125 144(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −25 72 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 567 80 ¯a(1)(1) 3 −315 64 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 192(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −833 320 ¯a(1)(1) 3 +245 ˆs1 ¯a(1)(1) (B + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) 833 ¯(1)(1) 245 ˆ ¯(1)(1) 49 ˆ4 ¯(1)(1) (B7) −80¯a(2)(1) 2 −760ˆs1 ¯a(2)(1) 2 −36(4 + 7ˆs1)¯a(1)(1) 3 + 1 720 ˆs 360¯a(0)(1) 2 + 900¯a(2)(1) 2 + 828¯a(1)(1) 3 + 441ˆs1 ¯a(1)(1) 3 −4 5 ˆs2 ¯a(1)(1) 3 , (B11) P(1) 3 (s) = 2 21 ¯a(0)(1) 2 −5 21 ¯a(2)(1) 2 −1 5 ¯a(1)(1) 3 −21 20 ˆs1 ¯a(1)(1) 3 + 3 2 ˆs ¯a(1)(1) 3 . (2017). arXiv:1701.07465 [hep-lat] 7. I. Caprini, G. Colangelo, H. Leutwyler, Phys. Rev. Lett. 96, 132001 (2006). arXiv:hep-ph/0512364 [hep-ph] (2006). arXiv:hep-ph/0512364 [hep-ph] 8. M. Albaladejo, J.A. Oller, Phys. Rev. D 86, 034003 (2012). arXiv:1205.6606 [hep-ph] References 1. A. Esposito, A. Pilloni, A.D. Polosa, Phys. Rep. 668, 1 (2016). arXiv:1611.07920 [hep-ph] 2. R.F. Lebed, R.E. Mitchell, E.S. Swanson, Prog. Part. Nucl. Phys. 93, 143 (2017). arXiv:1610.04528 [hep-ph] 2. R.F. Lebed, R.E. Mitchell, E.S. Swanson, Prog. Part. Nucl. Phys. 93 143 (2017) arXiv:1610 04528 [hep ph] 93, 143 (2017). arXiv:1610.04528 [hep-ph] 3. S.L. Olsen, T. Skwarnicki, D. Zieminska, Rev. Mod. Phys. 90, 015003 (2018). arXiv:1708.04012 [hep-ph] 015003 (2018). arXiv:1708.04012 [hep-ph] 4. M.T. Hansen, S.R. Sharpe, Phys. Rev. D 92, 114509 (2015). arXiv:1504.04248 [hep-lat] 5. H.W. Hammer, J.Y. Pang, A. Rusetsky, JHEP 10, 115 (2017). arXiv:1707.02176 [hep-lat] 6. R.A. Briceño, M.T. Hansen, S.R. Sharpe, Phys. Rev. D 95, 074510 (2017). arXiv:1701.07465 [hep-lat] 6. R.A. Briceño, M.T. Hansen, S.R. Sharpe, Phys. Rev. D 95, 074510 (2017) Xi 1701 07465 [h l ] Appendix B: Polynomials Mitchell, E.S. Swanson, Prog. Part. Nucl. Phys. 93, 143 (2017). arXiv:1610.04528 [hep-ph] 3. S.L. Olsen, T. Skwarnicki, D. Zieminska, Rev. Mod. Phys. 90, 015003 (2018). arXiv:1708.04012 [hep-ph] 4. M.T. Hansen, S.R. Sharpe, Phys. Rev. D 92, 114509 (2015). arXiv:1504.04248 [hep-lat] 5. H.W. Hammer, J.Y. Pang, A. Rusetsky, JHEP 10, 115 (2017). arXiv:1707.02176 [hep-lat] 6. R.A. Briceño, M.T. Hansen, S.R. Sharpe, Phys. Rev. D 95, 074510 (2017). arXiv:1701.07465 [hep-lat] 7. I. Caprini, G. Colangelo, H. Leutwyler, Phys. Rev. Lett. 96, 132001 (2006). arXiv:hep-ph/0512364 [hep-ph] 8. M. Albaladejo, J.A. Oller, Phys. Rev. D 86, 034003 (2012). arXiv:1205.6606 [hep-ph] 9. J.R. Peláez, Phys. Rep. 658, 1 (2016). arXiv:1510.00653 [hep-ph] 10. N.N. Khuri, S.B. Treiman, Phys. Rev. 119, 1115 (1960) 11. J.B. Bronzan, C. Kacser, Phys. Rev. 132, 2703 (1963) 12. I.J.R. Aitchison, Phys. Rev. 137, B1070 (1965) 13. I.J.R. Aitchison, R. Pasquier, Phys. Rev. 152, 1274 (1966) 14. R. Pasquier, J.Y. Pasquier, Phys. Rev. 170, 1294 (1968) 15. R. Pasquier, J.Y. Pasquier, Phys. Rev. 177, 2482 (1969) 16. A. Neveu, J. Scherk, Ann. Phys. 57, 39 (1970) 17. I.J.R. Aitchison, Unitarity, analyticity and crossing symme- try in two- and three-hadron ﬁnal state interactions (2015). arXiv:1507.02697 [hep-ph] 18. J. Stern, H. Sazdjian, N.H. Fuchs, Phys. Rev. D 47, 3814 (1993). arXiv:hep-ph/9301244 [hep-ph] 19. M. Zdrahal, J. Novotny, Phys. Rev. D 78, 116016 (2008). arXiv:0806.4529 [hep-ph] 20. J. Bijnens, K. Ghorbani, JHEP 11, 030 (2007). arXiv:0709.0230 [hep-ph] 21. J. Kambor, C. Wiesendanger, D. Wyler, Nucl. Phys. B 465, 215 (1996). arXiv:hep-ph/9509374 [hep-ph] 22. A.V. Anisovich, H. Leutwyler, Phys. Lett. B 375, 335 (1996). arXiv:hep-ph/9601237 [hep-ph] 23. P. Guo, I.V. Danilkin, D. Schott, C. Fernández-Ramírez, V. Appendix B: Polynomials (B6) (B6) (B6) + 5(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) −14 5 ¯a(1)(1) 3 −1 9 ˆs1 30¯a(2)(1) 0 −27¯a(1)(1) 1 P(0) 0 (s) = 5 3 ¯a(0)(0) 0 + 10 3 ¯a(2)(0) 0 −3¯a(1)(0) 1 + 10 9 ¯a(0)(0) 2 + 5¯a(2)(0) 2 −7 2 ¯a(1)(0) 3 + ˆs 9 4 ¯a(1)(0) 1 −20 9 (¯a(0)(0) 2 + 5¯a(2)(0) 2 ) + 105 8 ¯a(1)(0) 3 + 10(a( )( ) 2 + 5a( )( ) 2 ) + 2s1a( )( ) 3 + ˆs 2 3 ¯a(0)(1) 0 + 1 36 −60¯a(2)(1) 0 + 9(20 −9ˆs1)¯a(1)(1) 1 + ˆs 9 4 ¯a(1)(0) 1 −20 9 (¯a(0)(0) 2 + 5¯a(2)(0) 2 ) + 105 8 ¯a(1)(0) 3 + 10(−11 + 8ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 10(−11 + 8ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 10(−11 + 8ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) 3 Eur. Phys. J. Appendix B: Polynomials C (2018) 78 :574 + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 833 160 ¯a(1)(1) 3 −245 128 ˆs1 ¯a(1)(1) 3 −49 80 ˆs4 ¯a(1)(1) 3 , (B7 P(2) 0 (s) = 4 3 ¯a(0)(1) 0 −2 3 ˆs1 ¯a(0)(1) 0 + 2 3 ¯a(2)(1) 0 −4 3 ˆs1 ¯a(2)(1) 0 + 2¯a(1)(1) 1 −3 2 ˆs1 ¯a(1)(1) 1 −5 9(−1 + ˆs1)(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 7 5 ¯a(1)(1) 3 + 49 160 ˆs4 ¯a(1)(1) 3 −7 4 ˆs1 ¯a(1)(1) 3 + ˆs −1 3 ¯a(0)(1) 0 + 5 6 ¯a(2)(1) 0 −5 2 ¯a(1)(1) 1 + 9 8 ˆs1 ¯a(1)(1) 1 −55 36(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 10 9 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −133 20 ¯a(1)(1) 3 + 105 16 ˆs1 ¯a(1)(1) 3 + ˆs2 1 2 ¯a(1)(1) 1 + 125 144(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −25 72 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 567 80 ¯a(1)(1) 3 −315 64 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 192(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −833 320 ¯a(1)(1) 3 +245 256 ˆs1 ¯a(1)(1) 3 , (B8 P(1) 1 (s) = 4 9 ¯a(0)(1) 0 −10 9 ¯a(2)(1) 0 −3 2 ˆs1 ¯a(1)(1) 1 +1 9(−2 + 5ˆs1)¯a(0)(1) 2 −5 18(−2 + 5ˆs1)¯a(2)(1) 2 −7 60(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs 3 2 ¯a(1)(1) 1 + 1 9 ¯a(0)(1) 2 −25 36 ˆs1 ¯a(0)(1) 2 −5 18 ¯a(2)(1) 2 + 125 72 ˆs1 ¯a(2)(1) 2 + 7 20(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs2 1 8 ¯a(0)(1) 2 −5 16 ¯a(2)(1) 2 −49 320(−4 + 9ˆs1)¯a(1)(1) 3 , (B9 P(0) 2 (s) = 1 360 288¯a(1)(1) 1 −80¯a(0)(1) 2 −400ˆs1 ¯a(0)(1) 2 −400¯a(2)(1) 2 −200ˆs1 ¯a(2)(1) 2 + 36(4 + 7ˆs1)¯a(1)(1) 3 + 1 360 ˆs 540¯a(0)(1) 2 + 900¯a(2)(1) 2 −828¯a(1)(1) 3 −441ˆs1 ¯a(1)(1) 3 + 8 5 ˆs2 ¯a(1)(1) 3 , (B10 Eur. Phys. J. Appendix B: Polynomials C (2018) 78 :574 Page 13 of 14 574 + 133 10 ¯a(1)(1) 3 −105 8 ˆs1 ¯a(1)(1) 3 + ˆs2 1 72(−72¯a(1)(1) 1 −25(−5 + 2ˆs1)(¯a(0)(1) 2 + 5¯a(2)(1) 2 )) −567 40 ¯a(1)(1) 3 + 315 32 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 96(¯a(0)(1) 2 + 5¯a(2)(1) 2 ) + 833 160 ¯a(1)(1) 3 −245 128 ˆs1 ¯a(1)(1) 3 −49 80 ˆs4 ¯a(1)(1) 3 , (B7) P(2) 0 (s) = 4 3 ¯a(0)(1) 0 −2 3 ˆs1 ¯a(0)(1) 0 + 2 3 ¯a(2)(1) 0 −4 3 ˆs1 ¯a(2)(1) 0 + 2¯a(1)(1) 1 −3 2 ˆs1 ¯a(1)(1) 1 −5 9(−1 + ˆs1)(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 7 5 ¯a(1)(1) 3 + 49 160 ˆs4 ¯a(1)(1) 3 −7 4 ˆs1 ¯a(1)(1) 3 + ˆs −1 3 ¯a(0)(1) 0 + 5 6 ¯a(2)(1) 0 −5 2 ¯a(1)(1) 1 + 9 8 ˆs1 ¯a(1)(1) 1 −55 36(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 10 9 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −133 20 ¯a(1)(1) 3 + 105 16 ˆs1 ¯a(1)(1) 3 + ˆs2 1 2 ¯a(1)(1) 1 + 125 144(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −25 72 ˆs1(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) + 567 80 ¯a(1)(1) 3 −315 64 ˆs1 ¯a(1)(1) 3 + ˆs3 −25 192(2¯a(0)(1) 2 + ¯a(2)(1) 2 ) −833 320 ¯a(1)(1) 3 +245 256 ˆs1 ¯a(1)(1) 3 , (B8) P(1) 1 (s) = 4 9 ¯a(0)(1) 0 −10 9 ¯a(2)(1) 0 −3 2 ˆs1 ¯a(1)(1) 1 +1 9(−2 + 5ˆs1)¯a(0)(1) 2 −5 18(−2 + 5ˆs1)¯a(2)(1) 2 −7 60(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs 3 2 ¯a(1)(1) 1 + 1 9 ¯a(0)(1) 2 −25 36 ˆs1 ¯a(0)(1) 2 −5 18 ¯a(2)(1) 2 + 125 72 ˆs1 ¯a(2)(1) 2 + 7 20(−4 + 9ˆs1)¯a(1)(1) 3 + ˆs2 1 8 ¯a(0)(1) 2 −5 16 ¯a(2)(1) 2 −49 320(−4 + 9ˆs1)¯a(1)(1) 3 , (B9) −80¯a(2)(1) 2 −760ˆs1 ¯a(2)(1) 2 −36(4 + 7ˆs1)¯a(1)(1) 3 + 1 720 ˆs 360¯a(0)(1) 2 + 900¯a(2)(1) 2 + 828¯a(1)(1) 3 + 441ˆs1 ¯a(1)(1) 3 −4 5 ˆs2 ¯a(1)(1) 3 , (B11) P(1) 3 (s) = 2 21 ¯a(0)(1) 2 −5 21 ¯a(2)(1) 2 −1 5 ¯a(1)(1) 3 −21 20 ˆs1 ¯a(1)(1) 3 + 3 2 ˆs ¯a(1)(1) 3 . (B12) References 1. A. Esposito, A. Pilloni, A.D. Polosa, Phys. Rep. 668, 1 (2016). arXiv:1611.07920 [hep-ph] 2. R.F. Lebed, R.E. y arXiv:1507.02697 [hep-ph] 18. J. Stern, H. Sazdjian, N.H. Fuchs, Phys. Rev. D 47, 3814 (1993). arXiv:hep-ph/9301244 [hep-ph] 19. M. Zdrahal, J. Novotny, Phys. Rev. D 78, 116016 (2008). arXiv:0806.4529 [hep-ph] 20. J. Bijnens, K. Ghorbani, JHEP 11, 030 (2007). arXiv:0709.0230 [hep-ph] 21. J. Kambor, C. Wiesendanger, D. Wyler, Nucl. Phys. B 465, 215 (1996). arXiv:hep-ph/9509374 [hep-ph] 22. A.V. Anisovich, H. Leutwyler, Phys. Lett. B 375, 335 (1996). arXiv:hep-ph/9601237 [hep-ph] (B9) 23. P. Guo, I.V. Danilkin, D. Schott, C. Fernández-Ramírez, V. Mathieu, A.P. Szczepaniak, Phys. Rev. D 92, 054016 (2015). arXiv:1505.01715 [hep-ph] P(0) 2 (s) = 1 360 288¯a(1)(1) 1 −80¯a(0)(1) 2 −400ˆs1 ¯a(0)(1) 2 −400¯a(2)(1) 2 −200ˆs1 ¯a(2)(1) 2 + 36(4 + 7ˆs1)¯a(1)(1) 3 + 1 360 ˆs 540¯a(0)(1) 2 + 900¯a(2)(1) 2 −828¯a(1)(1) 3 −441ˆs1 ¯a(1)(1) 3 + 8 5 ˆs2 ¯a(1)(1) 3 , (B10 P(2) 2 (s) = 1 720 −288¯a(1)(1) 1 −160¯a(0)(1) 2 −80ˆs1 ¯a(0)(1) 2 P(0) 2 (s) = 1 360 288¯a(1)(1) 1 −80¯a(0)(1) 2 −400ˆs1 ¯a(0)(1) 2 −400¯a(2)(1) 2 −200ˆs1 ¯a(2)(1) 2 + 36(4 + 7ˆs1)¯a(1)(1) 3 + 1 360 ˆs 540¯a(0)(1) 2 + 900¯a(2)(1) 2 −828¯a(1)(1) 3 (1)(1) 8 2 (1)(1) 24. P. Guo, I.V. Danilkin, C. Fernández-Ramírez, V. Mathieu, A.P. Szczepaniak, Phys. Lett. B 771, 497 (2017). arXiv:1608.01447 [hep-ph] 25. G. Colangelo, S. Lanz, H. Leutwyler, E. Passemar, Phys. Rev. Lett. 118, 022001 (2017). arXiv:1610.03494 [hep-ph] 26. M. Albaladejo, B. Moussallam, Eur. Phys. J. C 77, 508 (2017). arXiv:1702.04931 [hep-ph] (B10) (B10) 27. F. Niecknig, B. Kubis, S.P. Schneider, Eur. Phys. J. C 72, 2014 (2012). arXiv:1203.2501 [hep-ph] 12 3 3 574 Page 14 of 14 Eur. Phys. J. C (2018) 78 :574 28. I.V. Danilkin, C. Fernández-Ramírez, P. Guo, V. Mathieu, D. Schott, M. Shi, A.P. Szczepaniak, Phys. Rev. D 91, 094029 (2015). arXiv:1409.7708 [hep-ph] 41. V. Meshcheryakov, D. Shirkov, V. Serebryakov, Dispersion Theo- ries of Strong Interactions at Low Energy (North-Holland Publish- ing Company, Amsterdam, 1969) 42. S. Lanz, Determination of the quark mass ratio Q from η →3π. Ph.D. thesis, Bern University (2011) 29. T. Isken, B. Kubis, S.P. Schneider, P. Stoffer, Eur. Phys. J. C 77, 489 (2017). arXiv:1705.04339 [hep-ph] 30. F. Niecknig, B. Kubis, JHEP 10, 142 (2015). arXiv:1509.03188 [hep-ph] 43. A.P. Szczepaniak, Phys. Lett. B 757, 61 (2016). arXiv:1510.01789 [hep-ph] 31. A. Pilloni, C. Fernández-Ramírez, A. Jackura, V. Mathieu, M. Mikhasenko, J. Nys, A.P. Szczepaniak, (JPAC), Phys. Lett. y arXiv:1507.02697 [hep-ph] B 772, 200 (2017). arXiv:1612.06490 [hep-ph] 44. M. Albaladejo, J.A. Oller, Phys. Rev. C 84, 054009 (2011). arXiv:1107.3035 [nucl-th] 45. M. Albaladejo, J.A. Oller, Phys. Rev. C 86, 034005 (2012). arXiv:1201.0443 [nucl-th] 32. F. Niecknig, B. Kubis, Phys. Lett. B 780, 471 (2018). arXiv:1708.00446 [hep-ph] 46. J.L. Basdevant, J.C. Le Guillou, H. Navelet, Nuovo Cim. A 7, 363 (1972) 33. S.M. Roy, Phys. Lett. 36B, 353 (1971) 47. J.L. Basdevant, C.D. Froggatt, J.L. Petersen, Nucl. Phys. B 72, 413 (1974) 34. G. Mahoux, S.M. Roy, G. Wanders, Nucl. Phys. B 70, 297 (1974) 34. G. Mahoux, S.M. Roy, G. Wanders, Nu 35. M. Knecht, B. Moussallam, J. Stern, N.H. Fuchs, Nucl. Phys. B 457, 513 (1995). arXiv:hep-ph/9507319 [hep-ph] 457, 513 (1995). arXiv:hep-ph/9507319 [hep-ph] 48. J.L. Basdevant, C.D. Froggatt, J.L. Petersen, Phys. Lett. 41B, 173 (1972) 36. M. Knecht, B. Moussallam, J. Stern, N.H. Fuchs, Nucl. Phys. B 36. M. Knecht, B. Moussallam, J. Stern, N.H. Fuchs, Nucl. P 471, 445 (1996). arXiv:hep-ph/9512404 [hep-ph] 36. M. Knecht, B. Moussallam, J. Stern, N.H. Fuchs , , , 471, 445 (1996). arXiv:hep-ph/9512404 [hep-ph] 49. J.L. Basdevant, C.D. Froggatt, J.L. Petersen, Phys. Lett. 41B, 178 (1972) 37. B. Ananthanarayan, D. Toublan, G. Wanders, Phys. Rev. D 51, 1093 (1995). arXiv:hep-ph/9410302 [hep-ph] 50. M.R. Pennington, S.D. Protopopescu, Phys. Rev. D 7, 1429 (1973) 38. B. Ananthanarayan, Phys. Rev. D 58, 036002 (1998). arXiv:hep-ph/9802338 [hep-ph] 51. A. Martin, Nuovo Cim. A 42, 930 (1965) 52. A. Martin, Nuovo Cim. A 42, 1219 (1966) 39. B. Ananthanarayan, G. Colangelo, J. Gasser, H. Leutwyler, Phys. Rep. 353, 207 (2001). arXiv:hep-ph/0005297 [hep-ph] 53. R. García-Martín, R. Kaminski, J.R. Peláez, J. Ruiz de 53. R. García-Martín, R. Kaminski, J.R. Peláez, J. Ruiz de Elvira, Phys. R L tt 107 072001 (2011) Xi 1107 1635 [h h] 53. R. García-Martín, R. Kaminski, J.R. Peláez, J. Ruiz de Elvira, Phys. Rev. Lett. 107, 072001 (2011). arXiv:1107.1635 [hep-ph] Rev. Lett. 107, 072001 (2011). arXiv:1107.1635 [hep-ph 54. J.A. Carrasco, J. Nebreda, J.R. Peláez, A.P. Szczepaniak, Phys. Lett. B 749, 399 (2015). arXiv:1504.03248 [hep-ph] 40. R. García-Martín, R. Kami´nski, J.R. Peláez, J. Ruiz de Elvira, F.J. Ynduráin, Phys. Rev. D 83, 074004 (2011). arXiv:1102.2183 [hep- ph] 123 123
https://openalex.org/W4387958365	https://drpress.org/ojs/index.php/fbem/article/download/12938/12589	English	null	The Effects of African Countries on Adopting Chinese Business Strategies	Frontiers in business, economics and management	2,023	cc-by	7,327	1.1. Statement of the problem The issue is described as follows: how Chinese business strategies helped Africa grow in the business world, whether African nations benefited from adopting them, and whether Africa should continue to do so. Iradukunda Ariane, Zaitun Chipembele Said, Karerangabo Tresor, Ying Wang* School of Economics and Business Management, Yunnan Technology and Business University, Kunming, Yunnan, * C di h * Corresponding author * Corresponding author Abstract: China’s business strategies have impacted many countries in the world with the global economy over the past three decades have significant economy impacts and political implications for African countries. This paper reviews the debate over whether these impacts on balance between positive and negative impacts for the continent of Africa and whether these strategies have impacted African countries how and in which ways, challenges and outcomes. Keywords: Business strategies, Business strategies of China, China-Africa Business. materials, footwear and plastic goods from China. 1. Background of the Study , p g The majority of these goods were imported machinery and electronic equipment. Some countries have seen a significant increase in imports, including Algeria, the Central African Republic, Chad, Ethiopia, Mali, Rwanda, Zambia and Zimbabwe. These countries also import trucks, tractors, motor vehicles for transportation, television receivers and electric line telephony. Additionally, countries like Benin, Gambia, Madagascar, Mauritania, Tanzania, and Togo imported textile and clothing goods. Mineral product, base metals, precious stones, and other unclassified primary Commodities made up the majority of the textiles and clothing that were exported from African countries to China. Africa's influence was relatively small ten years ago. Its diplomats were comparatively unskilled, and its aforementioned programs hardly had any impact. Many Chinese people also had doubts about their nation's standing as a global actor. China primarily defended core interests, such as the "one China" principle, in the majority of international forums. However, there has recently been continued economic growth, a younger generation of Chinese leaders who are more knowledgeable about China's role in the world, and better African scholarship. These strategies are now centered on Africa. China's' courtship of Africa serves as a resource heist in part. Due to its rapid economic growth and declining domestic petroleum and mineral reserves, Beijing is looking abroad for resources. 1.3. The study's significance The benefits of Chinese policies for African countries are what make this study significant. Chinese businesses have almost entirely entered African markets in an effort to expand them to China. More than 1000 of them are currently active in Africa, and many of these Chinese businesses there are privately held. Chinese companies have great degrees of freedom to operate in Africa on market terms.in addition to being backed by Chinese investment capital on favorable terms. In addition to being backed by Chinese investment capital on favorable terms. 1.2. The study's goal Trade and cultural exchanges between China and Africa have increased since the 7th century. China and Africa have begun direct contact via route in an effort to improve relations between Africa and China. A crucial component of China's independent foreign policy of peace has always been the growth and bolstering of friendly relations and cooperation with developing countries. Finding out whether and how Chinese business practices have impacted African countries are the aim of this study. 2.2. History Since the fifteenth century, China and Africa have been in contact. A wall painting of a giraffe was gifted to Beijing's Ming Imperial Tomb by the legendary Chinese admiral and sailor Zheng He on one of his numerous voyages to the Arab world and the east coast of Africa between 1413 and 1419. 2. Literature Review 2. Literature Review Frontiers in Business, Economics and Management ISSN: 2766-824X \| Vol. 11, No. 3, 2023 Frontiers in Business, Economics and Management ISSN: 2766-824X \| Vol. 11, No. 3, 2023 2.1. Overview of China and Africa Business China is Africa's largest trading partner and a significant source of foreign direct investment. The development of infrastructure and the growth of the economy have both benefited from its investment. Because Africa needs oil and other resources as well as a market for its goods, China has made investments there. Trade between African countries and China can be split into two categories: imports from China and exports to China. Over the past 20 years, African countries have primarily imported electronic equipment, machinery, clothing and textile goods, transportation Currently, Chinese businesses are actively investing in developing digital infrastructure in African nations. For instance, a Chinese telecommunication company began teaching an information and communication technology course with the goal of preparing local students for the workforce by fostering knowledge, understanding, and 9 interest in the field. selling their markets in China. even relatively small businesses, many more. will adhere to them. True Chinese strategy, however, is different. China is a central and integrated part of this one-world strategy and long-term development plan for conducting business as if it were its own country in the world, where China now plays an entirely different role than it did previously. interest in the field. Improve infrastructure right now you could say that any big project in African cities that is higher than three floors or roads that are longer than three kilometers are most likely being built and engineered by the Chinese. as Chinese started building rail roads, hospitals, airports, university, water supply throughout the continent. supp y t oug out t e co t e t. Since the 1950s, there has been a significant exchange of cultures between China and Africa. The two nations frequently send governmental cultural delegations to visit one another. Africa and China work together to train performers and artists. China sent academics to teach or research African art and culture while training acrobats for Tanzania and Sudan. Over the past ten years, Africa has undergone a significant economic transformation since China started making investments in African nations in the 1980s. The Continent has attracted new investments and projects, and its economic ties with developing countries have grown. One of them is how China's growing interest in Africa has sparked a lot of discussion among academics and journalists from both developed and developing nations. Since the 1950s, there has been a significant exchange of cultures between China and Africa. 2.4.1. Diversify across exports Other beneficial policies include removing trade barriers, improving trade facilitation, encouraging technology adoption through educational exchange programs, and funding digital economy-supporting communication technologies like broadband. African countries have done this because diversification of exports can, for example, help to combat export instability or the unfavorable effects of trade terms on primary products. Over the past ten years, Africa has undergone a significant economic transformation since China started making investments in African nations in the 1980s. The Continent has attracted new investments and projects, and its economic ties with developing countries have grown. One of them is how China's growing interest in Africa has sparked a lot of discussion among academics and journalists from both developed and developing nations. 2.4. Chinese business strategies A new strategy entails more than just a list of strategies for conducting business in China. The majority of large corporations are already vying with Chinese businesses by Figure 1. Conceptual model Figure 1. Conceptual model engages in unfair negotiations that exploit the relative weaknesses of African governments, encourage corruption, and encourage bad decision making. In their view, In order to maintain its Neo-colonial relationship, China imports manufactured goods from Africa in exchange for raw materials. Resentment of the nation's business practices has sparked widespread protests and violent acts against Chinese businessmen and immigrants in some countries. 2.4.2. Global expansion One of the strategies used by Chinese businesses to grow internationally is global expansion, which would begin with the belt road initiative's expanding ambition across Eurasia and Africa. China is at the center of a global network of trade and economic connections thanks to the construction and financing of its physical infrastructure. China is now poised to overtake Japan as the second-largest economy in the world, behind the United States. As well as experiencing rapid growth, China also became much more integrated with the global economy. Its share of global trade rose from less than 1 percent in 1980 to around 8 percent, and in 2009 it became the world’s largest exporter ahead of Germany. China’s growing trade surplus led to the accumulation of large foreign exchange reserves and to the country. The arrival of China is altering the region's way of life in a number of ways, Including employment opportunities. On the continent of China, several million African jobs had been created. A third of Chinese businesses had introduced a new Technology while nearly two thirds offered apprenticeships. More than any other nations in the world, China has employed close to 30000 people in Africa. 2.1. Overview of China and Africa Business The two nations frequently send governmental cultural delegations to visit one another. Africa and China work together to train performers and artists. China sent academics to teach or research African art and culture while training acrobats for Tanzania and Sudan. 2.5. Negative impact of Chinese Business in Africa 2.5. Negative impact of Chinese Business in Africa Africans accuse Chinese workers of refusing to integrate and working in unmarked offices and dormitories due to cultural barriers on the continent. Although businesses have quickly realized they must offer local employment if they want to remain in business, they are accused of hiring their own labor. The majority of Africans see Chinese presence as Neocolonialism, despite the Fact that African governments have seen it as a chance for Africa to grow predominately. In addition to the aforementioned, China's infrastructure projects have been the subject of countless worries as part of its extensive commitments and covert plans to create an unbreakable link for the use of Africa's natural resources. However, some in Africa criticize Chinese engagement. Chinese businesses are criticized by labor unions, civil society organizations, and other groups in African societies for their subpar working conditions, unsustainable environmental practices, as well as job displacement. Watchdogs for good governance issue a warning that China 10 Preconceptions of China, wholesomely serving nation that aims to preserve access to the global Pool of natural resources, have given rise to these worries. Preconceptions of China, wholesomely serving nation that aims to preserve access to the global Pool of natural resources, have given rise to these worries. instead let Africa pick a path for growth that is suitable for their circumstances. an example of collaboration and world development. In October 2012, the institute for peace and security studies held a China-Africa forum in Bishoftu, Ethiopia. The event was perhaps the first of it’s kind held in Ethiopia with academics, policy makers and stakeholders from the two regions participating in an open discussion about the economic, political and social relations between Africa and China. Most of discussion focused on the economic relationship between China and Africa. It has been argued that the sizable Chinese donations are a ruse to enter Africa and sway it to Chinese interests. Not only has China used this strategy with the US, but also in Africa. In order to boost the US dollar in favor of the Chinese Yuan, China has bought a lot of treasury bonds. This practice has been observed in Africa, where many African currencies have been inflated using debt provision to bring their local currencies' values up to par with the weak Yuan. When China-Africa collaboration is successful, South- South cooperation will also prosper. 2.5. Negative impact of Chinese Business in Africa Once China and Africa have reached their full potential, the world will be a better place. The only way for China and Africa to progress together in the new era is through cooperation. It will solidly provide the groundwork for China and Africa's future growth, as well as for the progress of all developing nations and more just international relations. Expanding economic cooperation quickly Because they behave improperly in the countries where they conduct business, Chinese citizens have come under fire. Reports on improper employee handling by Chinese private and state owned businesses were published by Human Rights Watch. The fatalities of 51 Zambian copper mine workers and Chinese factory worker are proof that, despite their claims to be making win-win investments in Africa, the Chinese have a history of providing subpar working conditions. It has also been stated that it is believe that Chinese companies have outsourced out of concern for breaking the law, which raises the question of why they are using win-win approach to help local Africans. This demonstrate how little Chinese investment value life over business. The trade and economic ties between China and Africa have grown quickly. China and Africa’s economic and trade cooperation reached a new level thanks to the 10 major cooperation plans, the 8 major initiatives, and FOCAC Johannesburg summit in 2015 and 2018 supporting development more and more. China encourages African countries to pursue their own growth while pursuing development and improving the lives of their citizens. In the new era , China has decided to boost its foreign aid to Africa , which between 2013 and 2018 totaled Rmb 270 billion. Interest free loans and concessional loans. Between 2000 and 2020, China helped African countries build more than 13,000km of roads, railways, ans sizable power plants. Additionally it provided funding more than 170 schools, 45 - sports venues, and 130 medical facilities. More than 160000 people were trained for Africa, and it also built several flagship projects, including the AU conference center. The only criticism leveled at the Chinese has been that they sell inferior goods. It has been acknowledged that these products are offered for incredibly low prices in an effort to drive out domestic and international companies. In Zimbabwe, locals have given names like Zhing-Zang in reference to the inexpensive Chinese goods. 2.5. Negative impact of Chinese Business in Africa In addition to being inexpensive, China's infrastructure and products have been built with durability and resilience in mind. As a result of their increased activity on the continent, the Chinese have increased their involvement in illegal activities like pouching and stealing African wildlife. Pinnock claims that the Chinese have been stealing African wildlife while falsely justifying it through dubious paperwork in Hwange National Park, where Zimbabwe exported 30 baby elephants to Maputo. Additionally, he asserts that the infants have little chance of ever reaching China. The Chinese claimed that Zimbabwe had too many elephant in its parks in 2014, but Walter Mzembi, the acting environment minister defended the claim despite the incident being made public. Growing commercial ties China has been Africa's main trading partner since 2009, and since then, it has contributed more and more to that trade. The number was higher than 21% in 2020. The trade between China and Africa is becoming more organized. China's exports to Africa have significantly increased in technology, with mechanical, electrical, and high-tech products now making up more than 50% of the total. In order to help more African agriculture and manufactured goods enter the Chinese market has increased its import of non resource products from Africa and provided zero tariff treatment to 97percent of taxable items exported to China by the 33 least developed countries in Africa. 2.6. Positive impact on how African countries adopted Chinese business strategies Through a number of scholarships, all of which are based the requirements of African countries for economic and social development, it encourages deserving African youth to study in China and supports the training of much needed professionals for African nations. As of 2012, both sides had implemented the 20+20 cooperation plan as a framework for intercollegiate cooperation for higher education institution in China and Africa. The digital economy's growing cooperation. China is assisting African nations in closing the digital divide. Building digital infrastructure, moving toward a digital society, and utilizing cutting-edge technologies like mobile finance and the Internet of Things, have all been successful in this field. They have worked with significant African, operators to provide Africa with complete, fundamental telecommunications Service, coverage. More than half of the continent's highspeed mobile broadband networks and wireless sites have been constructed more than 900 million locals have access to broadband internet thanks to the installation of more than 200,000 km of special fiber. The first 5G independent commercial networking in the area was also made available by the two sides. China and Africa’s commerce cooperation is becoming more and more extensive. Increasing social development cooperation In social areas like poverty reduction, health, education, science and technology, environmental protection, climate change, and exchanges between young people and women, China is fostering collaboration with Africa. China is assisting African nations in enhancing their comprehensive social development by fostering exchanges, offering assistance, and sharing experiences, which in turn drives their economic growth internally. The digital economy's growing cooperation. China is assisting African nations in closing the digital divide. Building digital infrastructure, moving toward a digital society, and utilizing cutting-edge technologies like mobile finance and the Internet of Things, have all been successful in this field. They have worked with significant African, operators to provide Africa with complete, fundamental telecommunications Service, coverage. More than half of the continent's highspeed mobile broadband networks and wireless sites have been constructed more than 900 million locals have access to broadband internet thanks to the installation of more than 200,000 km of special fiber. The first 5G independent commercial networking in the area was also made available by the two sides. China and Africa’s commerce cooperation is becoming more and more extensive. Increasing knowledge sharing and scientific and technological collaboration. In terms of technological innovation strategies, China actively improves coordination and communication with Africa. 2.6. Positive impact on how African countries adopted Chinese business strategies Respect for one another and progress Cooperation between developing nations in China and Africa focuses primarily on mutual support. China has developed a unique strategy for win-win cooperation with Africa over the years. It represents cooperation and solidarity between China and Africa and can be used to enhance international ties with Africa. Facilitating Africa's agricultural development. China has always been open to helping them develop their agricultural training there and sharing agricultural production and processing. More than 50,000 Africans have received training through initiatives like sending Chinese agricultural experts to Africa, and 23 agricultural demonstration centers have been constructed China has signed 72 bilateral and multilateral agreements for agricultural cooperation and has established mechanisms for cooperation with 23 African nations and regional organization. Maintaining equality, friendship, and sincerity. In order to achieve a common future, Chinese and African people have collaborated. Africa is respected, valued, and supported by China. Maintaining inclusivity and transparency. China is prepared to assist Africa and promote peace and prosperity alongside other international allies. All actions that enhance the interests of Africa are welcomed and supported by this organization. For the continent’s inclusive and sustainable development, industrialization is necessary in addition to job creation, poverty eradication, and raising living standards. Through its financial and investment cooperation with the continent, China will not meddle in the internal affairs of individual African countries, impose its will on those governments, or tack on any political benefits. China will Depending on their unique circumstances and development requirements, China assists African nations in improving their "soft" and "hard" investment environments. China uses 11 local Africans hail them as the white health messengers, role models of South cooperation, and warmly welcomed visitors. industrial alignment and capacity cooperation as its driving forces to advance Africa's industrialization and economic diversification. China has developed mechanisms for industrial capacity cooperation with 15 African countries. To entice businesses to Africa, China and African countries have collaborated to create special economic zones for trade, economic cooperation, industrial parks, and science parks. They have set up production and processing facilities and concentrated their operation Africa in order to boost local employment, tax revenues, technical cooperation, and industrial advancement. y The Africa Center for Disease Control and Prevention receives technical assistance from Chinese disease control specialists, which helps African nations better assess quarantine and border health. Increasing cooperation in education and human resources. China actively supported education in Africa. 2.6. Positive impact on how African countries adopted Chinese business strategies The China Africa Joint Research Center, an innovation cooperation center, and high level joint laboratories have all been established by China and African nations. Through initiatives like the Belt and Road Region Scholarship from the Alliance of International Science Organizations, Chinese government scholarships, the Talented Young Scientist Program, and the Innovative Talent Exchange Project, China has recently helped Africa develop a significant number of scientific and technological talents. Space cooperation has made progress. The two parties collaborate in the areas of precision agriculture, radio astronomy, satellite navigation and positioning, and disaster prevention and mitigation using China's Remote sensing data. Expanding Eco environmental protection and working together to combat climate change. A shared desire for a stunning environment and an improved quality of life unites the peoples of China and Africa. They collaborate to promote sustainable, recyclable, low carbon development and protect humanity's shared homeland as a result. They have collaborated on two events since 2012: a ministerial conference on China-African environmental cooperation, which encourages communication and coordination of environmental governance policies. The seminar on green cooperation will guide the future economy. In 2020, China and African countries joined the China-Africa environmental cooperation center, along with other international organizations, academic institutions, and companies. Increasing social development cooperation In social areas like poverty reduction, health, education, science and technology, environmental protection, climate change, and exchanges between young people and women, China is fostering collaboration with Africa. China is assisting African nations in enhancing their comprehensive social development by fostering exchanges, offering assistance, and sharing experiences, which in turn drives their economic growth internally. y Sharing knowledge of reducing poverty. China and Africa frequently struggle with poverty. The UN 2030 Agenda for Sustainable Development's main objective is to eradicate poverty. China has successfully eradicated poverty using Chinese characteristics, with hundreds of millions of impoverished people having managed to escape poverty, serving as a model for addressing the issue in Africa. China has successfully implemented the Program for Strengthening China-Africa Cooperation on Poverty Reduction. Through initiatives like the FOCAC Africa-China Poverty Reduction and Development Conference and the China-Africa Youth Exchange Program on Poverty Development, China has aided local governments, academics, businesses, youth organizations, and nongovernmental organizations in both China and Africa in carrying out various forms of exchange and pragmatic cooperation on poverty reduction. 2.6. Positive impact on how African countries adopted Chinese business strategies People-to-people interactions and cultural exchange Chinese culture has a saying that goes, "SteadyState relations flourish when there is friendship between the peoples." In China and Africa, conversations between young people and women as well as cultural, media, scientific, technological, and think tank exchanges have been encouraged. These initiatives have strengthened bonds between people and laid the groundwork for future advancements. Enhancing the relationship between medicine and health. People and lives have always been prioritized by the Chinese government. China has supported African nations in their efforts to combat various epidemics and plagues, establish a public health system ,and advance health in China and Africa. Chinese medical teams are dispatched as part of one of the longest and most successful projects cooperation involving the most African nations. China medical aid to Africa officially began with the first one, which was sent to Algeria in 1963. Over the past 58 years, China has treated 230 million patients in Africa, employing 23, 000 medical professionals. Nearly 1,000 Chinese medical professionals are currently employed by 98 medical centers in 45 African countries. The Expanding collaboration and interactions in the cultural and tourism sectors. The two sides continue to sign implementation plans for intergovernmental agreements on programs like the Year of China, Happy Spring Festival, Chinese and African Cultures in Focus, and Insight on China to further encourage cross cultural exchanges and cooperation. By December 2020, 346 of these implementation plans had been written and put into action. Since 2016, China has hosted hundreds of cultural seminars for African countries, with a total of almost 1,500 African participants. There are cultural 12 hubs for Chinese people in Morocco, Mauritius, Benin, Egypt, Nigeria, and Tanzania. The Chinese government has signed agreements with counterparts in Tunisia, Kenya, ivory coast, Senegal, Ethiopia, and Mozambique to establish Chinese cultural centers. Peace and security cooperation. and negative effects. And in my research paper I will use qualitative method which focuses on obtaining data through open minded and conversational communication, by gathering information using different ways like documents, case studies. Development is impossible without security and stability. China has always supported idea of Africans finding their own solutions to problems in Africa and has emphasize the necessity of addressing both the underlying causes and the symptoms, as well as the need for cooperation between all parties. China is a valuable player in African peace and security affairs. 3.3. Limitations The qualitative character of the study and the language barrier are its two main drawbacks. It is crucial for the researcher to interpret the data because qualitative research is by its very nature interpretive in nature. Thus, there is a chance for skewed interpretation, normative analysis, or value judgement. Second, because the researcher is not proficient in Mandarin, she may only use sources of data in English for her study. As a result, this study does not incorporate any further information or insightful conclusions that might have been drawn from texts and data that have been published in Mandarin. He pushed for a quicker implementation of the FOCAC Beijing summit’s goals, including improved public health, an economic reopening and greater consideration for people’s livelihoods. He also announced a number of anti pandemic measures that have gained widespread praise and support in Africa, such as a dept relief, debt service suspension, and economic reopening. the participants joint statement of the extraordinary China-Africa summit on solidarity against covid19, which was signed, reaffirmed their dedication to collaborating to combat the pandemic, promote cooperation and foster friendship. The two sides have worked closely together since the summit to reduce pandemics, rekindle economic activity, enhance China-Africa cooperation. 4.1. Strategies Adopted by China in Africa When China's Chairman, Mao, imparted information about the fear that may overtake the US if Asia, Africa, and Latin America were united, the China Africa strategy was born. The death of Mao, who was succeeded by Deng Xiaoping, however, delayed the relations and shifted China's emphasis to its own internal economic development. Despite the fact that this choice boosted China's productivity, China's later engagement with other African nations was prompted by the necessity for production resources and a market to supply its goods. The "low cost strategy" is one of the various strategies that have been advocated by China to enter the African market segment. The superior, cutting-edge technology that China has developed throughout the years and is still developing has allowed the Chinese to critique the interest of China doing business in Africa. 3.1. Nature of Research The primary descriptive aspect of this study is that it gives a thorough overview of China's business strategy in Africa during a specific time span. But the study also aims to analyze the nature of China's business interactions with Africa, which makes it exploratory. The research for this study is therefore mostly descriptive, but it also includes exploration and explanatory components. 3.2. Data gathering The study uses a primarily qualitative method to data collection. It mostly draws on books and periodicals that are considered to be scholarly literature. Publications from the institution and the center for Chinese studies contain more information. Additionally, publications are used, including pieces from print newspapers and online sites. An argument about the nature of China's commercial strategy involvement in Africa is formed by combining the facts, opinions, and arguments offered in the literature. Covid19 claims that by working together and supporting one another to end the pandemic, China and Africa have overcome a significant barrier. When the Extraordinary China-Africa summit on solidarity against covid19 was broadcast via video link in June 2020, China became the first country in the world to hold an anti pandemic summit with Africa. It has given international anti pandemic cooperation with Africa and given it new impetus. President Xi discussed crucial policies for enhancing China cooperation and bolstering global cooperation in the wake of the pandemic at the summit. 2.6. Positive impact on how African countries adopted Chinese business strategies China backs their initiatives to combat terrorism and advance peace and stability. In nations and regional organizations like the AU, China supports the "Silence the Guns in Africa" initiative. China is actively looking into way to positively foster and preserve Africa's peace and security, based on the principles of respecting African nations' wishes, refraining from interfering in their internal affairs, and adhering to the fundamental rules governing international relations. The reason why I choose to use qualitative method is because I’m already in the country that I’m working on which I think is easy for me to get some information from books, internet, interviews, questionnaire. 4. Strategies Adopted China received strong support and assistance from African countries and regional organizations like the AU during its most trying times to combat the epidemic. The AU Executive council supported China’s anti-pandemic efforts in a statement it released in February 2020. A significant regional organization or an entire continent had never before offered China such support. After covid19 struck Africa the largest scale program of its kind and the most difficult to implement since the PRC’s founding China immediately provided humanitarian aid. Since 2020, the central government of China has worked with regional governments, businesses, and social organizations to provide emergency anti-pandemic supplies to 53 African nations and the AU. Nearly every area of the continent has received these emergency supplies. 3. Methodology The objective of this research is to study the effects of Chinese business strategies has on African countries. So this chapter aims to collect secondary data, conceptual framework. It’s clear according to our second chapter there are positive China's Chairman, Mao, gave birth to the China Africa strategy. Later, China began interacting with more African nations due to a demand for production resource and an 13 giving it a competitive advantage over its rivals. The Chinese have opted for a strategy in which there are no conditions placed on their investments. This strategy has been so effective that it has taken the place of the West's benevolent approach, which was perceived as an effort to rule Africa. China was able to conduct in-depth research on Africa, which helped them comprehend the wide diversity of market segments that are present and how urgently the continent needed equitable investment conditions. China conducted a thorough examination of the market (Africa) and provided reasonable investment terms, in contrast to its previous trade partner. The worry that many of its states may collapse has been Africa's darkest dread. The concern that many of its nations will be over-exploited and profit from enterprises operating within their sphere of influence has long been Africa's biggest nightmare. However, as a result of China's provision of mutually beneficial accords, many countries have opened up to free trade and received large investments from the continually growing powerhouse. exclusive market to supply its output. The "low cost strategy" is one of the many strategies supported by China to enter the African market. Thanks to their cutting-edge, continually improving state-of-the-art technology, the Chinese have been able to make and sell things at afford able prices. In addition to the aforementioned, China has now gained access to Africa's natural resources, which are provided to its producers in the Peoples Republic of China. The West has previously rejected these projects. Chinese interest in conducting business in Africa has been sparked by their delight in the manufacturing sector's economies of scale. The Chinese have demonstrated innovation by applying a "low-cost strategy" to more than just manufactured goods. Chinese loans, which are much more accessible than those provided by the World Bank and the International Monetary Fund, have been used as a target for the strategy. Many African nations have openly welcomed China as a result of the affordability of its goods and non-conditional investments. 4.2. China's strategies' effectiveness Although most of China's strategies in Africa over the past ten years have been criticized as attempt to lubricate its imperialism there, they have proven to be far superior. Although the vivid actions demonstrate the need to create a dynamic business environment and the necessity of securing stable resource supply for China’s accelerating industry, there is currently little evidence to support the rumors of Chinese intentions over Africa. China's "low-cost strategy" for the majority of its export goods has come under fire as being dumped. Dumping is the practice of unloading a stock without considering its price. Due to the significant benefits that Chinese state-owned and private businesses provide to the government of their host nation, these criticisms have gotten little attention. This has made it possible to portray the low-cost strategy as predatory pricing. Chinese State-owned and privately-owned companies have been able to create goods at lower costs as a result of which they are not only able to be the leading suppliers but also producers, acquiring a greater market share by forcing the exit of competing companies as a result of either destructive or predatory pricing or low-cost strategy. p The dangerous effects of China's products on the domestic market in Africa have been the main charge leveled against it. Although governments have seen citizens benefit from the second-largest global economic power, it is thought that domestic business growth and development have been sharply declining. African businesses have suffered as a result of the impact of dumping on the domestic market, which has prompted Chinese retaliation. Other African nations have adopted anti-dumping policies and increased import duties of Chinese goods as a result of learning about Chinese customs. This is done to protect the local small and large industries from the severe pressure the Chinese products are putting on them. China's ability to provide loans has been successful both in western countries and in Africa. In order to secure resources at a fair price, China's ideal strategy in the West, where it seeks to equate the Yuan with the US dollar, has been to inflate the currencies of the countries it deals with. This has been made feasible by China's aggressive attempts to acquire US and African Treasury Bills utilizing inexpensive debts that are used to gain access to the host nations in an effort to influence and shape them to its growth goal. 3. Methodology In comparison to the West, the Chinese have an advantage because they can conduct business without interfering with the internal affairs of the host nation. Chinese embassies are established on the continent. Criticize China's desire to conduct business in Africa. What instances of this can be found on the continent of Africa? What is China using on the continent of Africa? How well suited are these tactics? What prevented China's interest in Africa from growing or being resisted? Some have criticized China's lack of effectiveness, despite the fact that it has successfully implemented strategies that have allowed it to grease its investment without restriction. Despite the fact that Chinese policies have given it a competitive advantage over its competitors, there has been a rise in opposition to Chinese growth in other African nations. Although there is some empirical support for the rumors, it is still thought that they were spread by journalists who were paid by the West in order to discredit China's superior commercial abilities. 4.3. Failure and Resistance to Chinese Expansion in Africa The US has viewed China's quick entry into Africa and expansion there through the application of wise strategic policies as a threat, despite the fact that Africans have welcomed Chinese generosity. The US thinks that China's entry into Africa could endanger security and reduce opportunities for US-Africa. The US and the rest of Europe may lose their former backyard gardens, which once provided the natural resources they needed to support their cutting-edge techno-scientific industries, as a result of extremely successful Chinese operations. These superpowers, who were once China's dominant trading partners, now view the expansion of China in Africa as a threat. 4.4. Analysis on The Effects of African Countries on Adopting Chinese Business Strategies How is African business after adopting Chinese business strategies; Africa is doing quite well after acquiring Chinese business because when you look back to African business before there’s a very big change which can be seen by everyone around the world, China have been working with African countries for long time in different fields like technology, construction, education, etc. The major trading partner and source of foreign investment on the African continent is China. Its investment has helped the economy grow and the construction of infrastructure. China has invested in Africa since the continent lacks markets for its exports and needs the continent's oil and other resources. What business lessons may be shared between China and Africa? Over time, China and Africa's economic links have become stronger. Africa's top trading partner at the moment is China, and throughout the 1990s, trade between the two continents surged by 700%. Another tactic is to establish a level playing field that will aid in lifting African and Chinese citizens out of poverty and transforming them into competitive global players. African people's economic and social policies have been neglected and have not been implemented quickly. g y As we have seen above Chinese business strategies has impacted Africa in so many ways which also shows the good relationship between them and as years goes by the relationship keeps getting stronger and stronger by investing more in Africa and we can see that since then Africa’s economy keep rising. criticize China's desire to conduct business in Africa. What instances of this can be found on the continent of Africa? The Chinese investment opportunities have a history of exploiting African resources, to put it mildly. Realizing the Chinese intentions has become difficult because of the way they conduct business. Since Chinese businesses typically engage in joint ventures that have no bearing on job creation and significant value addition to investment, the overall assessment of Chinese business has been observed to have little to no impact at all. Due to their involvement in both legal and illegal activities, China's operations have been erratic in some nations. Due to the fact that the Chinese investments' negative effects outweigh their positive ones, it has become difficult to distinguish between the good and bad. Most African governments have benefited most from Chinese investments at the expense of the general populace. 5.2. Conclusion It's crucial to remember that despite China's successful establishment of substantial investments from Cape Town to Cairo, it is still uncertain whether these investments would lead to a more sustainable development than past Western African investors. It should be highlighted that the Chinese investment has generated a great deal of buzz and attention, potentially paving the way for continued trade between China and Africa. As a result, significant investments were made, which has helped to Africa's slow growth, albeit this progress may be hampered by the continent's heavy reliance on exporting natural resources. This will suggest that China's voracious appetite for natural resources ignores the lack of corporate control on Chinese enterprises, the mismanagement, the limited diversification, the sensitivity to volatile commodity prices (raw materials), and other factors. 4.2. China's strategies' effectiveness China has made sizable investments in the continent, but labor abuses, such as illegitimate safety practices, hostility toward trade unions and risky shift extensions, have come under fire. It is thought that China is where many of these appalling working conditions were seen in Africa. It is thought that Chinese businesses put safety before productivity, forcing workers to work even in hazardous conditions. Chinese expansion has been hampered as a result, and occasionally they have had to stop operations in nations like Zambia where they consistently work in subpar conditions. By engaging with Africa through a non-conditional approach, China has displaced several Western countries, 14 These working conditions have been imposed by the Chinese on the continent not only in Zambia but also in Zimbabwe, the Democratic Republic of the Congo, Mozambique, and Angola. It has been reported that due to South Africa's stringent labour rules, the Chinese have maintained a perfect safety audit. capacity. capacity. 5.1. Summary y Furthermore, it should be noted that China places little value on people's well being in favor of its own operations. For instance, one of the largest buildings in the world is being built. The government did not take into account the effects of the infrastructure project on the displaced local population or potential natural disasters like earthquakes when building hydroelectric power plants. These have been seen in South Sudan's largest power plant, which was constructed by the Chinese. As a result, China's success in expanding its empire has been restricted to the investments it has already made in Africa due to its fear of the Chinese dark background. In order to safeguard its interests, industry, and citizens from improper trade and working practices, nations like South Africa have carefully monitored the Global Economic Giant as a result of China's practices. The major trading partner and source of foreign investment on the African continent is China. Its investment has helped the economy grow and the construction of infrastructure. China has invested in Africa since the continent lacks markets for its exports and needs the continent's oil and other resources. 4.4. Analysis on The Effects of African Countries on Adopting Chinese Business Strategies As a result, many African governments have been able to defend their citizens when they are caught engaging in illegal activity, as was the case in Zimbabwe, where Minister Walter Mzembi defended the Chinese for stealing elephants despite the fact that he lacked physical proof to support his assertions that the elephant population in parks exceeded the ideal 42000 Establishing a level playing field will help to free African and Chinese citizens from poverty and help them become competitive international players. Economic and social policies of the African people have been neglected and not swiftly put into practice. 5.3. Recommendation According to this research and other papers that are published, there is a good relationship between China and Africa. Even though there are some challenges and the research I did on this topic I would recommend China and Africa to keep their relationship because China has helped a lot in building Africa in different fields like agriculture, investment, infrastructure etc. And there are some challenges but that won’t stop the cooperation between China and Africa if they keep working together, they can solve whatever circumstances. And the positive impacts are way better than negative effects which also shows how African countries benefited from Chinese business strategies. References [1] What county is investing in African infrastructure ”Africa- China Relations-Taming the dragon: new frontiers of 15 cooperation?”(2021) London:The Economist Intelligence Unit, March 2021 cooperation?”(2021) London:The Economist Intelligence Unit, March 2021 27(110):208-223 (https://doi.org/10.1080/10670564.2018/1389016) 27(110):208-223 (https://doi.org/10.1080/10670564.2018/1389016) [6] [2] Baah,A.Y. and Jauch,H.(eds) (2009) Chinese investment in Africa:a labour perspective. Accra and Windhoek:Africa Labour Research Network. [6] Paul Nantulya, “Chinese Hard Power Supports Its Growing Strategic Interests In Africa”,Africa Center for strategic studies , spotlight, January17,2019. [3] Lee,C.K (2017) The specter of global China: politics, labor, and foreign investment in Africa. Chicago IL:University of Chicago press. [7] Eisenman ,J. and J. Kurlantzick, 2006,’China’s Africa Strategy’, Current History, May 2006, pp 219-24. [8] Van de Looy,J.,2006, China,’Africa and China: A Strategic Partnership? ASC Working Paper 67/2006, Africa Studies Centre, Leiden, The Netherlands. [4] Lin,J.Y.,Cai,F. and Li,Z. (1996) The China miracle: development strategy and economic reform. Sha Tin: Chinese University press. [9] Kaplinsky, R., D.McCormick, and M. Morris,2006, The Impact of China on sub saharan Africa. [5] Qobo,M. and le pere,G.(2018)’The role of China on Africa’s industrialization: the challenge of building global value chains 16
https://openalex.org/W3215332902	https://www.frontiersin.org/articles/10.3389/fpsyg.2021.746364/pdf	English	null	The Importance of Perceived Procedural Justice Among Defendants With a Non-Western Background Involved in Dutch Criminal Cases	Frontiers in psychology	2,021	cc-by	14,400	The Importance of Perceived Procedural Justice Among Defendants With a Non-Western Background Involved in Dutch Criminal Cases Lisa F. M. Ansems 1, Kees van den Bos 1,2 and Elaine Mak 1 1 School of Law, Utrecht University, Utrecht, Netherlands, 2 Department of Psychology, Utrecht University, Utrecht, Netherlands Keywords: perceived procedural justice, outcome judgments, perceived everyday discrimination, attenuation, reversal, critical test, criminal defendants, external attributions ORIGINAL RESEARCH published: 29 November 2021 doi: 10.3389/fpsyg.2021.746364 Correspondence: Lisa F. M. Ansems l.f.m.ansems@law.leidenuniv.nl Specialty section: This article was submitted to Forensic and Legal Psychology, a section of the journal Frontiers in Psychology Received: 23 July 2021 Accepted: 27 October 2021 Published: 29 November 2021 ORIGINAL RESEARCH published: 29 November 2021 doi: 10.3389/fpsyg.2021.746364 Edited by: Melissa de Vel-Palumbo, Flinders University, Australia Edited by: Melissa de Vel-Palumbo, Flinders University, Australia This study aims to put perceived procedural justice to a critical test in the context of Dutch criminal court hearings. To that end, we surveyed 198 criminal defendants to examine whether their perceptions of procedural fairness were significantly associated with trust in judges and intentions to protest against judicial rulings, among other variables. We also examine the possibility that sometimes unfair procedures may have nice aspects, because they offer opportunities to attribute negative outcomes to external causes. Previous studies conducted in different settings support this line of reasoning by showing that associations between perceived procedural justice and other variables are sometimes attenuated or even reversed, particularly when people feel strongly evaluated. The current study takes these insights into the novel context of Dutch criminal court hearings by focusing on defendants with a non-Western ethnic-cultural background. Some of these defendants may feel negatively evaluated by society, which can manifest as a high level of perceived discrimination. Thus, we examine whether the associations between perceived procedural justice and important other variables may be attenuated or reversed depending on respondents’ perceptions of everyday discrimination and their outcome judgments. Our results revealed significant associations between perceived procedural justice on the one hand and trust in judges and protest intentions on the other hand, which remained intact regardless of perceptions of everyday discrimination and outcome judgments. Hence, even in this real-life courtroom context, procedural justice was a relevant concern. Taken together, our findings support the importance of perceived procedural justice, even when it is put to a critical test. Reviewed by: Chris Martin, Georgia Institute of Technology, United States Jon Jackson, London School of Economics and Political Science, United Kingdom Correspondence: Lisa F. M. Ansems l.f.m.ansems@law.leidenuniv.nl 1Previous research suggests that criminal defendants sometimes find the term “fair” easier to understand and use than “just” and has therefore treated these terms as synonyms (Ansems et al., 2020). Given the similarity in research context, the current paper uses the terms “justice” and “fairness” interchangeably as well. INTRODUCTION Hence, one may wonder whether perceived procedural fairness is a relevant concern for defendants in these criminal cases, as they may be much more concerned with their case outcomes. Procedural Justice and the Fair Process Effect For instance, when people do not know whether or not they can trust authorities, they may look to whether or not these authorities treat them fairly and use this information as a heuristic substitute (e.g., Lind et al., 1993; Van den Bos and Lind, 2002).h To shed further light on this issue, the current paper focuses on the context of real-life criminal court hearings involving actual stakes and examines whether perceptions of procedural fairness are associated with trust in judges and intentions to protest against the judicial ruling, among other variables. More specifically, as our first hypothesis we propose the following: To shed further light on this issue, the current paper focuses on the context of real-life criminal court hearings involving actual stakes and examines whether perceptions of procedural fairness are associated with trust in judges and intentions to protest against the judicial ruling, among other variables. More specifically, as our first hypothesis we propose the following: Defendants with higher levels of perceived procedural justice report more trust in judges, more positive outcome judgments, lower intentions to protest against their outcomes, and higher levels of state self-esteem (i.e., self-esteem at the moment of filling out the questionnaire; Hypothesis 1). The current study puts procedural justice and the fair process effect to a critical test by assessing whether the associations between procedural justice perceptions and important other variables hold in the real-life courtroom context of Dutch criminal cases. Contrary to the psychological laboratories which provide the research context of many procedural justice studies, criminal court hearings involve actual stakes, with defendants risking sanctions ranging from fines to community service and imprisonment. In addition, convictions can have serious consequences for defendants’ positions on the job market. Citation: Ansems LFM, van den Bos K and Mak E (2021) The Importance of Perceived Procedural Justice Among Defendants With a Non-Western Background Involved in Dutch Criminal Cases. November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 1 Procedural Justice in Criminal Cases Ansems et al. Procedural Justice and the Fair Process Effect In an early critique of procedural justice research, Hayden and Anderson (1979; see also Anderson and Hayden, 1981) point to the use of simulation experiments, which necessarily involve a simplification of social situations, and the resulting limitations as to the conclusions that can be drawn from these studies. Others, too, note the importance of considering real- world complexities. For example, results from a study by Berrey et al. (2012) suggest that litigants involved in employment discrimination cases often did not distinguish between procedures and outcomes, and that they defined fairness in terms of whether or not the procedure benefited their own side. In line with this, Jenness and Calavita (2018) argue that, in their sample of incarcerated men in the United States, participants’ concerns about procedural fairness were largely subordinate to (or even defined by) their outcome concerns. Fair and just procedures are key aspects of law. Issues of procedural fairness, when viewed from a legal perspective, concern the extent to which legal procedures meet standards laid down in statutes, case law, and unwritten legal principles. In contrast, social psychologists empirically study the extent to which procedures correspond with citizens’ ideas about fairness and justice. These experiences of being treated fairly by decision-making authorities are referred to as perceived procedural justice1 (Lind and Tyler, 1988; Tyler and Lind, 1992). Fair and just procedures are key aspects of law. Issues of procedural fairness, when viewed from a legal perspective, concern the extent to which legal procedures meet standards laid down in statutes, case law, and unwritten legal principles. In contrast, social psychologists empirically study the extent to which procedures correspond with citizens’ ideas about fairness and justice. These experiences of being treated fairly by decision-making authorities are referred to as perceived procedural justice1 (Lind and Tyler, 1988; Tyler and Lind, 1992). When people perceive procedures as fair, they tend to be more satisfied with the outcomes of their cases and more inclined to accept those outcomes (Thibaut and Walker, 1975; Lind et al., 1993; Van den Bos et al., 2014). They also tend to report higher levels of self-esteem and trust in judges (Koper et al., 1993; Sedikides et al., 2008; Grootelaar, 2018). Other important attitudes and behaviors associated with perceived procedural justice are perceived legitimacy, cooperation with legal authorities, and compliance with the law (Paternoster et al., 1997; Tyler and Huo, 2002; Tyler, 2006). Procedural Justice and the Fair Process Effect Such favorable responses to perceived procedural justice are generally referred to as the fair process effect (Folger et al., 1979; Van den Bos, 2015).h When people perceive procedures as fair, they tend to be more satisfied with the outcomes of their cases and more inclined to accept those outcomes (Thibaut and Walker, 1975; Lind et al., 1993; Van den Bos et al., 2014). They also tend to report higher levels of self-esteem and trust in judges (Koper et al., 1993; Sedikides et al., 2008; Grootelaar, 2018). Other important attitudes and behaviors associated with perceived procedural justice are perceived legitimacy, cooperation with legal authorities, and compliance with the law (Paternoster et al., 1997; Tyler and Huo, 2002; Tyler, 2006). Such favorable responses to perceived procedural justice are generally referred to as the fair process effect (Folger et al., 1979; Van den Bos, 2015).h i In contrast, other studies found that perceptions of procedural fairness matter even to people involved in high stakes cases. For instance, Landis and Goodstein (1986) reported that inmates’ perceptions of outcome fairness were associated with both procedural and outcome issues, but that procedural characteristics were dominant in this regard. In addition, Casper et al. (1988) showed that perceived procedural justice was significantly associated with multiple measures of outcome satisfaction among defendants in felony cases. Other findings also indicate the importance of perceived procedural justice in criminal justice contexts (e.g., Tyler, 1984, 1988, 2006; Paternoster et al., 1997; Tyler and Huo, 2002). f There are various explanations for people’s concern with issues of procedural fairness (see, e.g., Van den Bos, 2005). For instance, in their pioneering research on this topic, Thibaut and Walker (1975) propose an instrumental explanation of the fair process effect by suggesting that people care about procedural justice because fair procedures are more likely to yield fair and favorable outcomes. Others argue that people care about procedural fairness for relational reasons, as being treated fairly communicates to them that they are valued members of society (e.g., Lind and Tyler, 1988; Tyler and Lind, 1992). Furthermore, information about procedural justice may help to make sense of uncertain situations. Attenuated or Reversed Fair Process Effects Our study also adds to current insights into procedural justice in criminal justice contexts by examining the possibility that sometimes unfair procedures may have nice aspects (Van den Bos et al., 1999). As explained by Brockner et al. (2009, p. 185), these “reductions in people’s desire for higher process fairness” may result in attenuated or even reversed fair process effects. That is, the associations between perceived procedural justice and relevant other variables may be weakened, possibly to the extent that they are no longer statistically significant, or reversed, such that people respond more favorably to perceived procedural unfairness than to perceived procedural fairness (Brockner et al., 2009). November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 2 Procedural Justice in Criminal Cases Ansems et al. One explanation for attenuated or reversed fair process effects is people’s self-enhancement motive. That is, people often want to feel good about themselves (Leary and Terry, 2013). When people perceive procedures as fair, they are more likely to view themselves as personally responsible for their outcomes and thus make more internal attributions (Leung et al., 2001; Brockner et al., 2009). Unfavorable outcomes may then harm their self-esteem (Weiner, 1985). To protect their self-esteem, people may look for opportunities to attribute these negative outcomes to external causes (Cohen 1982). Procedures that people perceive as unfair offer such external attribution opportunities (Brockner et al., 2003). That is, people may maintain their self-esteem by attributing negative outcomes to the perceived unfairness of procedures rather than to themselves. Hence, for people who receive negative outcomes unfair procedures can have nice aspects, at least under some circumstances (Van den Bos et al., 1999).h of these defendants may feel negatively evaluated by society, which can manifest as a high level of perceived discrimination (Huijnk and Andriessen, 2016). We propose that, as a result, these defendants might respond differently to perceptions of procedural fairness during their court hearings (Van den Bos et al., 1999). Hence, we assess whether the associations between perceived procedural justice and relevant other variables may be attenuated or reversed depending on how much discrimination defendants experience in their daily lives. In doing so, we also take into account research on discrimination which examines processes similar to the self- enhancement processes outlined earlier. Perceived discrimination can lead to various problems, including stress and reduced psychological well-being (Major et al., 2002). 2In this paper, a “non-Western ethnic-cultural background” refers to being born in a non-Western country, which according to Statistics Netherlands (2018) refers to countries in Africa, Latin America, and Asia (excluding Indonesia and Japan), or Turkey. We also use the term to refer to persons whose parents or other ancestors were born in a non-Western country. We included these latter respondents in our sample because they, too, might experience discrimination in their daily lives due to their ethnic-cultural background. Attenuated or Reversed Fair Process Effects At the same time, experiencing discrimination may enable people to maintain their self-esteem, as it reduces their sense of personal responsibility and deservingness of negative outcomes (Crocker and Major, 1989; Major, 1994). Hence, attributing negative events to discrimination rather than one’s personal qualities is a coping strategy people can use to counter the negative impact these events may otherwise have on their self-esteem (Major et al., 2002). Building on these insights, we formulate our second hypothesis as follows: This line of reasoning is supported by a number of empirical studies, which show that the positive association between perceived procedural justice on the one hand and self-esteem or related measures on the other hand may sometimes be attenuated or even reversed when outcomes are perceived as unfavorable in work contexts (e.g., Gilliland, 1994; Ployhart et al., 1999; Schroth and Shah, 2000; Brockner et al., 2008, 2009; Brockner, 2010). Some studies have found attenuated or reversed fair process effects when examining different kinds of dependent variables. Thus, possibly, people’s reduced threat to sense of self due to perceived procedural unfairness (or, vice versa, their heightened threat to sense of self due to perceived procedural fairness) translates into other kinds of reactions. For instance, one study found an attenuated fair process effect on trust in judges among research participants with relatively high external attribution ratings (see Ansems, 2021). In addition, Holmvall and Bobocel (2008) found a reversed fair process effect on measures of perceived outcome fairness and outcome satisfaction. Furthermore, Van den Bos et al. (1999) found a reversed fair process effect on protest intentions in the face of unfavorable outcomes manipulated in laboratory experiments. In their study, the reversal was triggered by the strength of the evaluative context: Participants who felt strongly evaluated during the decision-making procedure reported lower protest intentions when they perceived the procedure as unfair rather than fair. Thus, feeling strongly evaluated can play an important role in attenuating or reversing the fair process effect. There is a two-way interaction between perceived procedural justice and perceived everyday discrimination, such that defendants who experience relatively high levels of everyday discrimination show attenuated or reversed associations between perceived procedural justice and our other variables (i.e., trust in judges, outcome judgments, protest intentions, and state self-esteem; Hypothesis 2). Outcome Judgments g In addition to perceptions of everyday discrimination, we examine the potentially moderating role of defendants’ outcome judgments (i.e., how positively or negatively they judge their outcomes). According to Brockner and Wiesenfeld (1996), receiving negative outcomes triggers sense-making processes. As a result, people may pay more attention to issues of procedural fairness. Thus, the fair process effect may be strengthened when people receive negative outcomes. At the same time, people may look for opportunities to attribute these negative outcomes to external causes in order to protect their self-esteem (Cohen 1982), as explained earlier. Because procedures that are perceived as unfair offer such opportunities, the fair process effect may be attenuated or reversed (Brockner et al., 2009). Perceived Everyday Discrimination To examine these issues, the current research focuses on perceived procedural justice among defendants with a non-Western ethnic- cultural background2 involved in Dutch criminal cases. Some Research Context h Against this backdrop, the present study examines whether experiences of everyday discrimination and outcome judgments may moderate how defendants with a non-Western ethnic- cultural background involved in Dutch criminal cases react to perceived procedural justice during their court hearings. Our study helps to refine current insights into perceived procedural justice by focusing not only on the possible robustness of associations between perceived procedural justice and relevant other variables, but also on the potential attenuation or reversal of these associations. In doing so, we take insights from previous studies conducted in work settings and the psychological laboratory (e.g., Van den Bos et al., 1999; Brockner et al., 2009) and apply them to the novel context of Dutch criminal cases. To study these issues, we conducted a face-to-face survey among 198 defendants with a non-Western ethnic-cultural background involved in Dutch single judge criminal cases. In the Dutch court system, single judges handle relatively simple criminal cases in which the sanction demanded by the public prosecutor does not exceed 1 year of imprisonment. Typical cases handled by single judges include theft, simple assault, and traffic offenses such as driving under the influence. Defendants can be represented by a lawyer, or they can choose to defend themselves. Usually, single judge criminal court hearings last around 30 min and judges render a verdict directly afterward. The description below provides more details on the Dutch legal context and some of the main differences with the legal system of (for example) the United States. In addition, because of our focus on defendants with a non-Western ethnic-cultural background, this study sheds light on experiences of a relatively underinvestigated research population. Contrary to research participants in many other procedural justice studies, respondents in the current study generally have non-WEIRD (Western, Educated, Industrialized, Rich, and Democratic; Henrich et al., 2010) backgrounds. Taken together, we conducted our study in a real-life courtroom context, focusing on defendants with diverse ethnic-cultural backgrounds who might respond differently to perceived procedural justice. In this way, we critically examine the role of perceived procedural justice in Dutch criminal court hearings. “First, Dutch criminal proceedings take place largely “on paper”. That is, the emphasis is on the pretrial investigation rather than on court hearings, which generally last around 30 min in small criminal cases and 60–90 min in more severe ones. Second, the Dutch legal system does not have a plea-bargaining system like the United States. Perceived Everyday Discrimination Combining these possibilities, our third hypothesis suggests that the associations between perceived procedural justice and our other variables may be moderated by defendants’ outcome judgments. When defendants judge their outcomes more negatively, this may either strengthen the associations between perceived procedural justice and our other variables (Brockner and Wiesenfeld, 1996) or, alternatively, attenuate or even reverse these associations (Brockner et al., 2009). Thus, we formulate our third hypothesis as follows:h To examine these issues, the current research focuses on perceived procedural justice among defendants with a non-Western ethnic- cultural background2 involved in Dutch criminal cases. Some There is a two-way interaction between perceived procedural justice and outcome judgments, such that defendants who judge November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 3 Procedural Justice in Criminal Cases Ansems et al. that people with a non-Western migration background report relatively high levels of perceived discrimination (Huijnk et al., 2015; Huijnk and Andriessen, 2016; Andriessen et al., 2020). Indeed, there are signs that Dutch people with a migration background may be discriminated in important life domains (e.g., Thijssen et al., 2019). their outcomes more negatively show stronger, attenuated, or reversed associations between perceived procedural justice and our other variables (i.e., trust in judges, protest intentions, and state self-esteem; Hypothesis 3).3 Research Context Third, the administration of justice is entirely in the hands of professional judges; the Dutch legal system does not have bifurcated proceedings in which defendants’ guilt is determined by a jury and their sentences by a judge. Fourth, criminal court hearings in the Netherlands are less adversarial than in the United States. That is, Dutch hearings involve an active role for judges and traditionally treat defendants as subject of the investigation, whereas the United States legal system involves more passive judges and views the court hearing as a clash of parties.” (Ansems et al., 2020, p. 648) 3In this paper, we focus on the hypotheses that are central to the line of reasoning presented here. We also performed a small number of additional analyses. For example, we examined whether there was a significant three-way interaction between outcome judgments, perceived everyday discrimination, and perceived procedural justice. This analysis was conducted for exploratory purposes only, however, and is not reported in the current paper. After all, a power analysis (Faul et al., 2007) showed that, to achieve sufficient statistical power of 0.80 (Cohen et al., 2003) to detect the three-way interaction, with α = 0.05 and a relatively small effect size (f2 = 0.02), at least 387 respondents were needed. Complete details and results are available with the first author on request. 4Like some other Dutch courts, the court of the mid-Netherlands has buildings in multiple cities (including Utrecht, Lelystad, and Almere). Sample Our sample consisted of 198 defendants with a non-Western ethnic-cultural background who appeared before a single judge at the court of the Mid-Netherlands in Utrecht, Lelystad, and Almere.4 Table 1 details sample characteristics. Relatively, many defendants in Dutch criminal cases have a non-Western ethnic-cultural background. People with a Moroccan or Antillean background in particular are overrepresented in Dutch crime statistics, which could be due partly to negative stereotypes and ethnic profiling (Huijnk and Andriessen, 2016). Discrimination is a relevant issue in Dutch society more broadly, too, as several studies show Research Proceduret After gaining the court’s permission to conduct the study, we collected our data between January 21 and October 15, 2019. Except for the summer break, the first author went to the court almost every work day during this period to collect data and stayed there for the duration of the criminal court’s session that day (most often from 9 to 18 h, sometimes from 9 to 13 h, or from 13 to 18 h). Among the causes for the relatively long duration of data collection were our focus on defendants with a non-Western ethnic-cultural background, the fact that many defendants did not appear for their court hearings, and some defendants’ poor command of Dutch. November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 4 Procedural Justice in Criminal Cases Ansems et al. TABLE 1 \| Sample description. 5After having signed a confidentiality agreement, at the beginning of each week of data collection the first author received an overview of single judge criminal cases that would be heard that week, detailing times, charges, and defendants’ names. We used these overviews to decide in which hallway to wait for potential respondents when there were multiple court hearings taking place at different floors at the same time (to be able to approach as many eligible defendants as possible) and we shredded these overviews at the end of each week for reasons of confidentiality. Research Proceduret Categorical variables Variable Category N % Location of court hearing Utrecht 190 96.0 Lelystad 6 3.0 Almere 2 1.0 Gender Male 178 90.4 Female 19 9.6 Highest completed level of education Primary school 14 7.4 Secondary school 81 42.6 Secondary vocational education 62 32.6 Higher professional education 24 12.6 University 6 3.2 Other 3 1.6 Special needs education 1 0.5 None at all 2 1.1 Ethnic-cultural background Moroccan 85 42.9 Surinam 25 12.6 Turkish 20 10.1 Antillean 12 6.1 Other (e.g., Somalian, Iraqi, Afghan) 58 29.3 Offense Assault or violence 57 30.0 Theft, embezzlement, fencing, or breaking and entering 45 23.7 Traffic offense (e.g., driving under the influence) 43 22.6 Threatening someone 19 10.0 Drug offense 17 8.9 Insulting someone 13 6.8 Destruction 12 6.3 Scam or fraud 5 2.6 Case outcome Convicted with imposition of sanction or measure (conditional or unconditional) 152 79.2 Community service 108 65.1 Fine 51 30.7 Prison sentence 27 16.3 Acquitted 26 13.5 Found guilty without imposition of sanction or measure 13 6.8 Discharged from further prosecution 2 1.0 Legal assistance By lawyer 136 70.1 By someone else 5 2.6 None 53 27.3 Number of previous criminal court hearings None 69 35.0 One 48 24.4 Two to ten 66 33.5 More than ten 14 7.1 Continuous variables Variable Range (years) Average (years) SD Age 18–66 30.10 10.75 Defendants appeared before the court of the mid-Netherlands if they were accused of a crime that had been committed in that geographic region or if they were living there.hi The first author approached defendants in the court hallway after they had made their presence known at the counter to ask whether they were willing to participate in a study about how fairly and justly they felt they were treated during their court hearings, indicating that they would be thanked for their participation with a small token of appreciation. Seventeen respondents (8.6% of the sample) were approached by a research assistant. We approached respondents before the start of their court hearings as much as possible to ask whether they were willing to participate in the study once their court hearings had ended. When it was not possible to approach respondents before the start of their court hearings – for instance, because they appeared for their hearings only very last minute or because they were consulting with their lawyers – respondents were approached immediately after their court hearings. Research Proceduret pp yt g Our study procedures were approved by the ethical board of the Faculty of Law, Economics, Governance, and Organization at Utrecht University. Following these approved procedures, and because our study focused on how people with non-Western ethnic-cultural backgrounds would respond to issues of procedural justice, we approached defendants who appeared to have a non-Western ethnic-cultural background for participation in our study. Therefore, based on their names and physical appearance, we made an initial assessment whether people appeared to have a non-Western ethnic-cultural background and invited those people to take part in our study.5 At the start of the questionnaire, we informed respondents that we were interested in how people who were born in a different country than the Netherlands and people whose parents or other ancestors were born in a different country than the Netherlands would evaluate how they were treated during their court hearings, their trust in Dutch judges, and how they felt treated in their daily lives, among other things. We ensured that we always treated people respectfully throughout the entire study. In fact, while filling out the questionnaire or afterward, multiple respondents indicated that they appreciated studies like ours, as these are needed to help to understand discrimination in Dutch society. We also note that, due to our way of sampling, we may have missed people who were eligible for participation in our study but whose physical appearance or name was not clearly non-Western. Although we cannot rule out that this may have affected our results, we do not think this was a big problem in the current study. After all, these people may be less likely to feel discriminated against based on their ethnic-cultural background, whereas we were particularly interested in defendants November 2021 \| Volume 12 \| Article 746364 5 Procedural Justice in Criminal Cases Ansems et al. targeting respondents’ perceptions more generally (state self- esteem and perceived everyday discrimination).6 who experience relatively high levels of discrimination in Dutch society and might therefore respond differently to perceived procedural justice. Our impression is that we were successful in conducting our study in responsible and sound ways. In the Discussion section, we note limitations of our study that may inspire future research. Research Proceduret We measured perceived procedural justice with a six-item scale based on the findings of a recent qualitative interview study conducted in the same criminal courtroom context as the current study (Ansems et al., 2020). Our survey items corresponded with the six core components of perceived procedural justice among defendants as revealed by the interview study. Specifically, we asked respondents to indicate, on a Likert scale from 1 (totally disagree) to 7 (totally agree), to what extent they agreed with the following six statements: “During the court hearing, I was treated in a pleasant way,” “During the court hearing, I was treated in an unprejudiced manner,” “During the court hearing, I was sufficiently able to tell my side of the story,” “During the court hearing, my side of the story was listened to,” “During the court hearing, everything important has been taken into account,” and “During the court hearing, my case was treated in a careful manner.” Together, these items formed a reliable scale (α = 0.82) on which higher scores reflect higher levels of perceived procedural justice. Therefore, we report the results of our analyses without the additional 11 items which we included as backup in case the six-item scale would turn out to be unreliable and which were based on previous work in other courtroom settings (Grootelaar and van den Bos, 2018).7 In addition to having a non-Western ethnic-cultural background, our other inclusion criteria were that defendants had received the outcome of their case and that they had a sufficient command of Dutch. We immediately filtered out respondents who, when starting to fill out the questionnaire in the court hallway, turned out not to meet our inclusion criteria and thus turned out to be ineligible for participation in our study. In total, we approached 447 defendants (excluding defendants who, based on this initial screening, turned out to be ineligible for participation). Of those 447 defendants, 210 filled out our questionnaire, resulting in a response rate of 47.0%. In a later stage, before conducting our analyses, we filtered out the questionnaires that did not indicate the respondent’s ethnic-cultural background or that had a very large number of missing values. Thus, the final sample consisted of 198 respondents. 6Because we report all measures used in our questionnaire, we note that we also measured respondents’ external attribution ratings and the extent to which they identified with their ethnic-cultural subgroup. The items we used to measure respondents’ external attribution ratings yielded a very low Cronbach’s alpha (α = 0.17), rendering this variable unsuitable for analysis. We included the items on subgroup identification in our questionnaire for potential additional analyses, as previous work suggests that attributions to discrimination might not protect the self-esteem of people who strongly identify with their ethnic- cultural subgroup (McCoy and Major, 2003). In the end, we did not perform these analyses because of power issues. Hence, we decided to drop these variables from the current paper. Complete details and results are available on request. Research Proceduret A power analysis (Faul et al., 2007) showed that, to achieve sufficient statistical power of 0.80 (Cohen et al., 2003) to detect the two-way interaction between perceived everyday discrimination and perceived procedural justice, with α = 0.05 and a relatively small effect size (f2 = 0.04), at least 191 respondents were needed. We also assessed respondents’ outcome judgments, which in this study include outcome satisfaction, perceived outcome fairness, and perceived outcome favorability. Our outcome judgments scale was largely based on previous research in a similar context (Grootelaar and van den Bos, 2018) and consisted of six items: “I find this ruling fair,” “I find this ruling favorable,” “I am satisfied with the judge’s ruling,” “I find this ruling just,” “The judge’s ruling has positive consequences for me,” and “I agree with the judge’s ruling.” Again, respondents indicated on a scale from 1 to 7 to what extent they agreed with these statements, and for each respondent, we took the average of their scores on these items to calculate their scores on our outcome judgments scale (α = 0.97). Higher scores on this scale indicate that respondents judged their outcomes more positively. Most respondents completed the questionnaire directly. Six respondents (3.0% of the sample) filled it out at home and sent it to us in an envelope with prepaid postage stamps. The respondents who filled out the questionnaire directly often did so themselves, while 25 respondents (12.8% of the sample) preferred having the questions read out loud by the researcher. Before respondents filled out the questionnaire, we explained that the research focused on persons who were born in a different country and persons whose parents or other ancestors were born in a different country. In addition, we told respondents that participation was voluntary and anonymous, and that the research was conducted independently of the court and the Public Prosecution Service.t We examined perceived everyday discrimination with the 10-item version of the everyday discrimination scale (Williams et al., 1997, 2008). We asked respondents to indicate on a After they completed the questionnaire, we thanked respondents for their participation and offered to send them a summary of our research results, which we sent to interested respondents later. During the entire period of data collection, we kept an extensive logbook detailing relevant background information to the research, such as information obtained through informal conversations with defendants and defense lawyers. Frontiers in Psychology \| www.frontiersin.org Measures Our main variables were perceived procedural justice, outcome judgments, perceived everyday discrimination, trust in Dutch judges, protest intentions, and state self-esteem. The questionnaire started with those variables relating to the court hearing (perceived procedural justice, outcome judgments, protest intentions, and trust in Dutch judges) and then assessed variables 7We conducted all analyses involving perceived procedural justice with both this six-item scale and the entire 17-item scale. In the Results section, we explicitly note when these analyses yielded (slightly) different results with regard to our main findings. Frontiers in Psychology \| www.frontiersin.org November 2021 \| Volume 12 \| Article 746364 6 Procedural Justice in Criminal Cases Ansems et al. scale from 1 (never) to 6 (almost every day) how often they encountered the following events in their daily lives: “In my day-to-day life, I am treated with less courtesy than other people are,” “In my day-to-day life, I am treated with less respect than other people are,” “In my day-to-day life, I receive poorer service than other people at restaurants or stores,” “In my day-to-day life, people act as if they think I am not smart,” “In my day-to-day life, people act as if they are afraid of me,” “In my day-to-day life, people act as if they think I am dishonest,” “In my day-to-day life, people act as if they are better than I am,” “In my day-to-day life, I am called names or insulted,” “In my day-to-day life, I am threatened or harassed,” and “In my day-to-day life, I am followed around in stores.” Together, these items formed a reliable perceived everyday discrimination scale (α = 0.91). Higher scores on this scale reflect higher levels of perceived everyday discrimination. In addition, respondents who answered “a few times a year” (score 3 on the six-point scale) or more often to at least one question were asked to indicate what they thought was the main reason for these experiences: their gender, their age, their religion, their ethnic- cultural background, their level of education, their level of income, and/or some other reason (which they could then write down). In this way, we assessed perceived grounds of discrimination. 8There were missing values for perceived ground of discrimination (51 missing values), trust in judges (one missing value), grade for trust in judges (12 missing values), and self-esteem (one missing value). There were also some missing values for gender (one missing value), age (three missing values), highest completed level of education (eight missing values), offense (eight missing values), type of verdict (six missing values), sanction received (32 missing values), legal assistance (four missing values), and number of previous court hearings (one missing value). Measures to be proud of” (reverse coded), “Now, at this moment, I feel useless” (reverse coded), “Now, at this moment, I feel that I am a person of worth, at least on an equal plane with others,” “Now, at this moment, I wish I could have more respect for myself” (reverse coded), “Now, at this moment, I feel like I am a failure” (reverse coded), and “Now, at this moment, I take a positive attitude toward myself.” Respondents’ answers on these items were averaged into a reliable state self-esteem scale (α = 0.83) on which higher scores reflect higher state self-esteem. We also assessed relevant background variables, asking respondents to indicate whether they had legal assistance during their court hearings, their number of previous court hearings before a criminal judge, their highest completed level of education, their gender, and their age. At the end of the questionnaire, respondents could write down remarks or issues they deemed important that had not been the subject of our questions.8 Descriptive Statistics and Bivariate Correlations Higher scores on this scale represent stronger protest intentions. Descriptive Statistics and Bivariate Correlations All statistical analyses were conducted with IBM SPSS software. Table 2 presents means, standard deviations, and bivariate correlations for our main variables and background variables. As shown in Table 2, there were statistically significant relationships between some of our background variables (i.e., legal assistance, number of previous court hearings, and age) and some of our main variables (i.e., perceived procedural justice, trust in judges, and protest intentions). Hence, we controlled for legal assistance, number of previous court hearings, and age in the hierarchical regression analyses reported below by entering them in Step 1 of the analysis. Main effects were entered in Step 2, and two-way interactions were entered in Step 3. Following recommendations by Cohen et al. (2003), all continuous independent variables (including quasi-interval variables) were standardized before being entered into the equation when the equation involved an interaction effect. When reporting the results of these hierarchical regression analyses, we focus on the last step in the analysis that significantly added to the amount of explained variance in the dependent variables in our regression equations. We solicited their trust in Dutch judges with items that target this construct in a way that we deemed as direct and straightforward as possible (see also Grootelaar and van den Bos, 2018). Specifically, we asked respondents to indicate on a scale from 1 (totally disagree) to 7 (totally agree) to what extent they agreed with the following five statements: “I have faith in Dutch judges,” “I deem Dutch judges trustworthy,” “I trust Dutch judges,” “I do not trust Dutch judges” (reverse coded), and “I feel like Dutch judges cannot be trusted” (reverse coded). Respondents’ answers on these items were averaged into a reliable trust in Dutch judges scale (α = 0.90) on which higher scores reflect higher levels of trust. We also included an additional sixth item asking respondents to express their trust in Dutch judges with a grade between 1 (lowest) and 10 (highest), in line with the grading system used in Dutch schools. Following Stahl et al. (2008), we assessed protest intentions by asking respondents to indicate on a scale from 1 (not at all) to 7 (very much) to what extent they would like to criticize the ruling and to what extent they would like to protest against the ruling. Respondents’ answers on these two items were averaged into a reliable protest intentions scale (α = 0.85). The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). Highest completed level of education was measured on a nine-point scale, ranging fr (coded as 1) to university (coded as 9). Perceived grounds for discrimination included respondents’ ethnic-cultural background (N = 93; 63.3% of the sample), religion (N = 56; 38.1% of the sample), gender (N = 25; sample), age (N = 21; 14.3% of the sample), level of income (N = 17; 11.6% of the sample), and level of education (N = 17; 11.6% of the sample).p < 0.05; p < 0.01; p < 0.001. Adding Perceived Everyday Discrimination Hypothesis 2 proposed that there is a two-way interaction between perceived procedural justice and perceived everyday discrimination, such that defendants who experience relatively high levels of everyday discrimination show attenuated or reversed associations between perceived procedural justice and our other variables (i.e., trust in judges, outcome judgments, protest intentions, and state self-esteem). This hypothesis was not supported by our results, as we did not find significant interaction effects between perceived everyday discrimination and perceived procedural justice. We did find significant main effects of perceived procedural justice, sometimes in addition to significant main effects of perceived everyday discrimination. That is, perceived procedural justice was positively associated with respondents’ trust in judges (β = 0.48) and the grades they gave their trust in judges (β = 0.42), their outcome judgments (β = 0.58), and their state self-esteem (β = 0.21) and was negatively related to respondents’ protest intentions (β = −0.43). In addition, perceived everyday discrimination was negatively associated with trust in judges (β = −0.22) and the grades respondents gave their trust in judges (β = −0.16) as well as respondents’ state self-esteem (β = −0.17) and was positively associated with protest intentions (β = 0.24). Further details are shown in Tables 7–10. The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). Highest completed level of educ (coded as 1) to university (coded as 9). Perceived grounds for discrimination included respondents’ ethnic-cultural background (N = 93; 63.3% of the sample), r sample), age (N = 21; 14.3% of the sample), level of income (N = 17; 11.6% of the sample), and level of education (N = 17; 11.6% of the sample).p < 0.05; p < Reacting to Procedural Justice Finally, to measure respondents’ state self-esteem at the moment they filled out our questionnaire we adapted the global self-esteem scale of Rosenberg (1965) to measure state global self-esteem. Hence, respondents were asked to indicate on a scale from 1 (totally disagree) to 7 (totally agree) to what extent they agreed with the following 10 statements: “Now, at this moment, I am satisfied with myself,” “Now, at this moment, I think I am no good at all” (reverse coded), “Now, at this moment, I feel that I have a number of good qualities,” “Now, at this moment, I am able to do things as well as most other people,” “Now, at this moment, I feel like I do not have much Hypothesis 1 predicted that defendants with higher levels of perceived procedural justice report more trust in judges, more Frontiers in Psychology \| www.frontiersin.org November 2021 \| Volume 12 \| Article 746364 7 Procedural Justice in Criminal Cases Ansems et al. TABLE 2 \| Means, standard deviations, and correlations for the main variables and background variables. Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12 1. Procedural justice 5.38 1.27 – 2. Outcome judgments 4.55 2.18 0.59* – 3. Discrimination 2.48 1.16 −0.06 −0.12 – 4. Protest intentions 3.31 2.02 −0.45* −0.60* 0.26* – 5. Trust in judges 5.09 1.55 0.50* 0.42* −0.28* −0.34* – 6. Trust in judges grade 6.83 2.06 0.47* 0.40* −0.22 −0.34* 0.79* – 7. Self–esteem 5.62 1.09 0.19** 0.16* −0.22 −0.28* 0.26*** 0.15* – 8. Legal assistance (0 = no, 1 = yes) – – −0.19** −0.06 0.07 0.05 −0.14 −0.17* 0.00 – 9. Previous hearings 2.14 1.00 −0.06 0.02 0.11 −0.03 −0.15* −0.18* −0.13 0.11 – 10. Level of education 5.37 2.26 −0.05 −0.11 −0.08 0.05 0.02 0.02 −0.01 −0.01 −0.33*** – 11. Gender (0 = female, 1 = male) – – −0.03 −0.11 0.05 0.06 −0.05 0.03 −0.04 −0.09 0.17 −0.02 – 12. Age 30.10 10.75 0.10 0.10 −0.10 −0.15* 0.17* 0.14 0.02 −0.14 0.11 −0.19* −0.06 – The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). Highest completed level of education was measured on a nine-point scale, ranging from primary school (coded as 1) to university (coded as 9). Reacting to Procedural Justice Perceived grounds for discrimination included respondents’ ethnic-cultural background (N = 93; 63.3% of the sample), religion (N = 56; 38.1% of the sample), gender (N = 25; 17.0% of the sample), age (N = 21; 14.3% of the sample), level of income (N = 17; 11.6% of the sample), and level of education (N = 17; 11.6% of the sample).p < 0.05; p < 0.01; *p < 0.001. hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). Highest completed level of education was measured on a nine-point scale, ranging from primary school ed as 9). Perceived grounds for discrimination included respondents’ ethnic-cultural background (N = 93; 63.3% of the sample), religion (N = 56; 38.1% of the sample), gender (N = 25; 17.0% of the the sample), level of income (N = 17; 11.6% of the sample), and level of education (N = 17; 11.6% of the sample).p < 0.05; p < 0.01; p < 0.001. positive outcome judgments, lower intentions to protest against their outcomes, and higher levels of state self-esteem. This hypothesis was supported by our results. That is, respondents who felt treated more fairly during their court hearings showed more trust in judges (β = 0.50) and gave their trust in judges higher grades (β = 0.44), judged their outcomes more positively (β = 0.59), indicated lower protest intentions (β = −0.45), and reported higher state self-esteem (β = 0.22). Further details are presented in Tables 3–6. Different Cases and Contexts Different Cases and Contexts We note that there may also be other possible explanations for the lack of statistically significant interaction effects in our study. These explanations may relate, for instance, to the type of cases examined and the context of our research. In the organizational and performance-oriented settings of previous studies examining attenuated or reversed fair process effects (e.g., Van den Bos et al., 1999; Brockner et al., 2009), negative outcomes were likely to threaten people’s self-esteem and thus make them look for opportunities to attribute these outcomes to external causes. In the courtroom context of the current study, negative case outcomes may not have posed a similar threat to respondents’ sense of self-worth. Hence, the lack of significant interaction effects might be explained by the legal context of our study. This indicates, we think, that more research is needed into the operations of self-enhancement processes in relevant legal contexts, such as criminal court hearings. Our study provides an important first step in this regard. Similarly, the interaction between outcome judgments and perceived procedural justice has generally been found in work contexts or in other settings with different types of respondents than we examined in the current study (for overviews, see Brockner and Wiesenfeld, 1996; Brockner, 2010). Previous studies examining perceptions of actual defendants in criminal cases (Grootelaar and van den Bos, 2018) or undergraduates putting themselves in the position of defendants (Walker et al., 1974) did not find an interaction effect between outcomes and procedural justice. Our findings thus provide further support for the argument by Grootelaar and Van den Bos (2018), who did find interactions between perceived procedural justice and outcome favorability for motoring fine cases, that the type of case may play an important role in shaping people’s reactions to perceived procedural justice and outcome favorability in legal contexts. The present study critically examines the role of perceived procedural justice, and other important variables, in Dutch criminal court hearings. We think the message of what we learn from the reported findings is twofold. First, perceived procedural justice matters. That is, our findings showed robust associations between perceived procedural justice and trust in judges, outcome judgments, protest intentions, and state self-esteem. Second, processes of self-enhancement did not have the effects found by studies conducted in organizational contexts or laboratory settings. Adding Outcome Judgments g g Hypothesis 3 suggested that there is a two-way interaction between perceived procedural justice and outcome judgments, such that defendants who judge their outcomes more negatively show stronger, attenuated, or reversed associations between perceived procedural justice and our other variables (i.e., trust in judges, protest intentions, and state self-esteem). This hypothesis was not supported by the results, as our analyses did not reveal significant interaction effects between outcome judgments and perceived procedural justice. Our analyses did yield significant main effects of perceived procedural justice, sometimes in addition to significant main effects of outcome judgments. More specifically, we found a positive association between perceived procedural justice and trust in judges (β = 0.37) and the grade respondents gave their trust in judges (β = 0.32), a marginally significant association between perceived procedural justice and self-esteem (β = 0.18), and a negative association between perceived procedural justice and protest intentions (β = −0.15). We also found a positive association between outcome judgments and trust in judges (β = 0.22) and the grade respondents gave their trust in judges (β = 0.19) and a negative November 2021 \| Volume 12 \| Article 746364 8 Procedural Justice in Criminal Cases Ansems et al. associations between perceived procedural justice and our other variables remained intact regardless of the extent to which respondents experienced discrimination in their daily lives and how positively or negatively they judged their outcomes. This might be interpreted as an indication of the robustness of the fair process effect. These findings also fit with other studies, which show that people belonging to ethnic minorities respond equally favorably to perceived procedural justice as do people from majority groups (Tyler, 2001; Sunshine and Tyler, 2003; Higgins and Jordan, 2005; Johnson et al., 2017). association between outcome judgments and protest intentions (β = −0.50). Entering the entire 17-item perceived procedural justice scale into the regression equation rather than the six-item scale yielded the same (non-significant) result regarding the interaction between outcome judgments and perceived procedural justice, the only difference being that the association between outcome judgments and trust in judges was no longer statistically significant. Tables 11–13 present further details. Frontiers in Psychology \| www.frontiersin.org Different Cases and Contexts That is, outcome judgments and perceptions of everyday discrimination did not significantly moderate the associations between perceived procedural justice, on the one hand, and trust in judges, protest intentions, and state self-esteem on the other hand. In what follows, we deepen these conclusions. We then discuss the limitations of the present study, suggestions for future research that follow from these limitations, and practical implications of our findings. We note that there may also be other possible explanations for the lack of statistically significant interaction effects in our study. These explanations may relate, for instance, to the type of cases examined and the context of our research. In the organizational and performance-oriented settings of previous studies examining attenuated or reversed fair process effects (e.g., Van den Bos et al., 1999; Brockner et al., 2009), negative outcomes were likely to threaten people’s self-esteem and thus make them look for opportunities to attribute these outcomes to external causes. In the courtroom context of the current study, negative case outcomes may not have posed a similar threat to respondents’ sense of self-worth. Hence, the lack of significant interaction effects might be explained by the legal context of our study. This indicates, we think, that more research is needed into the operations of self-enhancement processes in relevant legal contexts, such as criminal court hearings. Our study provides an important first step in this regard The Importance of Fair Proceduresi p An important finding of this study is that respondents who felt treated more fairly during their court hearings reported higher levels of trust in judges, judged their outcomes more positively, showed lower protest intentions, and displayed higher state self-esteem. These favorable reactions to perceived procedural justice indicate that, even in the real-life courtroom context of our study in which respondents risked actual sanctions, respondents cared not only about their outcomes but also about the way they were treated during their court hearings. Thus, the current study contributes to the ongoing academic debate on the relative importance of perceived procedural justice in real-life cases. Our findings support the argument by Casper et al. (1988) that the positive associations between perceived procedural justice and relevant other variables represent real- world phenomena that can also be observed outside the artificial settings of psychological laboratories, including criminal justice contexts (see also, for instance, Tyler, 1984, 1988, 2006; Paternoster et al., 1997; Tyler and Huo, 2002). Conflicting Psychological Processesi l g y g Another potential explanation for not finding interactive effects of outcome judgments or perceived everyday discrimination and perceived procedural justice might be that conflicting psychological processes are at work. That is, the self- enhancement processes underlying attenuated or reversed fair process effects in other studies might play a role in the courtroom context of our study, but their effects may have been canceled out or overridden by other psychological processes (see also Brockner et al., 2009). Our study also complements current insights into perceived procedural justice in criminal justice contexts by examining whether defendants’ reactions to perceived procedural fairness may be moderated by experiences of everyday discrimination and outcome judgments. Our results indicated that none of the interaction effects we examined were statistically significant. In other words, rather than being attenuated or reversed, the November 2021 \| Volume 12 \| Article 746364 9 Procedural Justice in Criminal Cases Ansems et al. TABLE 3 \| Trust in judges regressed on procedural justice and relevant background variables. Variable Step 1 95% CI for b Step 2 95% CI for b b β t p b β t p Legal assistance (0 = no, 1 = yes) −0.37 −0.11 −1.48 0.141 −0.86, 0.12 −0.08 −0.02 −0.36 0.718 −0.51, 0.35 Previous hearings −0.21 −0.14 −1.91 0.058 −0.43, 0.01 −0.18 −0.12 −1.86 0.065 −0.36, 0.01 Age 0.03 0.18 2.50 0.013 0.01, 0.05 0.02 0.15 2.44 0.016 0.00, 0.04 Procedural justice 0.61 0.50 7.98 0.000 0.46, 0.76 df 3 4 F 4.31, p = 0.006 20.23, p = 0.000 F change 4.31, p = 0.006 63.62, p = 0.000 R2 0.07 0.31 Adjusted R2 0.05 0.29 N 189 189 The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). TABLE 3 \| Trust in judges regressed on procedural justice and relevant background variables. TABLE 4 \| Outcome judgments regressed on procedural justice and relevant background variables. Conflicting Psychological Processesi Variable Step 1 95% CI for b Step 2 95% CI for b b β t p b β t p Legal assistance (0 = no, 1 = yes) −0.27 −0.06 −0.75 0.455 −0.98, 0.44 0.21 0.04 0.72 0.475 −0.38, 0.80 Previous hearings 0.05 0.02 0.33 0.745 −0.26, 0.37 0.11 0.05 0.83 0.406 −0.15, 0.37 Age 0.02 0.08 1.08 0.282 −0.01, 0.05 0.01 0.05 0.77 0.442 −0.02, 0.03 Procedural justice 1.03 0.59 9.69 0.000 0.82, 1.24 df 3 4 F 0.72, p = 0.541 24.27, p = 0.000 F change 0.72, p = 0.541 93.85, p = 0.000 R2 0.01 0.34 Adjusted R2 −0.00 0.33 N 190 190 The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). TABLE 4 \| Outcome judgments regressed on procedural justice and relevant background variables. The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20) TABLE 5 \| Protest intentions regressed on procedural justice and relevant background variables. Variable Step 1 95% CI for b Step 2 95% CI for b b β t p b β t p Legal assistance (0 = no, 1 = yes) 0.16 0.04 0.48 0.630 −0.50, 0.82 −0.18 −0.04 −0.60 0.550 −0.78, 0.42 Previous hearings −0.03 −0.02 −0.20 0.842 −0.32, 0.26 −0.07 −0.04 −0.52 0.601 −0.33, 0.19 Age −0.02 −0.13 −1.71 0.088 −0.05, 0.00 −0.02 −0.10 −1.52 0.130 −0.04, 0.01 Procedural justice −0.72 −0.45 −6.74 0.000 −0.94, −0.51 df 3 4 F 1.21, p = 0.308 12.49, p = 0.000 F change 1.21, p = 0.308 45.46, p = 0.000 R2 0.02 0.21 Adjusted R2 0.00 0.20 N 190 190 The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). TABLE 5 \| Protest intentions regressed on procedural justice and relevant background variables. The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). respond even more favorably to perceived procedural justice than defendants who experience little everyday discrimination. Frontiers in Psychology \| www.frontiersin.org Conflicting Psychological Processesi In the current paper, we explored whether defendants who For instance, defendants who experience much discrimination in their daily lives may be pleasantly surprised by how fairly they feel treated during their court hearings, leading them to November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 10 Procedural Justice in Criminal Cases Ansems et al. TABLE 6 \| Self-esteem regressed on procedural justice and relevant background variables. Variable Step 1 95% CI for b Step 2 95% CI for b b β t p b β t p Legal assistance (0 = no, 1 = yes) 0.05 0.02 0.26 0.793 −0.31, 0.40 0.14 0.06 0.77 0.440 −0.21, 0.49 Previous hearings −0.14 −0.13 −1.78 0.077 −0.30, 0.02 −0.13 −0.12 −1.68 0.094 −0.28, 0.02 Age 0.00 0.03 0.39 0.696 −0.01, 0.02 0.00 0.02 0.23 0.821 −0.01, 0.02 Procedural justice 0.19 0.22 3.07 0.002 0.07, 0.32 df 3 4 F 1.06, p = 0.366 3.19, p = 0.015 F change 1.06, p = 0.366 9.41, p = 0.002 R2 0.02 0.07 Adjusted R2 0.00 0.04 N 189 189 The number of previous court hearings was measured on a five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). TABLE 6 \| Self-esteem regressed on procedural justice and relevant background variables. five-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). enabled us to robustly examine the three-way interaction between outcome judgments, perceived everyday discrimination, and perceived procedural justice. We would expect attenuated or reversed associations between perceived procedural justice and relevant other variables in particular among respondents with both relatively high levels of perceived everyday discrimination and relatively negative outcome judgments. Thus, future studies with larger samples are needed to better understand the issues examined in the current paper. Follow-up studies with larger samples could also examine, for instance, the possible role of being found guilty and seriousness of the sanction imposed. After all, more serious sanctions might pose a greater threat to defendants’ self-esteem, thereby making attenuated or reversed fair process effects more likely to occur. experience much discrimination in their daily lives might respond less favorably to perceived procedural justice because of self-enhancement processes. These potential moderating effects of perceived everyday discrimination may have canceled each other out, resulting in non-significance of the interaction effect. Conflicting Psychological Processesi f Defendants’ desire for fair treatment may also simply have overridden their self-enhancement motive. After all, perceived procedural justice may be desirable for various instrumental and non-instrumental reasons, as explained in the Introduction. These beneficial aspects of perceived procedural justice may have been stronger than defendants’ self-enhancement processes, resulting in favorable responses to procedures that defendants perceive as fair rather than unfair. We also note that we conducted our study at only one court and included only single judge criminal cases. Furthermore, the first author – who collected the bulk of the data – is a White and university-based researcher. As a result, interviewer effects may have played a role in our study. For instance, respondents may have concealed their levels of distrust in Dutch judges, as they may have considered the researcher as belonging to their outgroup (Hulst, 2017). Thus, we propose that it is important to replicate our study in other courts with different researchers and different types of court cases. Follow-up studies could also include defendants whom we were not able to include in our current sample, such as defendants in pre-trial detention. Levels of Perceived Everyday Discrimination Respondents’ relatively low levels of perceived everyday discrimination (M = 2.48, SD = 1.16, measured on a six-point scale) may be relevant as well. After all, respondents who scored one standard deviation above the mean level of perceived everyday discrimination (i.e., a score of 3.64) encountered negative treatment between a few times a year (score 3) and a few times a month (score 4). These experiences of discrimination may not have been sufficiently frequent to make defendants respond favorably to perceived procedural unfairness during their court hearings for self-enhancement reasons. Hence, we recommend that future studies examining these issues use samples in which levels of perceived everyday discrimination are likely to be higher. In addition, the correlational design of this study does not allow for conclusions about any causal relationships between our variables. Thus, although we believe the field work element to be a strength of the paper, this also has methodological limitations. For instance, perceptions of procedural justice may influence defendants’ outcome judgments, and vice versa, which renders the analysis pertaining to Hypothesis 3 difficult to interpret. Limitations An engaging aspect of our study, we think, is that we were able to study perceptions of actual defendants in single judge criminal cases after a 9-month period of data collection at the court of the mid-Netherlands. The flip side of this approach is that our sample is sufficiently large, yet smaller than we would have wanted ideally. For instance, a larger sample would have Furthermore, a possibility that cannot be ruled out in our correlational design is that some of the effects that we examined with this hypothesis might already be present in the variation in the independent variables, making the validity of the interaction Frontiers in Psychology \| www.frontiersin.org November 2021 \| Volume 12 \| Article 746364 11 r b 83 or b 38 05 38 95 0.16 30 Step 3 95% CI for b b β t p 0.24 0.05 0.79 0.430 −0.35, 0.83 0.13 0.06 1.01 0.315 −0.13, 0.39 0.08 0.04 0.58 0.565 −0.19, 0.34 1.29 0.58 9.56 0.000 1.02, 1.55 −0.21 −0.10 −1.56 0.122 −0.47, 0.06 0.04 0.02 0.29 0.774 −0.23, 0.31 6 16.62, p = 0.000 0.08, p = 0.774 0.35 0.33 190 b Step 3 95% CI for b b β t p 6 −0.04 −0.01 −0.18 0.858 −0.45, 0.38 4 −0.14 −0.09 −1.46 0.146 −0.32, 0.05 8 0.19 0.13 2.05 0.042 0.01, 0.38 4 0.76 0.49 8.00 0.000 0.57, 0.95 15 −0.35 −0.23 −3.71 0.000 −0.53, −0.16 0.11 0.07 1.14 0.258 −0.08, 0.30 6 16.70, p = 0.000 1.29, p = 0.258 0.36 0.33 189 b 2 9 4 for b 0.36 0.04 0.38 0.94 −0.15 mination, and their interaction on outcome judgments. 95% CI for b Step 2 p b β t p 0.455 −0.98, 0.44 0.23 0.05 0.78 0.438 0.745 −0.26, 0.37 0.13 0.06 0.99 0.322 0.282 −0.14, 0.49 0.08 0.04 0.60 0.552 1.29 0.58 9.58 0.000 −0.20 −0.09 −1.53 0.127 5 20.03, p = 0.000 48.44, p = 0.000 0.35 0.34 190 point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). mination, and their interaction on trust in judges. Limitations 95% CI for b Step 2 p b β t p 0.141 −0.86, 0.12 −0.05 −0.02 −0.25 0.8 0.058 −0.43, 0.01 −0.14 −0.09 −1.54 0.1 0.013 0.06, 0.50 0.20 0.13 2.11 0.0 0.75 0.48 7.94 0.0 −0.33 −0.22 −3.56 0.0 5 19.75, p = 0.000 40.17, p = 0.000 0.35 0.33 189 point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). p 0.43 0 32 0 0 0 0 0 tcom ve-po tcom p 0.6 0.8 0.0 r b 1 r b 34 13 12 0.63 76 13 b Step 3 95% CI for b b β t p 0 0.16 0.07 0.91 0.365 −0.19, 0.51 4 −0.11 −0.10 −1.40 0.164 −0.26, 0.05 5 −0.01 −0.01 −0.08 0.940 −0.16, 0.15 9 0.23 0.21 2.95 0.004 0.08, 0.39 04 −0.20 −0.18 −2.51 0.013 −0.35, −0.04 0.05 0.05 0.67 0.502 −0.10, 0.21 6 3.23, p = 0.005 0.45, p = 0.502 0.10 0.07 189 or b Step 3 95% CI for b b β t p 36 −0.24 −0.05 −0.82 0.414 −0.82, 0.34 4 −0.13 −0.06 −0.99 0.323 −0.38, 0.13 1 −0.14 −0.07 −1.09 0.277 −0.40, 0.12 .62 −0.89 −0.44 −6.73 0.000 −1.15, −0.6 73 0.50 0.25 3.82 0.000 0.24, 0.76 −0.13 −0.06 −0.98 0.328 −0.40, 0.13 6 11.35, p = 0.000 0.96, p = 0.328 0.27 0.25 190 mination, and their interaction on self-esteem. 95% CI for b Step 2 p b β t p 0.793 −0.31, 0.40 0.15 0.06 0.87 0.3 0.077 −0.30, 0.02 −0.11 −0.10 −1.45 0.1 0.696 −0.13, 0.19 −0.00 −0.00 −0.04 0.9 0.23 0.21 2.92 0.0 −0.19 −0.17 −2.43 0.0 5 3.80, p = 0.003 7.78, p = 0.001 0.09 0.07 189 point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). mination, and their interaction on protest intentions. 95% CI for b Step 2 p b β t 0.630 −0.50, 0.82 −0.22 −0.05 −0.76 0 0.842 −0.32, 0.26 −0.12 −0.06 −0.93 0 0.088 −0.55, 0.04 −0.15 −0.07 −1.15 0 −0.88 −0.43 −6.68 0 0.48 0.24 3.71 0 5 13.43, p = 0.000 31.17, p = 0.000 0.27 0.25 190 point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). b Step 3 95% CI for b b β t p −0.12 −0.04 −0.56 0.579 −0.54, 0.31 1 −0.20 −0.13 −2.08 0.039 −0.38, −0.01 0.22 0.15 2.33 0.021 0.03, 0.41 0.57 0.37 4.75 0.000 0.33, 0.81 0.34 0.22 2.94 0.004 0.11, 0.56 −0.02 −0.01 −0.14 0.888 −0.23, 0.20 6 15.42, p = 0.000 0.02, p = 0.888 0.34 0.32 189 Step 3 95% CI for b b β t p −0.09 −0.02 −0.32 0.751 −0.62, 0.45 −0.02 −0.01 −0.15 0.882 −0.25, 0.22 −0.15 −0.08 −1.26 0.210 −0.39, 0.09 −0.32 −0.16 −2.08 0.039 −0.62, −0.02 −1.01 −0.50 −7.01 0.000 −1.30, −0.73 −0.03 −0.01 −0.19 0.849 −0.29, 0.24 6 18.63, p = 0.000 0.04, p = 0.849 0.38 0.36 190 come judgments, and their interaction on trust in judges. 95% CI for b Step 2 p b β t p 0.141 −0.86, 0.12 −0.12 −0.03 −0.55 0.585 0.058 −0.43, 0.01 −0.20 −0.13 −2.09 0.038 0.013 0.06, 0.50 0.22 0.14 2.34 0.020 0.58 0.37 4.91 0.000 0.34 0.22 2.94 0.004 5 18.59, p = 0.000 37.47, p = 0.000 0.34 0.32 189 e-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). come judgments, and their interaction on protest intentions. 95% CI for b Step 2 9 p b β t p 0.630 −0.50, 0.82 −0.08 −0.02 −0.30 0.763 − 0.842 −0.32, 0.26 −0.02 −0.01 −0.16 0.875 − 0.088 −0.55, 0.04 −0.16 −0.08 −1.31 0.192 − −0.31 −0.15 −2.09 0.038 − −1.01 −0.50 −7.02 0.000 − 5 22.47, p = 0.000 53.33, p = 0.000 0.38 0.36 190 e-point scale (1 = 0, 2 = 1, 3 = 2–10, 4 = 11–20, 5 = more than 20). p 0.585 0.038 0.020 0.000 0.004 63 75 92 38 Procedural Justice in Criminal Cases Ansems et al. TABLE 13 \| Regression results for procedural justice, outcome judgments, and their interaction on self-esteem. Practical Implications While recognizing these limitations, we think the findings of our study can have some important practical implications. Trust in judges, for instance, is an issue that has the Dutch judiciary’s ongoing attention. Although the level of trust in the Dutch judiciary is relatively high compared to trust in other Dutch governmental institutions and judiciaries in other European countries (Ridder et al., 2019; Bovens, 2020), safeguarding this trust is considered important (Grimmelikhuijsen, 2018). Fair procedures, in terms of objective legal standards as well as people’s subjective perceptions, can play an important role in this regard. This is relevant not only with a view to maintaining and possibly increasing trust in judges as an end in itself, but also because trust in judges is related to other important attitudes and behaviors, such as perceived legitimacy and compliance with the law (Grootelaar and van den Bos, 2018).i Our finding that perceived procedural justice is negatively associated with protest intentions can be of interest to legal policymakers and judges as well. Although reporting protest intentions is not the same as actually appealing a verdict, the two are likely to be related. It is noteworthy in this regard that more than 90% of appeals to criminal verdicts are initiated by defendants (Croes, 2016). Promoting procedural justice could therefore be a way to decrease the number of appeals and the social costs associated therewith. These social costs may concern not only financial costs but also costs in terms of quality of adjudication, as judges’ workload is considered a threat for impartial adjudication by one out of five Dutch judges (Weijers, 2019). Taken together, we think our findings regarding the importance of procedural justice in the criminal courtroom context are relevant for both their contribution to procedural justice theory and their possible implications for legal practice. Variable Step 1 95% CI for b Step 2 95% CI for b Step 3 95% CI for b b β t p b β t p b β t p Legal assistance (0 = no, 1 = yes) 0.05 0.02 0.26 0.793 −0.31, 0.40 0.13 0.05 0.72 0.474 −0.22, 0.48 0.12 0.05 0.69 0.492 −0.23, 0.48 Previous hearings −0.14 −0.13 −1.78 0.077 −0.30, 0.02 −0.14 −0.13 −1.74 0.083 −0.29, 0.02 −0.14 −0.13 −1.72 0.086 −0.29, 0.02 Age 0.03 0.03 0.39 0.696 −0.13, 0.19 0.01 0.01 0.18 0.857 −0.14, 0.17 0.02 0.02 0.22 0.826 −0.14, 0.18 Procedural justice 0.19 0.18 1.96 0.051 −0.00, 0.39 0.19 0.17 1.85 0.066 −0.01, 0.38 Outcome judgments 0.09 0.08 0.92 0.360 −0.10, 0.28 0.09 0.08 0.91 0.363 −0.10, 0.28 Outcome judgments × Procedural justice −0.02 −0.02 −0.26 0.793 −0.20, 0.15 df 3 5 6 F 1.06, p = 0.366 2.72, p = 0.022 2.26, p = 0.039 F change 1.06, p = 0.366 5.12, p = 0.007 0.07, p = 0.793 R2 0.02 0.07 0.07 Adjusted R2 0.00 0.04 0.04 N 189 189 189 The number of previous court hearings was measured on a five point scale (1 0 2 1 3 2 10 4 11 20 5 more than 20) between the two terms difficult to assess. For example, after having received a negative case outcome, some defendants may have re-evaluated their sense of procedural fairness because of their need for self-enhancement. Thus, part of what might be going on in our analyses could be motivated reasoning regarding procedural fairness once defendants received a negative case outcome. This would fit with findings reported by Lilly and Wipawayangkool (2018) obtained in a non-courtroom setting. They found that external self-serving bias and self- threat following unfavorable outcomes were negatively related to procedural justice perceptions. Thus, studies using experimental control can clarify issues of causality and are therefore a viable avenue for future research into the issues examined here. REFERENCES Cohen, R. L. (1982). “Perceiving justice: An attributional perspective,” in Equity and justice in social behavior. eds. J. Greenberg and R. L. Cohen (New York: Academic Press), 119–160. Anderson, J. K., and Hayden, R. M. (1981). Questions of validity and drawing conclusions from simulation studies in procedural justice: a comment. Law Soc. Rev. 15, 293–303. doi: 10.2307/3053606 Cohen, J., Cohen, P., West, S. G., and Aiken, L. S.. (2003). Applied Multiple Regression/Correlational Analysis for the Behavioral Sciences. 3rd Edn. Mahwaw, NJ: Erlbaum. Andriessen, I., Dijkhof, J. H., van der Torre, A., van den Berg, E., Pulles, I., Iedema, J., et al. (2020). Ervaren discriminatie in Nederland II. Retrieved from the Netherlands Institute for Social Research. Available at: http://www. scp.nl (Accessed May 14, 2020). Crocker, J., and Major, B. (1989). Social stigma and self-esteem: the self-protective properties of stigma. Psychol. Rev. 96, 608–630. doi: 10.1037/0033-295X.96.4.608 Croes, M. T. (2016). De staat van het hoger beroep in het strafrecht. Factsheet Retrieved from the Research and Documentation Centre. Available at: http:// www.wodc.nl (Accessed May 14, 2020). Ansems, L. F. M. (2021). Procedural justice on trial: a critical test of perceived procedural justice from the perspective of criminal defendants. dissertation. Utrecht University. Ridderprint: Alblasserdam, Utrecht University. Faul, F., Erdfelder, E., Lang, A.-G., and Buchner, A. (2007). Gpower 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav. Res. Methods 39, 175–191. doi: 10.3758/BF03193146if Ansems, L. F. M., van den Bos, K., and Mak, E. (2020). Speaking of justice: a qualitative interview study on perceived procedural justice among defendants in Dutch criminal cases. Law Soc. Rev. 54, 643–679. doi: 10.1111/lasr.12499f Folger, R., Rosenfield, D., Grove, J., and Corkran, L. (1979). Effects of “voice” and peer opinions on responses to inequity. J. Pers. Soc. Psychol. 37, 2253–2261. doi: 10.1037/0022-3514.37.12.2253 Berrey, E., Hoffman, S. G., and Nielsen, L. B. (2012). Situated justice: a contextual analysis of fairness and inequality in employment discrimination litigation. Law Soc. Rev. 46, 1–36. doi: 10.1111/j.1540-5893.2012.00471.x Gilliland, S. W. (1994). Effects of procedural and distributive justice on reactions to a selection system. J. Appl. Psychol. 79, 691–701. doi: 10.1037/0021-9010.79.5.691 Grimmelikhuijsen, S. (2018). Van gegeven naar verdiend gezag: Hoe kan transparantere rechtspraak (blijvend) bijdragen aan legitimiteit? Rechtstreeks 15, 13–35. Bovens, M. (2020). “Het gezag van de juristen: Een empirische verkenning,” in Handelingen Nederlandse Juristen-Vereniging: De toekomst van de jurist, de jurist van de toekomst. eds. A. Berlee, M. FUNDING This research was funded by a Research Talent grant awarded by the Dutch Research Council (NWO, 406.16.558). This research was funded by a Research Talent grant awarded by the Dutch Research Council (NWO, 406.16.558). DATA AVAILABILITY STATEMENT The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found at: https://doi.org/10.24416/ UU01-74OU94. November 2021 \| Volume 12 \| Article 746364 15 Procedural Justice in Criminal Cases Ansems et al. ETHICS STATEMENT and edited the manuscript. EM provided input for the design and setup of the study and commented on several drafts of the manuscript, including editing of the manuscript. All authors contributed to the article and approved the submitted version. The studies involving human participants were reviewed and approved by the ethical assessment committee of the faculty of Law, Economics, and Governance at Utrecht University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. ACKNOWLEDGMENTS This paper is based on Chapter 3 of the first author’s dissertation (Ansems, 2021). We would like to thank the court of the mid-Netherlands for giving permission to conduct this research in their courthouse. We also thank Lamyae el Razouki for her assistance with data collection and data entry and Jan Fekke Ybema for his help with some of the analyses. AUTHOR CONTRIBUTIONS LA designed the study, including the questionnaire and the study’s procedures, organized approval by the court, organized and liaised about ongoing data collection within the court, collected most of the data and directed data collection by a research assistant, analyzed the data, interpreted results, and wrote the manuscript. KB provided conceptualization and theoretical input and aided in designing the study and questionnaire, provided conceptualization and theory used to integrate findings, co-interpreted results, REFERENCES Bovens, J. S. Buiting, A. R. Mackor, E. Mak, J. Silvis and E. T. T. Tai (Deventer: Wolters Kluwer), 31–55. Grootelaar, H. A. M. (2018). Interacting with procedural justice in courts. Dissertation, Amsterdam: Off Page. Brockner, J. (2010). A Contemporary Look at Organizational Justice: Multiplying Insult Times Injury. New York, NY: Routledge. Grootelaar, H. A. M., and van den Bos, K. (2018). How litigants in Dutch courtrooms come to trust judges: The role of perceived procedural justice, outcome favorability, and other socio-legal moderators. Law Soc. Rev. 52, 234–268. doi: 10.1111/lasr.12315 Brockner, J., de Cremer, D., Fishman, A. Y., and Spiegel, S. (2008). When does high process fairness reduce self-evaluations following unfavorable outcomes? The moderating effect of prevention focus. J. Exp. Soc. Psychol. 44, 187–200. doi: 10.1016/j.jesp.2007.03.002 Hayden, R. M., and Anderson, J. K. (1979). On the evaluation of procedural systems in laboratory experiments: a critique of Thibaut and Walker. Law Hum. Behav. 3, 21–38. doi: 10.1007/BF01039147 Brockner, J., Heuer, L., Magner, N., Folger, R., Umphress, E., van den Bos, K., et al. (2003). High procedural fairness heightens the effect of outcome favorability on self-evaluations: an attributional analysis. Organ. Behav. Hum. Decis. Process. 91, 51–68. doi: 10.1016/S0749-5978(02)00531-9 Henrich, J., Heine, S. J., and Ara Norenzayan, A. (2010). The weirdest people in the world? Behav. Brain Sci. 33, 61–83. doi: 10.1017/S0140525X0999152X Brockner, J., and Wiesenfeld, B. M. (1996). An integrative framework for explaining reactions to decisions: interactive effects of outcomes and procedures. Psychol. Bull. 120, 189–208. doi: 10.1037/0033-2909.120.2.189 Higgins, G. E., and Jordan, K. L. (2005). Race and gender: an examination of the models that explain evaluations of the court system for differences. Crim. Justice Stud. 18, 81–97. doi: 10.1080/14786010500071188 Brockner, J., Wiesenfeld, B. M., and Diekmann, K. A. (2009). Towards a “fairer” conception of process fairness: why, when and how more may not always be better than less. Acad. Manag. Ann. 3, 183–216. doi: 10.5465/194165 20903047228 Holmvall, C. M., and Bobocel, D. R. (2008). What fair procedures say about me: Self-construal and reactions to procedural fairness. Organ. Behav. Hum. Decis. Process. 105, 147–168. doi: 10.1016/j.obhdp.2007.09.001 Casper, J. D., Tyler, T., and Fisher, B. (1988). Procedural justice in felony cases. Law Soc. Rev. 22, 483–507. doi: 10.2307/3053626 Huijnk, W., and Andriessen, I. (2016). Integratie in zicht? De integratie van migranten in Nederland op acht terreinen nader bekeken. REFERENCES Trust in the Law: Encouraging Public Cooperation with the Police and Courts. New York: Russell Sage Foundation. Leary, M. R., and Terry, M. L. (2013). “Self-evaluation and self-esteem,” in Oxford Handbook of Social Cognition. ed. D. Carlston (New York: Oxford University Press), 534–547. Tyler, T. R., and Lind, E. A. (1992). “A relational model of authority in groups,” in Advances in Experimental Social Psychology, Vol. 25. ed. M. P. Zanna (San Diego, CA: Academic Press), 115–191.f Van den Bos, K. (2005). “What is responsible for the fair process effect?” in Handbook of Organizational Justice. eds. J. Greenberg and J. A. Colquitt (Mahwah, NJ: Lawrence Erlbaum Associates Publishers), 273–300. Leung, K., Steven, S., and Morris, M. W. (2001). When is criticism not constructive? The roles of fairness perceptions and dispositional attributions in employee acceptance of critical supervisory feedback. Hum. Relat. 54, 1155–1187. doi: 10.1177/0018726701549002 Van den Bos, K. (2015). “Humans making sense of alarming conditions: psychological insight into the fair process effect,” in Oxford Handbook of Justice in Work Organizations. eds. R. S. Cropanzano and M. L. Ambrose (New York: Oxford University Press), 403–417. Lilly, J. D., and Wipawayangkool, K. (2018). When fair procedures don’t work: a self-threat model of procedural justice. Curr. Psychol. 37, 680–691. doi: 10.1007/s12144-016-9555-7 Lind, E. A., Kulik, C. T., Ambrose, M., and de Vera, M. V. Park (1993). Individual and corporate dispute resolution: using procedural fairness as a decision heuristic. Adm. Sci. Q. 38, 224–251. doi: 10.2307/2393412h Van den Bos, K., Bruins, J., Wilke, H. A. M., and Dronkert, E. (1999). Sometimes unfair procedures have nice aspects: on the psychology of the fair process effect. J. Pers. Soc. Psychol. 77, 324–336. doi: 10.1037/0022-3514.77.2.324 Lind, E. A., and Tyler, T. R. (1988). The Social Psychology of Procedural Justice. New York: Plenum. f Van den Bos, K., and Lind, E. A. (2002). “Uncertainty management by means of fairness judgments,” in Advances in Experimental Social Psychology, Vol. 34. ed. M. P. Zanna (San Diego, CA: Academic Press), 1–60. Major, B. (1994). “From social inequality to personal entitlement: the role of social comparisons, legitimacy appraisals, and group membership,” in Advances in Experimental Social Psychology. Vol. 26. ed. M. P. Zanna (San Diego: Academic Press), 293–355. Van den Bos, K., van der Velden, L., and Lind, E. A. (2014). REFERENCES Retrieved from November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 16 Procedural Justice in Criminal Cases Ansems et al. the Netherlands Institute for Social Research Available at: http://www.scp.nl (Accessed May 14, 2020). Sunshine, J., and Tyler, T. R. (2003). The role of procedural justice and legitimacy in shaping public support for policing. Law Soc. Rev. 37, 513–548. doi: 10.1111/1540-5893.3703002 Huijnk, W., Dagevos, J., Gijsberts, M., and Iris Andriessen, I. (2015). Werelden van verschil. over de sociaal-culturele afstand en positie van migrantengroepen in Nederland. Retrieved from the Netherlands Institute for Social Research. Available at: http://www.scp.nl (Accessed May 14, 2020). Thibaut, J., and Walker, L. (1975). Procedural Justice: A Psychological Analysis. Hillsdale, NJ: Erlbaum. Thijssen, L., Coenders, M., and Lancee, B. (2019). Etnische discriminatie op de Nederlandse arbeidsmarkt. Verschillen tussen etnische groepen en de rol van beschikbare informatie over sollicitanten. Mens en Maatschappij 94, 141–176. doi: 10.5117/MEM2019.2.002.THIJ Hulst, L. (2017). Experimental legal studies on perceived procedural justice and trust in law and society. dissertation. Amsterdam: Off Page. 141–176. doi: 10.5117/MEM2019.2.002.THIJh f Jenness, V., and Calavita, K. (2018). It depends on the outcome: prisoners, grievances, and perceptions of justice. Law Soc. Rev. 52, 41–72. doi: 10.1111/ lasr.12312 Tyler, T. R. (1984). The role of perceived injustice in defendants’ evaluations of their courtroom experience. Law Soc. Rev. 18, 51–74. doi: 10.2307/3053480 Johnson, D., Wilson, D. B., Maguire, E. R., and Lowrey-Kinberg, B. V. (2017). Race and perceptions of police: experimental results on the impact of procedural justice. Justice Q. 34, 1184–1212. doi: 10.1080/07418825.2017.1343862 Tyler, T. R. (1988). What is procedural justice? Criteria used by citizens to assess the fairness of legal procedures. Law Soc. Rev. 22, 103–136. doi: 10.2307/3053563 Tyler, T. R. (2001). Public trust and confidence in legal authorities: what do majority and minority group members want from the law and legal institutions? Behav. Sci. Law 19, 215–235. doi: 10.1002/bsl.438 Koper, G., van Knippenberg, D., Bouhuijs, F., Vermunt, R., and Wilke, H. (1993). Procedural fairness and self-esteem. Eur. J. Soc. Psychol. 23, 313–325. doi: 10.1002/ejsp.2420230307 Tyler, T. R. (2006). Why People Obey the Law. New Haven: Yale University Press. Landis, J. M., and Goodstein, L. L. (1986). When is justice fair? An integrated approach to the outcome versus procedure debate. Am. Bar Found. Res. J. 11, 675–707. doi: 10.1111/j.1747-4469.1986.tb00262.x Tyler, T. R., and Huo, Y. J. (2002). REFERENCES On the role of perceived procedural justice in citizens’ reactions to government decisions and l Van den Bos, K., van der Velden, L., and Lind, E. A. (2014). On the role of perceived procedural justice in citizens’ reactions to government decisions and the handling of conflicts. Utrecht Law Rev. 10, 1–26. doi: 10.18352/ulr.287h Major, B., Quinton, W. J., and McCoy, S. K. (2002). “Antecedents and consequences of attributions to discrimination: theoretical and empirical advances,” in Advances in Experimental Social Psychology. Vol. 34. ed. M. P. Zanna (New York: Academic Press), 251–330.i Walker, L., Stephen, L. T., Allan Lind, E., and Thibaut, J. (1974). Reactions of participants and observers to modes of adjudication. J. Appl. Soc. Psychol. 4, 295–310. doi: 10.1111/j.1559-1816.1974.tb02809.x j Weijers, J. (2019). De opbrengst van de rechterlijke macht. Tréma 1, 5–14. McCoy, S. K., and Major, B. (2003). Group identification moderates emotional responses to perceived prejudice. Personal. Soc. Psychol. Bull. 29, 1005–1017. doi: 10.1177/0146167203253466 Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychol. Rev. 92, 548–573. doi: 10.1037/0033-295X.92.4.548 Williams, D. R., González, H. M., Williams, S., Mohammed, S. A., Moomal, H., and Stein, D. J. (2008). Perceived discrimination, race and health in South Africa: findings from the South Africa stress and health study. Soc. Sci. Med. 67, 441–452. doi: 10.1016/j.socscimed.2008.03.021 Paternoster, R., Brame, R., Bachman, R., and Sherman, L. W. (1997). Do fair procedures matter? The effect of procedural justice on spouse assault. Law Soc. Rev. 31, 163–204. doi: 10.2307/3054098 Sci. Med. 67, 441–452. doi: 10.1016/j.socscimed.2008.03.021 Ployhart, R. E., Ryan, A. M., and Bennett, M. (1999). Explanations for selection decisions: applicants’ reactions to informational and sensitivity features of explanations. J. Appl. Psychol. 84, 87–106. doi: 10.1037/0021-9010.84.1.87 Williams, D. R., Yu, Y., Jackson, J. S., and Anderson, N. B. (1997). Racial differences in physical and mental health: socioeconomic status, stress, and discrimination. J. Health Psychol. 2, 335–351. doi: 10.1177/135910539700200305 y Ridder, D., Josje, E. M., Huijnk, W., and van Rijnberk, S. (2019). Continu onderzoek burgerperspectieven 2019/4. Retrieved from the Netherlands Institute for Social Research. Available at: http://www.scp.nl (Accessed May 14, 2020). 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. REFERENCES 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. Rosenberg, M. (1965). Society and the Adolescent Self-Image. Princeton, NJ: Princeton University Press. Schroth, H. A., and Shah, P. P. (2000). Procedures: do we really want to know them? An examination of the effects of procedural justice on self-esteem. J. Appl. Psychol. 85, 462–471. doi: 10.1037/0021-9010.85.3.462 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. Sedikides, C., Hart, C. M., and David de Cremer, D. (2008). The self in procedural fairness. Soc. Personal. Psychol. Compass 2, 2107–2124. doi: 10.1111/j.1751-9004.2008.00156.x Stahl, T., Vermunt, R., and Ellemers, N. (2008). Reactions to outgroup authorities’ decisions: the role of expected bias, procedural fairness, and outcome favorability. Group Process. Intergroup Relat. 11, 281–299. doi: 10.1177/ 1368430208090643 Copyright © 2021 Ansems, van den Bos and Mak. 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. Statistics Netherlands (2018). Jaarrapport integratie 2018. Available at: https:// www.cbs.nl (Accessed May 14, 2020). November 2021 \| Volume 12 \| Article 746364 Frontiers in Psychology \| www.frontiersin.org 17
https://openalex.org/W3080538606	https://www.nature.com/articles/s41467-020-18116-1.pdf	English	null	On demand shape memory polymer via light regulated topological defects in a dynamic covalent network	Nature communications	2,020	cc-by	7,246	ARTICLE 7-020-18116-1 20-18116-1 ARTICLE NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications 1 State Key Laboratory of Chemical Engineering, College of Chemical and Biological Engineering, Zhejiang University, 310027 Hangzhou, China. 2 ZJU- Hangzhou Global Scientiﬁc and Technological Innovation Center, 311215 Hangzhou, China. ✉email: taoxie@zju.edu.cn Results S timuli-responsive soft materials with tailorable properties are key enablers for numerous emerging applications including soft actuators1–6, ﬂexible electronics7,8, and med- ical devices9. These properties are tied to the topological con- nectivity of covalent bonds, consequently locked into the materials during the synthesis/fabrication step. Dynamic covalent bonds, with their unique combination of adaptability and high bond strength, can be a paradigm shift. Indeed, dynamic covalent polymer network10–12 exhibits adaptive properties not present in typical covalent networks such as self-healing, reprocessing, and permanent shape reconﬁguration. In this context, two emerging opportunities are particularly noteworthy. The ﬁrst is related to recycling of otherwise intractable thermoset polymers13–22. Clas- sical thermoset polymers are widely used in high performance structural composites, but their chemically crosslinked nature makes them non-reprocessible. Triggering the dynamic char- acteristics of the covalent bonds, however, renders the network polymers reprocessible through covalent bond exchange within the network. The second opportunity is permanent shape recon- ﬁguration in a mold-free manner via solid-state plasticity23–26. It allows access to complex shapes that are unobtainable with molding, an attribute particularly relevant to shape morphing structures/devices. Of importance in the current context is that the topological rearrangement in the above cases does not result in different topologies, consequently, the material properties remain identical before and after the dynamic bond exchange. Network synthesis and mechanism of topological isomeriza- tion. We use a readily accessible chemistry to ensure the general applicability of the TIN design. Speciﬁcally, our network was synthesized via photoinitiated radical polymerization between polyethylene glycol diacrylate (PEGDA, polyethylene glycol with Mn of 3350) and N-hydroxyethylacrylamide (Fig. 1a). In this network, three features are noteworthy: polyethylene glycol (PEG) segments are crystallizable; PEGDA offers ester moieties; The N- hydroxyethylacrylamide comonomer provides pendent hydroxyls. In the presence of an organobase catalyst (1,5,7-triazabicyclo [4.4.0]dec-5-ene neutralized with two molar acetic acids, TBD), the ester bonds can be activated to undergo transesteriﬁcation with the pendent hydroxyls. As a result, the network topology can isomerize according to the routes schematized in Fig. 1b, c. The transesteriﬁcation can occur between a hydroxyl group and an ester attached on a different main chain, namely inter-chain transesteriﬁcation (Fig. 1b). Depending on the extent of the bond exchange, one or both ends of a PEG chain are released, resulting in isomeric states 2 and 3 (IS2 and IS3). Alternatively, the transesteriﬁcation can happen between a hydroxyl group and an ester on the same main chain (Fig. 1c). Results The result of this intra-chain transesteriﬁcation is that the one or both the PEG chain ends are also liberated, corresponding to IS4 and IS5, respectively. What is also intriguing is that, relative to the original isomeric state (IS1), the number of crosslinking points changes in different ways for the four isomeric states. The crosslinking points remain unchanged for IS2, but IS3 corresponds to doubling in crosslinking points. In contrast, the formation of intra-chain cycles for IS4 and IS5 leads to a reduction in the overall network crosslinking since the newly formed intra-chain cycles do not contribute to the crosslinking31. y g A counterintuitive opportunity arises when a dynamic covalent network is designed such that bond rearrangement leads to different topologies such as crosslinking densities, physical entanglements, and dangling chain length/distribution. For non-covalent supramolecular systems, achieving topological switching is more favorable given their more diverse connectivities27. However, supramolecular systems do not typically offer the mechanical robustness of their covalent counterparts. For dynamic covalent networks, topological transfor- mation is typically accomplished by directing the network reconﬁ- guration with external guiding molecules28. As elegant as the approaches are, the requirement for external molecules to participate the process limits it to non-crosslinked polymer solutions/polymer melts. The concept of topological isomerizable network (TIN)29, by contrast, allows topological shifting within a “fully enclosed” solid material, that is, without participation of external reagents. This isomerization mechanism typically requires highly delicate design to introduce network heterogeneity. For instance, a TIN consisting of a permanent network frame and dynamic grafted long chains can isomerize from a grafting network into a brush network29. From a design standpoint, such a hybrid network (permanent mainframe and dynamic grafted chain) is quite speciﬁc and unusual. Two questions naturally arise: Is the TIN principle intrinsically narrow based or is it applicable to common and widely accessible polymer networks? From a material standpoint, what might be the unusual opportunities not offered by traditional approaches? To answer the ﬁrst question, we realize that, besides inter-chain crosslinking, the universal topological features for real polymer networks are in fact a range of defects including intra-chain cycles (loops), dangling chains, and free chains (sol). These defects affect the macroscopic mechanical and thermomechanical properties in a non-constructive way, but are statistically unavoidable30,31. Consequently, minimizing these defects is often the goal in designing polymer networks. We note that Fig. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 ARTICLE On demand shape memory polymer via light regulated topological defects in a dynamic covalent network Wusha Miao1, Weike Zou1, Binjie Jin1, Chujun Ni1, Ning Zheng1, Qian Zhao 1,2 & Ta Wusha Miao1, Weike Zou1, Binjie Jin1, Chujun Ni1, Ning Zheng1, Qian Zhao 1,2 & Tao Xie 1,2✉ The ability to undergo bond exchange in a dynamic covalent polymer network has brought many beneﬁts not offered by classical thermoplastic and thermoset polymers. Despite the bond exchangeability, the overall network topologies for existing dynamic networks typically cannot be altered, limiting their potential expansion into unexplored territories. By harnessing topological defects inherent in any real polymer network, we show herein a general design that allows a dynamic network to undergo rearrangement to distinctive topologies. The use of a light triggered catalyst further allows spatio-temporal regulation of the network topology, leading to an unusual opportunity to program polymer properties. Applying this strategy to functional shape memory networks yields custom designable multi-shape and reversible shape memory characteristics. This molecular principle expands the design versatility for network polymers, with broad implications in many other areas including soft robotics, ﬂexible electronics, and medical devices. 1 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 We hereafter focus on PEG-2 for further investigation in thermomechanical properties before and after the isomerization. We next resort to a photobase generator that can provide spatio- temporal release of a transesteriﬁcation catalyst. Speciﬁcally, we employ a photobase generator that upon UV irradiation can release a strong organic base 1,5,7-triazabicyclo[4.4.0]dec-5-ene29. The catalyst amount can be controlled by the light irradiation time. Consequently, the thermally triggered bond exchange and the isomerization kinetics can be manipulated. Speciﬁcally, the thermal isomerization is conducted at 80 °C for 30 min hereafter. Accordingly, longer irradiation leads to network samples with progressively higher Tm (Fig. 2b). Figure 2c shows that Xc (cal- culated from DSC curves in Fig. 2b) increase with light irradia- tion, resulting in a similar increase for the storage modulus at 25 °C. When light irradiation is ﬁxed at 300 s, Supplementary Fig. 5 suggests the crystallinity increases progressively with thermal annealing and reaches a plateau value at an annealing time of 30 min. PEG-3. We hereafter focus on PEG-2 for further investigation in thermomechanical properties before and after the isomerization. We next resort to a photobase generator that can provide spatio- temporal release of a transesteriﬁcation catalyst. Speciﬁcally, we employ a photobase generator that upon UV irradiation can release a strong organic base 1,5,7-triazabicyclo[4.4.0]dec-5-ene29. The catalyst amount can be controlled by the light irradiation time. Consequently, the thermally triggered bond exchange and the isomerization kinetics can be manipulated. Speciﬁcally, the thermal isomerization is conducted at 80 °C for 30 min hereafter. Accordingly, longer irradiation leads to network samples with progressively higher Tm (Fig. 2b). Figure 2c shows that Xc (cal- culated from DSC curves in Fig. 2b) increase with light irradia- tion, resulting in a similar increase for the storage modulus at 25 °C. When light irradiation is ﬁxed at 300 s, Supplementary Fig. 5 suggests the crystallinity increases progressively with thermal annealing and reaches a plateau value at an annealing time of 30 min. These samples were denoted as PEG-X, with X representing the hydroxyl/ester molar ratio. With the presence of TBD as the transesteriﬁcation catalyst (2 wt%), all the samples were thermally annealed (120 °C, 60 min) for isomerization. The DSC results in Fig. 2a show that PEG-0, without the hydroxyl groups in the network, barely underwent any change in either melting tem- perature (Tm) or crystallinity (Xc). Results 1b, c are simpliﬁed to demonstrate qualitatively the trend in increasing topological defects. In reality, a small amount of defects (intra-chain cycles, dangling chains, and free chains) are expected for the starting topology (IS1) as they are unavoidable for any real networks. These defects are negligible for simple demonstration only. In addition, the inter- and intra-chain transesteriﬁcation can also occur in an inter- mixed fashion, resulting in more topologies beyond those shown in Fig. 1b, c. A more thorough discussion of this is provided in the supplementary information (see Supplementary Fig. 1 and related discussion). At this stage, direct and quantitative experimental characterization of the complex isomerization is difﬁcult, although this is an interesting subject to study in the future, most likely via theoretical calculation. For the purpose of the current study, we focus hereafter on the statistical outcome in terms of the impact of topological isomerization on the number of crosslinking points and liberated PEG chains, or defects. Overall, depending on the extent of dynamic bond exchange, the network isomerizes to a statistically mixed state (Fig. 1d) compromised of crosslink PEG chains (i), dangling PEG chains (ii), and free PEG chains (iii), inter-chain crosslink (iv), and intra- chain cycles (v). Of importance here is that the release of the PEG chain ends (ii and iii) introduces more segmental mobility that should favor the polymer crystallization. This is a mechanism that can be explored for programming thermomechanical properties of the network polymer. Additionally, the change in crosslinking offers an unusual opportunity in programming the rubbery modulus of the network material. In the present study, we aim to answer an intriguing question, that is, can the topological defects naturally present in real polymer networks be harnessed to enable a universal mechanism for topological isomerization? With this thought in mind, we hereafter illustrate such a TIN design and demonstrate its sur- prising beneﬁt via construction of shape memory polymers (SMPs) with on demand shape-shifting versatility not offered by existing SMP. Programming thermomechanical properties. We anticipate that the number of hydroxyl groups would strongly affect the network topological isomerization. Accordingly, three network samples with different hydroxyl contents were synthesized by varying the amount of N-hydroxyethylacrylamide comonomer. NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications 2 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 O O n O O O O OH n HO N H O HO TBD O O N H O H N O OH O H N O O O H N O O N H O HO O OH n O O TBD N H O HO O O n O O O TBD N H OH O NH O O O O OH n HO HN O O O NH O O O HO O O O n N H O HO TBD a Carbon–carbon chain Hydroxyl-terminated pendant chain Crystallizable PEG chain O O O O O n H N O OH + Photoinitiator hv b c Isomeric state 1 (IS1) Isomeric state 2 (IS2) Isomeric state 3 (IS3) Isomeric state 1 (IS1) Isomeric state 4 (IS4) Isomeric state 5 (IS5) i: Crosslink PEG chains ii: Dangling PEG chains iii: Free PEG chains iv: Inter-chain crosslink v: Intra-chain Cycles Isomerization d i v ii iii iv Fig. 1 Design of the dynamic covalent network and the mechanism of topological isomerization. a Polymer network synthesis. b Network isomerization via inter-chain transesteriﬁcation pathways. c Network isomerization via intra-chain transesteriﬁcation pathways. d Statistical outcome of the topological isomerization. a Carbon–carbon chain Hydroxyl-terminated pendant chain Crystallizable PEG chain O O O O O n H N O OH + Photoinitiator hv O O n O O O O OH n HO N H O HO TBD O O N H O H N O OH O H N O O O H N O O N H O HO O OH n O O TBD N H O HO O O n O O O TBD N H OH O NH O O O O OH n HO HN O O O NH O O O HO O O O n N H O HO TBD Carbon–carbon chain Hydroxyl-terminated pendant chain Crystallizable PEG chain b c Isomeric state 1 (IS1) Isomeric state 2 (IS2) Isomeric state 3 (IS3) Isomeric state 1 (IS1) Isomeric state 4 (IS4) Isomeric state 5 (IS5) Carbon–carbon chain Crystallizable PEG chain Crystallizable PEG chain b Isomeric state 2 (IS2) c d Isomerization Fig. 1 Design of the dynamic covalent network and the mechanism of topological isomerization. a Polymer network synthesis. b Network isomerization via inter-chain transesteriﬁcation pathways. c Network isomerization via intra-chain transesteriﬁcation pathways. d Statistical outcome of the topological isomerization. PEG-3. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 c Room temperature storage modulus (25 °C) and PEG crystallinities of the isomerized samples. d Gel fraction of the isomerized network samples. Error bars represent standard deviations calculated from three specimens. Hydroxyl increase a 20 30 40 50 60 70 Endo heat flow PEG-0 PEG-1 PEG-2 Temperature (oC) b Endo heat flow 20 30 40 50 60 0 s 50 s 150 s 300 s Temperature (oC) b a d 0 100 200 300 60 70 80 90 100 Gel fraction (%) Irradiation time (s) c 0 100 200 300 20 30 40 50 Crystallinity (%) Irradiation time (s) 0 100 200 300 Storage modulus (MPa) d c Fig. 2 Programming thermomechanical properties via network topological isomerization. a DSC curves of the network samples before (dotted lines) and after (solid lines) isomerization. b DSC curves for the isomerized PEG-2 samples obtained with different UV irradiation times. c Room temperature storage modulus (25 °C) and PEG crystallinities of the isomerized samples. d Gel fraction of the isomerized network samples. Error bars represent standard deviations calculated from three specimens. mostly localized in the non-isomerized region, which is reversed at 80 °C. This unusual reversal in modulus may provide future opportunities in mechanically relevant multifunctional devices. see that the rubbery modulus of the isomerized material before washing decreases with irradiation time and reaches plateau values at irradiation time of 150 s. This is consistent with the formation of various topological defects (dangling chain, loops, and free chains) that do not contribute positively to the network. The rubbery modulus of the isomerized material after washing follows a similar trend but with absolute modulus values consistently higher than the corresponding samples before washing. This is a clear evidence that free chains produced during the isomerization indeed reduce the rubbery moduli. Reversible and triple-shape memory performances. The ability to spatio-selectively program network crystallization is particu- larly relevant for functional shape memory properties. Due to its crystalline nature, PEG-2 (before isomerization) shows reversible shape memory performance32 when the temperature is switched between −20 and 80 °C, with 20% reversible strain under a constant stress of 0.8 MPa (Fig. 4a). The reversible strain is found to decrease linearly with the external stress (Fig. 4b), down to a negligible level when the stress is too small. To achieve stress-free reversible shape memory33–36, current approaches rely on the introduction of an internal stress. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 In contrast, both the Tm and Xc increase noticeably for PEG-1 and PEG-2, with the relative extent of increase being higher for PEG-2. We note that the starting PEG-2 has a slightly lower Tm than PEG-1. This is because the hydroxyl groups affect the overall crystallization of the network. The as-synthesized PEG-0, PEG-1, and PEG-2 have similar gel fractions between 95.8% and 94.1%. After the thermal isomerization, the gel fractions for PEG-1 and PEG-2 reduce to 68.6% and 53.0%, respectively, whereas the gel fraction for PEG-0 remains unchanged (Supplementary Fig. 2). Furthermore, 1H NMR analysis veriﬁed the existence of free PEG chains, which are extracted from PEG-2 after thermal isomerization (Supplemen- tary Fig. 3). The above results are consistent with the iso- merization mechanisms outlined in Fig. 1b, c. When the hydroxyl/ester ratio is increased further to 3, the resulting PEG-3 is characterized and compared to PEG-2. The results (Supple- mentary Fig. 4) shows that a higher hydroxyl/ester ratio of 3 led to a lower melting temperature, with no obvious change in crystallinity. However, the melting temperature differences before and after isomerization remain almost identical for PEG-2 and As for the gel fraction of an isomerized sample, it decreases with the irradiation time (Fig. 2d), consistent with Fig. 1b, c which suggest that the more extensive bond exchange results in more free PEG chains in the network. In determining gel fractions, a solvent washing step was required to remove the free chains in the network. Comparisons of the rubbery moduli before and after washing (Supplementary Fig. 6) can therefore shed more light on the isomerization process. From this ﬁgure, one can NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications 3 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 Hydroxyl increase a b c d 0 100 200 300 60 70 80 90 100 Gel fraction (%) Irradiation time (s) 0 100 200 300 20 30 40 50 Crystallinity (%) Irradiation time (s) 0 100 200 300 Storage modulus (MPa) 20 30 40 50 60 70 Endo heat flow Endo heat flow PEG-0 PEG-1 PEG-2 20 30 40 50 60 0 s 50 s 150 s 300 s Temperature (oC) Temperature (oC) Fig. 2 Programming thermomechanical properties via network topological isomerization. a DSC curves of the network samples before (dotted lines) and after (solid lines) isomerization. b DSC curves for the isomerized PEG-2 samples obtained with different UV irradiation times. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 This typically necessitates the use of two crystalline phases in which the high melting phase provides the internal stress whereas the low melting phase serves as the actuation phase33–35. In contrast to those macroscopically homogeneous microscopically phase separated systems, the abil- ity to pattern macroscopic material heterogeneity via topological isomerization provides a unique mechanism to introduce internal stresses required for external stress-free reversible shape memory. The operating principle is illustrated in Fig. 4c. The spatially patterned sample is linearly stretched at 80 °C, above the melting temperatures of both Rn and Ri regions. Cooling down to 0 °C under the stretching force ﬁxes the stretched shape. Upon removal of the force, the sample is heated to 37 °C. This selec- tively triggers the shape recovery in the Rn region. This recovery is constrained by the Ri region. The net effect is that the sample recovers to the 3D curved shape, corresponding to the stress balance between the two regions. In essence, internal stresses are created in both regions. When temperature is switched between Spatio-selective topological isomerization. The use of the pho- tobase generator sets up the stage for spatio-selective control of the network isomerization and corresponding macroscopic properties. Figure 3a shows that the use of different photomasks allows control of the crystallization as reﬂected in the optical property change at different temperatures. At 80 °C, all the crystals melt and the ﬁlms are transparent. At 0 °C, both the isomerized (Ri) and non-isomerized (Rn) regions crystallize, but the Ri appears more opaque due to the higher crystallinity. When the tempera- ture is raised to 37 °C, the contrast between the two regions become more evident due to the selective melting in the Rn region. In addition to the optical properties, the modulus contrast between the isomerized and non-isomerized regions can readily switch. At 37 °C, the Rn region is in its crystalline state and the Ri region in its rubbery state. Consequently, the modulus of the former is 25 times that of the latter (Fig. 3b). At 80 °C, both regions are at their rubbery states and their modulus contrast is reversed. The Ri region becomes 7 times softer than the Rn region because of its lower crosslinking density and that the free PEG chains do not contribute to the modulus. This reversal in modulus contrast is visually demonstrated in Fig. 3c. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 At 37 °C, stretching is NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications 4 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 37 °C 80 °C 0 1 2 3 50 60 80°C Elastic modulus (MPa) 0 s 300 s 37 °C 37 °C 0 °C 80 °C 37 °C 0 °C 80 °C a c b Stertch 30% Fig. 3 Spatio-selective topological isomerization. a Temperature induced change in optical contrast due to the patterned isomerization/crystallization. The letters and logo correspond to the isomerized region with light irradiation of 300 s. The rest of the sample corresponds to non-isomerized region with no light irradiation. b Comparison of the elastic moduli for the non-isomerized and the isomerized network. c Reversal of modulus contrast at different temperatures. All scale bars: 10 mm and error bars represent standard deviations calculated from three specimens. 37 °C 0 °C 80 °C 37 °C 0 °C 80 °C a a 37 °C 80 °C c Stertch 30% 0 1 2 3 50 60 80°C Elastic modulus (MPa) 0 s 300 s 37 °C b b c Fig. 3 Spatio-selective topological isomerization. a Temperature induced change in optical contrast due to the patterned isomerization/crystallization. The letters and logo correspond to the isomerized region with light irradiation of 300 s. The rest of the sample corresponds to non-isomerized region with no light irradiation. b Comparison of the elastic moduli for the non-isomerized and the isomerized network. c Reversal of modulus contrast at different temperatures. All scale bars: 10 mm and error bars represent standard deviations calculated from three specimens. 0 °C and 37 °C, the material shows reversible shape memory behavior with the low melting temperature Rn serving as the actuation region. For the model pattern in Fig. 4c, the bending curvature can be easily tuned by changing the area ratio between Ri and Rn (Supplementary Fig. 7). The light patterns dictate the spatial stress within the ﬁlm. Thus, employing diverse light pat- terns provide ample freedom to manipulate the reversible shape memory cycle (Fig. 4d). Importantly, the different geometric shapes involved in Fig. 4d are determined by the light patterns whereas the external deformation force is identical (i.e. simple linear stretching of 30%). This stands in sharp contrast to cur- rently known reversible SMP for which the shapes involved are dictated by the external programming force. For those materials, to create the 3D shape shifting in Fig. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 4d would require corre- sponding programming force much more complex than the linear stretching used for our material systems. This characteristic brings up two unique beneﬁts: A single programming force creates two temporary shapes that are non- linearly dependent; A simple stretching force allows access to complex 3D temporary shapes. Harnessing these attractive features for device applications may present interesting opportunities in the future. Discussion In summary, we demonstrate a generally applicable molecular design of a topology isomerizable network (TIN) and show its unique versatility toward designable SMP on demand. Relying on the transesteriﬁcation as the dynamic exchange reaction, this TIN can isomerize to various topological isomeric states with changes in topological defects. This in turn provides an opportunity to tailor the thermomechanical properties (crystallization and modulus). Coupled with spatio-temporal light regulation of the isomerization, the starting network polymer can evolve into many custom designable network composites. Based on this, we illus- trate unusual opportunities in designing reversible SMP and multi-SMP with shape-shifting versatility beyond those reported elsewhere. At ﬁrst glance, our approach may appear similar to what is achievable for two-stage polymerization in which network can be altered after the synthesis4,39–41. In reality, two-stage polymerization typically leads to higher crosslinking densities, without signiﬁcant other changes. By comparison, many other network topological features can be tuned via topological iso- merization. Consequently, the adjustable range of macroscopic properties is broader, as exempliﬁed in Fig. 3b, c. On a broad basis, the concept of TIN greatly extends the design space for dynamic covalent networks with implications for many other types of adaptive polymers beyond SMP. The opportunity to design the shape-shifting behavior on demand goes beyond the reversible shape memory. Figure 4e illustrates another possibility in manipulating triple-shape memory behavior. A polymer ﬁlm with an isomerization pattern is subjected to linear stretching (30%) at 80 °C. Cooling down to 0 °C ﬁxes this ﬁrst temporary shape. Upon heating to 37 °C under stress-free condition, the ﬁlm recovers to a second temporary 3D shape. Further heating to 80 °C recovers the original shape, the 2D ﬁlm. Again, depending on the pattern, the second temporary 3D shape can be quite different (Fig. 4f). This triple-shape behavior has a key distinction with currently known triple-shape examples for which both temporary shapes are typically determined by the programming force(s)37,38. For the current system, while the ﬁrst temporary shape is solely dependent on the programming force, the second temporary shape is dictated by both the pattern and programming force. Discussion 4 Reversible and triple-shape memory performances. a Cyclic reversible actuation of a non-isomerized material under a constan b The relationship between reversible strain and the external stress. c Schematic illustration of reversible shape memory process. d transformations obtained by patterned isomerization with the corresponding light patterns shown as the insets (light irradiation time: illustration of the triple-shape memory process. f The original shape and 2nd temporary shape in the triple-shape memory cycles, with light patterns shown as the insets (light irradiation time: 300 s). (All scale bars: 10 mm). c 0 °C Ri region Tm: 50 °C 37 °C 37 °C Rn region Tm: 31 °C Stretch at 80 °C Fix at 0 °C c c 37 °C Recover Recover 0 °C d e f 37 °C 37 °C 37 °C 1st temporary shape 2nd temporary shape 80 °C 0 °C Ri region Tm: 50 °C Rn region Tm: 31 °C Stretch at 80 °C Fix at 0 °C ersible and triple-shape memory performances. a Cyclic reversible actuation of a non-isomerized material under a consta tionship between reversible strain and the external stress. c Schematic illustration of reversible shape memory process. d ations obtained by patterned isomerization with the corresponding light patterns shown as the insets (light irradiation time: f th t i l h f Th i i l h d 2nd t h i th t i l h l it d e Fig. 4 Reversible and triple-shape memory performances. a Cyclic reversible actuation of a non-isomerized material under a constant stress (0.8 MPa). b The relationship between reversible strain and the external stress. c Schematic illustration of reversible shape memory process. d Reversible shape transformations obtained by patterned isomerization with the corresponding light patterns shown as the insets (light irradiation time: 300 s). e Schematic illustration of the triple-shape memory process. f The original shape and 2nd temporary shape in the triple-shape memory cycles, with the corresponding light patterns shown as the insets (light irradiation time: 300 s). (All scale bars: 10 mm). Fig. 4 Reversible and triple-shape memory performances. a Cyclic reversible actuation of a non-isomerized material under a constant stress (0.8 MPa). b The relationship between reversible strain and the external stress. c Schematic illustration of reversible shape memory process. d Reversible shape transformations obtained by patterned isomerization with the corresponding light patterns shown as the insets (light irradiation time: 300 s). Discussion e Schematic illustration of the triple-shape memory process. f The original shape and 2nd temporary shape in the triple-shape memory cycles, with the corresponding light patterns shown as the insets (light irradiation time: 300 s). (All scale bars: 10 mm). Methods poured into a glass mold separated by a silicone rubber spacer (0.3 mm thickness) and irradiated under UV light for 180 s (light source: IntelliRay 600 Flood UV, intensity: 60 mW/cm2). The obtained ﬁlm was vacuum dried at 70 °C for 24 h, then stored in a desiccator prior to testing. PEG-0 and PEG-1 were synthesized similarly except that the amount of N-hydroxyethylacrylamide was varied according to the hydroxyl/ester molar ratio. Materials. Polyethylene glycol (Mn = 3350) was purchased from Sigma-Aldrich. Irgacure 2959, N-hydroxyethylacrylamide, acryloyl chloride, trimethylamine, 1,5,7- triazabicyclo[4.4.0]dec-5-ene, and ketoprofen were acquired from TCI. Acetic acid, toluene, N,N′-dimethylformamide (DMF) were obtained from Guoyao chemical. All chemicals were used as received. The photobase generator was synthesized according to the literature29. Introduction of the photobase generator (PBG). PBG was introduced into a polymer network by soaking the material in 2 wt% PBG toluene solution for 20 min. With this procedure, 6 wt% of PBG (calculated from mass change) was incorporated into the network polymer. For light irradiation experiments, a polymer sample was irradiated equally on both sides and the irradiation time was taken as the sum of exposure times on both sides. Discussion 5 5 NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 Methods poured into a glass mold separated by a silicone rubber spacer (0.3 mm thic 0 40 80 120 0 10 20 30 40 Strain (%) Time (min) –20 0 20 40 60 80 Temperature (oC) 0.3 0.4 0.5 0.6 0.7 0.8 5 10 15 20 Actuation strain (%) Stress (MPa) a b c 0 °C 37 °C Recover Recover 0 °C d e f Ri region Tm: 50 °C 37 °C 37 °C 37 °C 37 °C 37 °C 1st temporary shape 2nd temporary shape 80 °C 0 °C Ri region Tm: 50 °C Rn region Tm: 31 °C Rn region Tm: 31 °C Stretch at 80 °C Stretch at 80 °C Fix at 0 °C Fix at 0 °C g. 4 Reversible and triple-shape memory performances. a Cyclic reversible actuation of a non-isomerized material under a constant stress (0.8 The relationship between reversible strain and the external stress. c Schematic illustration of reversible shape memory process. d Reversible sh ansformations obtained by patterned isomerization with the corresponding light patterns shown as the insets (light irradiation time: 300 s). e Sche ustration of the triple-shape memory process. f The original shape and 2nd temporary shape in the triple-shape memory cycles, with the correspo ht patterns shown as the insets (light irradiation time: 300 s). (All scale bars: 10 mm). 0 40 80 120 0 10 20 30 40 Strain (%) Time (min) –20 0 20 40 60 80 Temperature (oC) 0.3 0.4 0.5 0.6 0.7 0.8 5 10 15 20 Actuation strain (%) Stress (MPa) a b 0 40 80 120 0 10 20 30 40 Strain (%) Time (min) –20 0 20 40 60 80 Temperature (oC) a 0.3 0.4 0.5 0.6 0.7 0.8 5 10 15 20 Actuation strain (%) Stress (MPa) b 0 40 80 120 0 Time (min) –20 0.3 0.4 0.5 0.6 0.7 0.8 5 A Stress (MPa) c 0 °C 37 °C Recover Recover 0 °C d e f Ri region Tm: 50 °C 37 °C 37 °C 37 °C 37 °C 37 °C 1st temporary shape 2nd temporary shape 80 °C 0 °C Ri region Tm: 50 °C Rn region Tm: 31 °C Rn region Tm: 31 °C Stretch at 80 °C Stretch at 80 °C Fix at 0 °C Fix at 0 °C Fig. References 32. Chung, T., Romo-Uribe, A. & Mather, P. T. Two-way reversible shape memory in a semicrystalline network. Macromolecules 41, 184–192 (2008). 1. Ware, T. H., McConney, M. E., Wie, J. J., Tondiglia, V. P. & White, T. J. Voxelated liquid crystal elastomers. Science 347, 982–984 (2015). 1. Ware, T. H., McConney, M. E., Wie, J. J., Tondiglia, V. P. & White, T. J. Voxelated liquid crystal elastomers. Science 347, 982–984 (2015). 33. Behl, M., Kratz, K., Zotzmann, J., Nöchel, U. & Lendlein, A. Reversible bidirectional shape-memory polymers. Adv. Mater. 25, 4466–4469 (2013). q y 2. Xia, Y., Cedillo-Servin, G., Kamien, R. D. & Yang, S. Guided folding of nematic liquid crystal elastomer sheets into 3D via patterned 1D microchannels. Adv. Mater. 28, 9637–9643 (2016). 2. Xia, Y., Cedillo-Servin, G., Kamien, R. D. & Yang, S. Guided folding of nematic liquid crystal elastomer sheets into 3D via patterned 1D microchannels. Adv. Mater. 28, 9637–9643 (2016). 34. Song, H. et al. Synergetic chemical and physical programming for reversible shape memory effect in a dynamic covalent network with two crystalline phases. ACS Macro Lett. 8, 682–686 (2019). 3. Kowalski, B. A., Guin, T. C., Auguste, A. D., Godman, N. P. & White, T. J. Pixelated polymers: directed self assembly of liquid crystalline polymer networks. ACS Macro Lett. 6, 436–441 (2017). p 35. Turner, S. A., Zhou, J., Sheiko, S. S. & Ashby, V. S. Switchable micro-patterned surface topographies mediated by reversible shape memory. ACS Appl. Mater. Interfaces 6, 8017–8021 (2014). 4. Jin, B. et al. Programming a crystalline shape memory polymer network with thermo- and photo-reversible bonds toward a single-component soft robot. Sci. Adv. 4, eaao3865 (2018). f 36. Meng, Y., Yang, J., Lewis, C. L., Jiang, J. & Anthamatten, M. Photoinscription of chain anisotropy into polymer networks. Macromolecules 49, 9100–9107 (2016). 5. Qian, X. et al. Untethered recyclable tubular actuators with versatile locomotion for soft continuum robots. Adv. Mater. 30, 1801103 (2018). 5. Qian, X. et al. Untethered recyclable tubular actuators with versatile locomotion for soft continuum robots. Adv. Mater. 30, 1801103 (2018). 6. Zhang, X. et al. ACS Appl. Mater. Interfaces 11, 46212–46218 (2019). 37. Xie, T. Tunable polymer multi-shape memory effect. Nature 464, 267–270 (2010). locomotion for soft continuum robots. Adv. Mater. 30, 1801103 (2018). 6. Zhang, X. et al. ACS Appl. Mater. Interfaces 11, 46212–46218 (2019). 38. Behl, M., Bellin, I., Kelch, S., Wagermaier, W. & Lendlein, A. References One-step process for creating triple-shape capability of AB polymer networks. Adv. Funct. Mater. 19, 102–108 (2009). 7. Cao, Y. et al. Direct fabrication of stretchable electronics on a polymer substrate with process-integrated programmable rigidity. Adv. Funct. Mater. 28, 1804604 (2018). 39. Zhang, B., kowsari, K., Serjouei, A., Dunn, M. L. & Ge, Q. Reprocessable thermosets for sustainable three-dimensional printing. Nat. Commun. 9, 1831 (2018). 8. Kang, J. et al. Tough and water-insensitive self-healing elastomer for robust electronic skin. Adv. Mater. 30, 1706846 (2018). 9. Lendlein, A., Behl, M., Hiebl, B. & Wischke, C. Shape-memory polymers as a technology platform for biomedical applications. Expert Rev. Med. Devices 7, 357–379 (2010). 40. Luo, C. et al. Chemomechanics of dual-stage reprocessable thermosets. J. Mech. Phys. Solids 126, 168–186 (2019). 41. Nair, D. P. et al. Two-stage reactive polymer network forming systems. Adv. Funct. Mater. 22, 1502–1510 (2012). 10. Zou, W., Dong, J., Luo, Y., Zhao, Q. & Xie, T. Dynamic covalent polymer networks: from old chemistry to modern day innovations. Adv. Mater. 29, 1006100 (2017). 11. Zheng, N. & Xie, T. Thermadapt shape memory polymer. Acta Polymerica Sin 11, 46–55 (2017). ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 22. Liu, W. et al. Dynamic multiphase semi-crystalline polymers based on thermally reversible pyrazole-urea bonds. Nat. Commun. 10, 4753 (2019). Characterization. Differential scanning calorimetry (DSC) was conducted using aTA Q200 instrument. All samples were ﬁrst equilibrated at 80 °C and cooled to −50 °C at a rate of 5 °C/min. The DSC curves were obtained in a second heating run at a rate of 5 °C/min. Dynamic thermomechanical analysis (DMA) was con- ducted using aTA Q800 instrument. The reversible shape memory curves were obtained in a “controlled force” mode. The storage modulus was obtained in a “multi-frequency strain” mode with an amplitude of 20 µm at 1 Hz. Elastic moduli were determined by tensile tests conducted with a Zwick/Roell Z005 machine under a tensile speed of 10 mm/min. A minimum of three specimens in a standard dumbbell shape were tested. 23. Kloxin, C. J., Scott, T. F., Park, H. Y. & Bowman, C. N. Mechanophotopatterning on a photoresponsive elastomer. Adv. Mater. 23, 1977–1981 (2011). 24. Michal, B. T., Jaye, C. A., Spencer, E. J. & Rowan, S. J. Inherently photohealable and thermal shape-memory polydisulﬁde networks. ACS Macro Lett. 2, 694–699 (2013). 25. Zhao, Q., Zou, W., Luo, Y. & Xie, T. Shape memory polymer network with thermally distinct elasticity and plasticity. Sci. Adv. 2, e1501297 (2016). y y y 26. Ji, S., Fan, F., Sun, C., Yu, Y. & Xu, H. Visible light-induced plasticity of shape memory polymers. ACS Appl. Mater. Interfaces 9, 33169–33175 (2019). Acknowledgements This work was supported by the following programs: National Natural Science Foun- dation of China (No. 21625402, 51822307 and 51673169). The authors also thank Mrs. Li Xu for her assistance in performing DSC analyses at State Key Laboratory of Chemical This work was supported by the following programs: National Natural Science Foun- dation of China (No. 21625402, 51822307 and 51673169). The authors also thank Mrs. Li Xu for her assistance in performing DSC analyses at State Key Laboratory of Chemical ( ) 12. Scheutz, G. M., Lessard, J. J., Sims, M. B. & Sumerlin, B. S. Adaptable 12. Scheutz, G. M., Lessard, J. J., Sims, M. B. & Sumerlin, B. S. Adaptable crosslinks in polymeric materials: resolving the intersection of thermoplastics and thermosets. J. Am. Chem. Soc. 141, 16181–16196 (2019). Xu for her assistance in performing DSC analyses at State Key Laboratory of Chemical Engineering (Zhejiang University). Engineering (Zhejiang University). 13. Montarnal, D., Capelot, M., Tournilhac, F. & Leibler, L. Silica-like malleable materials from permanent organic networks. Science 334, 965–968 (2011). Received: 20 March 2020; Accepted: 29 July 2020; 29. Zou, W. et al. Light-triggered topological programmability in a dynamic covalent polymer network. Sci. Adv. 6, eaaz2362 (2020). p y 30. Zhong, M., Wang, R., Kawamoto, K., Olsen, B. D. & Johnson, J. A. Quantifying the impact of molecular defects on polymer network elasticity Science 353, 1264–1268 (2016). 31. Wang, J. et al. Counting loops in sidechain-crosslinked polymers from elastic solids to single-chain nanoparticles. Chem. Sci. 10, 5332–5337 (2019). Author contributions 14. Lu, Y., Tournilhac, F., Leibler, L. & Guan, Z. Making insoluble polymer networks malleable via oleﬁn metathesis. J. Am. Chem. Soc. 134, 8424–8427 (2012). T.X. conceived the concept. W.M. designed and conducted the experiments. W.M. and T.X. wrote the paper. W.M., W.Z., B.J., N.Z., and Q.Z. analyzed experimental results. W.M., W.Z., B.J., C.N., N.Z. Q.Z., and T.X. contributed to the discussion. 15. Wang, C. et al. Recyclable and repolymerizable thiol–X photopolymers. Mater. Horiz. 5, 1042–1046 (2018). 16. Taynton, P. et al. Heat- or water-driven malleability in a highly recyclable covalent network polymer. Adv. Mater. 26, 3938–3942 (2014). Synthesis of polyethylene glycol diacrylate (PEGDA). PEG (20.0 g, 5.97 mmol) and trimethylamine (2.78 g, 27.50 mmol) were dissolved in toluene (200 mL). Synthesis of polyethylene glycol diacrylate (PEGDA). PEG (20.0 g, 5.97 mmol) and trimethylamine (2.78 g, 27.50 mmol) were dissolved in toluene (200 mL). Acryloyl chloride (2.49 g, 27.50 mmol) was added dropwise at 0 °C. The mixture was kept at 80 °C for 20 h. The byproduct was removed by ﬁltration and PEGDA was obtained by precipitating the clear solution in hexane. The ﬁnal product was vacuum dried for 24 h at room temperature and its structure veriﬁed by 1H NMR analysis (Supplementary Fig. 8). Synthesis of polyethylene glycol diacrylate (PEGDA). PEG (20.0 g, 5.97 mmol) and trimethylamine (2.78 g, 27.50 mmol) were dissolved in toluene (200 mL). Acryloyl chloride (2.49 g, 27.50 mmol) was added dropwise at 0 °C. The mixture was kept at 80 °C for 20 h. The byproduct was removed by ﬁltration and PEGDA was obtained by precipitating the clear solution in hexane. The ﬁnal product was vacuum dried for 24 h at room temperature and its structure veriﬁed by 1H NMR analysis (Supplementary Fig. 8). Gel fraction tests. A specimen was weighed before soaking in CHCl3 for 24 h (replacing the solvent every 12 h). The swollen specimen was then vacuum dried for 24 h at room temperature and weighed. The gel fraction was calculated based on the ratio between the ﬁnal and initial weights. Synthesis of the polymer networks. For the synthesis of PEG-2, PEGDA (0.5 g, 0.15 mmol) was dissolved in DMF (0.5 g) at 60 °C. N-hydroxyethylacrylamide (0.07 g, 0.6 mmol) and photoinitiator (Irgacure 2959, 1.5 wt%) were added into the PEGDA solution and stirred for several minutes. The homogenous mixture was NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunicatio 6 Competing interests 17. Fortman, D. J., Brutman, J. P., Cramer, C. J., Hillmyer, M. A. & Dichtel, W. R. Mechanically activated, catalyst-free polyhydroxyurethane vitrimers. J. Am. Chem. Soc. 137, 14019–14022 (2015). Competing interests The authors declare no competing interests. Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-18116-1. Correspondence and requests for materials should be addressed to T.X. Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available. Reprints and permission information is available at http://www.nature.com/reprints Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 18. Delahaye, M., Winne, J. M. & Du Prez, F. E. Internal catalysis in covalent adaptable networks: phthalate monoester transesteriﬁcation as a versatile dynamic cross-linking chemistry. J. Am. Chem. Soc. 141, 15277–15287 (2019). Data availability 27. Gu, Y. et al. Photoswitching topology in polymer networks with metal–organic cages as crosslinks. Nature 560, 65–69 (2018). The data that support the ﬁndings of this study are available from the corresponding author on reasonable request. 28. Sun, H. et al. Macromolecular metamorphosis via stimulus-induced transformations of polymer architecture. Nat. Chem. 9, 817–823 (2017). Received: 20 March 2020; Accepted: 29 July 2020; Received: 20 March 2020; Accepted: 29 July 2020; © The Author(s) 2020 Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-18116-1. y g y 19. He, C., Shi, S., Wang, D., Helms, B. A. & Russell, T. P. Poly(oxime−ester) vitrimers with catalyst-free bond exchange. J. Am. Chem. Soc. 141, 13753–13757 (2019). 20. Li, L., Chen, X. & Torkelson, J. M. Reprocessable polymer networks via thiourethane dynamic chemistry: recovery of cross-link density after recycling and proof-of-principle solvolysis leading to monomer recovery. Macromolecules 52, 8207–8216 (2019). Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available. 21. Zhang, Z., Rong, M. & Zhang, M. Mechanically robust, self-healable, and highly stretchable “living” crosslinked polyurethane based on a reversible C-C bond. Adv. Funct. Mater. 28, 1706050 (2018). 7 NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18116-1 ARTICLE ARTICLE 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/. NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunications 8 NATURE COMMUNICATIONS \| (2020) 11:4257 \| https://doi.org/10.1038/s41467-020-18116-1 \| www.nature.com/naturecommunicatio
https://openalex.org/W4293771244	https://www.frontiersin.org/articles/10.3389/fevo.2022.980776/pdf	English	null	Casting a light on the shoreline: The influence of light pollution on intertidal settings	Frontiers in ecology and evolution	2,022	cc-by	8,207	Casting a light on the shoreline: The inﬂuence of light pollution on intertidal settings OPEN ACCESS EDITED BY Marcel Eens, University of Antwerp, Belgium REVIEWED BY Karen Manríquez, Andres Bello University, Chile María Bagur, CONICET Centro Austral de Investigaciones Cientíﬁcas (CADIC), Argentina CORRESPONDENCE Pedro A. Quijón pquijon@upei.ca SPECIALTY SECTION This article was submitted to Behavioral and Evolutionary Ecology, a section of the journal Frontiers in Ecology and Evolution RECEIVED 28 June 2022 ACCEPTED 12 August 2022 PUBLISHED 31 August 2022 OPEN ACCESS EDITED BY Marcel Eens, University of Antwerp, Belgium REVIEWED BY Karen Manríquez, Andres Bello University, Chile María Bagur, CONICET Centro Austral de Investigaciones Cientíﬁcas (CADIC), Argentina K. Devon Lynn and Pedro A. Quijón Department of Biology, University of Prince Edward Island, Charlottetown, PE, Canada Department of Biology, University of Prince Edward Island, Charlottetown, PE, Canada Light pollution is becoming prevalent among other coastal stressors, particularly along intertidal habitats, arguably the most exposed to anthropogenic light sources. As the number of light pollution studies on sandy beaches, rocky shores and other intertidal habitats raises, commonalities, research gaps and venues can be identiﬁed. Hence, the inﬂuence of light pollution on the behavior and ecology of a variety of intertidal macro-invertebrates and vertebrates are outlined by examining 54 published studies. To date, a large majority of the reported effects of light pollution are negative, as expected from the analysis of many species with circadian rhythms or nocturnal habits, although the severity of those effects ranges widely. Experimental approaches are well represented throughout but methodological limitations in measurement units and standardization continue to limit the proposal of general conclusions across species and habitats. In addition, studies targeting community variables and the explicit inﬂuence of skyglow are heavily underrepresented. Likewise, studies addressing the interaction between light pollution and other natural and anthropogenic stressors are critically needed and represent a key venue of research. The nature of those interactions (synergistic, additive, antagonistic) will likely dictate the impact and management of light pollution in the decades ahead. CITATION Lynn KD and Quijón PA (2022) Casting a light on the shoreline: The inﬂuence of light pollution on intertidal settings. Front. Ecol. Evol. 10:980776. doi: 10.3389/fevo.2022.980776 CITATION Lynn KD and Quijón PA (2022) Casting a light on the shoreline: The inﬂuence of light pollution on intertidal settings. Front. Ecol. Evol. 10:980776. doi: 10.3389/fevo.2022.980776 COPYRIGHT © 2022 Lynn and Quijón. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). Casting a light on the shoreline: The inﬂuence of light pollution on intertidal settings 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. light pollution, sandy beaches, rocky shores, tidal ﬂats, species responses light pollution, sandy beaches, rocky shores, tidal ﬂats, species responses TYPE Mini Review PUBLISHED 31 August 2022 DOI 10.3389/fevo.2022.980776 TYPE Mini Review PUBLISHED 31 August 2022 DOI 10.3389/fevo.2022.980776 Introduction (2018), these responses seem to be dictated by the strength of the stressor (light intensity), although there are also methodological diﬀerences, which in the case of Fanini et al. (2016) entailed the application of a recreation index (McLachlan et al., 2013) as a proxy of light pollution. Our review focuses on the inﬂuence of light pollution upon the behavior and ecology of marine intertidal species. We deﬁne intertidal settings broadly to include soft- (sandy beaches, sandy and muddy ﬂats, saltmarshes) and hard-bottoms (rocky shores, tide pools and man-made structures), and to encompass a diverse range of invertebrates and vertebrates, and the nature of their use of the intertidal (permanent or seasonal). Even though light pollution eﬀects may emerge from exposure to discrete light sources or the more diﬀuse and widespread skyglow (Kyba et al., 2015; Cox et al., 2022), most studies with intertidal species have focused on the former type of pollution. This is reﬂected in the results of this review. Nonetheless, our goals are threefold: (i) to establish commonalities on species’ responses to light pollution, (ii) to identify gaps emerging from those studies, and (iii) to propose key venues of research in relation to other co-occurring stressors. The geographic scope of our review is outlined in Table 1 and Figure 1 and includes 54 published articles (as of July 1st, 2022). These were searched using keywords such as “light pollution” “intertidal” and other relevant descriptors from databases (e.g., Academic Search Complete, BioOne), peer-reviewed and reputed sources, and cross-referencing. Various other behavioral and life history traits have been measured in beach amphipods, isopods and coleopteran insects (Phalerisida maculata), and their outcome is more consistent with respect to light pollution stress. Abundance declined in response to light pollution in T. spinulosus (Duarte et al., 2019), and in response to an urbanization index used as a proxy of light pollution (Legendre and Legendre, 1998) in P. maculata (Gonzalez et al., 2014). Similarly, light pollution was detrimental to feeding rates in O. tuberculata (Luarte et al., 2016) and A. longicornis (Lynn et al., 2021a) and closely related variables such as food absorption eﬃciency in A. longicornis (Lynn et al., 2021a), RNA:DNA ratios in P. maculata and in T. spinulosus (Quintanilla-Ahumada et al., 2021, 2022) as well as growth rates in O. tuberculata (Luarte et al., 2016). Introduction Natural ecosystems are facing an unprecedented increase in the number of human- related stressors (Halpern et al., 2008; Bellard et al., 2012). So, it is critical to identify what aspects of the species’ behavior and ecology are being altered and what changes can be expected in the short- and long-term (Griﬀen et al., 2016). An examination of the published evidence also helps to identify research gaps and venues, particularly in emerging ﬁelds of study (Dicks et al., 2014). Arguably, one of such ﬁelds is the inﬂuence of artiﬁcial light pollution at night (hereafter light pollution). Compared to other equally pervasive stressors, light pollution has only recently gathered considerable attention Frontiers in Ecology and Evolution Frontiers in Ecology and Evolution 01 frontiersin.org 10.3389/fevo.2022.980776 Lynn and Quijón Lynn and Quijón (Davies and Smyth, 2017) despite its fast growth worldwide (Sánchez de Miguel et al., 2021). Historically, most research on light pollution has focused on terrestrial settings and species (Vaz et al., 2020), while studies on aquatic systems (Baz et al., 2022) and on marine systems have lagged behind (Longcore and Rich, 2004). Among the latter, intertidal settings are of interest because they are already exposed to a wide range of other stressors (Gunderson et al., 2016), and are arguably the most directly exposed to the main sources of light pollution. (Davies and Smyth, 2017) despite its fast growth worldwide (Sánchez de Miguel et al., 2021). Historically, most research on light pollution has focused on terrestrial settings and species (Vaz et al., 2020), while studies on aquatic systems (Baz et al., 2022) and on marine systems have lagged behind (Longcore and Rich, 2004). Among the latter, intertidal settings are of interest because they are already exposed to a wide range of other stressors (Gunderson et al., 2016), and are arguably the most directly exposed to the main sources of light pollution. Bregazzi and Naylor, 1972) or mid-level light pollution (60 lx; Luarte et al., 2016; Lynn et al., 2021a), it remained only slightly altered in other cases (Fanini et al., 2016; Dhouha et al., 2019). The same can be said about burrowing isopods such as Tylos europaeus in Tunisia, T. spinulosus in Chile and A. bipleura in Australia. While activity declined in the former two species, it remained unaltered in the latter one (Fanini et al., 2016). As suggested by Dhouha et al. Sandy beaches and resident invertebrates A wide range of light pollution (0.005–200 lx) has been directly or indirectly measured in sandy beaches by thirteen articles nearly equally split in laboratory and ﬁeld studies. Often referred to as semi-terrestrial species, talitrid amphipods were among the ﬁrst model-species used in the study of circadian rhythms (see Scapini et al., 2005) and light pollution. Among them, Talitrus saltator was the ﬁrst to be used to assess the inﬂuence of skyglow changes on the species’ orientation across the intertidal (Torres et al., 2022). T. saltator on beaches in Wales and Tunisia, Orchestoidea tuberculata in Chile, Americorchestia longicornis in Canada and Platorchestia smithi in Australia have all been used to measure responses to more direct sources of light pollution. Locomotor activity, a variable intuitively expected to be sensitive to light pollution, has been measured across most studies but has oﬀered inconsistent results. While the activity of three of these species was drastically reduced in the presence of high (200 lx; Introduction Although the trends are consistent across species, there have been exceptions: Not all the diﬀerences between controls and light pollution treatments have been signiﬁcant. Only two studies have addressed the eﬀects of light pollution on sandy beach communities located in Wales (Garratt et al., 2019) and Uruguay (Orlando et al., 2020). While species diversity increased with light pollution in some sites (Garratt et al., 2019) it declined in others (Orlando et al., 2020). Garratt et al. (2019) also documented a handful of other responses to light pollution stress: an increase in overall biomass, a change in species composition but a lack of eﬀects on overall abundance and dominance. Frontiers in Ecology and Evolution Sandy beaches and seasonal vertebrates Seasonal vertebrates such as loggerhead (Caretta caretta), leatherback (Dermochelys coriacea), green (Chelonia mydas), hawksbill (Eretmochelys imbricata), and ﬂatback turtles (Natator depressus) are among the species that have gathered the most public awareness about light pollution impacts (Mazor et al., 2013; Pendoley and Kamrowski, 2016). Seventeen studies to date, most of them ﬁeld-oriented, have addressed the inﬂuence of a wide range of light pollution (0.01–480 lx) on the nesting and hatching of turtles. Studies on the behavior of female turtles attempting to nest have measured light pollution with diﬀerent methodologies, from direct in situ measurements Frontiers in Ecology and Evolution 02 frontiersin.org (country), study setting (L, laboratory; F, ﬁeld), light pollution intensity reported (in lux unless otherwise speciﬁed; n.d.: unclear), Setting Light (lx) Measured response Light pollution inﬂuence # L 140 Activity No inﬂuence Signiﬁcant reduction 1 L 200 Activity Activity suppression 2 F Skyglow Orientation Changes in skyglow alter orientation 3 L 5–140 Survival/activity Mild reduction/light-dependent 4 L/F 60 Activity feeding/growth rate absorption eﬃciency At least 50% reduction 2x/3x reduction No diﬀerence 5 L/F 60 Activity/feeding rate absorption eﬃciency growth rate Alteration/2x reduction 3x reduction Mild reduction 6 F Index Activity Mild reduction 7 L/F 120 Abundance/size activity rhythm 5x reduction/reduction Reduction and alteration 8 L 0–100 RNA:DNA Abs. eﬃciency/Growth Dose response (decline) Dose response (rise)/No eﬀect 9 L 60–120 RNA:DNA 10–20% Reduction 10 F Index Abundance Decrease with urbanization level 11 F 0.005–5.12 Species composition Species richness/biomass Dominance/total density Species abundance Altered by light intensity Enhanced (interact w/org. matter) No change Variable (some up, some down) 12 F Satellite Species richness Signiﬁcant reduction 13 F 0.5–480 Nesting density Signiﬁcantly reduction 14 F n.d. Hatchling sea-ﬁnding Signiﬁcantly disruption 15 F 0.01–0.08 Hatchling orientation Signiﬁcantly disruption 16 F n.d. Sandy beaches and seasonal vertebrates Hatchling orientation Impairing after long exposure 17 F Satellite Nesting density Nest persistence Signiﬁcant reduction Signiﬁcant reduction 18 F Satellite Nesting density Signiﬁcant reduction 19 F 1.9–29.2 Hatchling orientation Signiﬁcant disruption 20 (Continued) (country), study setting (L, laboratory; F, ﬁeld), light pollution intensity reported (in lux unless otherwise speciﬁed; n.d.: unclear), Setting Light (lx) Measured response Light pollution inﬂuence # L 140 Activity No inﬂuence Signiﬁcant reduction 1 L 200 Activity Activity suppression 2 F Skyglow Orientation Changes in skyglow alter orientation 3 L 5–140 Survival/activity Mild reduction/light-dependent 4 L/F 60 Activity feeding/growth rate absorption eﬃciency At least 50% reduction 2x/3x reduction No diﬀerence 5 L/F 60 Activity/feeding rate absorption eﬃciency growth rate Alteration/2x reduction 3x reduction Mild reduction 6 F Index Activity Mild reduction 7 L/F 120 Abundance/size activity rhythm 5x reduction/reduction Reduction and alteration 8 L 0–100 RNA:DNA Abs. eﬃciency/Growth Dose response (decline) Dose response (rise)/No eﬀect 9 L 60–120 RNA:DNA 10–20% Reduction 10 F Index Abundance Decrease with urbanization level 11 F 0.005–5.12 Species composition Species richness/biomass Dominance/total density Species abundance Altered by light intensity Enhanced (interact w/org. matter) No change Variable (some up, some down) 12 F Satellite Species richness Signiﬁcant reduction 13 F 0.5–480 Nesting density Signiﬁcantly reduction 14 F n.d. Hatchling sea-ﬁnding Signiﬁcantly disruption 15 F 0.01–0.08 Hatchling orientation Signiﬁcantly disruption 16 F n.d. Hatchling orientation Impairing after long exposure 17 F Satellite Nesting density Nest persistence Signiﬁcant reduction Signiﬁcant reduction 18 F Satellite Nesting density Signiﬁcant reduction 19 F 1.9–29.2 Hatchling orientation Signiﬁcant disruption 20 (Continued) Setting Light (lx) Measured response Light pollution inﬂuence # F n.d. Hatchling/sea-ﬁnding Disruption with bright light 21 F 14.5 (%) Nesting activity model forecast Signiﬁcant reduction extinction within 80 year. 22 L 4.6–4.9 × 1011 Photons/cm2/s Hatchling orientation Signiﬁcant disruption 23 F 250–500 W Hatchling Sea-ﬁnding ability Signiﬁcant disruption when lights nearby (<200 m) 24 F n.d. Hatchling survival At least 50% reduction 25 F Rad. octagons Hatchling sea-ﬁnding Signiﬁcant reduction 26 F Satellite Nesting activity Signiﬁcant reduction 27 F n.d. Hatchling Sea-ﬁnding ability Signiﬁcant disruption Disruption 28 F n.d. Nesting density Preference for shaded areas 29 F Various Nesting behavior Predation risk (ghost crab) 20% Less nesting attempts Risk increase 30 L 50–100 Phototaxis settlement behavior Light avoidance by cyprids Shade-seeking by B. Balanoides 31 F n.d. Sandy beaches and seasonal vertebrates Phototaxis behavior Light seeking in 82% of larvae 32 F 212–16 Settlement rates Late settlement rate reduction 33 F 95–16 Settlement rates Late settlement rate reduction 34 L 80 W Larval release Increase of density with light 35 L 0.5 Larval release Timing and rhythm disruption 36 L 25 µ Einsteins × m2 × s2 Expansion activity Expansion of tentacle activity constant/added risks 37 L ∼25 Behavior risk perception Less refuge seeking at waterline More complex response to prey and predator’s cues 38 L 0–50 Interaction with full moon Interaction with moon at high lux 39 L/F 330–490 Prey search/self-righting Metabolism/density Lower/2–3x slower 2x increase/4–5x lower in lit areas 40 (Continued) g g p g g F 14.5 (%) Nesting activity model forecast Signiﬁcant reduction extinction within 80 year. 22 L 4.6–4.9 × 1011 Photons/cm2/s Hatchling orientation Signiﬁcant disruption 23 F 250–500 W Hatchling Sea-ﬁnding ability Signiﬁcant disruption when lights nearby (<200 m) 24 F n.d. Hatchling survival At least 50% reduction 25 F Rad. octagons Hatchling sea-ﬁnding Signiﬁcant reduction 26 F Satellite Nesting activity Signiﬁcant reduction 27 F n.d. Hatchling Sea-ﬁnding ability Signiﬁcant disruption Disruption 28 F n.d. Nesting density Preference for shaded areas 29 F Various Nesting behavior Predation risk (ghost crab) 20% Less nesting attempts Risk increase 30 L 50–100 Phototaxis settlement behavior Light avoidance by cyprids Shade-seeking by B. Balanoides 31 F n.d. Phototaxis behavior Light seeking in 82% of larvae 32 F 212–16 Settlement rates Late settlement rate reduction 33 F 95–16 Settlement rates Late settlement rate reduction 34 L 80 W Larval release Increase of density with light 35 L 0.5 Larval release Timing and rhythm disruption 36 L 25 µ Einsteins × m2 × s2 Expansion activity Expansion of tentacle activity constant/added risks 37 L ∼25 Behavior risk perception Less refuge seeking at waterline More complex response to prey and predator’s cues 38 L 0–50 Interaction with full moon Interaction with moon at high lux 39 L/F 330–490 Prey search/self-righting Metabolism/density Lower/2–3x slower 2x increase/4–5x lower in lit areas 40 (Continued) tting Light (lx) Measured response Light pollution inﬂuence Model Light spectrum Snail conspicuousness Broader spectrum raises conspicuousness in color morphs L 70 Activity metabolic rate/weight Increased probability and freq. 20% increase/no change F n.d. g Ligh el Light s n ∼ 0 2 n ∼ 4 0.18 Sat n n 2018); 5Luarte et al. (20 3); 16Dimitriadis et al. ( 5Harewood and Horroc 1); 35Amano (1986); 36 45Simons et al., 2022; 4 Sandy beaches and seasonal vertebrates 26 15 7 52 24 53 54 35 36 43 41 32 3/39 2 12 31 4738 44 18 16 1 4 30 21 51 6 33 46 28 1419,20 29 16 23 22 27 25 37 45 11/34 40 10 8/42 9 5 13 48 49 50 FIGURE 1 Outline maps of the main geographic regions where the study of light pollution on (macro) intertidal organisms has taken place this far. The maps are for illustrative purposes only, so locations are all approximate and not necessarily at scale. Numbers refer to the 54 studies listed and detailed in Table 1. (Witherington, 1992; Mazor et al., 2013) to satellite imagery (Strobl et al., 2016). These studies have shown rather consistent responses to light pollution: leatherback, green and hawksbill turtle nests have declined in lit areas (Strobl et al., 2016; Silva et al., 2017; Hu et al., 2018). While some species avoided illuminated beaches (Brei et al., 2016), others such as leatherback turtles nesting in Israel used these sites less persistently (Mazor et al., 2013). Furthermore, some females took longer to choose and complete their nests in lit than in dark beaches (Silva et al., 2017). Numerical models have also predicted that permanent light exposure may cause local-scale extinction in some species within 80 years (Brei et al., 2016). Light color (Silva et al., 2017) or source (mercury vs. low pressure sodium; Witherington, 1992) have been also suggested to inﬂuence the eﬀects of light pollution on nesting. These should be carefully considered when attempting the management of this source of stress (Gaston et al., 2012). 2005; Berry et al., 2013; Kamrowski et al., 2014). Similar results were found for leatherback hatchlings in Gabon (Bourgeois et al., 2009), which in the presence of light pollution, could not use natural cues such as backshore silhouettes to orient themselves. Regarding another stressor, predation, Silva et al. (2017) found that ghost crab’s activity and predation on loggerhead turtle nests was exacerbated under light pollution in beaches of Cabo Verde (NW Africa). In contrast, Harewood and Horrocks (2008) found that predation on hatchlings crawling over the beach was unaltered, but these authors also suggested that disoriented hatchlings incurred on a harmful waste of energy, that later imposed limitations to their ability to swim. Sandy beaches and seasonal vertebrates Foraging Increased foraging time but no eﬀect on physiological factors F ∼40 Predation Cannibalism/density 44–61% Reduction 30% increase/5x increase F 0–6 Feeding rate/concealment Burrow maintenance Reduction/reduction Increase F 2–6 Spatial aggregation Increase F n.d. Fish diversity Larger ﬁsh/numbers under light F Fish predation rates Prey communities Increase of predation rates Alteration prey communities L ∼30 Egg fertilization/spawning Hatching success No changes Signiﬁcantly reduced F 4.3 Survival rates Growth rates 36% Reduction 44% reduction F 0.18–0.71 Choice of site Foraging/prey intake Increase in visual predators Increase/Increase in 4 spp. F Satellite Plover density/roosting Reduction in roosting density F n.d. Foraging, patch preference Signiﬁcant reduction near lights L n.d. Crawling activity/feeding Signiﬁcantly reduction t al. (2018); 5Luarte et al. (2016); 6Lynn et al. (2021a); 7Fanini et al. (2016); 8Duarte et al. (2019); 9/10Quintanilla-Ahumada et al. (2022, 2021) (2013); 16Dimitriadis et al. (2018); 17Lorne and Salmon (2007); 18Mazor et al. (2013); 19Hu et al. (2018); 20Witherington and Bjorndal (1991); 6); 25Harewood and Horrocks (2008); 26Kamrowski et al. (2014); 27Strobl et al. (2016); 28Witherington and Martin (2000); 29Salmon et al. (1 (2021); 35Amano (1986); 36Saigusa and Kawagoye (1997); 37Shick and Dykens (1984); 38Underwood et al. (2017); 39Tidau et al. (2022); 40Ma 010); 45Simons et al., 2022; 46Bird et al. (2004); 47Orvain and Sauriau (2002); 48/49Nuñez et al. (2021a,b); 50Quiñones-Llópiz et al. (2021); 51 2, 202 (1991 t l 10.3389/fevo.2022.980776 Lynn and Quijón 26 15 7 52 24 53 54 35 36 6 43 41 32 3/39 2 12 31 4738 44 18 16 1 4 30 21 51 33 46 28 1419,20 29 16 23 22 27 25 37 45 11/34 40 10 8/42 9 5 13 48 49 50 FIGURE 1 Outline maps of the main geographic regions where the study of light pollution on (macro) intertidal organisms has taken place this far. The maps are for illustrative purposes only, so locations are all approximate and not necessarily at scale. Numbers refer to the 54 studies listed and detailed in Table 1. Frontiers in Ecology and Evolution Rocky shores and related hard-bottom species (2022) also found that Western snowy plovers (Charadrius nivosus) in California were less likely to roost in areas exposed to light pollution due to higher predation risk. late settling stages (spats) suﬀered a signiﬁcant reduction in density with respect to (naturally dark) controls. The literature oﬀers two other examples of light pollution’s detrimental eﬀects at even earlier life stages: Light pollution stress delayed the release of larvae in the sponge Halichondria panicea (Amano, 1986) and a prominent crab species (Hemigrapsus sanguineus) (Saigusa and Kawagoye, 1997) both in Japan. Established stages (juveniles-adults) have reacted rather consistently to light pollution. In an artiﬁcially built jetty in California, light pollution changed the diel activity of sea anemones (Anthopleura elegantissima; Shick and Dykens, 1984). This prompted these anemones to keep expanding their tentacles into the night hours, which at low and high tide conditions exposed them to higher desiccation and predation risk, respectively. Avoidance of light pollution by slow moving marine snails has been reported in the UK (dogwhelks; Nucella lapillus; Underwood et al., 2017) and Chile (large predatory “locos,” Concholepas concholepas; Manríquez et al., 2019). In dogwhelks, light exposure caused a reduction in their response to predator’s olfactory cues, and has also been shown to interact with natural moonlight cycles on the foraging rates of this species on acorn barnacles (Tidau et al., 2022). Light pollution also slowed down the ability of locos to up-right their bodies from an upside-down position, a key antipredator ability in this species (Manríquez et al., 2019). Another anti- predator strategy (camouﬂage) has been assessed on Littorina obtusata and L. fabalis in the UK (McMahon et al., 2022). Testing the light spectra perceived by three predators of these snails, these authors concluded that light pollution makes some littorinid’s color morphs more conspicuous, removing the beneﬁts originally provided by camouﬂage. The inﬂuence of light pollution on a direct indicator of stress, metabolic rates, has been also measured in the locos referred above and small tide-pool rockﬁsh (Girella geofrans; Pulgar et al., 2019). In both species, light pollution increased metabolic rates, adding energy costs and demands in both species. In the case of the snails, this was accompanied by reduced feeding rates while avoiding lit areas (Manríquez et al., 2019), whereas in the rockﬁsh light pollution altered circadian and circatidal rhythms (Pulgar et al., 2019). Bird et al. Rocky shores and related hard-bottom species (2004) reported light pollution eﬀects on nocturnal beach mice on tidal ﬂats of Florida, United States: These mice foraged less and reduced food patch preference in areas closer to light sources. Similarly, Orvain and Sauriau (2002) studying mud snails (Hydrobia ulvae) on French mud ﬂats found a reduction in snails’ crawling activity (and growth) in response to light. The impact of light included reduced survival rates, as this stressor entailed enhanced predation rates on the snails by their visual predators. The inﬂuence of light pollution on a more mobile invertebrate, the burrowing crab Neohelica granulata has been reported in saltmarsh habitats from Argentina (Nuñez et al., 2021a,b). Using tethering and cage experiments, these authors showed that juvenile crab survival decreased by 30–60% when exposed to light pollution, while the abundance of adult crabs increased up to ﬁve times (Nuñez et al., 2021a). Light exposure also reduced crab feeding rates and increased burrow maintenance (Nuñez et al., 2021a). The eﬀects of light pollution on the activity and spatial distribution (aggregation) of this key bioturbator species were also found to be site (context) dependent (Nuñez et al., 2021b; Quiñones- Llópiz et al., 2021). Additional studies have addressed the inﬂuence of light pollution on ﬁsh living in borderline habitats, deﬁned here as shallow subtidals closely connected to the intertidal. Since these are not tidal habitats, the studies below illustrate impacts but are not meant to be a comprehensive list (see Bassi et al., 2022 for studies on light pollution and ﬁsh). Becker et al. (2013) found that shallow water ﬁsh diversity in a South African estuary changed in the presence of light pollution: Large predatory and small shoaling ﬁsh aggregated near light sources at night. Given that they were primarily visual predators, these authors suggested that this would enhance feeding rates on nearby prey. Likewise, Bolton et al. (2017) studied ﬁsh predation and behavior under a wharf in Australia and found that light pollution reduced ﬁsh abundance but increased their predation rates on sessile invertebrates. Also in Australia, but in coral Frontiers in Ecology and Evolution Rocky shores and related hard-bottom species The second main impact of light pollution is upon turtle hatchling success (Witherington and Martin, 2000; Lorne and Salmon, 2007). Although light pollution remains detrimental, the precise eﬀects are less consistent than those described above. Light pollution disrupts orientation and/or local survival of small hatchlings emerging in sandy beaches of Australia, Greece and Florida, United States (e.g., Witherington and Bjorndal, 1991; Salmon et al., 1995; Lorne and Salmon, 2007). For example, in loggerhead hatchlings exposed to this stressor in Greece, nearly 50% were led astray from a direct path to the water (Dimitriadis et al., 2018). Such eﬀects are mitigated by the distance to light sources or the presence of physical shields (Pendoley and Kamrowski, 2016) or by the inﬂuence of full moons (Tuxbury and Salmon, These studies (12) have used a wide range of light pollution intensities (0.5–490 lux), mainly on experimental studies conducted in laboratory settings. These studies were preceded by seminal articles focusing on the inﬂuence of natural light levels (e.g., moon cycles) on the establishment of barnacles (e.g., Thorson, 1964; Crisp and Ritz, 1973). They were followed by recent studies addressing the diﬀerential inﬂuence of light pollution on the settlement of acorn barnacles (Semibalanus balanoides) in Atlantic Canada (Lynn et al., 2021b) and Jellhius cirratus and Notochthamalus scabrosus in northern Chile (Manríquez et al., 2021). In both studies, early settling stages (cyprids) were only mildly aﬀected by light pollution, whereas Frontiers in Ecology and Evolution 06 frontiersin.org 10.3389/fevo.2022.980776 10.3389/fevo.2022.980776 Lynn and Quijón Lynn and Quijón colonies (Oro et al., 2005) or on seabird collisions with light- emitting structures (buildings, platforms or wharves; Jones and Francis, 2003). Reports on this stressor’s eﬀects directly linked to the intertidal zone are more limited (ten, primarily ﬁeld- oriented are listed here), and for shorebirds they stem from the study of foraging at daylight vs. night hours (e.g., Rojas et al., 1999). For example, a study of the common redshank (Tringa tetanus) in estuarine ﬂats in Scotland showed that birds exposed to light pollution turned to sight-based foraging and fed well into the night hours (Dwyer et al., 2013). Similarly, among six shorebirds in Portugal, visual foragers used light pollution to their advantage and ended up feeding longer and heavier in lit areas (Santos et al., 2010). Tactile foragers faced a disadvantage and where feasible they switched to visual foraging. Simons et al. Estuarine and vegetated bottoms used by shorebirds and other species Tidal ﬂats and vegetated bottoms are used as stopovers by many shorebirds (Murray and Fuller, 2015) which become exposed to the same light pollution sources outlined above. Still, most light-related studies continue to focus on urban skylights altering the orientation of migratory birds (Cabrera- Cruz et al., 2018), on the mortality of seabirds at or near Frontiers in Ecology and Evolution 07 frontiersin.org 10.3389/fevo.2022.980776 Lynn and Quijón light intensities are needed to allow more meaningful comparisons among experiments. The use of lux units has allowed studies across species and habitats in intertidal and terrestrial habitats (e.g., Grunst et al., 2022). However, species’ spectral detection ranges change among distinct types of organisms (Dominoni et al., 2020) and lux measurements do not necessarily capture those diﬀerences. Alternative units and methodologies further dissecting light characteristics and species perception ranges need to be consistently applied to describe this stressor’s impacts (Gaston et al., 2012). Likewise, more dose-response analyses are needed to better understand the reaction of species to current and predicted light-stress scenarios (see Quintanilla-Ahumada et al., 2022). At the landscape scale, the use of remote sensing tools is growing in popularity and applicability (Hu et al., 2018). However, as in other habitats (e.g., Raap et al., 2017) and disciplines, a combination of rigorous exploratory and experimental approaches is likely to provide the best prospects for understanding how species respond to light pollution. With regards to the interplay between light pollution and other natural and anthropogenic stressors, it remains to be seen whether additive, non-additive or antagonistic eﬀects predominate in these interactions. Their nature will likely dictate the impact and management of light pollution in the decades ahead. reefs habitats, Fobert et al. (2019) found that the reproductive success (egg hatching) of a clownﬁsh (Amphiprion ocellaris) was reduced under light pollution. Another study from coral reefs in the French Polynesia examined the long-term eﬀects of light pollution on juvenile anemoneﬁsh (Amphiprion chrysopterus; Schligler et al., 2021) and found that this stressor reduced ﬁsh survival and growth. While light pollution has positive eﬀects on some species, these beneﬁts often extend across trophic levels to their own predators, a balance (or unbalance) that warrants further study. Acknowledgments We thank the handling editor and two reviewers for their comments and feedback on an earlier version of the manuscript. Funding This work was supported by the Natural Sciences and Engineering Research Council, Canada (NSERC) through Discovery grants (PQ) and CGS-D scholarships (KDL). Frontiers in Ecology and Evolution Commonalities, gaps, venues, and co-occurring stressors To date, a large majority of the reported eﬀects of light pollution is negative. To some degree, this is related to the targets of most of these studies, which have included species with marked circadian rhythms (e.g., talitrid amphipods), nocturnal habits (e.g., foraging shorebirds), or tradeoﬀs involving visual predators (e.g., rocky shore snails). The severity of these eﬀects ranges widely, from mild responses to the complete suppression or disruption of animal activity. From the literature reviewed, a ﬁrst important gap has been identiﬁed already: The limited number of studies focusing explicitly on community-level eﬀects, which to date include only Garratt et al. (2019) and Orlando et al. (2020). A second gap is the virtual absence of studies explicitly assessing the response of intertidal organisms to the inﬂuence of skyglow, or the brightening of the night sky resulting from the reﬂection and large-scale scattering of light emissions (see Kyba et al., 2015). Skyglow is often mentioned, but with one exception (Torres et al., 2022), most studies continue to focus on discrete light emissions (either direct sources or their reﬂection). A third outstanding gap is the virtual absence of studies addressing the combined eﬀects of light pollution and other stressors (see Miller et al., 2017 for examples in terrestrial insects). Exceptions include a few studies referred above: Orvain and Sauriau (2002) and Garratt et al. (2019) found that the eﬀects of light pollution on community biomass and mud snails, respectively, were modulated by the availability of organic matter in the habitat. Similarly, the inﬂuence of light pollution on dogwhelk foraging (Tidau et al., 2022) and sea turtle hatchlings (a few studies) interacted with natural moonlight cycles. These studies mark the beginning of the examination of potential synergistical eﬀects (see Halfwerk and Slabbekoorn, 2015) between light pollution and other natural or anthropogenic stressors and should likely escalate in complexity as new research becomes available. Author contributions KDL and PQ conceived and conducted the review of existing literature and drafted and edited the full manuscript. Both authors contributed to the article and approved the submitted version. References (2019). Artiﬁcial light at night causes reproductive failure in clownﬁsh. Biol. Lett. 15:20190272. doi: 10.1098/rsbl.2019.0272 Bird, B. L., Branch, L. C., and Miller, D. L. (2004). Eﬀects of coastal lighting on foraging behavior of beach mice. Conserv. Biol. 18, 1435–1439. doi: 10.1111/j. 1523-1739.2004.00349.x Garratt, M. J., Jenkins, S. R., and Davies, T. W. (2019). Mapping the consequences of artiﬁcial light at night for intertidal ecosystems. Sci. Total Environ. 691, 760–768. doi: 10.1016/j.scitotenv.2019.07.156 Bolton, D., Mayer-Pinto, M., Clark, G. F., Daﬀorn, K. A., Brassil, W. A., Becker, A., et al. (2017). Coastal urban lighting has ecological consequences for multiple trophic levels under the sea. Sci. Total Environ. 576, 1–9. doi: 10.1016/j.scitotenv. 2016.10.037 Gaston, K. J., Davies, T. W., Bennie, J., and Hopkins, J. (2012). Reducing the ecological consequences of night-time light pollution: Options and developments. J. Appl. Ecol. 49, 1256–1266. doi: 10.1111/j.1365-2664.2012.02 212.x Bourgeois, S., Gilot-Fromont, E., Viallefont, A., Boussamba, F., and Deem, S. L. (2009). Inﬂuence of artiﬁcial lights, logs, and erosion on leatherback sea turtle hatchling orientation at Pongara National Park, Gabon. Biol. Conserv. 142, 85–93. doi: 10.1016/j.biocon.2008.09.028 Gonzalez, S. A., Yanez-Navia, K., and Munoz, M. (2014). Eﬀect of coastal urbanization on sandy beach coleoptera Phaleria maculata (Kulzer, 1959) in northern Chile. Mar. Pollut. Bull. 83, 265–274. doi: 10.1016/j.marpolbul.2014.03. 042 Bregazzi, P. K., and Naylor, E. (1972). The locomotor activity rhythm of Talitrus saltator (Montagu) (Crustacea, Amphipoda). J. Exp. Biol. 57, 375–391. doi: 10. 1242/jeb.57.2.375 Griﬀen, B. D., Belgrad, B. A., Cannizzo, Z. J., Knotts, E. R., and Hancock, E. R. (2016). Rethinking our approach to multiple stressor studies in marine environments. Mar. Ecol. Prog. Ser. 543, 273–281. doi: 10.3354/meps1 1595 Brei, M., Perez-Barahona, A., and Strobl, E. (2016). Environmental pollution and biodiversity: Light pollution and sea turtles in the Caribbean. J. Environ. Econ. Manage. 77, 95–116. Cabrera-Cruz, S., Smolinsky, J. A., and Buler, J. J. (2018). Light pollution is greatest within migration passage areas for nocturnally-migrating birds around the world. Sci. Rep. 8:3261. doi: 10.1038/s41598-018-21577-6 Grunst, M. L., Grunst, A. S., Pinxten, R., Eens, G., and Eens, M. (2022). An experimental approach to investigating eﬀects of artiﬁcial light at night on free- ranging animals: Implementation, results, and directions for future research. J. Vis. Exp. e63381. ∗∗volQ. doi: 10.3791/63381 Cox, D. T. C., Sánchez de Miguel, A., Bennie, J., Dzurjak, S. A., and Gaston, K. J. (2022). Majority of artiﬁcially lit earth surface associated with the non-urban population. Sci. References Amano, S. (1986). Larval release in response to a light signal by the intertidal sponge Halichondria panicea. Biol. Bull. 171, 371–378. doi: 10.2307/1541679 Dimitriadis, C., Fournari-Konstantinidou, I., Sourbes, L., Koutsoubas, D., and Mazaris, A. D. (2018). Reduction of sea turtle population recruitment caused by nightlight: Evidence from the Mediterranean region. Ocean Coast. Manage. 153, 108–115. doi: 10.1016/j.ocecoaman.2017.12.013 Bassi, A., Love, O. P., Cooke, S. J., Warriner, T. R., Harris, C. M., and Madliger, C. L. (2022). Eﬀects of artiﬁcial light at night on ﬁshes: A synthesis with future research priorities. Fish Fish. 23, 631–647. doi: 10.1111/faf.12638 Dominoni, D. M., Halfwerk, W., Baird, E., Buxton, R. T., Fernández-Juricic, E., Fristrup, K. M., et al. (2020). Why conservation biology can beneﬁt from sensory ecology. Nat. Ecol. Evol. 4, 502–511. doi: 10.1038/s41559-020- 1135-4 Baz, E.-Z., Hussein, A. A. A., Vreeker, E. M. T., Soliman, M. F. M., Tadros, M. M., El-Shenawy, N. S., et al. (2022). Consequences of artiﬁcial light at night on behavior, reproduction, and development of Lymnaea stagnalis. Environ. Pollut. 307:119507. doi: 10.1016/j.envpol.2022.119507 Duarte, C., Quintanilla-Ahumada, D., Anguita, C., Manriquez, P. H., Widdicombe, S., Pulgar, J., et al. (2019). Artiﬁcial light pollution at night (ALAN) disrupts the distribution and circadian rhythm of a sandy beach isopod. Environ. Pollut. 248, 565–573. doi: 10.1016/j.envpol.2019.02.037 Becker, A., Whitﬁeld, A. K., Cowley, P. D., Jarnegren, J., and Naesje, T. F. (2013). Potential eﬀects of artiﬁcial light associated with anthropogenic infrastructure on the abundance and foraging behaviour of estuary-associated ﬁshes. J. Appl. Ecol. 50, 43–50. doi: 10.1111/1365-2664.12024 Dwyer, R. G., Bearhop, S., Campbell, H. A., and Bryant, D. M. (2013). Shedding light on light: Beneﬁts of anthropogenic illumination to a nocturnally foraging shorebird. J. Anim. Ecol. 82, 478–485. doi: 10.1111/1365-2656. 12012 Bellard, C., Bertelsmeier, C., Leadley, P., Thuiller, W., and Courchamp, F. (2012). Impacts of climate change on the future of biodiversity. Ecol. Lett. 15, 365–377. doi: 10.1111/j.1461-0248.2011.01736.x Fanini, L., Hughes, L. E., Springthorpe, R., Tosetto, L., and Lowry, J. K. (2016). Surface activity patterns of macrofauna on pocket, tidal beaches: Insights into the role of wrack and artiﬁcial lighting. Reg. Stud. Mar. Sci. 7, 63–71. doi: 10.1016/j. rsma.2016.05.007 Berry, M., Booth, D. T., and Limpus, C. J. (2013). Artiﬁcial lighting and disrupted sea-ﬁnding behaviour in hatchling loggerhead turtles (Caretta caretta) on the Woongarra coast, southeast Queensland, Australia. Aust. J. Zool. 61, 137–145. doi: 10.1071/ZO13028 Fobert, E. K., Burke Da Silva, K., and Swearer, S. E. 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. Research on light pollution has not been free of limitations. Experimental approaches are well represented, but a standardization of methodologies and more consistent 08 frontiersin.org 10.3389/fevo.2022.980776 10.3389/fevo.2022.980776 Lynn and Quijón Lynn and Quijón Publisher’s note 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. All claims expressed in this article are solely those of the authors and do not necessarily represent those of their aﬃliated References L., Limpus, C., Pendoley, K., and Hamann, M. (2014). Inﬂuence of industrial light pollution on the sea-ﬁnding behaviour of ﬂatback turtle hatchlings. Wildl. Res. 41, 421–434. doi: 10.1071/WR14155 Pendoley, K., and Kamrowski, R. L. (2016). Sea-ﬁnding in marine turtle hatchlings: What is an appropriate exclusion zone to limit impacts of industrial light at night? J. Nat. Conserv. 30, 1–11. doi: 10.1016/j.jnc.2015.12.005 Kyba, C. M., Tong, K. P., Bennie, J., Birriel, I., Birriel, J. J., Cool, A., et al. (2015). Worldwide variations in artiﬁcial skyglow. Sci. Rep. 5:8409. doi: 10.1038/srep08409 Pulgar, J., Zeballos, D., Vargas, J., Aldana, M., Manriquez, P., Manriquez, K., et al. (2019). Endogenous cycles, activity patterns, and energy expenditure of an intertidal ﬁsh is modiﬁed by artiﬁcial light at night (ALAN). Environ. Pollut. 244, 361–366. doi: 10.1016/j.envpol.2018.10.063 Legendre, P., and Legendre, L. (1998). Variation partitioning of species data matrices: Estimation and comparison of fractions. Ecology 87, 2614–2625. doi: 10.1890/0012-9658(2006)87[2614:VPOSDM]2.0.CO;2 Quiñones-Llópiz, J. D., Ribeiro, P. D., Luppi, T. A., Chiaradia, N. M., and Nuñez, J. D. (2021). Artiﬁcial light at night (ALAN) mediates transient spatial aggregation of an ecosystem engineer, the crab Neohelice granulata (Dana, 1851) (Decapoda: Brachyura: Varunidae), under diﬀerent ecological contexts. J. Crustac. Biol. 41:ruab060. doi: 10.1093/jcbiol/ruab060 Longcore, T., and Rich, C. (2004). Ecological light pollution. Front. Ecol. Environ. 2, 191–198. doi: 10.1890/1540-9295(2004)002[0191:ELP]2.0.CO;2 Lorne, J. K., and Salmon, M. (2007). Eﬀects of exposure to artiﬁcial lighting on orientation of hatchling sea turtles on the beach and in the ocean. Endang. Species Res. 3, 23–30. doi: 10.3354/esr003023 Quintanilla-Ahumada, D., Quijón, P. A., Manríquez, P. H., Pulgar, J., García- Huidobro, M. R., Miranda, C., et al. (2022). Artiﬁcial light at night (ALAN) causes variable dose-responses in a sandy beach isopod. Environ. Sci. Pollut. Res. Int. 29, 35977–35985. doi: 10.1007/s11356-021-17344-2 Luarte, T., Bonta, C. C., Silva-Rodriguez, E. A., Quijón, P. A., Miranda, C., Farias, A. A., et al. (2016). Light pollution reduces activity, food consumption, and growth rates in a sandy beach invertebrate. Environ. Pollut. 218, 1147–1153. doi: 10.1016/j.envpol.2016.08.068 Quintanilla-Ahumada, D., Quijón, P. A., Pulgar, J., Manriquez, P. H., Roberto Garcia-Huidobro, M., and Duarte, C. (2021). Exposure to artiﬁcial light at night (ALAN) alters RNA:DNA in a sandy beach coleopteran insect. Mar. Pollut. Bull. 165:112132. doi: 10.1016/j.marpolbul.2021.112132 Lynn, K. D., Quintanilla-Ahumada, D., Anguita, C., Widdicombe, S., Pulgar, J., Manriquez, P. H., et al. (2021a). References Artiﬁcial light at night alters the activity and feeding behaviour of sandy beach amphipods and pose a threat to their ecological role in Atlantic Canada. Sci. Total Environ. 780:146568. doi: 10.1016/j.scitotenv. 2021.146568 Raap, T., Pinxten, R., and Eens, M. (2017). Rigorous ﬁeld experiments are essential to understand the genuine severity of light pollution and to identify possible solutions. Glob. Change Biol. 23, 5024–5026. doi: 10.1111/gcb.13843 Lynn, K. D., Tummon Flynn, P., Manríquez, K., Manríquez, P. H., Pulgar, J., Duarte, C., et al. (2021b). Artiﬁcial light at night alters the settlement of acorn barnacles on a man-made habitat in Atlantic Canada. Mar. Pollut. Bull. 163:111928. doi: 10.1016/j.marpolbul.2020.111928 Rojas, L. M., McNeil, R., Cabana, T., and Lachapelle, P. (1999). Diurnal and nocturnal visual capabilities in shorebirds as a function of their feeding strategies. Brain Behav. Evol. 53, 29–43. doi: 10.1159/000006580 Saigusa, M., and Kawagoye, O. (1997). Circatidal rhythm of an intertidal crab Hemigrapsus sanguineus: Synchrony with unequal tide height and involvement of a light-response mechanism. Mar. Biol. 129, 87–96. doi: 10.1007/s002270050149 Manríquez, K., Quijón, P. A., Manríquez, P. H., Miranda, C., Pulgar, J., Quintanilla-Ahumada, D., et al. (2021). Artiﬁcial light at night (ALAN) negatively eﬀects the settlement success of two prominent intertidal barnacles in the southeast Paciﬁc. Mar. Pollut. Bull. 168:112416. doi: 10.1016/j.marpolbul.2021. 112416 Salmon, M., Tolbert, M. G., Paynter, D. P., Goﬀ, M., and Reiners, R. (1995). Behavior of loggerhead sea turtles on an urban beach. II. Hatchling orientation. J. Herpetol. 29, 568–576. doi: 10.2307/1564740 Manríquez, P. H., Jara, M. E., Díaz, M. I., Quijón, P. A., Widdicombe, S., Pulgar, J., et al. (2019). Artiﬁcial light pollution inﬂuences behavioural and physiological traits in a keystone predator species Concholepas concholepas. Sci. Total Environ. 661, 543–552. doi: 10.1016/j.scitotenv.2019.01.157 Sánchez de Miguel, A., Bennie, J., Rosenfeld, E., Dzurjak, S., and Gaston, K. J. (2021). First estimation of global trends in nocturnal power emissions reveals acceleration of light pollution. Remote Sens. 13:3311. doi: 10.3390/rs13163311 Mazor, T., Levin, N., Possingham, H. P., Levy, Y., Rocchini, D., Richardson, A. K., et al. (2013). Can satellite-based night lights be used for conservation? The case of nesting sea-turtles in the Mediterranean. Biol. Conserv. 159, 63–72. doi: 10.1016/j.biocon.2012.11.004 Santos, C. D., Miranda, A. C., Granadiero, J. P., Lourenco, P. M., Saraiva, S., and Palmeirim, J. M. (2010). Eﬀects of artiﬁcial illumination on the nocturnal foraging of waders. Acta Oecol. 36, 166–172. doi: 10.1016/j.actao.2009.11.008 Scapini, F., Rossano, C., Marchetti, G. References Total Environ. 841:156782. doi: 10.1016/j.scitotenv.2022.156782 Gunderson, A. R., Armstrong, E. J., and Stillman, J. H. (2016). Multiple stressors in a changing world: The need for an improved perspective on physiological responses to the dynamic marine environment. Annu. Rev. Mar. Sci. 8, 357–357. doi: 10.1146/annurev-marine-122414-033953 Crisp, D. J., and Ritz, D. A. (1973). Responses of cirripede larvae to light. I. Experiments with white light. Mar. Biol. 23, 327–335. doi: 10.1007/BF00389340 Halfwerk, H., and Slabbekoorn, H. (2015). Pollution going multimodal: The complex impact of human-altered sensory environment on animal perception and performance. Biol. Lett. 11:20141051. doi: 10.1098/rsbl.2014.1051 Davies, T. W., and Smyth, T. (2017). Why artiﬁcial light at night should be a focus for global change research in the 21st century. Glob. Change Biol. 24, 872–882. doi: 10.1111/gcb.13927 Halpern, B. S., Walbridge, S., Selkoe, K. A., Kappel, C. V., Micheli, F., D’Agrosa, C., et al. (2008). A global map of human impact on marine ecosystems. Science 319, 948–952. doi: 10.1126/science.1149345 Dhouha, B. A., Raja, J., Elfed, M., and Karima, N. A. (2018). Eﬀect of light intensity on the locomotor activity rhythm of Talitrus saltator (Montagu 1808) from Korba beach. Biol. Rhythm Res. 49, 735–745. doi: 10.1080/09291016.2017. 1413764 Harewood, A., and Horrocks, J. (2008). Impacts of coastal development on hawksbill hatchling survival and swimming success during the initial oﬀshore migration. Biol. Conserv. 141, 394–401. doi: 10.1016/j.biocon.2007. 10.017 Dhouha, B. A., Raja, J., Elfed, M., and Karima, N. A. (2019). Locomotor activity rhythm of two sympatric species Talitrus saltator (Amphipoda) and Tylos europaeus (Oniscidea) from Tunisian beaches under continuous light. Biol. Rhythm Res. 50, 180–191. doi: 10.1080/09291016.2018.1433471 Hu, Z., Hu, H., and Huang, Y. (2018). Association between nighttime artiﬁcial light pollution and sea turtle nest density along Florida coast: A geospatial study using VIIRS remote sensing data. Environ. Pollut. 239, 30–42. doi: 10.1016/j. envpol.2018.04.021 Dicks, L. V., Walsh, J. C., and Sutherland, W. J. (2014). Organizing evidence for environmental management decisions: A ‘4S’ hierarchy. Trends Ecol. Evol. 29, 607–613. doi: 10.1016/j.tree.2014.09.004 Frontiers in Ecology and Evolution 09 frontiersin.org Lynn and Quijón Lynn and Quijón 10.3389/fevo.2022.980776 Jones, J., and Francis, C. M. (2003). The eﬀects of light characteristics on avian mortality at lighthouses. J. Avian Biol. 34, 328–333. doi: 10.1111/j.0908-8857.2003. 03183.x Orvain, O., and Sauriau, P. G. (2002). Environmental and behavioural factors aﬀecting activity in the intertidal gastropod Hydrobia ulvae. J. Exp. Mar. Biol. Ecol. 272, 191–216. doi: 10.1016/S0022-0981(02)00130-2 Kamrowski, R. References M., and Morgan, E. (2005). The role of biological clock in sun compass orientation of free-running individual of Talitrus saltator. Anim. Behav. 69, 835–843. doi: 10.1016/j.anbehav.2004.08.008 McLachlan, A., Defeo, O., Jaramillo, E., and Short, A. D. (2013). Sandy beaches conservation and recreation: Guidelines for optimizing management strategies for multipurpose use. Ocean Coast. Manage. 71, 256–268. doi: 10.1016/j.ocecoaman. 2012.10.005 Schligler, J., Cortese, D., Beldade, R., Swearer, S. E., and Mills, S. C. (2021). Long- term exposure to artiﬁcial light at night in the wild decreases survival and growth of a coral reef ﬁsh. Proc. R. Soc. B 288:20210454. doi: 10.1098/rspb.2021.0454 McMahon, O., Smyth, T., and Davies, T. W. (2022). Broad spectrum artiﬁcial light at night increases the conspicuousness of camouﬂaged prey. J. Appl. Ecol. 59, 1324–1333. doi: 10.1111/1365-2664.14146 Shick, J. M., and Dykens, J. A. (1984). Photobiology of the symbiotic sea anemone Anthopleura elegantissima: Photosynthesis, respiration, and behaviour under intertidal conditions. Biol. Bull. 166, 608–619. doi: 10.2307/1541166 Miller, C. R., Barton, B. T., Zhu, L., Radeloﬀ, V. C., Oliver, K. M., Harmon, J. P., et al. (2017). Combined eﬀects of night warming and light pollution on predator–prey interactions. Proc. R. Soc. B 284:20171195. doi: 10.1098/rspb.2017. 1195 Silva, E., Marco, A., da Graça, J., Pérez, H., Abella, E., Patino-Martinez, J., et al. (2017). Light pollution aﬀects nesting behavior of loggerhead turtles and predation risk of nests and hatchlings. J. Photochem. Photobiol. B Biol. 173, 240–249. doi: 10.1016/j.jphotobiol.2017.06.006 Murray, N. J., and Fuller, R. A. (2015). Protecting stopover habitat for migratory shorebirds in East Asia. J. Ornithol. 156, S217–S225. doi: 10.1007/s10336-015- 1225-2 Simons, A. L., Martin, K. L. M., and Longcore, T. (2022). Determining the eﬀects of artiﬁcial light at night on the distributions of western snowy plovers (Charadrius nivosus nivosus) and California grunion (Leuresthes tenuis) in southern California. J. Coast. Res. 38, 302–309. doi: 10.2112/JCOASTRES-D-21-00107.1 Nuñez, J. D., Bas, C., Peréz García, M., Ocampo, E. H., Ribeiro, P. D., and Luppi, T. A. (2021a). Artiﬁcial light at night increases the predatory pressure in a salt marsh keystone species. Mar. Environ. Res. 167:105285. doi: 10.1016/j.marenvres. 2021.105285 Strobl, E., Brei, M., and Perez-Barahona, A. (2016). Environmental pollution and biodiversity: Light pollution and sea turtles in the Caribbean. J. Environ. Econ. Manage. 77, 95–116. doi: 10.1016/j.jeem.2016.02.003 Nuñez, J. D., Sbragaglia, V., Ribeiro, P. D., Spivak, E., and Luppi, T. A. (2021b). Lynn and Quijón Tuxbury, S. M., and Salmon, M. (2005). Competitive interactions between artiﬁcial lighting and natural cues during seaﬁnding by hatchling marine turtles. Biol. Conserv. 121, 311–316. doi: 10.1016/j.biocon.2004.04.022 Underwood, C. N., Davies, T. W., and Queirós, A. M. (2017). Artiﬁcial light at night alters trophic interactions of intertidal invertebrates. J. Anim. Ecol. 86, 781–789. doi: 10.1111/1365-2656.12670 Vaz, S., Manes, S., Gama-Maia, D., Silveira, L., Mattos, G., Paiva, P. C., et al. (2020). Light pollution is the fastest growing potential threat to ﬁreﬂy conservation in the Atlantic Forest hotspot. Insect Conserv. Divers. 14, 211–224. doi: 10.1111/ icad.12481 Witherington, B. E., and Martin, R. E. (2000). Understanding, assessing, and resolving light pollution problems on sea turtle nesting beaches. 2nd ed. rev. Florida Marine Research Institute Technical Report TR-2 73. St. Petersburg, FL: Florida Fish and Wildlife Conservation Commission, Marine Research Institute. References The magnitude of behavioral responses to artiﬁcial light at night depends on the ecological context in a coastal marine ecosystem engineer. Mar. Environ. Res. 165:105238. doi: 10.1016/j.marenvres.2020.105238 Thorson, G. (1964). Light as an ecological factor in the dispersal and settlement of larvae of marine bottom invertebrates. Ophelia 1, 167–208. doi: 10.1080/ 00785326.1964.10416277 Orlando, L., Ortega, L., and Defeo, O. (2020). Urbanization eﬀects on sandy beach macrofauna along an estuarine gradient. Ecol. Indic. 111:106036. doi: 10. 1016/j.ecolind.2019.106036 Tidau, S., Whittle, J., Jenkins, S. R., and Davies, T. W. (2022). Artiﬁcial light at night reverses monthly foraging pattern under simulated moonlight. Biol. Lett. 18:20220110. doi: 10.1098/rsbl.2022.0110 Oro, D., de León, A., Minguez, E., and Furness, R. W. (2005). Estimating predation on breeding European storm-petrels (Hydrobates pelagicus) by yellow-legged gulls (Larus michahellis). J. Zool. 265, 421–429. doi: 10.1017/ S0952836905006515 Torres, D., Tidau, S., Jenkins, S., and Davies, T. (2022). Artiﬁcial skyglow disrupts celestial migration at night. Curr. Biol. 30, R677–R697. doi: 10.1016/j.cub. 2020.05.002 10 Frontiers in Ecology and Evolution frontiersin.org Lynn and Quijón 10.3389/fevo.2022.980776 10.3389/fevo.2022.980776 Witherington, B. E. (1992). Behavioural response of nesting sea turtles to artiﬁcial lighting. Herpetologica 48, 31–39. Witherington, B. E. (1992). Behavioural response of nesting sea turtles to artiﬁcial lighting. Herpetologica 48, 31–39. Witherington, B. E., and Bjorndal, K. A. (1991). Inﬂuences of artiﬁcial lighting on the seaward orientation of hatchling loggerhead turtles Caretta caretta. Biol. Conserv. 55, 139–149. doi: 10.1016/0006-3207(91)90 053-C Witherington, B. E., and Martin, R. E. (2000). Understanding, assessing, and resolving light pollution problems on sea turtle nesting beaches. 2nd ed. rev. Florida Marine Research Institute Technical Report TR-2 73. St. Petersburg, FL: Florida Fish and Wildlife Conservation Commission, Marine Research Institute. Frontiers in Ecology and Evolution frontiersin.org 11
https://openalex.org/W4393126494	https://tribology.khnu.km.ua/index.php/ProbTrib/article/download/941/1394	Russian	null	Tribotechnical coatings	Problemi tribologìï	2,024	cc-by	2,954	Introduction The problem of durability and reliability of moving joints of machine parts stands out against the background of general technical and industrial achievements. The unshakable existence of the problem is based on the phenomenon of friction, which is associated with one of the most pressing situations in modern technology - wear and tear of machine parts and structures. Wear, as a universal phenomenon of degradation due to friction, limits the capabilities and shortens the service life of modern technical systems. Hence the importance and necessity of considering issues aimed at reducing friction forces and increasing wear resistance. Objective The purpose of the work is to generalize the theoretical and applied results of studies of tribological resistance of detonation coatings of the Nb-Zr-V-Si-C-MgC2 system under constant load conditions in a sliding velocity field. Tribotechnical coatings A.O. Zemlyanoy1, S.S. Bys2, V.V. Shchepetov1, S.D. Kharchenko1, O.V. Kharhenko3 1General Energy Institute of National Academy of Sciences of Ukraine, Ukraine 2Khmelnytskyi National University, Ukraine 3National Aviation University, Ukraine E-mail:serhiibys@gmail.com Received: 15 February 2024: Revised 5 March 2024: Accept 14 March 2024 Received: 15 February 2024: Revised 5 March 2024: Accept 14 March 2024 Copyright © 2024 A.O. Zemlyanoy, S.S. Bys, V.V. Shchepetov, S.D. Kharchenko, O.V. Kharhenko.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. Abstract Wear and tear limits the possibilities and shortens the operational life of modern technical systems. Therefore, the importance and necessity of consideration of issues aimed at reducing frictional forces and increasing wear resistance cannot be doubted. The paper summarizes the theoretical and applied results of triboresistance studies of detonation coatings of the Nb-Zr-V-Si-C-MgC2 system under conditions of constant loading in the field of sliding velocities. It has been established that the ratio of the quality of the components that make up the surface modified structures changes. It is noted that at the initial test speeds the presence of lower metal carbides that are part of the coating dominates, with an increase in speed under the current load due to solid- phase and diffusion processes, higher ones are formed in the graphite matrix carbides with enhanced thermodynamic properties. Key words: detonation coatings, wear intensity, structural-phase composition, graphitization. Problems of Tribology, V. 29, No 1/111-2024, 61-65 Problems of Tribology Website: http://tribology.khnu.km.ua/index.php/ProbTrib E-mail: tribosenator@gmail.com DOI: https://doi.org/10.31891/2079-1372-2024-111-1-61-65 Problems of Tribology, V. 29, No 1/111-2024, 61-65 Problems of Tribology Website: http://tribology.khnu.km.ua/index.php/ProbTrib E-mail: tribosenator@gmail.com DOI: https://doi.org/10.31891/2079-1372-2024-111-1-61-65 Problems of Tribology, V. 29, No 1/111-2024, 61-65 Research results The main factors determining the course of friction and wear processes are external influences, which determine the degree and gradients of elastoplastic deformation, temperature, level of activation, a number of derivative phenomena and ultimately determine the leading type of wear. Based on the test results, averaged graphical dependences of wear intensity and friction coefficients in the sliding velocity field are presented at a constant load of 15 MPa (fig. 1). With a monotonous increase in sliding speed under constant load conditions throughout the entire operating test range, the lowest values of wear intensity (curve 1) and friction coefficients (curve 1'), corresponding to normal mechanochemical wear, are found in coatings of the Nb-Zr-V-Si-C-MgC2 system. Fig. 1. Dependence of wear intensity (1, 2, 3) and friction coefficient (1', 2', 3') on the sliding speed of coatings: 1, 1' – Nb-Zr-V-Si-C-MgC2 systems; 2, 2' – VK15 (WC-Co); 3, 3' – based on nichrome (Ni-Cr-Al-B). At P = 15 MPa. Fig. 1. Dependence of wear intensity (1, 2, 3) and friction coefficient (1', 2', 3') on the sliding speed of coatings: 1, 1' – Nb-Zr-V-Si-C-MgC2 systems; 2, 2' – VK15 (WC-Co); 3, 3' – based on nichrome (Ni-Cr-Al-B). At P = 15 MPa. Studies of the composition and structure of coatings have made it possible to establish that their structure consists mainly of a finely dispersed mixture, in which strengthening formations are presented in the form of carbide and silicide phases and intermetallic compounds, which are distributed in a solid solution matrix and, thus, represent a structure close to structure of dispersion-strengthened materials. The nature of the interaction of the coating components was determined using the Lanscan program, and it was found that the distribution spectra of Nb, Zr and V along the scan line correlate with the phases of silicon and carbon, which confirms their interaction with the production of the corresponding carbides and silicides, and also determines the possibility of the formation of both solid solutions and ternary compounds. Analysis of oxygen scanning lines with elements included in the nanocoating showed their phase alignment, which indicates their chemical interaction, causing passivation of the working surface due to the formation of secondary oxide structures. Materials and methods of research The preparation of the powder mixture was carried out by mechanochemical synthesis using a laboratory attriter of the “IES-1-0.5” type. To prevent sticking of powders to the chamber walls and to optimize the spheroidization process, an anti-adhesive additive in the form of zinc stearite was added to the mixture. The study of the features of friction surfaces in which activation processes occur, affecting the intensity of mechanochemical reactions, was carried out using a complex method of physical and chemical analysis, including metallography (optical microscope "Neofot-32"), durometric analysis (hardness tester M-400 from LECO), scanning electron microscopy (scanning electron microscope JSM-840), X-ray structural phase analysis (diffractometer DRON-UM1). Copyright © 2024 A.O. Zemlyanoy, S.S. Bys, V.V. Shchepetov, S.D. Kharchenko, O.V. Kharhenko.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. 62 Problems of Tribology The tribological properties of the coatings were assessed by friction of model samples according to the end pattern under conditions of distributed contact. The tests were carried out in continuous sliding mode at a load of 15 MPa, the thickness of the coatings after finishing was 0,750-0,80 mm, roughness Ra = 0,32-0,63. g g g Tests of detonation coatings were carried out according to a program in which, under similar conditions, coatings of the VK15 type and coatings dusted with alloyed nichrome powder were tested for comparison. Research results The contact zone, which directly forms the friction surface, is a thin plastically deformable layer, which, according to micro-X-ray spectral analysis performed on a MAP-3 device (probe diameter 1 μm), is a conglomerate of dispersed phases of complex and simple oxides and compounds of silicides and carbides components included in coating composition. contact area, reduce the specific load due to filling and smoothing surface microroughnesses, microcracks, and causes the fixation of particles graphite in microcavities of contact interfaces and ensures a reduction and stabilization of the friction coefficient, in addition, it helps to lower the temperature, level out and reduce deformation fields. The contact zone, which directly forms the friction surface, is a thin plastically deformable layer, which, according to micro-X-ray spectral analysis performed on a MAP-3 device (probe diameter 1 μm), is a conglomerate of dispersed phases of complex and simple oxides and compounds of silicides and carbides components included in coating composition. a b Fig. 2. Topography of friction surfaces coated with a graphite film P = 15 MPa: a - V=0,5 m/s; b - V=1,5 m/s b Fig. 2. Topography of friction surfaces coated with a graphite film P = 15 MPa: a - V=0,5 m/s; b - V=1,5 m/s In figure 3 shows an electron diffraction pattern from the surface of the contact layer. The presence of diffusion halos with textured maxima indicates that the structure corresponds to a directional orientation in the friction field. According to the authors, the pattern of its formation can be represented as a process of amorphization and mechanochemical alloying, including dispersion, grinding of a dispersoid with particles of oxides, silicides and their transformation under the influence of local temperatures and pressures into an oriented ultradisperse structure, which represents a substance with liquid-like properties. Fig. 3. Electron diffraction pattern from the surface of the contact layer at V=0,5m/s and Р=15 MPa. Fig. 3. Electron diffraction pattern from the surface of the contact layer at V=0,5m/s and Р=15 MPa. In figure 4 shows the qualitative distribution of chemical components over the working surface area. The results of physicochemical phase analysis of secondary structures in the form of surface films based on graphite made it possible, by analyzing the diffraction pattern, to identify carbide, silicide and oxide phases with a significant degree of probability. Research results Thus, the obtained research results, supplemented by X-ray diffraction analysis data, allow us to conclude that the initial coatings are an ultradisperse conglomerate of almost uniformly distributed particles of silicides NbSi, NbSi2, ZrSi, Zr2Si, VSi2, V5Si3 and extensive colonies of carbides NbC, Nb2C, ZrC, VC, V2C in a fine-grained matrix of solid solutions of Nb in β-Zr, β-Zr in Nb, V in Nb and Zr. Under conditions of structural activation under friction loading, the passivating role of secondary structures is played by surface films of oxides and, first of all, SiO2 and V2O5, which are characterized by increased density and adhesive strength, as well as high-quality compact oxides that form Nb and Zr. In addition, the formation of solid solutions of oxygen in niobium and zirconium occurs. Thus, according to X-ray diffraction analysis, a slight increase in the density and lattice period of zirconium has been established, which is, in our opinion, a consequence of the introduction of oxygen atoms into the octahedral pores of the hexagonal lattice and, as a result, the formation of a solid solution. However, the main factor that minimizes molecular adhesion interaction, preventing setting and destruction, is the formation of a surface film of structurally free α-graphite, the formation of which is caused by the thermal decomposition of magnesium carbide [1]. Moreover, the higher the temperature, the greater the amount of carbon converted into graphite and the more graphite is formed. The process is similar to graphitizing annealing. Thus, under friction loading, the running- in of surfaces whose composition includes magnesium carbides, in a certain sense, can be considered as a specific type of heat treatment, accompanied by graphitization. In figure 2 shows the topography of friction surfaces coated with a graphite film. In fig. 2.b (at V = 1,5 m/s and P = 15 MPa) the graphite film occupies almost the entire working surface, which helps to increase the actual Problems of Tribology 63 contact area, reduce the specific load due to filling and smoothing surface microroughnesses, microcracks, and causes the fixation of particles graphite in microcavities of contact interfaces and ensures a reduction and stabilization of the friction coefficient, in addition, it helps to lower the temperature, level out and reduce deformation fields. Research results At the same time, it was possible to establish the patterns of structural changes in the identified phases during their reorganization, which is accompanied under friction conditions by both the transformation of the structural-phase composition of secondary structures, caused by varying external influences (sliding speed, load, temperature), and the stability of the dynamic sequence of the processes of their destruction and formation during friction. Thus, the ratio of the quality of the components that make up the surface structures changes, but their general order (the dynamics of formation and destruction) remains constant. Thus, in the testing area up to 2,0 m/s, the dominant factor is the presence of low metal carbides included in the coating composition and the presence of their oxide reagents. With an increase in sliding speed under the influence of increased shear strains, temperatures and effective load, as a result of solid-phase tribochemical and diffusion processes in the graphite matrix, a balanced presence of higher carbides is formed, characterized by increased thermodynamic stability, which corresponds to the manifestation of the structural adaptability of materials during friction [2]. 64 Problems of Tribology Fig. 4. Qualitative distribution of chemical components over the working surface area. Fig. 4. Qualitative distribution of chemical components over the working surface area. The nature of the change in the friction coefficient (fig. 1, curve 1') is consistent with the established wear pattern. The decrease in its value and stability with increasing sliding speed confirm the high performance of the studied coatings of the Nb-Zr-V-Si-C-MgC2 system. In accordance with the presented test results, the developed niobium-based coatings meet high tribological characteristics. The possibility of increasing the operational reliability of coatings through the use of magnesium carbide in their composition, which promotes the formation of carbide graphite under friction loading, has been theoretically substantiated and experimentally confirmed. It should be noted that the presence of magnesium carbide is an effective means of regulating wear and ensuring service life through the formation of modified surface films based on carbide graphite, which, in cooperative mutual consistency with secondary structures, ensure a stable manifestation of minimizing the tribological properties of coatings. Based on this, the amount of magnesium carbide in the coating material may influence the stability of normal mechanochemical wear. Also tested under similar test conditions were VK15 coatings (fig. Research results 1, curve 2), which are the most widely used tribological material for wear protection of a large range of critical parts of various designs and purposes. However, under the conditions of these tests at sliding speeds of more than 1,7 m/s, the temperature factor begins to influence the tendency to reduce wear resistance. Coatings based on nichrome (fig. 1, curve 3) alloyed with aluminum and boron, starting from 0,4 m/s, are characterized by a monotonous increase in wear intensity with increasing speed. The structure of the coatings is represented by a nickel-based solid solution and a dispersed mixture of strengthening phases of nickel aluminides, chromium borides, and complex boride compounds. The microhardness was about 10,6 GPa. The protective functions of secondary structures representing heterophase oxide thin-film objects are suppressed by the development of plastic deformation with increasing sliding speed. Niobium-based coatings containing magnesium carbide and not containing expensive and scarce components meet environmental safety requirements and, due to their operational capabilities, have the prospect of widespread use for the production of competitive tribotechnical equipment. The most effective use of the studied coatings is to improve the quality of operation of friction units when strengthening and restoring moving joints of control mechanisms, hinges of guide surfaces, cams, sliding supports, lever parts, high-speed and heavily loaded units in which the use of traditional oils and lubricants is undesirable. It should be noted that the development of coatings containing modified films of carbide graphite, the results of their tests in various friction and wear modes will significantly expand the arsenal of achievements of modern tribological technology. The studied niobium-based coatings can be used to restore and strengthen worn parts using any technological methods using powder materials. 4. The test results form the basis for the development of composite coatings with improved tribological characteristics, which will expand the arsenal of achievements of modern tribological technology. In addition, the developed powders, which do not contain scarce and expensive components, can be used for restoration and strengthening by any technological methods using powder materials. Conclusions 1. Through theoretical premises and experimental studies, the optimal structural-phase composition of coatings of the Nb-Zr-V-Si-C-MgC2 system was realized. A high level of performance characteristics of coatings was noted, achieved by optimizing the chemical composition and spraying technology. 1. Through theoretical premises and experimental studies, the optimal structural-phase composition of coatings of the Nb-Zr-V-Si-C-MgC2 system was realized. A high level of performance characteristics of coatings was noted, achieved by optimizing the chemical composition and spraying technology. 2. Using modern methods of physicochemical analysis, the structural-phase composition of coatings was studied, the structure of the surface modified layer based on carbide graphite, which prevents molecular adhesion interaction and minimizes friction parameters, was determined. 2. Using modern methods of physicochemical analysis, the structural-phase composition of coatings was studied, the structure of the surface modified layer based on carbide graphite, which prevents molecular adhesion interaction and minimizes friction parameters, was determined. p 3. It has been established that the ratio of the quality of the components that make up the surface modified structures changes, it is noted that at the initial test speeds the presence of lower metal carbides that are part of the coating dominates, with an increase in speed under the current load due to solid-phase and diffusion processes, higher ones are formed in the graphite matrix carbides with enhanced thermodynamic properties. Problems of Tribology 65 1. Babak V.P., Fialko N.M., Shchepetov V.V., Gladky Ya.M., Bis S.S. Self-strengthening composite nanocrysts // Physico-chemical mechanics of materials. – 2023. – p. 37-43. y y p 2. Kostetsky B.I., Nosovsky I.G., Karaulov A.K. and others. Surface strength of materials during friction // Kiev: Tekhnika, 1976, 296 p. 1. Babak V.P., Fialko N.M., Shchepetov V.V., Gladky Ya.M., Bis S.S. Self-strengthening composite nanocrysts // Physico-chemical mechanics of materials. – 2023. – p. 37-43. 2. Kostetsky B.I., Nosovsky I.G., Karaulov A.K. and others. Surface strength of materials during friction // Kiev: Tekhnika, 1976, 296 p. References Зємляной А.О., Бись С.С., Щепетов В.В., Харченко С.Д., Харченко О.В. Триботехнічні покриття Зношування обмежує можливості та скорочує строки експлуатації сучасних технічних систем. Тому важливість та непохідність розгляду та вирішення питань зниження сил тертя та підвищення опру зношуванню не викликає жодного сумніву. В роботі зроблено узагальнення теоретичних та практичних результатів досліджень трибостійкості детонаційних покрить системи Nb-Zr-V-Si-C-MgC2 в умовах постійного навантаження ковзанням. Встановлено, що змінюється співвідношення якості компонентів, що входять до складу поверхнево модифікованих структур. Відмічено, що на початкових швидкостях випробування домінує наявність нижчих карбідів металів, які входять до складу покриття, при збільшенні швидкості під струмовим навантаженням за рахунок твердофазних і дифузійних процесів вищі утворюються в карбідах графітової матриці. з підвищеними термодинамічними властивостями. Ключові слова: детонаційні покриття, інтенсивність зношування, структурно-фазовий склад, графітизація Ключові слова: детонаційні покриття, інтенсивність зношування, структурно-фазовий склад, графітизація
https://openalex.org/W2902615143	https://www.dependability.ru/jour/article/download/294/502	Russian	null	Ensuring an efficient transportation infrastructure security system by means of solutions that enable detection of intrusions into protected areas	Nadëžnostʹ	2,018	cc-by	3,418	Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Резюме. Ввиду специфики работы транспорт сам по себе является потенциальным источником опасности. А если вторгнуться в его сферу незаконными, агрессивными действиями, опасность становится реальной, грозящей тяжелейшими последствиями. Статистика последних 10–15 лет показывает, что от 50 до 70% совершаемых терро- ристических актов связано с транспортом. Какими–то отдельными мерами обеспечить транспортную безопасность невозможно. К этой проблеме надо подходить комплексно и системно. От обеспечения транспортной безопасности существенно зависит нацио- нальная безопасность Российской Федерации. Федеральный закон «О транспортной безопасности» от 9 февраля 2007 г. № 16-ФЗ впервые в отечественной практике по- ставил вопрос об обеспечении безопасности всей транспортной отрасли Российской Федерации, установил правовые основы деятельности по обеспечению безопасности объектов транспортной инфраструктуры и транспортных средств от актов незаконного вмешательства, в том числе террористической направленности. Впервые предусматри- вается единый для всех видов транспорта системный подход к антитеррористической защите. Транспорт – весьма уязвимый объект для террористических атак. Имеются в виду сами транспортные средства, транспортные коммуникации, помещения вокзалов, транспортные средства с опасными грузами. Уязвимость транспорта связана с возмож- ностью повреждения средств сигнализации, устройств автоматики и связи, охрана кото- рых с учетом масштабов и протяженности, железных дорог РФ затруднена. Несмотря на проблемные вопросы и объективные трудности в законодательных вопросах при реали- зации мер по обеспечению транспортной безопасности, работники железных дорог РФ прилагают усилия для обеспечения защищенности объектов транспортной инфраструк- туры (ОТИ) и транспортных средств (ТС) от актов незаконного вмешательства. Незамед- лительно реагируя на другие вызовы и угрозы, обеспечивают надежное функционирова- ние работы транспортного комплекса, сохраняя тем самым спокойствие и безопасность наших граждан. В данной статье рассмотрены вопросы эффективной системы безопас- ности ОТИ. Большая роль отводится системам, позволяющим зафиксировать факт про- никновения посторонних лиц в охраняемую зону объекта. Наталья А. Кузьмина Ключевые слова: транспортная безопасность, акт незаконного вмешательства, объ- ект транспортной инфраструктуры, система безопасности, охрана объекта, нарушитель, датчики, извещатели. Формат цитирования: Кузьмина Н.А., Обеспечение эффективной системы безопасно- сти объектов транспортной инфраструктуры посредством систем, позволяющих зафик- сировать факт проникновения посторонних лиц в охраняемую зону // Надежность. 2018. Т. 18, № 4. С. 51-55. DOI: 10.21683/1729-2646-2018-18-4-51-55 Федеральный закон № 16-ФЗ «О транспортной безопасности» обязывает защищать ОТИ от актов не- законного вмешательства [2]. Несмотря на принимаемые меры субъектами транс- портной инфраструктуры и перевозчиками по повышению уровня защищенности транспортного комплекса, угроза совершения актов незаконного вмешательства (АНВ) остается. Сухая статистика констатирует, что к сожалению, пока полностью исключить возможность совершения АНВ не предоставляется возможным. УДК 656.2-027.45:343.326 Оригинальная статья DOI: 10.21683/1729-2646-2018-18-4-51-55 УДК 656.2-027.45:343.326 Оригинальная статья DOI: 10.21683/1729-2646-2018-18-4-51-55 Надежность, том 18, №4, 2018 Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия В этой связи первоочеред- ной задачей является максимально затруднить попытки совершения АНВ в деятельность объектов транспортной инфраструктуры (ОТИ) и транспортных средств (ТС) и нарушить коварные планы нарушителей. Под защищенностью, как правило понимают наличие соответствия технических средств защиты, которые со- ответствуют категории объекта. Такой подход не предусматривает оценку эффектив- ности принятых мер и способности всей системы про- тивостоять реальным угрозам. И нет никакой гарантии того, что в нештатной ситуации реагирование на нее будет своевременным и корректным. Какой должны быть эффективная система безопас- ности объекта и как обеспечить его защищенность? Этот вопрос актуален сегодня для многих субъектов транспорт- Это возможно лишь при консолидированном под- ходе государства и общества к вопросам обеспечения безопасности [1]. 51 Надежность, том 18, №4, 2018. Сообщение ной инфраструктуры, перед которыми стоит непростая задача обеспечения транспортной безопасности. рующих места возможного проникновения на объект, фиксацию факта несанкционированного проникновения и передачу сигнала тревоги на пульт управления систе- мой. Системы пультовой охраны часто интегрируются с системами пожарной сигнализации и имеют во многом схожую с ними архитектуру. В начале 2014 г. в силу вступил федеральный закон № 15-ФЗ «О внесении изменений в отдельные законодатель- ные акты Российской Федерации по вопросам обеспе- чения транспортной безопасности». При помощи этого документа были устранены пробелы Федерального закона № 16-ФЗ «О транспортной безопасности», действовавше- го еще с 2007 года. Данный закон, являясь по своей сути базовым актом, имел чрезвычайно широкое толкование, не содержал конкретных требований к силам и средствам обеспечения транспортной безопасности, а по отдельным направлениям лишь формализовал процесс [3]. В состав системы охраны объекта входит обору- дование централизованного управления, приемно- контрольные приборы (ПКП) или панели (ПКН) и датчики контроля охраняемого объекта. Централизо- ванное управление системой обеспечивает компьютер или заменяющая его в небольших системах специальная панель управления. Приемно-контрольные приборы осуществляют питание шлейфов с устанавливаемыми на объекте датчиками, принимают и анализируют пере- даваемые датчиками сообщения, формируют и передают сигнал тревоги на пункт централизованного управления, управляют оповестительными устройствами и другими системами обеспечения безопасности объекта. На не- больших объектах управление всеми системами может осуществляться непосредственно ПКП без передачи информации на пункт централизованного управления. р ф р р [ ] При комплексном подходе к обеспечению безопасности на ОТИ причины, повлекшие возможность совершения террористического акта, должны рассматриваться как сим- птомы неудовлетворительной организации эксплуатации данных объектов, на совершенствование которой требуется направлять основные усилия, людские и материальные ресурсы. Согласно предложенной профессором Колорад- ского университета (США) Д. Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Датчики разбития стекла (рисунок 1, б) реагируют на звук бьющегося стекла и предназначены для защиты Датчики разбития стекла (рисунок 1, б) реагируют на звук бьющегося стекла и предназначены для защиты а б в г д Рисунок 1 – Охранные извещатели: а – извещатель охранный магнитоконтактный ИО-102-14 (СМК-14, геркон); б – датчик раз- биения стекол ИО329-3 Арфа; в – извещатель охранный поверхностный вибрационный «Шорох-2» (ИО-313-5/1); г – четырехлу- чевой датчик-барьер инфракрасных лучей NR110QS; д – схема использования лучевых датчиков б а б а в Рисунок 1 – Охранные извещатели: а – извещатель охранный магнитоконтактный ИО-102-14 (СМК-14, геркон); б – датчик раз- биения стекол ИО329-3 Арфа; в – извещатель охранный поверхностный вибрационный «Шорох-2» (ИО-313-5/1); г – четырехлу- чевой датчик-барьер инфракрасных лучей NR110QS; д – схема использования лучевых датчиков 52 Обеспечение эффективной системы безопасности объектов транспортной инфраструктуры посредством систем, позволяющих зафиксировать факт проникновения посторонних лиц в охраняемую зону ной системы безопасности объектов транспортной инфраструктуры посредством систем, позволяющих зафиксировать факт проникновения посторонних лиц в охраняемую зону окон охраняемого объекта. Принцип работы датчика основан на анализе спектра фиксируемого шума и его сравнении с записанными в память датчика эталонными звуковыми сигналами. Наиболее совершенные типы датчиков формируют сигнал тревоги при наличии двух факторов: удара по стеклу и звука бьющегося стекла. а а б Рисунок 2 – Датчики движения: а – извещатель инфра- красный пассивный «Фотон-9»; б – охранный оптико- электронный извещатель потолочной установки «Фотон-21» Вибрационные датчики (рисунок 1, в) фиксируют на- личие вибрации и ударов, возникающих при попытке раз- рушения защищаемого объекта. Принцип действия таких датчиков основан на использовании пьезоэлектрического эффекта или электромагнитной индукции для преобразо- вания сигналов вибрации, подвергающихся разрушению, в анализируемые электрические сигналы. б Фотоэлектрические (лучевые) датчики инфракрас- ного (ИК) излучения (рисунок 1, г, д) используют для защиты коридоров, лестничных пролетов, ворот и входных дверей, а также в системе периметральной охраны объекта. б а Рисунок 2 – Датчики движения: а – извещатель инфра- красный пассивный «Фотон-9»; б – охранный оптико- электронный извещатель потолочной установки «Фотон-21» Действие датчиков основано на образовании барьера из формируемых специальными излучателями для защи- ты охраняемой зоны объекта системы модулированных инфракрасных лучей. При попадании нарушителя в зону действия барьера выдается сигнал тревоги. сигнала тревоги на пульт охраны, управление световой и звуковой оповестительной и пожарной сигнализа- цией, а также другими техническими устройствами и системами обеспечения безопасности объекта. В ПКП обычно совмещаются функции охранной и пожарной сигнализации. Примером ПКП могут служить приемно- контрольные охранно-пожарные приборы «Кварц», «Радиус», «Сигнал», «Риф» и т. п. Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Питерсоном теории мно- жественности причин возникновения любой нештатной ситуации, можно не только спрогнозировать возможность ее возникновения, но и выявить обстоятельства, сделавшие возможным ее появление. Следовательно, безопасностью, в том числе и транспортной, не только можно, но и необ- ходимо управлять так же, как и любой другой областью транспортной системы. обеспечение транспортной безопасности должно представлять собой одну из неотъем- лемых, ежедневных функций руководящих и оперативно- управленческих работников различного уровня, такую же, как снижение стоимости, обеспечение требуемого объема и качества перевозок грузов и пассажиров. В системах пультовой охраны для контроля возмож- ных мест проникновения на охраняемый объект исполь- зуются различные, отличающиеся по типу и принципу действия и решаемой задаче, датчики-извещатели, такие как: магнитные, вибрационные, фотоэлектрические, микроволновые, ультразвуковые датчики, шлейфы, датчики разбития стекла, датчики движения. Наиболее простыми, дешевыми и широко используе- мыми являются магнитоконтактные датчики, устанав- ливаемые на окна, двери охраняемого объекта. Данные датчики состоят из закрепляемого на подвижной части окна или двери магнитоуправляемого контакта – геркона и постоянного магнита, закрепляемого на раме окна или дверной коробке. При решении задачи обеспечения безопасности на объектах транспортной инфраструктуры большая роль отводится системам, позволяющим зафиксировать факт проникновения посторонних лиц в охраняемую зону объекта. Для этой цели используются различные систе- мы пультовой охраны, решающие задачи оповещения о факте проникновения на объект с указанием места и времени нарушения границы объекта. Данные системы обеспечивают сбор информации с датчиков, контроли- Примером такого датчика может служить накладной малогабаритный магнитоконтактный охранный извеща- тель ИО-102-14 (СМК-14, геркон), используемый для защиты дверей и окон (рисунок 1, а). Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Шлейфы систем охранной сигнализации представляют собой наклеиваемую на защищаемый элемент конструк- ции (стекло, дверь и т. п.) ленту из алюминиевой фольги, включенную в цепь охранной сигнализации. При разру- шении конструкции вместе с лентой разрывается цепь охранной сигнализации и формируется сигнал тревоги. Прибор «Кварц» (рисунок 3, а) является довольно простым ПКП и позволяет обеспечить питание и кон- троль одного шлейфа с включенными в него охранными или пожарными извещателями. Датчики движения позволяют распознать движение в охраняемой зоне по изменению уровня инфракрасного излучения при перемещении излучающего тепло объекта в зоне чувствительности датчика. По мере движения объекта ИК-излучение попадает на разные сегменты оптической системы, что приводит к формированию ряда импульсных сигналов, фиксируемых электронной системой датчика. Для защиты от ложных срабатываний в современных датчиках движения используется микропроцессорная обработка фиксируемых сигналов. В качестве примера датчиков движения представлены два типа инфракрасных пассивных извещателя семейства «Фотон» (рисунок 2). Прибор также управляет оповещателями и передает на пульт центрального наблюдения сигнал тревоги. а б Рисунок 3 – Приемно-контрольные приборы: а – «Кварц», б – «Радиус-4И» б Наряду с рассмотренными выше типами извещателей для охранной сигнализации также могут использоваться радиоволновые, ультразвуковые, микроволновые, ем- костные и другие датчики, принцип действия которых основан на анализе отраженного от объекта сигнала или изменения поля и емкости охраняемого объекта. Однако означенные датчики получили гораздо меньшее рас- пространение, так как весьма критичны к влиянию из- менений окружающей среды, типу охраняемого объекта и наличию различных дестабилизирующих факторов. а Рисунок 3 – Приемно-контрольные приборы: а – «Кварц», б – «Радиус-4И» Применяемые на железнодорожном транспорте приемно-контрольные охранно-пожарные приборы «Радиус-4И» (рисунок 3, б) и «Радиус-6И» могут ис- пользоваться как при централизованной, так и при автономной охране объектов от несанкционированного доступа и пожаров. От извещателей информация передается в другую обязательную структурную составляющую любой охранной системы, роль которой выполняют ПКП, а на небольших объектах – ПКН. Задачей ПКП является проверка исправной работы оборудования, сбор и анализ поступающей от извещателей информации, передача Прибор «Радиус-4И» обеспечивает возможность контроля четырех, а «Радиус-6И» шести шлейфов с включенными в них охранными и пожарными извещате- лями. При получении сигнала от любого из извещателей устройство по каналам связи передает сигнал тревоги, 53 Надежность, том 18, №4, 2018. Сообщение ектов с большой протяженностью, таких как железные и автомобильные дороги, трубопроводы и т. п. ектов с большой протяженностью, таких как железные и автомобильные дороги, трубопроводы и т. п. а также сигнализирует о проникновении на объект или пожаре световыми и звуковыми сигналами. В обоих приборах предусмотрена индикация состояния шлей- фов с использованием светодиодов. Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Диапазон рабочих температур прибора составляет от минус 10 до плюс 50°С, что обеспечивает возможность его широкого ис- пользования при охране различных объектов. Периметральные системы позволяют зафиксировать нарушение границы охраняемой территории и выдать сигнал тревоги намного раньше, чем нарушитель попадет на расположенные на данной территории особо важные объекты. Данные системы довольно часто конструктивно связаны с физическим ограждением охраняемого объек- та, поэтому их работа напрямую зависит от физических характеристик ограждения, его параметров, наличия вибраций, используемого материала, а также от выбора места установки оборудования, качества выполнения монтажных работ и ряда других факторов. р р р Для охраны распределенных по территории стацио- нарных объектов используют системы радиоохраны с передачей извещения о проникновении на объект или по- жаре по радиоканалу. Одним из примеров такой системы является «Риф Стринг-202». Данная система включает в себя размещаемую в центре охраны базовую станцию, пульт наблюдения и компьютер, а также размещаемые на объектах приемно-контрольные приборы с передаю- щим устройством. Система позволяет контролировать с одного пульта до 600 устанавливаемых на объектах пере- дающих устройств мощностью 10 мВт. В зависимости от местности система обеспечивает дальность связи от 25 до 50 и более километров, что достигается исполь- зованием сверхузкополосных каналов связи и помехоу- стойчивого кодирования. В системе применена цифровая фильтрация и одновременная, параллельная обработка всех принятых по различным каналам связи сигналов от контролируемых объектов. Передача каждого нового сообщения с объектного приемно-контрольного прибора осуществляется на новой частоте, случайным образом выбираемой из 1024 заранее запрограммированных частот связи. Для проверки исправности оборудова- ния каждый объектный передатчик один раз в минуту передает на базовую станцию контрольные сигналы. Благодаря использованию лицензируемых частот и передатчиков малой мощности эксплуатация системы «Риф Стринг-202» не предусматривает получения раз- решительных документов. Для надежной охраны объекта периметральные си- стемы должны отвечать определенным требованиям, таким как: - полный контроль всего периметра охраняемого объ- екта и отсутствие «мертвых зон»; - полный контроль всего периметра охраняемого объ- екта и отсутствие «мертвых зон»; - высокая чувствительность обнаружения нарушителя; - низкая вероятность ложных срабатываний; - надежная работа в условиях электромагнитных по- мех от работающего оборудования и расположенных вблизи индустриальных объектов; - надежная работа в различных климатических условиях. В настоящее время для обеспечения безопасности объектов наиболее часто используются лучевые, радио- волновые и емкостные периметральные системы. Данные системы имеют высокую эффективность обнаружения и позволяют надежно защитить охраняемую территорию. Однако лучевые системы, например, инфракрасные барьеры, могут эффективно контролировать только пря- молинейные участки периметра. Библиографический список зарегистрировано 670 АНВ в деятельность железнодо- рожного транспорта. Среди них отмечены случаи с при- знаками террористического характера, в том числе 68 со- общений об угрозах совершения террористических актов. Также произошло 142 наложения посторонних предметов на рельсы, 79 случаев разоборудования железнодорожных путей и 269 случаев разукомплектования средств сигна- лизации, централизации и блокировки (СЦБ). Старовойтов А.С. Защищенность объектов транс- 1. порта необходимо поддерживать // Транспортная и техно- логическая безопасность. – 2017. – № 2. – С. 100-101. Федеральный закон от 09.02.2007 г. № 16-ФЗ 2. «О транспортной безопасности». общений об угрозах совершения террористических актов. Также произошло 142 наложения посторонних предметов на рельсы, 79 случаев разоборудования железнодорожных путей и 269 случаев разукомплектования средств сигна- лизации, централизации и блокировки (СЦБ). Федеральный закон от 03.02.2014 № 15-ФЗ «О вне- 3. сении изменений в отдельные законодательные акты Российской Федерации по вопросам обеспечения транс- портной безопасности». В течение 2017 года на объектах железнодорожного транспорта зафиксировано 5 случаев обнаружения взрывчатых веществ, 665 бесхозных подозрительных предметов. Изъято 20 единиц огнестрельного оружия, 1292 патрона различного калибра, 49 взрывоопасных предметов [6]. Постановление правительства РФ от 26.04.2017 г. 4. № 495 «Об утверждении требований по обеспечению транспортной безопасности, в том числе требований к антитеррористической защищенности объектов (территорий), учитывающих уровни безопасности для различных категорий объектов транспортной инфра- структуры и транспортных средств железнодорожного транспорта». Между тем, опыт США, Канады и ряда европейских стран говорит о том, что забота об укреплении транс- портной безопасности находится в центре внимания не только государства. Этой проблеме уделено огромное внимание и в ее разрешении активно участвуют обще- ственные организации. Так, например, в США сотни частных компаний и фирм, под эгидой соответствую- щих общественных фондов, систематически выделяют в рамках своих годовых бюджетов огромные средства на исследования и разработки проблем обеспечения транспортной безопасности. Отмеченное выше свиде- тельствует в пользу того, что забота о кардинальном и неотложном повышении уровня безопасности на транс- порте не может быть лишь уделом государства, равно как и не должны только государством изыскиваться необходимые для решения этой задачи финансовые, организационные и кадровые ресурсы. Кроме того, даже самая хорошо отлаженная система транспортной безопасности не может эффективно функционировать без широкой опоры на понимание и поддержку всех структур гражданского общества. Natalia Kuzmina. Ensuring transport security in the 5. transport infrastructure and means of transport railway transport facilities. / Natalia Kuzmina, Tatyana Odudenko // Proceedings of the XV International Academic Congress «Fundamental and Applied Studies in the Modern World» (United Kingdom, Oxford, 06–08 September 2016). Volume XV. «Oxford University Press», 2016. Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия Радиоволновые системы критичны к рельефу местности и плохо работают при наличии в охраняемой зоне деревьев или кустарников, что часто имеет место вблизи железных и автомобильных дорог. Емкостные системы требуют наличия физическо- го ограждения объекта, так как фиксируют изменение электрического поля и емкости при приближении или прикосновении нарушителя к данному ограждению. Однако лучевые системы, например, инфракрасные барьеры, могут эффективно контролировать только пря- молинейные участки периметра. Радиоволновые системы критичны к рельефу местности и плохо работают при наличии в охраняемой зоне деревьев или кустарников, что часто имеет место вблизи железных и автомобильных дорог. Емкостные системы требуют наличия физическо- го ограждения объекта, так как фиксируют изменение электрического поля и емкости при приближении или прикосновении нарушителя к данному ограждению. Для полного контроля больших территорий исполь- зуют системы радиоохраны, также обеспечивающие сбор информации с распределенных по территории объекта датчиков с радиоизвещением. Датчики в данном случае располагаются по всей территории охраняемого объекта таким образом, чтобы расстояние между ними не превышало радиуса действия датчика, при котором обеспечивается обнаружение постороннего объекта и не образуются так называемые мертвые зоны. При попада- нии в зону действия датчика человека или постороннего предмета датчик фиксирует факт возникновения не- штатной ситуации и посылает по радиоканалу на пульт управления системой сигнал тревоги. Для охраны протяженных объектов также могут ис- пользоваться проводно-радиоволновые системы, пред- ставляющие собой два расположенных по охраняемому периметру параллельных фидера. Один фидер выпол- няет роль приемной, а другой передающей антенны. При попадании в зону действия поля данных фидеров постороннего предмета происходит возмущение поля, сопровождающееся изменением параметров принимае- мого сигнала, постоянно фиксируемых соответствую- щим оборудованием. Во многих случаях для обеспечения безопасности любой территории достаточно надежно контролировать только ее периметральные границы, чтобы полностью исключить факт несанкционированного проникнове- ния на данную территорию. Такие системы получили название периметральных. Данные системы просто незаменимы при охране объектов с большими защи- щаемыми территориями, таких как аэродромы, базы, транспортные логистические терминалы, а также объ- Рассматриваемые системы радиоохраны обычно интегрируются с системами видеонаблюдения, так как требуют идентификации нарушителя, пытающегося проникнуть на охраняемый объект. Конечно только человек остается центром принятия решения в системе безопасности. Он следит за состояни- ем технических средств, получает тревожные сигналы, 54 Обеспечение эффективной системы безопасности объектов транспортной инфраструктуры посредством систем, позволяющих зафиксировать факт проникновения посторонних лиц в охраняемую зону управляет силами реагирования. Несмотря на четкий и понятный алгоритм действий, неукоснительное соблю- дение требований по обеспечению транспортной безо- пасности на ОТИ и ТС железнодорожного транспорта, в нестандартной ситуации может сработать естественный человеческий фактор: утомляемость, усталость, рас- терянность и даже халатность. Библиографический список – 718 p. www.roszeldor.ru.[Электронный ресурс] – Режим 6. доступа http://doc.rzd.ru. Процесс эволюции не остановить // Транспортная и 7. технологическая безопасность. – 2017. – № 3. – С. 34. Процесс эволюции не остановить // Транспортная и 7. технологическая безопасность. – 2017. – № 3. – С. 34. Наталья А. Кузьмина, Дальневосточный государственный университет путей сообщения, Хабаровск, Россия И тогда ценой ошибки или промедления в реагировании на угрозы может стать человеческая жизнь [4]. ность – одно из важнейших условий развития личности, общества, государства. Это всегда требовало и требует от людей компетентности в мире опасностей и способах защиты от них. Применение средств физической защиты в сочетании с организационными мероприятиями и действиями подразделений транспортной безопасности является важнейшим фактором, обеспечивающим обнаружение и противодействие попыткам совершения АНВ в отно- шении имущества, груза и физических лиц на объектах железнодорожного транспорта. За последние годы немалое количество террористиче- ских актов было совершено на объектах транспорта или с использованием транспортных средств. Сухая статист- ка неумолима, о чем свидетельствуют отечественные и международные источники. Вот почему защита ОТИ и ТС от актов незаконного вмешательства, особенно совершенных умышленно, в форме террористических актов и диверсий, в последние годы вошла в число самых актуальных проблем во всем мире [5]. Эффективная система управления безопасностью – как хороший солдат: и в мирное, и в боевое время отвечает за точное и своевременное исполнение воз- ложенных обязанностей и задач. При этом позволяет видеть ситуацию цельно и максимально снижает веро- ятность ошибки [7]. По итогам 2017 года на сети железных дорог России зарегистрировано 670 АНВ в деятельность железнодо- рожного транспорта. Среди них отмечены случаи с при- знаками террористического характера, в том числе 68 со- общений об угрозах совершения террористических актов. Также произошло 142 наложения посторонних предметов на рельсы, 79 случаев разоборудования железнодорожных путей и 269 случаев разукомплектования средств сигна- лизации, централизации и блокировки (СЦБ). Сведения об авторе Наталья А. Кузьмина – кандидат педагогических наук, доцент кафедры «Организация перевозок и без- опасность на транспорте» Дальневосточного государ- ственного университета путей сообщения, Хабаровск, Россия, e-mail: kuzminaprepodavatel@mail.ru Подводя итог, хотелось бы напомнить, что одна из основных задач субъекта транспортной инфраструктуры или перевозчика – это безопасность человека. Безопас- Поступила: 10.01.2018 Поступила: 10.01.2018 55
W4249922453.txt	https://figshare.com/articles/thesis/Persian_bloggers_exile_nostalgia_and_diasporic_nationalism/14656185/1/files/28138029.pdf	en		Persian bloggers : exile, nostalgia and diasporic nationalism	null	2,021	cc-by	17,882	Ryerson University Digital Commons @ Ryerson Theses and dissertations 1-1-2009 Persian bloggers : exile, nostalgia and diasporic nationalism Maryam Aghvami Ryerson University Follow this and additional works at: http://digitalcommons.ryerson.ca/dissertations Part of the Sociology Commons Recommended Citation Aghvami, Maryam, "Persian bloggers : exile, nostalgia and diasporic nationalism" (2009). Theses and dissertations. Paper 505. This Major Research Paper is brought to you for free and open access by Digital Commons @ Ryerson. It has been accepted for inclusion in Theses and dissertations by an authorized administrator of Digital Commons @ Ryerson. For more information, please contact bcameron@ryerson.ca. PERSIAN BLOGGERS: EXILE, NOSTALGIA AND DIASPORIC NATIONALISM by Maryam Aghvami, BA, English Translation and Literature, Allameh University, School of Literature and Foreign Languages, Tehran, Iran, 1990 A Major Research Paper presented to Ryerson University in partial fulfillment of the requirements for the degree of Master of Arts in the Program of Immigration and Settlement Studies Toronto, Ontario, Canada, 2009 1 Maryam Aghvami 2009 UBRABV Author's Declaration I hereby declare that I am the sole author of this major research paper. I authorize Ryerson University to lend this paper to other institutions or individuals for the purpose of scholarly research. Maryam AghVami I farmer authorize Ryerson University to reproduce this paper by photocopying or by other means, in totak>r in part, at the request of other institutions or individuals for the purpose of scholarly research. Maryam Aghvami 11 PERSIAN BLOGGERS: EXILE, NOSTALIA AND DIASPORIC NATIONALISM Maryam Aghvami Master of Arts, 2009 Immigration and Settlement Studies Ryerson University ABSTRACT Exiled Iranian bloggers employ the cyber sphere to create an imaginary national community to remain connected to the homeland and to encourage a sense of belonging to a nation they were forced to leave behind. This constructed cyber community carries the nostalgic impulse of the exiled bloggers. The imaginary virtual Iranian community enables them to share past memories and their anticipation of a future return to Iran. This nostalgia is also located in the engagement in activism as they express nationalism while living in the diaspora. What tends to be absent when bloggers are engaging in a diasporic nationalism and the nostalgic impulse is an engagement of their diasporic experience, because they assume their exile is a temporary state. As a result of detachment from the diasporic experience in the blogs and a total disengagement from life in the diaspora, their integration into the new country is kept at bay. Key words: Persian Blogging; Iranian diaspora, exile; nostalgia; nation; nationalism. in Acknowledgments First and foremost, I would like to thank my supervisor, Professor Nima Naghibi, who has been extremely supportive and encouraging and made the process of writing this MRP a smooth and joyful experience. I would also like to thank my second reader, Professor Tariq Amin Khan for his insightful comments and questions during the oral review. My sincere heartfelt gratitude goes to Professor Vappu Tyyska, ISS program director for her friendship, lively presence, genuine support and ongoing inspiration. Finally and most importantly, I would like to thank my husband and my love, Nima Birgani, who has been my rock and my motivation throughout the entire process of my studies. IV Table of Contents I. Introduction page 1-6 II. Iranian diaspora - Blogging Revolution page 7-16 III. Blogging, Exile and Nostalgia page 17-26 IV. Blogging and Diasporic Nationalism page 27 - 50 V. Conclusions page 51-52 VI. Works Cited page 53 - 55 v I. Introduction Since the 1979 Islamic revolution, Iran has faced the biggest brain drain of its history. In the past three decades, under the rule of the Islamic theocracy in Iran, waves of Iranian activists, writers and journalists have migrated from their homeland and are living in exile in the diaspora. Some exiled Iranians have employed transnational cyber-media communications, but more specifically the blogsphere, in order to remain connected to the homeland and to enjoy a sense of belonging to a nation they were forced to leave behind. Under the shadow of censorship and media crackdown of the ultra-conservative rule of the Islamic government in Iran, blogging has become a phenomenon that has enabled Iranians to find new ways to challenge the political oppression of the theocratic rule and to communicate with one another effectively. John Kelly and Bruce Etling (2008) suggest: "Given the repressive media environment in Iran today, blogs may represent the most open public communications platform for political discourse" (2). For this exiled population, blogging has emerged as a way to transform the physical distance, which has enabled an engagement in long-distance political activism and expressions of nationalism while remaining in the diaspora. These writers are vigorously engaged in a global discourse with the homeland, expressing national sentiments and raising their voices about political crackdown and social injustice in the home country. This research project will examine the blog entries of four exiled bloggers in the Western diaspora, their strong sense of nationalism and belonging to the homeland in the Persian blogosphere and their disengagement from the social and political development, for instance, federal elections, provincial strikes and immigration and settlement issues in their host countries. Further in this research, I will address the reasons and the issues related to the selection of these four specific exiled bloggers. Drawing on my experience as an exiled Iranian journalist in Canada, the initial motivation for this study originated from my personal interest in reading the weblogs of Iranians living in the diaspora. The main goal of this research project is that cyber space and the virtual existence of a community of Iranians in the cyber sphere facilitates the exiled bloggers' belonging to a national community and engagement in the discourse over national issues and provides an opportunity to continue political activism on-line and long-distance, which speaks to their loyalty and commitment to political change in the homeland. According to Thomas Eriksen (2006), "immigrants develop transnational networks with people from their own country as an alternative or a supplement to full membership in one nation or the other" (1). This raises the question, what is it about this "imagined" community in the cyber sphere that echoes more commitment and supports more concern for national issues of the homeland than a concern for their diasporic experience in the host country? In this research I am arguing that the blogsphere assists exiles to create an imaginary nation that facilitates expressions of diasporic nationalism through a nostalgic lens (Said, Anderson & Boym). According to Ernst Renan (1990), a nation is a soul, a spiritual principle [...] the desire to live together, the will to perpetuate the value of the heritage that one has received in an undivided form (19). These exiled writers echo their sense of loss and despair and create an intimate and familiar space that frees them from the isolation and solitude of living in the host country. By looking to the past memories of the homeland and by projecting a hope for a future return, the cyber-space created by the exiled bloggers engages the past and future; as a result, their lives in the present time in the diaspora is absent in their weblogs. In the imagination of the exiled bloggers, a lost home that existed in the past is restored by a virtual Iran in this cyber sphere. According to Eriksen, "The Internet can be instrumental in creating and re-creating a shared, collective past among its users" (1). This imagined community, what I am naming a community of Iranianness, echoes the pain and misery of the loss of a homeland they were forced to leave behind forever (Anderson 56). By mourning the loss and despair of the past and ultimately engaging the nostalgic impulse, the exiled bloggers invest their emotional resources to restoring the memory of an ideal homeland, in a sense disengaging them from the reality of their lives in the diaspora, which constitutes opportunities for integration and construction of new lives in a more secure society than that of their homelands. Diasporic nationalism and the nostalgic impulse in the cyber imaginary community creates an intimate sphere, a familiar home where these exiled authors long to remain connected to a homeland disengaged from the social and political life of the host country. As a result of an absence of living in the diaspora through their blogging and a complete detachment from their lives beyond the virtual borders of the cyber sphere, the integration of the exiled authors into the new country is kept at bay. While exile and the reality of solitude and isolation have transformed the diasporic experiences of these exiled Iranian bloggers in the Western countries, this research acknowledges that the diasporic reality of these non-white and non-European migrants in the Western countries also put them in challenging positions. In fact, it is important to recognize that the isolation and solitude of the exiled bloggers may have not entirely emanated from within and the condition of migrancy such as race, racism, the colonial encounter and the marking of the 'other' may have played out in their access to opportunities as well as their social and political integration in the Western diasporas. However, given the scope of this research project, I will not be able to address the issue of exiled bloggers' isolation and onus of integration in the host countries. While the exiled bloggers assume their displacement is a temporary state, they project their hopes for a return to the homeland and ultimately remain distant from their lives in the diaspora. What is certain is the diasporic existence of the bloggers remains absent from their imaginary Iranian community in the cyber sphere. By remembering and mourning the past and hoping for a future return to the homeland, the bloggers create a space between an imagined homeland and their lives in the diaspora. As a result, these bloggers float between two domains: the past, crucial for maintaining their national identity, and the future, a promise of a return to the native land. For these exiled populations, the past is vital for preserving their roots to the homeland's culture, language and heritage. The future, on the other hand, promises a possible social and political transition in the homeland. As a result, living and preserving the past dissociates the bloggers from their present life in the diaspora. According to Anderson (1984), "[njation is an imagined political community - and imagined as both inherently limited and sovereign" (48). I am examining here the ways that four exiled Iranian bloggers create their own imagined community in their weblogs. I argue that cyber space makes available the opportunity to create a national community of 'Iranianness' that signifies to these specific bloggers an opportunity to express nationalistic sentiments while living in the diaspora. This space provides an opportunity to look at the past, through a nostalgic lens, and to restore the home country they were forced to leave behind (Boym xv). These exiled bloggers have made claim to a virtual cyber space in their weblogs and have created an online "presence" as if they are still living in Iran and can play a role in the process of change in Iran. This research analyzes the content of weblogs as the principal text for an exploration of the Persian blogsphere in the diaspora. I will be examining the blog posts of four exiled bloggers, Mehrangiz Kar, Masoud Behnoud, Nikahang Kowsar and Sibiltala in a period leading to the June 2009 presidential election and its aftermath in Iran. In addition to focusing on the weblogs of four bloggers, I will analyze the cyber communication between the bloggers and their readers and compare the two texts. Mehrangiz Kar is an exiled blogger, journalist and activist living in the United States. Kar was an outstanding lawyer and women's rights activist in Iran but the increasing pressures of the Iranian government forced her to flee her homeland in the early 1990s. Mehrangiz Kar pursued her activism through writing in her blog and for the Persian on-line media in the diaspora. Kar represents the first generation of exiled Iranian women in the diaspora who maintained a vibrant social and professional life before and after the 1979 revolution. Masoud Behnoud is one of the most prominent writers and journalists both before and after the 1979 Islamic revolution in Iran. Behnoud experienced imprisonment in solitary confinement in Tehran's notorious Evin prison, state-sponsored crackdowns on journalists and media censorship which forced him to flee Iran for Europe where he has since been living. Along with his continued cooperation with the Persian-speaking media in the diaspora, blogging has enabled him not only to remain in Iran's national memory and maintain his popularity but he is also free to express a sense of nationalism and belonging to the homeland. Behnoud belongs to the first generation of exiled Iranians that has experienced both monarchy and theocracy during his adult life in Iran. Behnoud's blog enjoys a wide range of readership, among them are a diverse group of activists, politicians and three generations of Iranians, whether in Iran or in the diaspora. The success of Behnoud's blog is his lively interaction and/or tension with his readers, as if instead of a cyber interaction, he is actually engaged in a face-to-face real discourse with Iranians in the homeland. Sibiltala, a pseudonym to protect her identity, belongs to a 30- year- old exiled Iranian woman living in Canada. Similar to the younger generation of Iranian bloggers in Iran and in the diaspora, Sibiltala's weblog employs cyber-media not only to express her views freely but also to closely monitor and to speak about national political, social and cultural issues. Sibiltala belongs to the second generation of Iranian migrants living in the diaspora and to a group of Iranians who are too young to remember the 1979 revolution in Iran. Nikahang Kowsar is an internationally renowned cartoonist and blogger living in exile in Canada. Kowsar owes his popularity and success to a short-lived era of flourishing media during the reform government of the former Iranian president Mohammad Khatami in early 2000. Following a period of media crackdown in early 2000, Kowsar was detained for drawing controversial cartoons about Iranian government officials. He was later accused of acting against national security by Iranian government's hard line judiciary. In order to save his life, he fled the country in 2003 and left his family and homeland. Kowsar's blog enjoys wide readership among Iranians in Iran and in diaspora. My goal in selecting these particular Persian weblogs was an effort to echo Iranian exiled writers' national concerns and long-distance political engagements with the homeland. In addition to the attention of this research to the diversity of the bloggers in terms of age, gender and the political period they left Iran, my main objective is to explore their common interests, themes and attention to the issues related to their lives at home and in the diaspora. These bloggers share common reasons for living in exile in the diaspora; their lives have been interrupted by the aftermath of the 1979 Islamic revolution in Iran and such drastic transformations in the social and political life has resulted in their migration to the diaspora. These exiled bloggers are all currently residing in developed and democratic states in North America and Europe; they enjoy freedom of speech and have access to the most advanced technologies and transnational media to facilitate their communication with the homeland. II. Iranian Diaspora - Blogging Revolution The diasporic Iranian community is a Vibrant and multi-faceted one. Comparable to the population of Iran, the fabric of the Iranian diaspora is religiously and ethnically diverse. An ethnically and culturally distinct population, historically and geopolitically exposed to the revolution and war, Iranians have established their lives as global citizens. The existing scholarship of the history of emigration of Iranians to Western diasporas recognizes three major waves of exodus. According to the Migration Information Source Sept. 2006 report ("Iran: A Vast Diaspora Abroad and Millions of Refugees at Home.") Shirin Hakimzadeh has observed that the first wave of migration during the period of 1950 to 1979 consists of Iranians who migrated abroad, a move that was triggered by Iran's slow economic recovery and resumption of oil production after World War II. The second major wave of migration of Iranians took place after the 1979 Islamic revolution and directly corresponds with the post-revolution aftershocks and implementation of restrictive Islamic rules in Iran. The mass exodus of Iranians after the revolution included a large population of religious and ethnic minorities, social and political activists and former affiliates of the monarchy that fled from the Iranian government's persecution and crackdown. In addition, the 8-year Iran-Iraq war (1980 - 1988) greatly contributed to the migration of thousands of Iranian refugees and migrants abroad. The mass migration of skilled and professional Iranians took place in this era and resulted in the biggest brain drain of Iran's history. The third wave of migration during the mid-1990s has continued to date. Thousands of Iranian professionals have shaped the fabric of the population who fled Iran during this period. The third wave of Iranians migrants were significantly motivated by violations of human rights and intensified oppression of religious and ethnic minorities by the Iranian government and its state-sponsored strict control of social, cultural and political public and private spheres of Iranians in the country. The beginning of the Internet era and a short-lived period of media openness and freedom of expression during the presidency of reformist Mohammad Khatami in early 2000 in Iran was followed by increasing frustration and disappointment towards the increasing crackdown, imprisonment and torture of media activists; this led to the mass migration of journalists, writers and bloggers in the West. The exiled writers who were forced to abandon their professional lives in Iran and sought refuge in a new country, employed cyber media, particularly blogging, in order to bridge the distance with a homeland they left behind. They found a virtual status in their weblogs to remain connected to the developments in their homeland. A significant characteristic shared by the four bloggers I am examining here is a deep and broad understanding of the situation in Iran and its ongoing social and political change, as well as an extensive familiarity with the powerful players in the ruling Islamic government in Iran. In addition, these bloggers represent four displaced Iranians whose personal and social lives were transformed by the aftermath of the 1979 Islamic revolution in Iran. The impact of the socio political changes of the post-1979 revolution in Iran on the lives of these four bloggers reflects the painful episodes of exilic life experienced by thousands of Iranians in the diaspora. In spite of relative familiarity with the official languages of their host countries, the exiled bloggers have chosen to write in Persian. Although language is a barrier and also a factor involved in the authors' decision to write in Persian, by publishing their blog posts in a language that is familiar and intimate for their audience they are able to effectively communicate with a wider population of Persian speakers across the globe and also to express their sense of nostalgia and nationalism in exile in the diaspora in a candid and spontaneous manner (Boym, Said).1 In addition, writing - In order to facilitate this research I will be providing the English translation. 8 in Persian also supports the exiled bloggers' attempt to expand the realms of their cyber presence and enables them to articulate their nostalgia through the romantic language of Persian literature and poetry. More importantly, since Persian is the official language of Iran (Persia), the four bloggers' intent to write in Persian may also be an indication of authentic Iranianness and a protection of national identity and cultural heritage in the diaspora. Women, both in Iran and in the diaspora, play a crucial role in shaping the diversity and vitality of the Persian blogsphere. According to Saeid Golkar (2008), 76% percent of weblog authors are men and 24 percent are women (53). In spite of the fact that the existing scholarship on the balance of male and female bloggers indicates less activity of Iranian women on the Internet compared to Iranian men, it is important to note that women in Iran and in the diaspora have an active presence in the Persian blogsphere but tend to maintain a more cautious presence in the virtual sphere. In an attempt to protect their security, many female bloggers hide their real identities; adopt pseudonyms and blog anonymously. More importantly, the virtual existence and active presence of the Iranian women in the blogsphere remains a strong indication of their critical focus on gender-related issues and their determination to employ cyber-media in order to promote the cause of Iranian women. This is also one of the reasons why I have chosen to focus on two female (Kar and Sibiltala) and two male bloggers (Kowsar and Behnoud) living in exile in the diaspora. In Iran and in the diaspora, blogging has liberated Iranian women like Sibiltala and Kar from social and cultural restrictions and has enabled them to express their views freely and without fear of persecution and discrimination. More importantly, by writing in their blogs, Kar and Sibiltala have been able to facilitate information exchange with like-minded Iranians who are active in the women's movement in the homeland, bridging the geographical gaps and surpassing the censorship on the flow of specific information on foreign organizations that actively pursue women's rights while living in the diaspora. The use of the pseudonym by female bloggers, such as Sibiltala, is a political decision so they are able to express opinions on subject matters that are considered taboo in Iranian culture. The interesting aspect of these particular bloggers is that each author develops a personalized virtual space which maintains an intimate and familiar nationalist literature that suits his/her exilic experience and each has different approaches to expressing their sense of loss and despair over separation from the homeland; however, what they have in common is an undivided and unanimous message of Iranianness in their weblogs. These bloggers, particularly Sibiltala who left Iran at a very young age, share a common desire to maintain their Iranian roots and cultural heritage and unanimously express a desire for the restoration of the Iranian nation as they imagine it. Boym (2001) suggests that the "[inability to return home is both a personal tragedy and an enabling force. This doesn't mean that there is no nostalgia for the homeland; only that this kind of nostalgia precludes restoration of the past" (252). The cyber space these bloggers have created and share is driven by a nationalistic and nostalgic impulse while often adopting different approaches and holding different views on issues such as personal and social values, politics and perspectives on the ways the transition of the homeland to a democracy should occur. For instance, Sibiltala has not lived her adult life under the ultra-orthodox Islamic system in Iran and believes in unrestricted freedom of speech that facilitates the construction of an Iranian society that tolerates opposing voices and democratic values not dictated by the state. On the other hand, Behnoud and Kar belong to a generation that witnessed the transition of Iran under the rule of monarchy to the establishment of the Islamic government in 1979. They believe in resilience and tolerance towards the democratization process. Kowsar has experienced the disappointment and frustration of the failure of the reform movement in Iran and is critical 10 of the lost opportunities of establishing a civil society in Iran. These four exiled bloggers create a virtual Iranian identity in their weblogs and this familiar space separates them from the isolation and solitude of the exile in the new country. It is important to recognize that the nous of this diasporic sense of separation should not entirely fall on the exiled Iranian bloggers and the structural migration systems in the host countries also play a key role in isolation of the non- European migrants and their integration into Western societies. It is in this imagined sphere of Iranianness that the bloggers relive their past memories and maintain a sense of belonging to a national community that does not really exist in the diaspora. In spite of the fact that the four selected bloggers share a nationalist impulse, they have different ways of expressing their nostalgic sentiments. All four bloggers adopt diverse ways of expressing their sense of loss and pain of the forced migration, however nostalgia is a collective impression that the exiled bloggers share in their virtual existence in the cyber space. Kowsar's frustration with exile and inability to contribute to the cause of Iranians in the homeland is reflected in the following post: Neda's2 name comes up at least once or twice when I speak about current crisis in Iran with American and Canadian friends. In this part of the world, we have seen the video clip (of Neda's murder) hundreds of times, we have cried tens of times, we have repeatedly thought about what we would have done, if we were living there. What could we do...? Neda symbolizes innocence of a large group of people that are seeking justice from these oppressive groups. Neda's blood will one day haunt these tyrannical groups down. Revenge is not the answer! Never! [...] For me, Neda is a symbol of peace and serenity and denial of violence. Neda is the voice of a nation that is exhausted by the 2 Neda is the name of a girl whose death during the post-election protests was captured by a cell phone camera. 11 violence and is in quest of freedom and democracy. (July 30,2009) As expressions of nationalism, nostalgia and human dignity are central to the virtual existence of the exiled bloggers on the net, the diasporic experience of the authors is absent from their Iranian cyber communities. These virtual communities facilitate the bloggers' existence in a familiar sphere, ease their pain of exile in the foreign land and assist them to relive the past memories and promises hope for a future return. In a birthday note to his father in Iran, Kowsar's pain of separation from his family is reflected in this blog post: I have been constantly thinking about you these days. Six years of separation. Some day, we may meet again in the South Pole, years from now. I hope there comes a day that everyone appreciates you [...] a day will come that wise people will realize what you have achieved. Throughout the years, these individuals were rare to find. But their small numbers demonstrated their true values. Happy Birthday Dear Dad. (May 29,2009) The present however is absent from this virtual national sphere in the diaspora and the four selected weblogs resonate a total disengagement of the bloggers' with their lives in the new country. Such detachment in their diasporic existence impedes seizing the opportunity of embracing the new life and engaging with the social and political life in the host country. Instead of moving forward with their life in their adopted country, by looking backward through a nostalgic lens and engaging in long-distance political activism in an imaginary national cyber sphere, these exiled authors have chosen to embrace their cyber existence. Kowsar explains: Why did I launch this blog? During the years (in Canada) that a part of my memory was lost, I had forgotten a period of my life (in Iran), my psychotherapist suggested that I should remember a phase of my life and start writing about my memories and express my feelings. (May 24,2009) 12 While the exiled bloggers remain content in the comfort of an intimate cyber Iranian community, they tend to overlook the challenges and opportunities that the diasporic experience in a developed and democratic country may offer. In addition, far from the oppression and crackdown of the Islamic government, the exiled Iranian bloggers take advantage of the freedom of expression and technological advancements in their host countries; however, on rare occasions they will publish announcements of an event that is related to the Iranians in the diaspora, distribute petitions or announce a press release that condemns violations of human rights and imprisonment, it is unlikely that the exiled bloggers will incorporate their diasporic existence in their on-line virtual discourses. Unlike some weblogs that deal with personal issues, these particular weblogs and bloggers actually place themselves in the public sphere and address a general audience and social and political topics of interest. While these bloggers share their personal experiences in the host country, such as meetings with interesting individuals and attending music concerts, the majority of posts focuses on the ongoing challenges of Iranians in the homeland and expresses their personal views about a perspective of change in Iran. Moreover, the authors and the readers communicate with each other through the cyber media and the interaction between the two creates a dialogue surrounding the debates on certain national issues. The communication and tension, facilitated by physical distance, between the bloggers and their audience establishes a long-distance relationship on common national concerns such as absence of personal and social freedoms, violations of human rights and persecution of activists in Iran. What is interesting is that the exiled authors not only assume they are fulfilling a national responsibility, they also believe they have an existence in the homeland. In an imaginary letter written by a young girl in Iran, Behnoud echoes the struggles of the young generation in the homeland: 13 My generation is seeking peace, a better atmosphere to live. (My generation) is tired of lies, empty and rusted words. (My generation) It has been exposed to bullet, blood and death. But (this generation) is determined not to turn into a blind rebel and fighter to eternally live with these words. We demanded the revolutionary songs from (Iran) the state broadcasting when we re-lived childhood memories in our peaceful underground lives, when the children of the tortured and the torturers gathered under the peace flags of 2nd of Khordad3 and when we demonstrated our affection to the homeland with love, and not with blood. (June 22,2009) Regardless of the authors' views on the political group that may have taken the helm of power in Iran, the posts reflect the bloggers' collective support for implementation of social and political change through civil and democratic means. Fascinated by the courage and determination of the young generation in Iran that seeks non-violence democratic change in the country, exiled Behnoud expresses his nationalist message in this imaginary letter sent from a young Iranian girl: My generation is engaged in making history and is responsible for it too now. Even today, these words that are written by my hands are a part of the responsibility that my generation is taking. Now it is our turn to make a history, a history that avoids the curse of the next generation, a generation that had not welcomed this history. The blood that has unjustly been spilled will not be forgotten. This blood will make history, it will make a future, a future that is peaceful and belongs to silent fighters, those that have raised their hands to show victory. The future belongs to the unique image of an oppressed nation that have reacted with tolerance and silence and a nation that has put the (1979) 3 - 2nd of Khordad represents the reform movement of former President Khatami in early 2000 in Iran 14 revolution behind. A nation, that writes its slogans on small pieces of paper. This is a nation that doesn't believe in death and is full of life. (June 22,2009) In Iran, freedom of expression and the press are regulated by the constitution of the Islamic Republic. Due to the Iranian government's strict press policies and the crackdown on the media critical of the Iranian government and dissident journalists, hundreds of publications have been closed by Iran's hard-line judiciary and media activists were imprisoned, tortured and forced to confess to acting against national security. In July 12,2009, Reporters Without Borders (RSF) described Iran, with a total of 41 journalists in jail, as the world's biggest prison for journalists and cyber-dissidents and expressed concerns about Iran becoming the world's most dangerous place for them to operate. The OpenNet Initiative (ONI) recently reported on the media crackdown and internet censorship in Iran by noting that because the Iranian press is under the control of religious conservatives who sit above elected officials in Iran's peculiar hybrid political system, and because that conservative control is used to silence dissent, Iranians who think differently go online to express their views (5). As a result of intense crackdown and repression of the Iranian government against the independent press, Internet and social networks such as Facebook have played a vital role in the flow of information from Iran and cyber media and blogs represent the most open public communication tool for social and political discourse among Iranians that take initiatives to surpass the Iranian government's censorship and crackdown on the media and freedom of expression. The flourishing of blogs, in Iran and in the diaspora, created a new sphere for Iranians to freely express their ideas on social and political life, and weblogs in particular have revolutionized the way Iranians engage with the world and one another. According to Farid Shirazi (2008), "Iran is the third largest country of bloggers [...] and one of the world's most 15 vibrant blogospheres, surpassing the rest of the Middle East" (303). According to Golkar (2006), Persian is the fourth largest language community of Weblogs in the world (52). The diversity of Iranian bloggers include Iranian politicians, secular and religious groups, dissidents, exiles, human rights activists, ethnic and religious groups, writers and artists, youth, and women and students in Iran and in the diaspora to communicate with like-minded activists. ONI (2009) reports: "The Persian blogosphere has been heralded as one of the largest and most active in the world. The number of active Persian blogs is estimated to be approximately 60,000" (2). The content of blogs consists of a wide range of opinions representing diverse topics ranging from politics, human rights, and poetry to religion and culture. Although the blogs are filtered or blocked by the Iranian government, the activities of Iranians on the Internet continue to grow at a fast rate and cyber media and blogs continue to be the most influential means for democratic discourse and information exchange under an Islamic government. Kelly and Etling (2008) note that "[e]ven in a restrictive media environment blogs can provide a more open political discussion space than is otherwise allowed in authoritarian regimes" (23). The Iranian government has sensed the influence that blogging has on rejecting its legitimacy on public opinion and has considered cyber media a threat to its existence. The Iranian government has launched widespread suppression of open discourse, has imposed punitive actions against Internet activists, filtered the blogs, intensified imprisonment and torture of bloggers who are considered an enemy of the Iranian government and bloggers have been accused of acting against national security and staging velvet revolutions. Following the Iranian government's crackdown on media and Internet activists, bloggers were sent to jail and websites were blocked. 16 III. Blogging, Exile and Nostalgia Under the rule of the ultra-conservative theocratic Iranian regime, hundreds of activists, in search of freedom have fled Iran's state-sponsored persecutions and are living in exile in the diaspora. The exiled populations of Iranian activists have suffered imprisonment, torture and prosecution and have been forced to flee their homes. For the displaced population of Iranian citizens who are unable to revisit their homeland, the exilic experience is an open wound. Edward Said (2000) characterizes exile as the "unhealable rift forced between a human being and a native place, between self and its true home: its essential sadness can never be surmounted" (171). Hammed Shahidian (2000) suggests that "[e]xile means a mind torn asunder, pieces missing, pieces extra, memories convoluted" (72). Exiled Iranian writers who long for a past of their homeland and constantly hope for a future return seek a sense of belonging to a national community in the diaspora. inn I Nostalgia is an inseparable part of the exilic experience in the diaspora. A sense of loss Ml IC!J and despair and a desire to return to a homeland left behind confines the exile's mind and soul in the alien land, hi order to alleviate the pain of exile and reconnect their emotional bonds with a homeland they were forced to leave behind, Iranian bloggers reconstruct their past memories in an imaginary homeland in their weblogs, a national community, as Boym suggests, fantasizes a home that does not really exist in the new country. Blogging has helped Iranian authors create imaginary transnational and trans-political spaces to replace an exilic experience in the diaspora, however, by creating this imagined national community, the myth of an idealized homeland, what Boym refers to as 'the romance of one's own fantasy', is replaced with virtual nationalistic and nostalgic sentiments. Boym (2001) suggests two kinds of nostalgia: "the restorative and the reflective. Restorative nostalgia stresses 17 the nostos [return home] and attempts a trans-historical reconstruction of the lost home. Reflective nostalgia thrives on algia, the longing itself, and delays the homecoming - wistfully, ironically, desperately" (xviii). These exiled bloggers in particular epitomize the restorative nostalgia with a looking backward to a territorial land that is rooted in their history and cultural heritage. In their cyber communities, Iranian bloggers assume the responsibility of restoring a past in their homeland while remaining in exile. Boym (2001) suggests that "restorative nostalgia returns and rebuilds one's homeland with paranoiac determination" (354). While the exiled are deprived from physically returning to Iran, they create a national community of Iranianness in their community in the blogsphere, a virtual national belonging that is recreated through a longing for a memory of what they have left behind in the homeland. The exiled bloggers' cyber existence assists them to remember the time that they belonged to a national community and grants them the freedom to relive their past in an imaginary space that resonates Iranianness in the foreign land. Reconstructing and reviving the past memories restores the image of a lost homeland and liberates the bloggers from the alienation of exile and the prospects of future return appear real and possible. Restorative nostalgia allows exiled bloggers' cyber presence to renew the past memories, and reflect on the life that existed in the past. For the exiled populations, the longing to return and the mourning for the loss of the homeland becomes a part of the new existence and an enduring ritual of everyday life in the diaspora. Kowsar explains the pain of separation in this post: This morning, I felt distressed when I remembered the incidents and several experiences of the past 12 years, (a period) that fortunately or unfortunately I have spent half of it outside Iran [...]. Safety of many, including myself was not secured (in Iran) so I was forced to migrate [...]. I was separated from my wife and child for four years and many may not be 18 able to comprehend this (dilemma) but you should go through the consequences of such loss and despair on both sides in order to understand this experience. [...] One that has been forced to (exile) and is separated from one's child and wife does not have similar experiences with the one that has abandoned his/her family to divorce. (May 24,2009) Frustrated over the inability to return and motivated to remain connected to the social and political life in the homeland, exiled activists have generated a powerful and dynamic force to employ transnational cyber media to create an imagined community of Iranianness in the blog sphere. Boym (2001) suggests that exiles "develop a peculiar kind of diasporic intimacy, a survivalist aesthetics of estrangement and longing in imagined homelands of the exiles who never returned" (xix). The determination to create a cyber community in the blogsphere is rooted in the authors' desire to restore the memory of an imaginary homeland in their blogs and liberate their souls from the isolation of exile and alienation in the foreign land. Boym (2001) argues that "ordinary exiles often become artists of their lives, remaking themselves and then- second homes with great ingenuity" (253). Exiled bloggers' initiative to create an imagined community of Iranianness in their weblogs while living in the diaspora is a coping mechanism to diminish the isolation of the strange and alien land. In the 5th anniversary of launching his blog, exiled Kowsar writes a thank you letter to his readers: Dear Patient Readers of "Memoirs of an Angry Exiled" Weblog, Thank you for reading a large number of my posts and thanks to those who have continued reading my blog. I have lived with this weblog (and lived) with its readers and with those who wrote comments. I have lived with those that I made jokes about and pulled their legs (on-line). (May 24,2009) The reconstructed imagined national community in these blogs carries a nostalgic impulse 19 of authentic 'Iranianness' as it shares common culture, language and heritage and a strong sense of attachment to the homeland. While exilic experience deprives the bloggers from participation in national debates, the authors claim a virtual presence and engage in a debate with the homeland. In a letter addressed to I.R.4 Sibiltala writes Hello I.R., After eight years, I saw my bedroom in Tehran today. I looked out through my bedroom window and I looked into the window of my dead grandfather. I have missed him immensely. I wasn't there when he died. They informed me fifteen days after his death, I damn them with poison of sorrow. I used to get up and look into his window before I sleep in my bed. I could tell he is still up and (he is) reading a book when the light was on. And then I could go to sleep. (May 15,2009) The nostalgic impulse in Sibiltala's blog post is an expression of despair and agony for childhood memories in a homeland. The author's exilic isolation in the diaspora and projections of hope for change and return to the homeland ignites the longing of Sibiltala to articulate her sense of loss in a letter to her homeland's authorities: Dear I. R., How would you understand separation? It (separation) is something that elucidates the continental distance with the very first teardrop that falls from one's eyes from this part of the world to the other part of the world. One would not imagine that a space in between a small television, the computer monitor and the camera would resonate so much sense of nostalgia and separation. With the very first teardrop, one would feel it (nostalgia). In between two cameras, dear daddy, could you move the computer, a bit further [...] Oh daddy, how much the trees have grown! When did you install the pond? Was it always 4 Islamic Republic 20 there? When the tears are falling down, in between crying out loud and extreme mourning and grief, one would decide to write you a letter. (May 15,2009) Through access to advanced cyber media in the diaspora, Sibiltala brings life to her imagined Iranian community in the cyber sphere. The nostalgic urge inspires the author to reconstruct a cyber presence in the homeland located in the past. Similar to the other three bloggers, Sibiltala shares a sense of nostalgia for a homeland that she left at an early age. Her virtual existence in the weblog is an opportunity to look backwards, her life in the diaspora is focused on a time that she may be able to return to Iran, which undermines her presence in the host country. For these particular Iranian bloggers, migration to a new country and looking forward to a new home in the diaspora is not an alternative solution; instead by looking back to the past through memories and restoring the lost home in an imaginary national community they seek to heal the despair and pain of solitude in the foreign land. An imagined nation in the cyber sphere of the weblogs assists these exiled bloggers to travel to the past, an imagined time and space that resemble living at home, intimate and familiar. The notion of stepping out of one's national heritage and culture makes the exiled population's diasporic experience a complicated and multifaceted phenomenon. Boym (2001) suggests "[e]xile is both about suffering in banishment and springing into a new life. The leap is also a gap, often unbridgeable; it reveals an incommensurability of what is lost and what is found" (256). Moreover, survival tactics to reconcile with the new environment in the diaspora do not seem to diminish the anguish of the displacement and separation. Past memories while living in exile restore the life that existed in the past. The intimacy of this on-line community is a complex flow of two zones of time and space reflecting their presence and absence from the homeland, fluctuating between nostalgic feelings of despair and hope, proximity and separation. Boym (2001) suggests that nostalgia is 21 longing for a place but also yearning for a different time (xv). The intimacy of the imagined cyber community relieves the isolation of exilic life. The familiar sphere enables the bloggers to dedicate the virtual space to mourning the homeland they left behind and indulging in a shared nostalgia for a by-gone era. Reflecting upon memories of the day Kowsar left his homeland in exile, he writes: It is now six years that I have been separated from my homeland. I am extremely nostalgic, but [...] I don't claim I will return home in a few months. When I was hit by my last moments in Tehran's Mehrabad airport, I anticipated that a day would come that an oppressive force will prevail (in Iran) and no one will be able to stop them. (June 26,2009) By reflecting on the experiences that landed Kowsar in prison, the author acknowledges the impossible notion of return to the homeland and as a result he lives within virtual territories of his imagined Iranian community in the diaspora. For Kowsar, his cyber activity acts as a coping mechanism that alleviates his feeling of loss and echoes his nostalgia for the homeland. Situated in between the past and the future, Kowsar is positioned in a third surreal virtual space that enables the author to alleviate the isolation and alienation of life in the diaspora. Because of his active participation in the virtual community, he maintains an active presence in Iran. In his cyber imaginary community of Iranianness, Kowsar remembers the familiar past and is concerned that prospects for change in the homeland remain uncertain: It appears the mottos3 that we promoted in the early revolution are about to vanish: we now have neither the republic nor the Islamic values that they had claimed (to establish). Our dependence has also been violated by our nuclear dependence to the Russians. And we have forgotten about freedom long ago. (June 12,2009) ' Independence, Freedom and the Islamic Republic 22 The existence of the exiled bloggers in the diaspora is the shuttling between two layers of living, those of homeland memories and those of imagining their return to their homeland in the future. Nostalgic sentiments create the impression of living in a temporary state in the host country. By existing in a transient state, the exiled authors are living in between past and future. For these exiled authors, the past is not going to be forgotten, it is restored and remembered and makes the present alienation in the diaspora endurable. The tendency for the central discourse of the bloggers is focused around the events that happened in the past or what they hope takes place in the future. While restoring the past will not secure a future return of the authors to the homeland, the impact that such nostalgic nationalism has on the bloggers will transform their exilic experience in the diaspora. Behnoud's writing skills and his extensive use of metaphors express his concerns for the homeland in his blog as he remembers Iran's historical milestones through a nationalistic lens. Behnoud writes: My generation is the offspring of the revolution. For the same very reason, this generation decided to dedicate a glorious celebration to the next generation, to establish peace. A peace that doesn't eliminate the enemy, but it fosters friendship for its generation and for the future generations. My generation has struggled for peace. Outside the safe make up of the family, my generation has exercised friendship, a friendship that is free from any ideology. My generation has exercised peace but it has not distributed weapons and political statements in its underground life. My generation practiced music, danced and created joyful moments. My generation has been contented; it made jokes and memories and laughed at the oppression. My generation has chosen to live and it has eliminated the slogans of "Down with..." (June 22,2009) In his exilic isolation and in his imagined national community, Behnoud remembers 23 episodes of Iran's history when the popular uprising turned into the 1979 Islamic revolution and the monarchy was overthrown. The author restores the past memories in a contemporary context and imagines how the younger generation of Iranians may take a similar path in the future. In his virtual Iranian community, unable to return to the homeland, he relives 30-year old memories of his presence in the political development in Iran and recreates an imaginary role for the young generation of Iranians, discouraging them from revolutionary sentiments in the future. By focusing on the homeland, these exiled bloggers reconstruct a sense of national belonging in their blogs, thereby protecting their cultural roots. Boym (2001) states: "It is the promise to rebuild the ideal home that lies at the core of many powerful ideologies of today, tempting us to relinquish critical thinking for emotional bonding. The danger of nostalgia is that it tends to confuse the actual home with the imaginary one" (xvi). By creating an imagined national community through a nostalgic lens, the uprooted bloggers are looking for ways to validate their authentic Iranianness and maintain their Iranian roots in the diaspora. The imagined community of Iranianness in the cyber sphere assists these bloggers to alleviate the pain and loss of exile and renew their emotional bonds with a nation that existed in the past. By expressing the pain and despair of their alienated and uprooted exilic lives in the diaspora, exiled writers establish their national belonging and cultural roots in an imagined national community in their blogsphere. In a letter to the Iranian government, Sibiltala insists on the emotional bonds with the homeland and the importance of maintaining roots in the diaspora. Dear I. R., And what do you know about root? You would not need to fly from one place to the other in order to uproot yourself and displace your deep roots. The root is much stranger than these claims. It is impossible to liberate yourself from your origin. Root is something that 24 generates a sarcastic definition of freedom [...] what is it in the roots that confine you when you step into "freedom"? (May 16,2009) Banned from geographical territories of the homeland, the virtual community of the blogsphere is the authors' imaginary license to escape the realms of exile and return to a time that they were actually living in Iran, a revisiting of the past that in reality is impossible. Shahidian defines exile as a "redefining and remapping borders, homeland and host land. Borders become the sites of difference between interiority and exteriority; they are points of infinite regressions" (72). Disengagement with the life in the diaspora makes the present experience of these exiled bloggers an exhausted effort to join the forces of a time and space that no longer exists. Frustrated with her exilic life, Sibiltala is thriving for a familiar sense of belonging and intimacy in the diaspora. While the national community that allows her to preserve and protect her Iranian roots does not exist in the host country, she has created an imaginary native land in her weblog. Sibiltala writes: Roots and love of one's native land is "something" that one cannot describe. I assume that r in (i our despair is more rooted than these claims. If this is something that is impossible to define, it is also an eternal feeling [...] The moment one is connected to his/her homeland through improved technology and is able to look into one's bedroom window and the dead grandfather's bedroom window, a space that is "in between" one's mind and the advanced virtual technology, in between this space, tears fall down and one would comfort oneself. (May 15,2009) This joyful return to the bedroom window is associated with a sense of nostalgia and mourning for what has been lost or left behind. The acts of remembrance, expressions of regret and remorse drive the exiled subject to take action to restore an ideal homeland and to bridge the 25 gap between the past and the future. By writing about the past and remembering a Utopian homeland through a nostalgic lens, the blogs have become the meeting place for organizing virtual demonstrations and expressing solidarity and unity with the Iranians at home. The dual existence of the bloggers in exile in the cyber community and constant shift between past and present generate a state of living in between two time zones, neither of which exist in the present. The challenge of establishing new roots and a sense of belonging within a diasporic community beyond their cyber existence remains a real one for these exiled bloggers. Their hyperreal existence remains remote both from their homeland and from the new country in which they physically reside. 26 IV. Blogging and Diasporic Nationalism For these particular exiled Iranian bloggers, long distance nationalism in the diaspora is central to the debates in their blogsphere. Blogs have assisted exiled bloggers to invent a virtual community of like-minded Iranians. Such a cyber community of Iranians has facilitated the bloggers to engage, communicate and express nationalism with members of a nation that share a similar language, culture and roots in the homeland. The displaced bloggers construct a virtual national community in exile to restore their membership in a nation they were forced to leave behind. Moreover, belonging to a community of Iranianness is a significant component of the virtual community of the exiled bloggers. In their cyber community, the exiled authors claim a long distance political activist space in the diaspora on the net that is engaged in a diasporic nationalism. Benedict Anderson (1984) states that: It is quite possible to conceive of the emergence of the new imagined national communities without any one, perhaps all of them being present. What, in a positive sense, made the new communities imaginable was a half-fortuitous, but explosive, interaction between a system of production and productive relations (capitalism), a technology of communications (print) and the fatality of human linguistics diversity. (56) These exiled bloggers, as Anderson suggests employ the blogsphere and create an individualized imagined nation in their cyber existence while living in the diaspora. Detached from a national community to which they belonged in the past and uprooted from their authentic Iranian culture and heritage in the homeland, they long to join a familiar Iranian community and establish a surreal existence within the sphere of their weblogs. Said (2000) argues "[nationalism is an assertion of belonging in and to a place, a people, a heritage. It affirms the home created by a community of language, culture, and customs; and, by so doing, it fends off 27 exile, fights to prevent its ravages" (176). By writing about the homeland, its social and political life and its ongoing struggles for change, the exiled Iranian bloggers express a sense of nationalism through a nostalgic lens and feel they belong to an Iranian virtual nation in the cyber sphere, an instrumental sphere that is emotionally empowering and provides them with an effective means to look to the past in a homeland where they share common culture and heritage. As Erikson has argued: "Nations thrive in cyberspace, and the Internet has in the space of only a few years become a key technology for keeping nations (and other abstract communities) together. Nations which have lost their territory [...] nations which are for political reasons dispersed [...], nations with large temporary diasporas overseas [...] or nations where many citizens work temporarily or permanently abroad [...] appear in many guises on the Internet - from online newspapers and magazines to semi-official information sites and "virtual community" homepages. (77) By writing about national issues and expressing solidarity with the people and feeling the suffering of the people that they were forced to leave behind, these exiled bloggers assume the task of creating a virtual imaginary community of Iranian society and collective nationalism in the diaspora. In their imaginary national existence in the cyber space, these bloggers desperately seek to contribute to a homeland that existed in the past and does not exist in the new country. While cyber-nationalism in the diaspora has effectively altered the positioning of these exiled bloggers by including them in a national community, engaging in the ongoing process of transformation in the homeland and living a fantasy of participating in social and political life in Iran, they see themselves as involved and invested in crucial ways as if they are still residing in Iran. By providing a critique of the Western lifestyle as complacent and indicating that a certain conventional characterization of life is defined and dictated to her in the diaspora, Sibiltala 28 supports the notion that despite claims that the West is democratic, she claims it is not. Instead, Sibiltala expresses her nationalism for Iran and belonging to a community of Iranians. Unable to return to the homeland, Sibiltala claims a virtual space of Iranianness in her blogs. I live with Iran. This is the truth. I am happy that I live with Iran. If my life wasn't engaged with Iran, I could be a relatively happy and lifeless human being. I was like those that live in the West, those that are slaves of daily routines. Certain conventional characterization of life are defined and dictated to them: This is freedom, these are rights, this is democracy, and this is prosperity. (June 22,2009) As the author restores an imaginary existence in the homeland, she recreates a dual virtual presence that is neither in Iran nor in the diaspora. The exiled blogger overlooks her existence in the diaspora, and as a result, she confines her presence to a blissful but surreal longing for a time she belonged to a national community. What is it about nationhood and a homeland that existed somewhere in the past that has become a platform for expression of nationalism and Iranianness in the weblogs? The existing scholarship offers different interpretations of nation and nationalism. Benedict Anderson (1984) argues that nation "is imagined because the members of even the smallest nation will never know most of their fellow- members, meet them, or even hear of them; yet in the minds of each lives the image of their communion" (48). While these exiled bloggers long for membership in a national community and belonging to Iranian nation, they compensate their physical absence from the homeland by creating an imaginary community of Iranianness in the virtual sphere. Diasporic nationalism reflected in the cyber national community enables the exiled bloggers to fulfill their lives on an alien soil in exile, share concerns and aspirations with members of a community of Iranians and relive their past memories in a nation that is imagined in their virtual existence and doesn't exist in the host 29 country. In the case of the exiled Iranian bloggers, Sibiltala's diasporic nationalism and praise for Iranians exercising their democratic rights in the June 2009 presidential election is reflected in the following post. I am going to vote because I don't want to be alienated from this spirited nation (Iranians), and the good and bad uniqueness. I want my life and my culture to be tied to this nation. I am going to vote and while I am browsing at the news, newspapers and Internet, no matter how unfortunate we have been, I am grateful to God that at least we are alive. We are extraordinarily thriving. (June 12,2009) By writing in her virtual imagined community of Iranianness, Sibiltala expresses her emotional nationalist sentiments. By constructing a virtual community of Iranianness, she wants desperately to hold on to an imaginary nation to express nationalism, while remaining unconcerned about her life in the diaspora. The concepts of "home" and "the Iranian People" mobilized by these four bloggers become symbols of nationalism expressed through a nostalgic view of events such as solidarity with politically active Iranians at home, and condemnations of brutal state-sponsored crackdowns on peaceful protests in the homeland. Through the expression of a diasporic nationalism, the exiled bloggers form a united nationalist front that engages in the social and political life of the homeland through their cyber sphere. The presidential elections in 2009 led to an increase in online political organizing, which provided a further impetus for increasingly contentious controls on the web sites used by legitimate opposition contenders. These particular bloggers were constantly engaged with the election fallout in Iran and reacted towards the violent crackdown on peaceful protests of Iranian people by security agents of the Iranian government, engaging in these discussions as if they were in the homeland. As a result, the debates in the blogsphere focused on what was considered 30 Iranian and related to Iran, and life in the diaspora existed where there was an event and/or public announcement organized to express solidarity with the people of the homeland. In a testament to her strong sense of nationalism and commitment to the process of change in the homeland, Kar writes: We keep asking each other where in the moment of history we are standing. Responses are inspiring and they promise "change". Each of us have personal angles, we have particular judgments and share our views: It's time for the (Iranian) government to genuinely assess its shaken legitimacy [...] It is time to support all political prisoners [...] It is time for all Iranians, in Iran and in the diaspora to contemplate the criticisms and defiance that deferred solidarity and unity. (July 12,2009) The bloggers' constant reflections on the pain and suffering of Iranians and the expression of solidarity with like-minded activists in the homeland are manifestations of their cyber presence. It solidifies the idea that they are still living in Iran. Distressed and saddened by the violent crackdown of the Iranian government security agents against Iranians who had protested the election results and frustrated from his inability to contribute to the cause of his nation, Behnoud writes: Look, my Iran has turned Green. The streets represent the redness of its flag. And its people demonstrate the whiteness of denying violence. My Iran, Look... Green, White and Red Look, I have demanded the glory and respect for my Iranianness. (June 22,2009) In his cyber surreal existence, Behnoud's imagination is drawing the flag of his homeland. In the cyber imagined Iranian community, the author's creative mind is playing with the colors of the 31 flag: Green reminds his readers of a current civil green movement6 that was ignited by the Iranian people during the June 2009 presidential election in Iran. Red symbolizes the blood of people in the author's homeland who were killed by security agents of the Iranian government during the massive protests in June and July 2009 of the election fallout. White represents the peaceful nature of Iranians' uprising. In his imagined cyber presence, Behnoud seeks his Iranianness and claims a metaphorical authentic Iran as if he is carrying the flag in the streets of his homeland. The determination to invent a cyber community in the blogs is rooted in the exiled authors' desire to take steps towards preserving their national identity through a nostalgic lens. The bloggers' emotional bonds with the homeland along with the projected hopes for a future return generate a kind of virtual nostalgic nationalism that is not only an affirmation of Iranianness in exile but also a personal space to relive and remember the memories of homeland. "Exile is predicated on the existence of, love for, and bond with, one's native place; what is true of all exiles is not that home and love of home are lost, but that loss is inherent in the very existence of both" (Said 185). Such an intimate virtual national community however exists in a time that is constantly shifting between past memories and future aspirations of the homeland. "A life of exile moves according to a different calendar, and is less seasonal and settled than life at home. Exile is life led outside habitual order. It is nomadic, de-centered, contrapuntal; but no sooner does one get accustomed to it than its unsettling force erupts anew" (Said 186). In the virtual Iranian community, the persistent pain and inability to contribute to the cause of the nation is replaced with cyber expressions of unity and solidarity with the Iranians that are trapped under state-sponsored violent crackdowns in the homeland. In the aftermath of the 6 Green symbolizes supporters of Mir Hussein Mousavi, reformist contender in 2009 presidential election in Iran. 32 crackdown on reform supporters in Iran, Kar wrote in her post election commentary: Iranian diaspora is not missing a moment in the cyber sphere. Solidarity and unity with the post-election political determination have never been so widespread and unconditional. [...]. Since me and like-minded friends are forced out to exile, we have no other choice but to live "on-line" and respond to questions of the younger generation, for which we often don't have clear answers [...] everyone is busy with on-line exchange in the Facebook [...]. In one side of the scene (in Iran), hopeful young players have exposed their fragile bodies to oppression and violence, in the other side (in the diaspora), young players denounce the violence and inform the world [...]. I feel that the positive and negative energy of the huge events that is going on in Tehran's streets has captured my space in exile. (June 2009) Cyber-communications has assisted these particular exiled writers to pursue national discourse over developments in Iran and freely exercise an online dissent while remaining within the safe virtual borders of their weblogs in the diaspora. During the June 2009 presidential election campaign, in the virtual Iranian community of their weblogs the exiled bloggers expressed support for their candidate of choice, recreated the joyful street celebrations, shared views after the election fallout, expressed solidarity with the protesters that accused the Iranian government of vote rigging and condemned the loss of lives during crackdown of the Iranian government against peaceful protests. Similar to other public virtual spheres, the weblogs carry the tension of differentiating between the best methods to protect the Iranians from the violent crackdown and the best contributing forces to the social and political movement in the homeland. Interestingly, similar approaches were adopted by the bloggers during the election fall out and massive protest demonstrations in Iran. Behnoud reflects on participation of Iranians in the election and speaks as if he is actually participating in the process in Iran: 33 Good for you that waited in the long lines and voted and protected your vote from vigilantes that want to hide behind few votes of their supporters. I was hoping I would not see such a scene in my lifetime. You witnessed that we did everything within our abilities to persuade people to participate and vote and have the final say. The June 12th vote turned into the most politically motivated vote of (Iran's) contemporary history. Whatever the consequences bring about depends on the nation's initiatives and statesmen's dignity and pragmatism. (June 12,2009). Behnoud's blog exhibits the author's constant engagement with the national issues and promotion of nationalism among Iranians. For Behnoud, blogging has emerged as an instrumental emotional mechanism to physically remain in the diaspora and maintain ties with an imaginary Iran in his weblog. Karim (2006) states that "there appears to be an attempt by diasporic participants in cyberspace to create a virtual community that attempts to eliminate the distance that separate them in the real world" (10). Through his meticulous attention to the development in the homeland in his virtual Iranian community, Behnoud expresses nationalistic enthusiasm over pre-election civil participation in Iran in his virtual Iranian community: In an atmosphere of deceit, lie and hostility; "Green hope, Green truth, Green enthusiasm and Green friendship" (are born). Friendship and intimacy had faded away for so many years. And once again, you (youth) reminded us of the sweet taste of Green hope in the blue sky of truth and honesty. I praise your Green faithfulness because even the emperor of lies and its fraudulent and deceitful system were unable to steal your Green dignity. I admired your defiance when you challenged the emperor of deceit and (you) confronted him in order to attest that your presence signify denial of his ideology. You spoke for us. You and your presence embody the Green security in the garden of aspiration. I 34 wholeheartedly believe in your truthfulness and innocence. (June 20,2009) Through a romantic and nostalgic nationalism, Behnoud claims an Iranian space and a belonging to a nation. Behnoud is remembering the nation, as it existed in the past. Edward Said (2000) notes that "exiles feel an urgent need to reconstitute their broken lives, usually by choosing to see themselves as a part of a triumphant ideology or a restored people" (177). Through his virtual presence, Behnoud seeks to compensate his absence from the homeland and restore his lost homeland in his weblog. Behnoud speaks with and for Iranians and as a result not only engages with his audience but also feels that he is still a member of the Iranian nation. Interestingly, this cyber discourse frees Behnoud from isolation and alienation of exile in a foreign land. More importantly, this virtual community has been able to foster the bloggers' existence in the remembering, reliving and recreating non-physical home borders and express a cyber-nationalism as if they still have a presence there. In Behnoud's blog, an anonymous reader writes to the author as if they live in the same country: Dear Behnoud, I am very optimistic and hopeful for the future of our beloved Iran. The existence of these brutal enemies rests in the darkness of human's ignorance. This darkness has long begun to leave this Arian land [...] my fellow citizen: If I rise, if you rise, we can take revenge of beloved Neda's blood that raises our call for freedom from these enemies. (July 6,2009) The ongoing interaction of the exiled bloggers with their readers and constant long distance dialogue over the issues related to social and political process in Iran demonstrate the complicated layers of a diasporic nationalism in exile in the diaspora that is fascinating though problematic for the authors' life in the diaspora since it totally disengages them from the reality of the society they are currently residing. It is important to note that Behnoud's disengagement 35 with the host society may be involuntary and the European country may have disengaged the non- European and non-white exiled biogger. This raises the important question is the onus of such disengagement on the migrant or the host society, although addressing this issue will not be within the scope of this research. More importantly, the safe existence of these particular bloggers in the diaspora has secured them the privilege of benefiting from advanced cyber technologies to mobilize support for the homeland and causes of Iranians in times of social and political crisis. During the June 2009 presidential election fever, the weblogs were filled with expressions of long-distance nationalism and the responsibility the authors felt in helping Iranians in their struggles for democratic change in Iran. In a blog post before the June 2009 presidential election, Kar writes: Despite the existing cynicism, I am going to vote. Under the current circumstances, I wonder whether the candidates would be able to fulfill their promises. I am fascinated with their claims and over-statements. For me, voting in the current situation is a sort of pronouncement on Iranian identity. It is like renewing your Iranian passport. This campaign has attracted my attention. (June 2009) The constructed diasporic nationalism in the blogsphere is an attempt to express a sense of belonging to the homeland and to corresponding national affiliations. Eriksen (2006) notes that "in a "global era" of movement and deterritorialisation, the Internet is typically used to strengthen, rather than weaken, national identities" (77). The national identity and the sense of belonging to a nation that these exiled bloggers seek in their cyber imaginary Iranianness community position them in an in-between state that restore the past that does not exist in the diaspora and diverts their attention and dedication to their new life in the diaspora. Said (2000) suggests that "exile, unlike nationalism is fundamentally a discontinuous state of being" (177). 36 The temporary state of exile and the fascination with the memory of homeland in a different time and space haunts the imagination of the exiled bloggers and manifests itself as a sense of nationalism in the virtual Iranian community. In anticipation of participation in the June 2009 presidential election in Iran, Sibiltala expresses her national sentiments for Iranians living globally: Our people have high spirits. Five buses are traveling to Ottawa to vote. It will take these folks twelve hours to get to the embassy [...]. The people of our country are very cool. I hope they always remain cool. Their prosperity lies in this morale. (June 12,2009) Blogging has opened avenues for Sibiltala to engage in the homeland's political process while living in the diaspora. Karim (2003) notes that "by employing new communications, migrant communities endeavor to make homes (even if "temporarily") in milieus that are away from the home" (9). Assuming her exile as a temporary state, Sibiltala has created a virtual community of Iranianness and is constantly engaged with a cyber presence that is obsessed with national issues and the process of democratization and civic participation of Iranians in Iran and in the diaspora. More importantly, while the presence of Iranianness in Sibiltala's blog is overwhelmingly evident, she does not reflect similar attachment to the social and political advancement of the host country that has facilitated such an independent and liberating national existence in the cyber sphere. We are left to ponder the nature and extent of this obsessive commitment and loyalty to an imaginary national community in the cyber space, so that a blogger such as Sibiltala who has spent a greater part of her adult life in the diaspora prefers to contribute her virtual existence in promoting civic participation and democratization in the homeland than pursuing socio-political life in the host country. Before Iran's presidential election, Behnoud expresses his fascination with the civic 37 participation of Iranians in the election campaigns and their engagement in a democratic process in the homeland. Behnoud's diasporic nationalism and engagement with the process of change in the homeland appears in this blog post: The events of these days represent the capacity of Iranian society and this generation. It also affirms the regime's ability to tolerate democracy, freedom and vitality. Such characteristics make Iran a unique model for the Middle East. This is how power will be established in the country, not with the slogans and heroic and epic propaganda. Showcasing this capacity petrifies the opponents. (June 09,2009) By writing in his weblogs, Behnoud is constantly shifting from his exilic solitude in the diaspora to membership in a wider social sphere of Iranians, whether at home or abroad, a population that shares the same past and desires the same changes in the future. According to Said (2000) nationalisms are about groups, but in a very acute sense exile is a solitude experienced outside the group: the deprivations felt from at not being with others in the communal habitation (177). The pain and loss of separation from membership in a national community and the gap between the actual homeland and the exilic existence in the diaspora is occupied by constant engagement with a life in Iran that the authors are unable to claim. For Behnoud, belonging to a community of Iranianness, in Iran and in the diaspora is compensation for separation from a nation that he mourns. When the bloggers engage in cyber political activism, the isolation of exile and the alienation of the new country disappear and a familiar sphere of intimate presence on the net appears. As a result, the diasporic experiences of the exiled bloggers are shaped by national concerns, looking to the past and mourning the life they have left behind. Unable to return to the homeland and deprived from actual physical presence in Iran, these 38 bloggers launched an intense long-distance political activism to assist the information flow to a homeland that was entangled in a circle of censorship and internet filtering and offered their solidarity and sympathy for their colleagues and Mends that were detained or beaten by the government's security agents. In their virtual community of Iranianness, the exiled bloggers assume the responsibility of healing the pain and suffering of Iranians and not only express solidarity with the nation but also delegate support to assist the transition in the homeland. In his cyber sphere, Kowsar is constantly engaged in a long-distance solidarity with his colleagues in Iran and expresses a sense of responsibility towards a nation that he was forced to leave behind. Leaving his homeland, family and a flourishing profession behind, blogging facilitated Kowsar to enjoy freedom of speech in the new country and to support the cause of Iranians that seek change. In order to condemn the imprisonment of journalists and activists in the aftermath of June 2009 election fall out and to publicize the misery of his jailed colleagues in prison in Iran, Kowsar launched a cyber campaign and published the news releases, statements and petitions of the international media organizations in his weblog. While Kowsar is deprived from physical presence in the homeland, he expresses concern over imprisonment of his friends and criticizes violations of freedom of speech in Iran in this post: Naturally, I am worried about my friends and journalists in Tehran. Activities of the foreign media are banned, Iranian journalists could have provided good resources for the foreign correspondents; confirm or deny the truth behind the events. But it seems that the regime is not even tolerating this small assistance. This regime it seems is based on sheer lies and fears the truth. (June 22,2009) For Kowsar, forced to abandon his flourishing profession in the homeland, blogging in exile in a democratic country is a fundamental means to create an imagined Iranian community 39 in this blog, to engage with like-minded Iranians and to promote his national aspirations. Through his cyber sphere, Kowsar enjoys new opportunities in exile to promote his views about freedom of speech in the homeland, while his weblog does not reflect the author's view on similar issues in the host country. What is the nature of this diasporic nationalism that echoes more loyalty towards an imaginary homeland in the cyber sphere than virtual commitment to the new country that has provided safety and security for them? Kowsar's concerns over national issues and lack of efficient leadership in Iran are reflected in this post: Our county has drastic disparities with India. The most important difference is that we don't have a "Gandhi"! But even if we had one, how efficient and well was he able to denounce adoption of violence while viciously harassing and manipulating the nation? (June 14,2009) As a result of living in an imaginary community, the prospects of integrating into the host society in the diaspora remain vague. By writing in their virtual Iranian community, these exiled writers seek belonging to a familiar space, an intimate home to which they used to belong. Instead they assume their exile is temporary and they take their diasporic experience in the new country for granted. Gokcen Karanfil (2008) suggests that "transnational media consumption promotes communications and frees the migrant from the pincers of a 'frozen image' of the homeland" (29). As a result, the present life in the diaspora is obscured in the blogs and remains outside the authors' cyber existence. Sibiltala's virtual exilic existence in the diaspora resonates with her euphoric memories of the homeland that she left behind long ago. She candidly speaks out about her exile and nationalism in the diaspora in her imagined community of Iranianness in the virtual sphere. In her cyber sphere, Sibiltala writes of Iranians' spirit: Iran reminds me of a different world, a world that its people are struggling with so many 40 1 challenges, they are relentlessly besieged, yet, they are alive. They survive and thrive. Iranians are not depressed; they are not happy slaves of a conventional and predictable system. Some may argue that by presenting life (in the diaspora) in this way, I am not making any sense. Some may even claim that life is rosy here. I realize it is not. I swear to your life that I live with Iran every day. But I still can't figure out where the human being's prosperity lies in. I am sure that human being's prosperity is not what I am witnessing everyday in this free, democratic and wealthy country that I live. (June 12,2009) Far from alienation and isolation in the diaspora, this imagined Iranian community on the net is a safe, intimate and familiar cyber presence for Sibiltala and allows her to look into the familiar past in Iran, offers common interests and often represents symbols of Iranianness such as the Persian language, culture and heritage. Said (2000) states "[fjor an exile, habits of life, expression, or activity in the new environment inevitably occur against the memory of these things in another environment. Thus both the new and old environments are vivid, actual, occurring together contrapuntally" (186). But Sibiltala's cyber space is a community of Iranianness and she is not concerned about her position in the diaspora and belonging to the new society. While exiled bloggers enjoy the luxury of having access to the advanced communication tools and cyber media in a developed country in the diaspora, the constructed imagined community of Iranianness in their weblogs is only engaged with their Iranian identity and belonging to their native land. It is important to note that the responsibility of the host society in the migrants' social and political engagement is an important issue that needs future research. More importantly, while the weblogs of these particular displaced writers proves that national heritage and cultural roots are inseparable element of their cyber existence, they do not demonstrate similar interest and attention to their new hybrid identities in the diaspora. The on- 41 line community has become a meeting place of Iranians and a virtual examination of Iranianness and expressions of Iranian nationalism in the diaspora. The imagined community of exiled bloggers has created a virtual front for political debates and exchanging views on what is taking place in the homeland. Kowsar writes: One of the mistakes that some of my colleagues make is to have little expectation from of the politicians. Basically, the problem is rooted in assuming given that our country is not following Switzerland (political) model, the journalist should not expect to be able to question the politicians' accountability [...] We are responsible for allowing the serial killings to remain unsolved [...]. We are the ones who forget. We, the journalists are not doing our true professional work. Our politicians do whatever they like and they know that we are not going to question them. (June 10,2009) Cyber-presence has facilitated Kowsar's claim to a space and a sense of virtual national presence in Iran. Kowsar's imagined community is a virtual sphere where he is able to articulate his nationalism in a foreign land. It is, however, important to note that most of these exiled bloggers are arguably living in safer conditions in the new country than in their homeland and their safety and security is not at stake as it was in Iran. Similar to some of the other bloggers in this research, Kowsar has been enjoying the safety and comfort of diasporic life while employing his cyber presence to communicate with Iranians and participate in socio-political developments in the homeland. While Kowsar expresses concerns for the political future of Iran, he does not acknowledge that the time and the actual place that he was forced to leave has changed and is not going to be the same again. This reluctance to leave the past behind, and to forge a connection with the new country suggests that the exiled blogger is desperately trying to hold on to an authentic Iranian identity in the diaspora. 42 In the aftermath of the June 2009 presidential election fallout and during the massive crackdown of Iranian government's security agents against peaceful protests of Iranians, these particular bloggers revolutionized their virtual on-line community and became the voice of the Iranians who had been silenced. They praised the bravery and resistance of Iranians in opposing the election fraud and promoted non-violent dissent. The authors' expressions of solidarity, frustration, loss and pain for their compatriots in Iran filled cyber space as if the bloggers' cyber Iranian front was fighting for the same cause as those in the homeland. Behnoud writes: The eventful Khordad ended with a Red and Green Saturday. 'Death', a word that I always avoided and 'bullet' had become a worn out word [...]. Years have passed since 'blood' was a part of the dusty revolutionary songs and 'martyr' had been purged from the nation's daily conversations, once again (they) have emerged from our literature. (June 22,2009) The exiled bloggers in the diaspora are not only engaged in political activism in an on-line sphere but also are taking part in a wider communication with their readers. The debates in the blogs are not neutral and more importantly, the readers also play an active role in driving the bloggers' national sentiments. Some readers argue that the exilic life in the diaspora has transformed the authors' political views and thus they question their arguments; others feel close to the bloggers' ideology and welcome their solidarity and sympathy. The aftermath of the June 2009 presidential election in Iran and massive crackdown of Iranian government against the peaceful protests, similar to other exiled bloggers, Behnoud expresses solidarity and unity with Iranians in the homeland in this post: I whisper in your ears: You are not alone. Our fears disappeared. [...] We are not alone [...] we are Iran [...] Give me your hand. Tell me your name. (Aug. 06,2009) In response to Behnoud's blog post on solidarity with the Iranians in the homeland, an 43 anonymous reader comments: How long do I have to read about unity of this nation and shed remorseful tears to seek solidarity with Iranians? [...] I think about those in exile and how they painfully live through (the uprising) through YouTube videos and Facebook photos and how they agonizingly seek one second of solidarity with this beautiful Green wave, let alone shaking the hands of fellow citizens or knowing the names of these compatriots. (Aug. 06,2009) Engaging in the blogsphere in a discourse on social and political life with the Iranians inside Iran has enabled Behnoud to maintain his membership to an Iranian community and to assume he is bearing the long-distance responsibility of uniting and encouraging a nation that doesn't exist in the diaspora. As Renan (1990) argues, "[s]uffering in common unifies more than joy does. Where national memories are concerned, grief is of more value than triumphs, for they impose duties, and require a common effort" (19). The comments and the on-going cyber dialogues over the social and political life in Iran represents these bloggers' persistent desire to belong to a community of Iranianness and to play a role in the process of change in the homeland. The bloggers' invented national community results in claiming a virtual territory that has captured the past, ignoring their present existence in the diaspora. Moreover, the cybercommunity of the exiled bloggers resonates with their perseverance in re-living past memories, feeling nostalgic for what they have left behind in the homeland. While the exiled bloggers invent a familiar feeling of home through their memories and hopes for a return to the homeland, they are wandering in a virtual territory that is located neither at home nor in the diaspora. Kar's constant focus on national issues and expressions of nationalism indicate her desire to recreate her past in her present life in exile in the diaspora: Past frequent opposition to a regime that has banned my participation and engagement of 44 1 like-minded individuals has been documented. I am still opposing those ideas (of the government). The (presidential) candidates do not represent us. But during the years that I have lived in the diaspora, I have reached to this recognition that the existing democracies' path to where they are now has been a long one. Perhaps, under current circumstances, participation and voting in an election that the players are trying to bring the process closer to the global standards are the sole responsibilities of every Iranian. (June 2009) Unable to return to her cultural roots and heritage, Kar focuses on recreating an imaginary Iran in the blogsphere, her membership to a national community and engages in a political activism in the diaspora. Said (2000) argues that "a state of exile [...] is designed to reassemble an exile's broken history into a new whole - is virtually unbearable, and virtually impossible in today's modern world" (177). For Kar, diasporic nationalism appears to be a coping mechanism to confront the experiences of isolation in exile and alienation in the diaspora. Cyber presence of the exiled bloggers echoes an active virtual presence in Iran as if they still live in the homeland though this personal sphere fails to engage in a discourse over the diasporic experience in the host country. Such diasporic nationalism can construct images of an ideal home in an imaginary virtual space that lacks a real emotional and physical bond with the present and instead is entangled in a web of past memories that cannot be framed within the image of the homeland that they have left behind. While Sibiltala is far from the geographical borders of her homeland, she is expressing sense of nationalism while living in Canada. By making a distinction between homeland and nationalism and implying the political responsibility of nationalism and love of her homeland, exiled Sibiltala writes: Dear I. R., Every one, our parents, schools and leftist universities taught us to maintain an 45 I !i ?l "internationalist" position. And we are still preoccupied with our country. Sometimes I try 1 to justify it as a sense of nationalism. But it is not. It is one's love of the homeland! Dear I.R. Do you know what one's love of the homeland is? (May 15,2009) In her blog, the author proudly defends her national identity, indicating a strong sense of nationalism and belonging to a homeland that she left behind at an early age. These exiled authors have assumed the responsibility of echoing common goals and shared concerns over the fate of a nation they were forced to leave behind. Sibiltala's blog is a virtual community that puts numerous national subject matters in an open discussion forum and engages in social and political life in Iran. Unlike the other three bloggers that have been exposed to Iran's statesponsored censorship and imprisonment, Sibiltala has not experienced similar restrictions and has been able to enjoy the safety and security of expressing her views on national concerns with the protection of democratic values and advanced cyber media communications in the host country. These exiled bloggers create an imaginary Iranian community in their weblogs to restore the memory of the homeland and to express nationalism in a virtual time and space, which existed in the past. While time and space may have remained static in the lives of the exiled bloggers and nationalistic sentiments may have not been diminished in the course of time, it is a fact that Iran, Iranians and their social and political demands have changed and rather transformed in the course of time. While providing migrants with equal opportunities and the role that the host society plays are essential means in their engagement and socio-political integration ants into the receiving country, ultimately, while this research acknowledges that the host countries play a key role in the migrants' integration process, the exiles' nostalgic view of the homeland and diasporic nationalism is problematic since those who exist in this realm 46 1 cannot move forward and integrate into the new society in the diaspora. As a result, they create a new space, in-between the created imagined cyber community and their lives in the diaspora. This temporary existence in the diaspora along with cyber presence in a virtual imagined national community challenges social and political integration into the host society. Looking to the past through a nostalgic lens and hoping for an end to the temporary residence in the new country and return to the homeland shuttles the exiled bloggers between the past and the future and as a result, the present remains absent from their lives in diaspora. At this extreme the exile can make a fetish of exile, a practice that distances him Or her from all connections and commitments. To live as if everything around you were temporary and perhaps trivial is to fall prey to petulant cynicism as we; as querulous lovelessness. (Said 183) For the exiled blogger, residence in the new country appears to be a transitional period that enables them to hope for a future return to the homeland. Ultimately, such obsession with the homeland and expressions of diasporic nationalism in a cyber sphere position the exiled bloggers in a dual existence: the cyber community of their blogs and their diasporic experience in the new country. Assessing the possibilities of change in the homeland disengages the bloggers from the developments of life in the new country and isolates them from attention to their new social identity in the diaspora. More importantly, the exiled bloggers' dual imaginary existence in exile poses challenges for their settlement and integration into the new society. Said (2000) suggests that "no matter how well they will do, exiles are always eccentrics who feel their difference (even as they frequently exploit it) as a kind of orphanhood" (182). In this imagined national on line community, traces of the bloggers' existence in the diaspora have been eliminated in order 47 to maintain the virtual space as Iranian as possible. The relationship between the exiled bloggers and the host country is significant in understanding the dynamics of the authors' contribution to the host society and their commitment to the social and political process in the new country. While the bloggers' creative ways of communicating with the Iranians through their cyber communities are significant in restoring nationalist sentiments and expressing nationalism, recognizing the impact of such communities on active participation of this population in the process of change in the host society is of distinct importance. While the bloggers claim a cyber existence in the community of Iranianness and assume that they have preserved their cultural roots and have treasured their national heritage, what seem to be at stake is a total disengagement of the exiled bloggers from life in the new country and the actual absence of diaspora in the imaginary virtual sphere. Said (2000) suggests that "the achievements of exile are permanently undermined by the loss of something left behind forever" (173). While most of the existing scholarship praises the instrumental role of transnational communication technologies, claiming that they "have effectively changed the positioning of the migrants by including them in the ongoing process of change in the homeland", (Eriksen 1), it is important to note that transnational networks, especially these weblogs may act as effective means to impede the integration and settlement process of migrant communities. The risk of diasporic nationalism through a nostalgic lens is that it distracts the authors from reality of living in the diaspora and complicates the dual existence and belonging of the exiled migrants and results in total disengagement from the diasporic experience and ultimately delays the integration process into the host society. It appears that forced migration and absence from Iran is the priority of these exiled bloggers and there is little reference to life in their blogs. Outside the virtual boundaries of the cyber community of Iranianness and the reality of life in the diaspora, a gap emerges 48 between their past life and future return. The terrains of this imagined community do not share borders with the new country in which the exiled bloggers reside. Instead, the kind of long-distance nationalism that is expressed in this sphere posits a nostalgic view of life in the past, glossing over the present. Thus we need to question whether such dedication to diasporic nationalism and virtual long-distance activism is an urgent desire to return to a homeland that seeks transition to democracy or is this nostalgic impulse in these particular weblogs a coping mechanism to alleviate the pain of exile and create a sense of belonging for subjects who feel a drift. More importantly, the self-claimed personal space of the blogsphere that reconstructs an idealized home creates an emotional gap between the migrant and the diaspora. The exiled bloggers have drawn national borders within their imagined cyber presence, well protected by a relentless sense of nationalism that will not penetrate the diasporic experience. In addition, in the imagined virtual community, the diaspora is the strange foreign land and the virtual blogsphere is the intimate heritage and national land. Shahidian (2000) suggests that "exiles are physically in one land, and intellectually in another" (72). Their obsessive recalling of past memories and their active participation in debates over the social and political future of the homeland in the blogsphere engages the authors so much so that their cyber existence overshadows their presence in the diaspora. Although the term nostalgia may not explicitly appear in the blogs, the authors' expressions of nationalism and fascination with Iran and Iranianness are clear indications of the fact that they are constantly dwelling in a longing for the homeland and its memories within the boundaries of their cyber imagined communities. Said (2000) argues: "The exile knows that in a secular and contingent world, homes are always provisional. Borders and barriers, which enclose us within the safety of familiar territory, can also become prisons, and are often defended beyond reason and necessary. Exiles cross borders and break barriers of thought and experience" (185). This cyber-imagined community is a place that frees them from their exilic alienation in the diaspora. While the onus of such separation and isolation is not entirely on the exiled bloggers and the host countries can impede and/or motivate their engagement into the society, they however appear to be totally disengaged from their lives in the diaspora. It seems as though the present life in the new country is withdrawn from their Iranian existence, in which they live in their blogs' invented nation. By disengaging with the civil society and democracy in the new country, the community of the Iranian bloggers in exile is actively promoting similar achievements in the homeland. 50 V. Conclusions Blogs have transformed the way exiled Iranian authors express nationalist sentiments towards their homeland while living in the diaspora. The exiled bloggers, on whom I've focused here, Behnoud, Kar, Kowsar and Sibiltala, have created a cyber community of Iranianness in order to engage in the social and political development in the homeland and to express concerns for a homeland they were forced to leave behind. These particular exiled authors echo nationalistic sentiments through a nostalgic impulse. The cyber community of Iranianness, as a virtual sphere, assists these exiled bloggers to maintain a national belonging to their native land: it facilitates social and political debate through longing of the past by sharing memories with the community and a hope for a future return. By creating a community of Iranianness in a cyberspace of the blogsphere, one of the main objectives for these authors is to create a sense of belonging to a national community so they can cope with the isolation and alienation of living in a foreign country. What I have identified in this research is that with the creation of this cyber community comes a great deal of long-distance political activism and expressions of nationalism while living in the diaspora. The focus for these exiled bloggers on the future and their return to Iran assumes their state of exile is a temporary one as they project an anticipated hope for a future change for Iran as a nation and for the bloggers' ultimate return to the homeland. I argue here that the diasporic experience of the exiled bloggers is absent from this virtual existence, this community of Iranianness. The fascination with remembering and reliving the past by recalling memories and projecting hope for future return makes invisible their present existence of living in the diaspora and overlooks the possibilities and opportunities of establishing new identities and communities of belonging in the diaspora. More importantly, constant engagement with national issues and concerns about the future of Iran results in total 51 disengagement with their lives in the diaspora and challenges the process of social and political integration in the new country. Given the scope of this research project, I have not been able to address the role in which the host countries play in the disengagement of the exiled bloggers from the new society, however, I have acknowledged that the sense of isolation and alienation that the exiled bloggers experience in the Western diasporas may be the outcome of conditions of migrancy and with issues such as how race, racism, the colonial encounter and the marking of the 'other' have played out in their diasporic reality. As a result the onus of integration is not entirely on the migrants and the host country plays a key role in their integration process. An assessment of the exiled bloggers' commitment to the construction of an online national community and to restoring an imaginary homeland in their weblogs is a key factor in understanding the dynamics of emigre diasporic nationalism and its impact on migrants' exilic experience in the host country. 52 1 VL Works Cited Anderson, Benedict. Imagined Communities: Reflections on the Origin and Spread of Nationalism. Revised ed. New York: Verso, 1984. Print. Behnoud, Masoud. "Don't Worry." Weblog post. Collection ofEssays. N.p., 09 June 2009. Web. 14 June 2009. <www.masoudbehnoud.com>. Behnoud, Masoud. "Who Is Responsible for the Coup?." Weblog post. Collection of Essays. N.p., 13 June 2009. Web. 17 June 2009. <www.masoudbehnoud.com>. Behnoud, Masoud. "Green Hope." Weblog post. Collection ofEssays. N.p., 20 June 2009. Web. 25 June 2009. <www.masoudbehnoud.com>. Behnoud, Masoud. "Golrokh! Draw Your Tears Green." Weblog post. Collection ofEssays. N.p., 22 June 2009. Web. 24 June 2009. <www.masoudbehnoud.com>. Behnoud, Masoud. "Don't Worry." Weblog post. Collection ofEssays. N.p., 21 June 2009. Web. 27 June 2009. <www.masoudbehnoud.com>. Behnoud, Masoud. "Thinking: Hawaiian's Crime." Weblog post. Collection of Essays. N.p., 06 July 2009. Web. 20 July 2009. < http://www.masoudbehnoud.com>. Behnoud, Masoud. "Your Eyes Say You Have Cried." Weblog post. Collection of Essays. N.p., 09 July 2009. Web. 20 July 2009. < http://www.masoudbehnoud.com>. Behnoud, Masoud. "I Write For You." Weblog post. Collection of Essays. N.p., 06 Aug. 2009. Web. 08 Aug. 2009. <www.masoudbehnoud.com>. Boym, Svetlana. The Future of Nostalgia. New York: Basic Books, 2001. Print Eriksen, Thomas Hylland. "Nations in Cyberspace". Media-Anthropology.Net. 27 March 2006. Web. 14 May 2009. "Global Internet Filtering-Iran." Opennet.com. OpenNet Initiative, 16 June 2009. Web. July 01, 2009. <http://opennet.net>. Golkar, Saeid. "Politics in Weblogs: A Safe Space For Protest." Iran Analysis Quarterly A Publication of the Iranian Studies Group at the Massachusetts Institute of Technology 2.3 9 (2005): 49-59. Web. 18 May 2009. Hakimzadeh, Shirin. "Iran: A Vast Diaspora Abroad and Millions of Refugees at Home." migrationinformation.org. Migration Information Source, Sept. 2006. <http://www.migrationinformation.org>. Web. 14 July 2009. Kar, Mehrangiz. "The Election and I." Weblog post. Mehrangiz Kar. N.p., June 2009. Web. 12 53 July 2009. <http://www.mehrangizkar.net>. Kar, Mehrangiz. "View from the Distance." Weblog post. Mehrangiz Kar. N.p., June 2009. Web. 28 June 2009. < http://www.mehrangizkar.net >. Kar, Mehrangiz. "Transition Time." Weblog post. Mehrangiz Kar. N.p., 12 July 2009. Web. 22 July 2009. < http://www.mehrangizkar.net >. Karanfil, Gokcen. "The Message of Transnational Media: Changing Notions of 'Threat' and Opportunities for Cultural Diversity" 3.1(2008): 24-33. Global Media Journal: Mediterranean Edition. Web. 8 July 2009. Karim, H. Karim. Mapping Diasporic Mediascapes. The Media of Diaspora. Ed. Karim H. Karim. 1st. ed. Montreal: Routledge, (2003): 1-17. Print. Karim, Karim H. "Virtual Diasporas and Their Communication Networks". Virtual Diasporas and Global Problem Solving Project Papers. The Nautilus Institute. 2006. Web. 13 June 2009. Kelly, John and Etling, Bruce. "Mapping Iran's Online Public: Politics and Culture in the Persian Blogosphere." Berkman Centre Research Publication 2008-01 (2008): 2-29. Web. 05 June 2009. Kowsar, Nikahang. "Post Number 6000." Weblog post. Nikahang Kowsar Memoir. N.p., 24 May 2009. Web. 27 July 2009. < http://nikahang.blogspot.com>. Kowsar, Nikahang. "Twelve Years Passed." Weblog post. Nikahang Kowsar Memoir. N.p., 24 May 2009. Web. 27 July 2009. <http://nikahang.blogspot.com>. Kowsar, Nikahang. "Happy Birthday Dad!." Weblog post. Nikahang Kowsar Memoir. N.p., 25 May 2009. Web. 26 July 2009. < http://nikahang.blogspot.com>. Kowsar, Nikahang. "The Winners and The Losers." Weblog post. Nikahang Kowsar Memoir. N.p., 10 June 2009. Web. 26 June 2009. <http://nikahang.blogspot.com>. Kowsar, Nikahang. "Let's Denounce Violence." Weblog post. Nikahang Kowsar Memoir. N.p., 14 June 2009. Web. 18 June 2009. <http://nikahang.blogspot.com>. Kowsar, Nikahang. "Big Shock." Weblog post. Nikahang Kowsar Memoir. N.p., 12 June 2009. Web. 18 June 2009. <http://nikahang.blogspot.com>. Kowsar, Nikahang. "I Am Tired." Weblog post. Nikahang Kowsar Memoir. N.p., 22 June 2009. Web. 25 June 2009. <http://nikahang.blogspot.com>. Kowsar, Nikahang. "Six Years of Separation." Weblog comment. Nikahang Kowsar Memoir. N.p., 26 June 2009. Web. 27 June 2009. < http://nikahang.blogspot.com>. 54 1 Kowsar, Nikahang. "40 Days After Neda." Weblog comment. Nikahang Kowsar Memoir. N.p., 30 July 2009. Web. 1 Aug. 2009. < http://nikahang.blogspot.com>. Renan, Ernst. What is a Nation! Nation and Narration. Homi Bhabha ed. London: Routledge, 1990. Print. Said, Edward. Reflections on Exile. Cambridge: Harvard University Press, 2000. Print Shahidian, Hammed. "Sociology and Exile: Banishment and Tensional Loyalties." Current Sociology 48.2 (2000): 71-79. Web. 13 July 2009. Shirazi, Farid. Social Networks Within Filtered ICT Networks: A Case Study Of The Growth Of Internet. International Federation for Information Processing. Toronto: Ryerson University, 2008: 299-317. Web. 10 July 2009. Sibiltala. "What Do I Do But to Bear the Oppression of the Fate!" Weblog post. Sibiltala. N.p., 16 May 2009. Web. 20 July 2009. <http://sibiltala.blogspot.com>. Sibiltala. "We Are Alive! We Are Very Alive!" Weblog post. Sibiltala. N.p., 12 June 2009. Web. 27 June 2009. < http://sibiltala.blogspot.com>. "41 journalists in prison after a month of unrest." Reporters Without Borders. 12 July 2009. Web. 26 July 2009. <http://www.rsf.org>. 55
https://openalex.org/W2970537246	https://escholarship.org/content/qt4h37z2tm/qt4h37z2tm.pdf?t=qan4xz	English	null	The “rights” of precision drug development for Alzheimer’s disease	Alzheimer's research & therapy	2,019	cc-by	14,487	UC San Diego UC San Diego Previously Published Works Title The “rights” of precision drug development for Alzheimer’s disease Permalink https://escholarship.org/uc/item/4h37z2tm Journal Alzheimer's Research & Therapy, 11(1) ISSN 1758-9193 Authors Cummings, Jeffrey Feldman, Howard H Scheltens, Philip Publication Date 2019-12-01 DOI 10.1186/s13195-019-0529-5 Peer reviewed UC San Diego UC San Diego Previously Published Works Title The “rights” of precision drug development for Alzheimer’s disease Permalink https://escholarship.org/uc/item/4h37z2tm Journal Alzheimer's Research & Therapy, 11(1) ISSN 1758-9193 Authors Cummings, Jeffrey Feldman, Howard H Scheltens, Philip Publication Date 2019-12-01 DOI 10.1186/s13195-019-0529-5 Peer reviewed UC San Diego REVIEW Open Access © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Keywords: Alzheimer’s disease, Drug development, Clinical trials, Biomarkers Keywords: Alzheimer’s disease, Drug development, Clinical trials, Biomarkers financial burden of AD in the USA will increase from its current $259 billion US dollars (USD) annually to more than $1 trillion USD by 2050 [5]. The cost of AD to the US economy currently exceeds that of can- cer or cardiovascular disease [6]. The “rights” of precision drug development for Alzheimer’s disease Jeffrey Cummings1, Howard H. Feldman2 and Philip Scheltens3 Jeffrey Cummings1, Howard H. Feldman2 and Philip Scheltens3 Powered by the California Digital Library University of California eScholarship.org (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 https://doi.org/10.1186/s13195-019-0529-5 Cummings et al. Alzheimer's Research & Therapy https://doi.org/10.1186/s13195-019-0529-5 Abstract There is a high rate of failure in Alzheimer’s disease (AD) drug development with 99% of trials showing no drug- placebo difference. This low rate of success delays new treatments for patients and discourages investment in AD drug development. Studies across drug development programs in multiple disorders have identified important strategies for decreasing the risk and increasing the likelihood of success in drug development programs. These experiences provide guidance for the optimization of AD drug development. The “rights” of AD drug development include the right target, right drug, right biomarker, right participant, and right trial. The right target identifies the appropriate biologic process for an AD therapeutic intervention. The right drug must have well-understood pharmacokinetic and pharmacodynamic features, ability to penetrate the blood-brain barrier, efficacy demonstrated in animals, maximum tolerated dose established in phase I, and acceptable toxicity. The right biomarkers include participant selection biomarkers, target engagement biomarkers, biomarkers supportive of disease modification, and biomarkers for side effect monitoring. The right participant hinges on the identification of the phase of AD (preclinical, prodromal, dementia). Severity of disease and drug mechanism both have a role in defining the right participant. The right trial is a well-conducted trial with appropriate clinical and biomarker outcomes collected over an appropriate period of time, powered to detect a clinically meaningful drug-placebo difference, and anticipating variability introduced by globalization. We lack understanding of some critical aspects of disease biology and drug action that may affect the success of development programs even when the “rights” are adhered to. Attention to disciplined drug development will increase the likelihood of success, decrease the risks associated with AD drug development, enhance the ability to attract investment, and make it more likely that new therapies will become available to those with or vulnerable to the emergence of AD. Introduction Alzheimer’s disease (AD) is rapidly increasing in fre- quency as the world’s population ages. In the USA, there are currently an estimated 5.3 million individ- uals with AD dementia, and this number is expected to increase to more than 13 million by 2050 [1, 2]. Approximately 15% of the US population over age 60 has prodromal AD and nearly 40% has preclinical AD [3]. Similar trends are seen globally with an antici- pated worldwide population of AD dementia patients exceeding 100 million by 2050 unless means of delay- ing, preventing, or treating AD are found [4]. The Amplifying the demographic challenge of the rising numbers of AD victims is the low rate of success of the development of AD therapies. Across all types of AD therapies, the failure rate is more than 99%, and for dis- ease-modifying therapies (DMTs), the failure rate is 100% [7, 8]. These numbers demand a re-examination of the drug development process. Success in other fields such as cancer therapeutics can be helpful in guiding better drug discovery and development practices of AD treatments. For example, 12 of 42 (28%) drugs approved by the US Food and Drug Administration (FDA) in 2017 * Correspondence: cumminj@ccf.org 1Department of Brain Health, School of Integrated Health Sciences, UNLV and Cleveland Clinic Lou Ruvo Center for Brain Health, 888 West Bonneville Ave, Las Vegas, NV 89106, USA Full list of author information is available at the end of the article Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 2 of 14 Cummings et al. Alzheimer's Research & Therapy Fig. 1 The rights of AD drug development were oncology therapies (www.fda.gov); this contrasts with 0% of AD drugs in development. There are cur- rently 112 new molecular entities in clinical trials in AD, whereas there are 3558 in cancer trials [9, 10]. Success in cancer drug development attracts funding and leads to more clinical trials, accelerating the emergence of new therapies. This model can assist in improving AD drug development. g p Patient care increasingly demands precision medicine with the right drug, in the right dose, administered to the right patient, at the right time [11–13]. Precision medicine requires precision drug development. Effective medications, delivered in a correct dose, to a patient in the stage of the illness that can be impacted by therapy requires that these precision treatment characteristics be determined in a disciplined drug development program [14]. Drug development sponsors have developed sys- tematic approaches to drug testing including the “rights” of drug development [15, 16], the “pillars” of drug devel- opment [17], model-based drug development [18, 19], and a translational medicine guide [20]. These ap- proaches are appropriate across therapeutic areas, and none have been applied specifically to AD drug develop- ment. Building on these foundations, we describe a set of “rights” for AD drug development which are aligned with precision drug development. We consider lessons derived from drug development across several fields as well as learnings from recent negative AD treatment tri- als [14, 17, 21, 22]; we note the areas where success in the “right” principles is pursued. These “rights” for drug development are not all new innovations, but recent re- views of the AD drug pipeline show that they are often not implemented [16, 23, 24]. We consider how the “rights” will strengthen the AD drug discovery and de- velopment process, increase the likelihood of success, de-risk investment in AD therapeutic research, and spur interest in meeting the treatment challenges posed by the coming tsunami of patients. Fig. 1 The rights of AD drug development therapy of two approved agents in AD [29]. Cholinester- ase inhibitors have shown benefit in mild, moderate, and severe AD dementia [26]; memantine is effective in moderate and severe AD dementia [30]. No agent has shown benefit in prodromal AD (pAD), mild cognitive impairment (MCI), or preclinical AD [31]. Figure 1 provides an overview of the “rights of AD drug development.” The right drug Clinical drug development is guided by defining a target product profile (TPP) describing the desirable and ne- cessary features of the candidate therapy. The TPP es- tablishes the goals of the development program, and each phase of a program is a step toward fulfilling the TPP [50, 51]. Drugs with TPP-driven development plans have a higher rate of regulatory success than those with- out [50]. Characterizing a candidate therapy begins with screen- ing assays of the identified target in preclinical discovery campaigns, identifies a lead candidate or limited set of related candidates, continues through establishing the pharmacokinetic (PK) and pharmacodynamic (PD) fea- tures in non-clinical animal models, gains refined PK and safety information with first-in-human (FIH) expos- ure in phase 1 clinical trials, and accrues greater PD and dose-response information in phase 2 trials. Finally, fully powered trials for clinical efficacy are undertaken in phase 3 with efficacy confirmation [52]. Safety data are collected throughout the process. g p Preliminary characterization of the molecule as a treat- ment candidate showing the desired effect in the screen- ing assay starts by determining that it has drug-like properties including molecular weight of ≤500 Da, bond features that support membrane penetration including the blood-brain barrier (BBB), no “alerts” that predict toxicity [53, 54], and chemical properties that suggest scalable manufacture and formulation [55, 56]. If the molecule has these encouraging properties, its absorp- tion, distribution, metabolism, excretion, and toxicity (ADMET) are determined in non-clinical models [57]. BBB penetration must be shown in humans in the course of the drug development program during phase 1 [53]. The human BBB has p-glycoprotein transporters and other mechanisms that may not be present in rodents, and central nervous system (CNS) penetration in animal models of AD is not a suffi- cient guide to human CNS entry [58]. Measurement of CNS levels in non-human primates more closely reflects the human physiology, but direct measures of cerebrospinal fluid (CSF) levels in phase 1 human Analysis of predictors of success in drug development programs shows that agents linked to genetically defined targets have a greater chance of being advanced from one phase to the next than drugs that address targets having no genetic links to the underlying disease [15, 21]. Transgenic (tg) animal models and knockout and knockin models of disease can add to the genetic evi- dence for a target. The right target No other target has been validated by successful therapy; all agents currently in development are unval- idated at the level of human benefit. Several targets are partially supported by biological and behavioral ef- fects in animal models, and some agents have shown beneficial effects in preliminary clinical trials [32]. The lack of validation of a target by a specific trial does not disprove its worthiness for drug development; val- idation depends on concurrent conduct of other “rights” in the development program. AD biology is complex, and only one target—the cholin- ergic system—has been fully validated through multiple successful therapies. Four cholinesterase inhibitors have been found to improve the dual outcomes of cognition plus function or cognition plus global status in patients with AD dementia [25, 26]. The successful development of memantine supports the validity of the N-methyl-D- aspartate (NMDA) receptor as a viable target, although only one agent has been shown to exert a therapeutic ef- fect when modulating this receptor [27, 28]. A combin- ation agent (Namzaric) addressing these two targets has been approved, establishing a precedent for combination For an agent to be a DMT, the candidate drug treat- ment must meaningfully intervene in disease processes leading to nerve cell death [33] and be druggable (e.g., Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 3 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy Defining the “right target” (or combination of targets) is currently the weakest aspect of AD drug discovery and development. The absence of a deep understanding of AD biology or focus on inappropriate targets will re- sult in drug development failures regardless of how well the drug development program is conducted. This em- phasizes the importance of investment by the National Institutes of Health (NIH), non-US basic biology initia- tives, foundations, philanthropists, and others in the fun- damental understanding of AD biology and identifying druggable targets and pathways [49]. modifiable by a small molecule agent or immunotherapy [34, 35]). Viable targets must represent critical non-re- dundant pathways necessary for neuronal survival. The right target Ideal targets have a proven function in disease pathophysi- ology, are genetically linked to the disease, have greater representation in disease than in normal function, can be assayed using high-throughput screening, are not uni- formly distributed throughout the body, have an associ- ated biomarker, and have a favorable side effect prediction profile [36]. Druggability relates to proteins, peptides, or nucleic acids with an activity that can be modified by a treatment [35]. y A current National Institute of Health (NIH) ontology of candidate targets in AD includes amyloid-related mechanisms, tau pathways, apolipoprotein E e-4 (ApoE- 4), lipid metabolism, neuroinflammation, autophagy/pro- teasome/unfolded protein response, hormones/growth factors, dysregulation of calcium homeostasis, heavy metals, mitochondrial cascade/mitochondrial uncoup- ling/antioxidants, disease risk genes and related path- ways, epigenetics, and glucose metabolism [37, 38]. Other mechanisms may emerge; highly influential nodes in networks may be identified through systems pharma- cology approaches; and opportunities or requirements for combination therapies may be discovered. Genetic editing techniques are increasingly used in experimental treatment paradigms, and RNA interference approaches show promise in non-AD neurodegenerative disorders [39]. With the recognition that late-life sporadic AD fre- quently has multiple contributing pathologies, identify- ing a single molecular therapeutic target whose manipulation is efficacious in all affected individuals may not be forthcoming [40–43]. The right drug Genes can help prioritize drug candi- dates as well as support target validation [44]. Genes implicate potentially druggable pathways and networks involved in AD pathogenesis [45, 46]. Genetic linkages to amyloid precursor protein (APP), beta-site amyloid precursor protein cleavage enzyme (BACE), gamma- secretase, ApoE, tau metabolism, and immune function are elements within the pathophysiology of AD with identified genetic influences [47]. A coding mutation in the APP gene, for example, results in a 40% reduction in amyloid beta protein (Aβ) formation and a substantial reduction in the risk of AD [48]. This observation sup- ports exploring the use of APP-modifying agents for the treatment and prevention of AD. Preliminary characterization of the molecule as a treat- ment candidate showing the desired effect in the screen- ing assay starts by determining that it has drug-like properties including molecular weight of ≤500 Da, bond features that support membrane penetration including the blood-brain barrier (BBB), no “alerts” that predict toxicity [53, 54], and chemical properties that suggest scalable manufacture and formulation [55, 56]. If the molecule has these encouraging properties, its absorp- tion, distribution, metabolism, excretion, and toxicity (ADMET) are determined in non-clinical models [57]. BBB penetration must be shown in humans in the course of the drug development program during phase 1 [53]. The human BBB has p-glycoprotein transporters and other mechanisms that may not be present in rodents, and central nervous system (CNS) penetration in animal models of AD is not a suffi- cient guide to human CNS entry [58]. Measurement of CNS levels in non-human primates more closely reflects the human physiology, but direct measures of cerebrospinal fluid (CSF) levels in phase 1 human Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 4 of 14 Page 4 of 14 Cummings et al. Alzheimer's Research & Therapy was administered at a too-low dose. In some cases, re- ceptor occupancy studies with positron emission tomog- raphy (PET), saturation of active transport mechanisms, physical limits on the amount of drug that can be ad- ministered, or dose-response curves that remain flat above specific doses obviate the need or the ability to demonstrate an MTD. In all other circumstances, an MTD should be established during phase 1 [72]. MTDs have been difficult to establish for monoclonal anti- bodies (mAbs), and decisions are often based on feasibil- ity rather than established PK/PD relationships [5]. The right drug The decision to increase the doses of mAbs by several folds in recent trials after phase 2 or 3 trials showed no drug- placebo difference (e.g., solanezumab, crenezumab, gan- tenerumab, aducanumab) demonstrates the difficulty of establishing dose and PK/PD relationships of mAbs; the absence of understanding of PK/PD for mAbs may have contributed to the failure of development programs for these agents. Formulation issues should be resolved prior to evaluating the MTD to ensure that formulation challenges do not prevent the assessment of a full range of doses. studies are required in a disciplined drug develop- ment program. CSF levels allow the determination of plasma/CSF ratios and help establish whether periph- eral levels predict CNS exposures and whether CSF levels are compatible with those showing therapeutic effects in animal models of AD [59, 60]. CSF levels are an acceptable proxy for brain levels but leave some aspects of brain entry, neuronal penetration, and target exposure unassessed [61]. Understanding the PK/PD principles at the site of exposure of the agent to the target is one of the three pillars of drug development proposed by Morgan et al. [17]. Challen- gesin achievingtarget exposure is one reason for drug development failures in otherwise well-conducted pro- grams. Tarenflurbil, for example, was shown to have poor BBB penetration after the development program was completed [62]. The “right drug” has shown efficacy in non-clinical models of AD. These models have not predicted success in human AD but advancing an agent to human testing without efficacy in animal models would add additional risk to the development program. A common strategy in- volves using genetic technologies to establish tg species bearing one or more human mutations leading to the overproduction of Aβ [63, 64]. These animals develop amyloid plaques similar to those of human AD but lack neurofibrillary tangles or cell death and are only partial simulacra of human AD [65]. They more closely resemble autosomal dominant AD with mutation-related overpro- duction of Aβ than typical late-onset AD where clearance of Aβ is the principal underlying problem [66, 67]. Activity in several AD models should be demonstrated to increase confidence in the robustness of the mechanism of the can- didate agent [68]. There are recent efforts to more closely model human systems biology using human induced pluripotent stem cell (IPSC) disease models for drug screening [69–71]. Phase 2 studies establish dose and dose-response rela- tionships. The right drug Showing a dose-response association increases confidence in the biological effects of an agent and de- risks further development. The response may be a clinical outcome or a target engagement biomarker linked to the mechanism of action (MOA) of the agent [73–75]. An ac- ceptable dose-response approach includes a low dose with no or little effect, a middle dose with an acceptable bio- logical or clinical outcome, and a high dose that is not well tolerated or raises safety concerns. After the exploration of the dose-response range in phase 2, one or two doses are advanced to phase 3 and will include the final dose(s) of the package insert of information for prescribers and patients. Using a Bayesian dose-finding approach to decide which of 5 BAN2401 doses to advance to phase 3 is an ex- ample of dose-finding in phase 2 of a development pro- gram [76]. Demonstration that the agent has neuroprotective ef- fects is critical to the definition of DMT [33, 52], and interference in the processes leading to cell death should be established prior to human exposure. Many programs have shown effects on Aβ without documenting an im- pact on neuroprotection; more thorough exploration and demonstration of neuroprotection in non-clinical models may result in agents that exert greater disease modification in human trials. The “right drug” has acceptable toxicity. Safety assess- ment begins with a review of structural alerts of the molecule predictive of toxicity such as hepatic injury assessed as part of lead candidate nomination and pro- ceeds through evaluations of target organ toxicity in sev- eral animal species—typically a rodent species and a dog species [77, 78]. Given an acceptable non-clinical safety profile, the agent is advanced to phase 1 for a FIH as- sessment of safety in the clinical setting with the deter- mination of the MTD. Safety and tolerability data continue to accrue in phase 2 and phase 3 trials. The number of human exposures remains relatively low until phase 3, and important toxicity observations may be de- layed until the late phases of drug development. Semage- cestat, avagecestat, and verubecestat were all in phase 3 Phase 1 establishes the PK features and ADMET char- acteristics of the candidate compound in humans. Sev- eral drug doses are assessed, first in single ascending dose (SAD) studies and then in multiple ascending dose (MAD) studies. The right drug A maximum tolerated dose (MTD) should be established in phase 1; without this, failure to show efficacy in later stages of development will invari- ably raise the question of whether the candidate agent Page 5 of 14 Page 5 of 14 Page 5 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy The “right” biomarker varies by the type of information needed to inform a development program and the spe- cific phase of drug development. Despite their import- ance, no biomarker has been qualified by the FDA for use across development programs [89]. before cognitive toxicity was identified as an adverse event [79–81]. Some toxicities may not be identified until after approval and widespread human use. Vigi- lance for toxic effects of agents does not stop with drug approval and continues through the post-approval and marketing period [82]. AD is a fatal illness and—like life-extending cancer therapies—side effects of treatment may be an acceptable trade-off for slowing cognitive de- cline and maintaining quality of life [83]. before cognitive toxicity was identified as an adverse event [79–81]. Some toxicities may not be identified until after approval and widespread human use. Vigi- lance for toxic effects of agents does not stop with drug approval and continues through the post-approval and marketing period [82]. AD is a fatal illness and—like life-extending cancer therapies—side effects of treatment may be an acceptable trade-off for slowing cognitive de- cline and maintaining quality of life [83]. The amyloid (A), tau (T), and neurodegeneration (N) framework provides an approach to diagnosis and moni- toring of AD and helps guide the choice of biomarkers for drug development [90, 91]. “A” biomarkers (amyloid positron emission tomography [PET], CSF Aβ) support the diagnosis of AD; “A” and “T” (tau PET; CSF phos- pho-tau) biomarkers are pharmacodynamic biomarkers that can be used to demonstrate target engagement with Aβ or tau species; and “N” (magnetic resonance imaging [MRI], fluorodeoxyglucose PET, CSF total tau) bio- markers are pharmacodynamic markers of neurodegen- eration that can provide evidence of neuroprotection and disease modification [33]. Additional markers for “N” are evolving, including neurofilament light (NfL) chain, which has shown promise in multiple sclerosis (MS) trials and preliminary AD trials [92]. The right drug Markers of synaptic degeneration such as neurogranin may also contribute to the understanding of therapeutic impact on “N” in AD. Emerging biomarkers are gaining credibil- ity and will add to or amplify the ATN framework ap- plicable to drug development [93]. The “right drug” at the end of phase 3 has demon- strated the specified features of the TPP, including effi- cacy and safety, and meets all the requirements for approval by the FDA, the European Medicines Agency (EMA), and other regulatory authorities as an AD ther- apy [50]. From an industry perspective, the “right” drug has substantial remaining patent life, is competitive with other agents with similar mechanisms, and will be ac- ceptable to payers with reimbursement rates that make the development of the agent commercially attractive [15, 21]. The “right” features of the candidate agent can be scored with a translatability score that allows com- parison and prioritization of agents for their readiness to proceed along the translational pathway to human test- ing and through the phases of clinical trials [84, 85]. Greater use of translational metrics may enhance the likelihood of drug development success [86]. In AD trials, biomarkers are needed to support the diagnosis. In prevention trials involving cogni- tively normal individuals, genetic trait biomarkers are used to establish the risk state of the individual or state biomarkers are employed to demonstrate the presence of AD pathology. In trials of treat- ments for autosomal dominant AD, demonstration of the presenilin 1, presenilin 2, or APP mutation is required in the trial participants [94, 95]. Similarly, in trials involving ApoE-4 homozygotes or The right biomarker Biomarkers play many roles in drug development and are critical to the success of development programs (Table 1) [48]. Including biomarkers in development plans has been associated with greater success rates across therapeutic areas [15, 21, 87]. The use of several types of biomarkers (predictive, prognostic) in develop- ment programs is associated with higher success rates in trials compared to trials with no or few biomarkers [88]. Table 1 Role of biomarkers in AD drug development Table 1 Role of biomarkers in AD drug development Role in trial Examples of biomarker used Identification of trial population Presence of presenilin 1 (PS1), presenilin 2 (PS2), or amyloid precursor protein (APP) mutations; ApoE-4 plu TOMM40; trisomy 21 Confirmation of diagnosis; exclude non- AD diagnoses Amyloid imaging; CSF AD signature Prognosis and course projection In MCI, ApoE-4 carriers progress more rapidly Amyloid production and clearance (target engagement) Stable isotope-labeled kinetics (SILK); BACE activity reduction with BACE inhibitor; CSF Aβ reduction by BACE inhibitor or gamma-secretase inhibitor Impact of therapy on brain circuit and network function fMRI; EEG Impact of therapy on intermediate targets Amyloid imaging; CSF amyloid; tau PET; CSF phospho-tau Disease modification MRI atrophy; CSF total tau; FDG PET; neurofilament light Stratification for trial analysis ApoE-4 genotype Side effect monitoring MRI surveillance for amyloid-related imaging abnormalities (ARIA); liver function tests; complete blood counts; electrocardiography Page 6 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy phosphodiesterase inhibitors [108], and increased Aβ fragments in the plasma and CSF with gamma-secretase inhibitors and modulators [109]. Candidate target en- gagement/proof-of-pharmacology biomarkers include peripheral indicators of inflammation and oxidation for use in trials of anti-inflammatory and antioxidant com- pounds. Sponsors of drug development should advance markers of target engagement in concert with the candi- date therapy; these may be used after regulatory ap- proval as companion or complementary biomarkers [110, 111]. Demonstration of target engagement does not guarantee efficacy in later stages of development, but target engagement shown by the “right” biomarker provides important de-risking of a candidate treatment by showing biological activity that may translate into clinical efficacy. The right biomarker With en- richment based on amyloid abnormalities, the decline was more rapid and the predicted sample size per arm to show a drug-placebo difference was 148 [103]. Target engagement biomarkers are the “missing link” in many development programs. Having shown that the candidate agent affects the target pathology in preclinical models and is safe in phase 1, sponsors have sometimes advanced through minimal phase 2 studies or directly to phase 3 [22] without showing that the drug treatment has meaningfully engaged the target in humans. Well- conducted phase 2 studies are a critical element of prin- cipled drug development and will provide two key pieces of information: target engagement and doses to be assessed in phase 3 [73, 74]. Phase 2 provides the plat- form for deciding if the candidate agent is viable for fur- ther development [75]. Target engagement may be shown directly, for example, with PET receptor occu- pancy studies or indirectly through proof-of-pharmacol- ogy [104, 105]. Examples of proof-of-pharmacology in AD drug development include the demonstration of re- duced Aβ production using stable isotope-labeled kinet- ics (SILK) [106], reduced CSF Aβ with BACE inhibitors [107], glutaminyl cyclase enzyme activity with g gy A substantial number of individuals with a clinical diagnosis of AD have been shown to lack amyloid plaque deposition when studied with amyloid imaging. Forty percent of patients diagnosed clinically with prodromal AD and 25% of those diagnosed with mild AD dementia lack evidence of amyloid pathology when studied with amyloid PET [52, 101]. Those with suspected non-amyloid pathology (SNAP) have un- determined underlying pathology and may not re- spond to proposed AD therapies. SNAPs may not decline in the expected manner in the placebo group, compromising the ability to demonstrate a drug-pla- cebo difference [102]. SNAPs should be excluded from AD trials; the “right” biomarker for this includes amyloid imaging, the CSF AD signature, or tau im- aging in patients with the AD dementia phenotype. In the idalopirdine development program, no enrichment strategies were used and power calculations showed that more than 1600 participants per arm would be needed to show a drug-placebo difference. With en- richment based on amyloid abnormalities, the decline was more rapid and the predicted sample size per arm to show a drug-placebo difference was 148 [103]. The right biomarker Semagecestat’s effect on Aβ production in the CSF and aducanumab’s plaque-lowering effect are examples where target engagement was demonstrated in phase 2 or phase 1B, and the agents still failed to show a beneficial drug-placebo difference in later-stage trials [32, 109]. Target engagement and proof-of-pharmacol- ogy are “pillars” of successful drug development [17]. heterozygotes or AD in Down syndrome, appropri- ate testing of chromosome 19 polymorphisms or chromosome 21 triplication is required [96]. A com- bination of ApoE-4 and TOMM-40 has been used to attempt to show the risk and age of onset of AD [97]. State biomarkers useful in preclinical diagnosis include amyloid PET and the CSF Aβ/tau signature of AD [98, 99]. Tau PET may be useful in identify- ing individuals appropriate for tau-targeted interven- tions or for measuring success in reducing the propagation of tau pathology [100]. heterozygotes or AD in Down syndrome, appropri- ate testing of chromosome 19 polymorphisms or chromosome 21 triplication is required [96]. A com- bination of ApoE-4 and TOMM-40 has been used to attempt to show the risk and age of onset of AD [97]. State biomarkers useful in preclinical diagnosis include amyloid PET and the CSF Aβ/tau signature of AD [98, 99]. Tau PET may be useful in identify- ing individuals appropriate for tau-targeted interven- tions or for measuring success in reducing the propagation of tau pathology [100]. p p g p gy [ ] A substantial number of individuals with a clinical diagnosis of AD have been shown to lack amyloid plaque deposition when studied with amyloid imaging. Forty percent of patients diagnosed clinically with prodromal AD and 25% of those diagnosed with mild AD dementia lack evidence of amyloid pathology when studied with amyloid PET [52, 101]. Those with suspected non-amyloid pathology (SNAP) have un- determined underlying pathology and may not re- spond to proposed AD therapies. SNAPs may not decline in the expected manner in the placebo group, compromising the ability to demonstrate a drug-pla- cebo difference [102]. SNAPs should be excluded from AD trials; the “right” biomarker for this includes amyloid imaging, the CSF AD signature, or tau im- aging in patients with the AD dementia phenotype. In the idalopirdine development program, no enrichment strategies were used and power calculations showed that more than 1600 participants per arm would be needed to show a drug-placebo difference. The right biomarker gy p g p Changes in the basic biology of AD—amyloid gener- ation, tau aggregation, inflammation, oxidation, mitochon- drial dysfunction, neurodegeneration, etc.—are linked to human cognition through neural circuits whose integrity is critical to normal memory and intellectual function [112]. Two techniques of assessing neural networks are electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). In cognitively normal individ- uals with positive amyloid PET and low levels of tau as shown by tau PET, fMRI measures of the default mode network (DMN) reveal hyperactive circuit functions. In those with elevated amyloid and elevated tau levels, the circuits become hypoactive compared to age-matched controls [113, 114]. Decline in circuit function predicts progressive cognitive impairment [115]. Disrupted DMN function is present in prodromal AD and in AD dementia [116, 117]. Assessment of DMN integrity may be an im- portant biomarker with predictive value for the impact of the intervention on clinical outcomes [112]. EEG is dependent on the intact network function and may have applications in AD drug development similar to, but more robustly, than those of fMRI [108, 118, 119]. Both EEG and fMRI require procedural and interpretative standardization to be implemented in multi-site trials. A recent alternative for the assessment of circuit integrity in AD is SV2A PET, targeting and visualizing the synaptic network and currently under study as a possible measure of target engagement for drugs aiming to influence synap- tic function [120]. g p Target engagement biomarkers are the “missing link” in many development programs. Having shown that the candidate agent affects the target pathology in preclinical models and is safe in phase 1, sponsors have sometimes advanced through minimal phase 2 studies or directly to phase 3 [22] without showing that the drug treatment has meaningfully engaged the target in humans. Well- conducted phase 2 studies are a critical element of prin- cipled drug development and will provide two key pieces of information: target engagement and doses to be assessed in phase 3 [73, 74]. Phase 2 provides the plat- form for deciding if the candidate agent is viable for fur- ther development [75]. Target engagement may be shown directly, for example, with PET receptor occu- pancy studies or indirectly through proof-of-pharmacol- ogy [104, 105]. The right participant dependent plaque reduction. In a phase 1B trial, aduca- numab achieved both significant plaque reduction and benefit on some clinical measures with evidence of a dose-response relationship [32]. The beneficial effect was not recapitulated in a phase 3 trial. Bapineuzumab and gantenerumab decreased plaque Aβ but had no cor- responding impact on cognition or function in the doses studied [121, 122]. Removal of plaque amyloid may be necessary but not sufficient for a therapeutic benefit of anti-amyloid agents or may be a coincidental marker of engagement of a broad range of amyloid species includ- ing those required for a therapeutic response. Tau PET assesses target engagement by anti-tau therapeutics; re- duced tau burden or reduced tau spread would indicate a therapeutic response [123]. Aβ and tau signals do not measure neuroprotection and are not necessarily evi- dence of disease modification (DM). AD progresses through a spectrum of severity from cog- nitively normal amyloid-bearing preclinical individuals, to those with prodromal AD or prodromal/mild AD de- mentia and, finally, to those with more severe AD de- mentia [136, 137] (Fig. 2). Based on this model, trials can target primary prevention in cognitively normal in- dividual with risk factors for AD but no state biomarkers indicative of AD pathology, secondary prevention in pre- clinical AD participants who are cognitively normal but have positive state biomarkers (positive amyloid PET, low CSF Aβ), and treatment trials aimed at slowing dis- ease progression in prodromal or prodromal/mild AD dementia or mild, moderate, and severe AD dementia (Fig. 2). Although AD represents a seamless progression from unaffected to severely compromised individuals, participants can be assigned to the progressive phases based on genetic markers, cognitive and functional as- sessments, amyloid imaging or CSF Aβ and tau mea- sures, tau imaging, and MRI [52, 136, 137]. The ATN Framework is designed to guide the identification of the “right” participant for clinical trials [90, 91]. Early inter- vention has proven to be associated with better out- comes in other disorders such as heart failure [138] suggesting that early intervention in the “brain failure” of AD may have superior outcomes compared to later- phase interventions. However, available cognitive-enhan- cing agents have been approved for mild, moderate, and severe AD and have failed in trials with predementia participants; some DMT mechanisms may require use earlier in the disease process before pathologic changes are extensive [139–141]. The right participant Biomarkers play a critical role in demonstrating DM in DMT development programs. Evidence of neuro- protection is essential to support DM, and structural magnetic resonance imaging (MRI) is the current bio- marker of choice for this purpose. Hippocampal atro- phy has been linked to progressive disease and to nerve cell loss [124–126]. In clinical trials, MRI has often not fulfilled expectations, and atrophy has sometimes been greater in the treatment groups than in the placebo controls [127, 128]. Recent studies have shown drug-placebo differences on MRI in the anticipated direction suggesting that MRI may be an important DM marker depending on the underlying MOA of the agent. As noted, serum and CSF bio- markers of neurodegeneration such as NfL and synap- tic markers have promise to assess successful DMTs but have been incorporated into relatively few AD tri- als [129]. CSF measures of total tau may be closely related to neurodegeneration and provide useful evi- dence of the impact on cell death [130, 131]. The right participant also relates to the MOA of the agent being assessed. Cognitive enhancing agents will be examined in patients with cognitive abnormalities; agents reducing amyloid production may have the optimal chance of success in primary or secondary prevention; tau prevention trials may focus on the preclinical participants; tau removal agents might be appropriate for prodromal AD or AD dementia; combinations of agents may be assessed in trials with participants with corresponding bio- marker changes. Experience with a greater array of agents in a variety of disease phases will help inform the match between the “right” participant and specific agent MOAs. Development of more biomarkers such as those indicating CNS inflammation, excessive oxidation, or the presence of concurrent pathologies such as TDP-43 or alpha-synu- clein may assist in matching treatment MOA to the pathological form of AD. Biomarkers could eventually have a role as surrogate outcomes for AD trials if they are shown to be pre- dictive of clinical outcomes. Currently, no AD bio- marker has achieved surrogate status, and biomarkers are used in concert with clinical outcomes as measures of treatment effects. Biomarkers have a role in monitoring side effects in the course of clinical trials. Liver, hematologic, and cardiac ef- fects are monitored with liver function tests, complete blood counts, and electrocardiography, respectively. Ata- becestat, for example, is a BACE inhibitor whose develop- ment was interrupted by the emergence of liver toxicity [132]. The right participant Amyloid-related imaging abnormalities (ARIA) of the effusion (ARIA-E) or hemorrhagic (ARIA-H) type may occur with MAbs and are monitored in trials with serial MRI [133]. ARIA has been observed with bapineuzumab, gantenerumab, aducanumab, and BAN2401 [32, 134, 135]. The right biomarker Examples of proof-of-pharmacology in AD drug development include the demonstration of re- duced Aβ production using stable isotope-labeled kinet- ics (SILK) [106], reduced CSF Aβ with BACE inhibitors [107], glutaminyl cyclase enzyme activity with Amyloid imaging is a target engagement biomarker es- tablishing reduction of plaque amyloid [111]. Several monoclonal antibodies have shown a dose and time- Page 7 of 14 Page 7 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 7 of 14 The right trial The “right trial” is a well-conducted clinical experiment that answers the central question regarding the superior- ity of the drug over placebo at the specified dose in the Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 8 of 14 (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy Fig. 2 Spectrum of AD and the corresponding cognitive and biomarker state of trial participants (A, amyloid abnormalities; T, tau abnormalities; N, neurodegeneration) Alzheimer Preclinical Cognitive Composite (APCC) used in the Alzheimer’s Prevention Initiative, for ex- ample, are used as outcomes in studies of preclinical AD [137, 144, 145]. The Clinical Dementia Rating- Sum of Boxes (CDR-sb) is commonly used as an out- come in prodromal AD [146]. The AD Assessment Scale-Cognitive subscale (ADAS-cog) [147] or the neuropsychological test battery (NTB) [148] and the CDR-sb or Clinical Global Impression of Change with Caregiver Input (CIBIC+) are common dual out- comes in trials of mild-moderate AD dementia [40, 146]. The AD Composite Score (ADCOMS) is an analytic approach including items from the CDR-sb, ADAS-cog, and Mini-Mental State Examination (MMSE) that is sensitive to change and drug effects in prodromal AD and mild AD dementia [149]. The severe impairment battery (SIB) is the outcome as- sessment most commonly used in severe AD [150]. Having tools with sufficient sensitivity to detect drug-placebo differences in predementia phases of AD is challenging. Commonly used tools such as the ADAS-cog were developed for later stages of the dis- ease. Newer instruments such as the PACC and APCC detect changes over time in natural history studies, but their performance in trials is unknown. Alzheimer Preclinical Cognitive Composite (APCC) used in the Alzheimer’s Prevention Initiative, for ex- ample, are used as outcomes in studies of preclinical AD [137, 144, 145]. The Clinical Dementia Rating- Sum of Boxes (CDR-sb) is commonly used as an out- come in prodromal AD [146]. The AD Assessment Scale-Cognitive subscale (ADAS-cog) [147] or the neuropsychological test battery (NTB) [148] and the CDR-sb or Clinical Global Impression of Change with Caregiver Input (CIBIC+) are common dual out- comes in trials of mild-moderate AD dementia [40, 146]. The AD Composite Score (ADCOMS) is an analytic approach including items from the CDR-sb, ADAS-cog, and Mini-Mental State Examination (MMSE) that is sensitive to change and drug effects in prodromal AD and mild AD dementia [149]. The right trial Advocacy groups throughout the world strive to overcome this attitude; success in engaging par- ticipants in trials will become more pressing as more preclinical trials involving cognitively normal individuals are initiated. Sample size is related to the magnitude of the detectable effect which is in turn related to the effect size of the agent and the sensitivity of the measurement tool (clinical instruments or biomarkers); these factors require optimization to allow the conduct of trials with feasible sample sizes. pp Globalization of clinical trials with the inclusion of trial sites in many countries is a common response to slow recruitment of trial participants. By increasing the number of trial sites, recruitment can be acceler- ated and drug efficacy demonstrated more promptly. Globalization, however, increases the number of lan- guages and cultures of participants in the trials as well as increasing the heterogeneity of background experience among the trial sites and investigators. These factors may increase measurement variability and make it more difficult to demonstrate a drug-pla- cebo difference [159–161]. The “right trial” will limit these factors by minimizing the number of regions, languages, and trial sites involved. Within diverse countries such as the USA, the inclusion of minority participants is key to insuring the generalizability of the findings from trials [162]. Hallmarks of poorly designed or conducted trials in- clude failure of the placebo group to decline in the course of a trial (assuming an adequate observation period), failure to show separation of the placebo group from an active treatment arm such as donepezil, exces- sive measurement variability, or low levels of biological indicators of AD such as the percent of ApoE-4 carriers or the presence of fibrillar amyloid on amyloid imaging [22]. The right trial Table 2 summarizes the instruments currently used in trials of each major phase of AD. The right trial will include the right doses selected in phase 2 and the right biomarkers as noted above. The biomarker will be chosen to match the questions to be answered for each trial phase. Target engagement bio- markers are critical in phase 2, and DM biomarkers are critical in phase 3 of DMT trials. The trial duration may vary from 12 months to 8 years for DMTs or 3–6 months for symptomatic agents based on the anticipated duration of exposure needed to demonstrate a drug-placebo difference. Preclinical trials may involve observing patients for up to 5 years to allow sufficient decline in the placebo group to be able to demonstrate a drug-placebo difference. These trial duration choices are arbitrary; a basic biological understanding linking the changes in the pathology to the duration of drug exposure is lacking. Using an adaptive design approach, it is possible to adjust trial durations based on emerging patterns of efficacy [76, 154]. Adaptive designs may be used to optimize sample size, trial duration, and dose selection and have been successful in trials of chemotherapy and in trials for treatments of diabetes [155]. Adaptive designs are cur- rently in use in the European Prevention of AD (E- PAD), the Dominantly Inherited Alzheimer Network- Treatment Unit (DIAN-TU), and a study of oxytocin in frontotemporal dementia [156]; broad exploration of the approach is warranted [157, 158]. The right trial is also efficiently conducted with rapid start-up, certified raters, a central institutional review board (IRB), and timely recruitment of appro- priate subjects. Programs such as the Trial-Ready Co- hort for Prodromal and Preclinical AD (TRC-PAD), Global Alzheimer Platform (GAP), and the EPAD ini- tiative aim to enhance the efficiency with which trials are conducted [157, 163]. Development of online registries and trial-ready cohorts may accelerate trial recruitment and treatment evaluation [164–166]. Registries have been helpful in trial recruitment to non-AD disorders [167]. Inclusion of the right number of the right participants is of key importance in successfully advancing AD thera- peutics. Compared to other fields, there is a reluctance by patients and physicians to participate in clinical trials for a disease that is considered by some to be a part of normal aging. The right trial The severe impairment battery (SIB) is the outcome as- sessment most commonly used in severe AD [150]. Having tools with sufficient sensitivity to detect drug-placebo differences in predementia phases of AD is challenging. Commonly used tools such as the ADAS-cog were developed for later stages of the dis- ease. Newer instruments such as the PACC and APCC detect changes over time in natural history studies, but their performance in trials is unknown. time frame of observation in the defined population. Poorly conducted or underpowered trials do not resolve the central issue of drug efficacy and should not be con- ducted since they involve participant exposures and po- tential toxicity without the ability to provide valid informative scientific data. Trial sponsors incur the re- sponsibility to report the results of trials to allow the field to progress by learning from the outcome of each experiment. Participants have accepted the risks of un- known drug effects and placebo exposure, and honoring this commitment requires that the learnings from the trial be made available publically [142]. A key element includes a sample size based on thor- oughly vetted anticipated effect sizes. Trial simulations are available to model the results of varying effect sizes and the corresponding required population size [143]. Participation criteria critical to the trial success include defining an appropriate population of preclinical, pro- dromal, or AD dementia using biomarkers as noted above [136, 137]. Other key participation criteria include the absence of non-AD neurologic diagnoses, physical illness incompatible with trial requirements, or use of medications that may interact with the test agents. Fewer exclusions from trials lead to more generalizable results. Inclusion of diverse populations representative of the populations to which the agent will be marketed en- hances the generalizability of trial results. The Alzheimer’s Disease Cooperative Study (ADCS) Activities of Daily Living (ADL) scale is commonly used to assess daily function in patients with MCI and mild to severe AD dementia [151]. The Amsterdam Instrumen- tal Activities of Daily Living (IADL) Questionnaire is Clinical outcomes will be chosen based on the spe- cific population included in the trial. The Preclinical Alzheimer Cognitive Composite (PACC) and the Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 9 of 14 increasingly employed for this purpose in MCI/pro- dromal AD and mild AD dementia [152, 153]. The right trial Trials with these features would not be expected to Table 2 Instruments appropriate as the outcome assessments in different phases of AD Domain Prevention trials Prodromal AD trials AD dementia trials Cognition PACC; APCC NTB ADAS-cog in mild to moderate AD; SIB in moderate to severe AD Global/composite None CDR-sb; ADCOMS; iADRS CIBIC+ in shorter trials; CDR-sb in longer trials Function None ADCS ADL MCI scale; Amsterdan IADL scale ADCS ADL scale Behavior NPI NPI NPI ADAS-cog Alzheimer’s Disease Assessment Scale-cognitive subscale, ADCOMS Alzheimer’s Disease Composite Scale, Alzheimer’s Disease Cooperative Study Activities of Daily Living scale, APCC Alzheimer’s Prevention Initiative (API) Composite Cognitive, CDR-sb Clinical Dementia Rating-Sum of Boxes, CIBIC+ Clinical Interview-Based Impression of Change with Caregiver Input, IADL Instrumental Activities of Daily Living, iADRS Integrated Alzheimer’s Disease Rating Scale, NPI Neuropsychiatric Inventory, NTB neuropsychological test battery, PACC Preclinical Alzheimer Cognitive Composite, SIB severe impairment battery Table 2 Instruments appropriate as the outcome assessments in different phases of AD ADAS-cog Alzheimer’s Disease Assessment Scale-cognitive subscale, ADCOMS Alzheimer’s Disease Composite Scale, Alzheimer’s Disease Cooperative Study Activities of Daily Living scale, APCC Alzheimer’s Prevention Initiative (API) Composite Cognitive, CDR-sb Clinical Dementia Rating-Sum of Boxes, CIBIC+ Clinical Interview-Based Impression of Change with Caregiver Input, IADL Instrumental Activities of Daily Living, iADRS Integrated Alzheimer’s Disease Rating Scale, NPI Neuropsychiatric Inventory, NTB neuropsychological test battery, PACC Preclinical Alzheimer Cognitive Composite, SIB severe impairment battery ADAS-cog Alzheimer’s Disease Assessment Scale-cognitive subscale, ADCOMS Alzheimer’s Disease Composite Scale, Alzheimer’s Disease Cooperative Study Activities of Daily Living scale, APCC Alzheimer’s Prevention Initiative (API) Composite Cognitive, CDR-sb Clinical Dementia Rating-Sum of Boxes, CIBIC+ Clinical Interview-Based Impression of Change with Caregiver Input, IADL Instrumental Activities of Daily Living, iADRS Integrated Alzheimer’s Disease Rating Scale, NPI Neuropsychiatric Inventory, NTB neuropsychological test battery, PACC Preclinical Alzheimer Cognitive Composite, SIB severe impairment battery Page 10 of 14 Page 10 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Page 10 of 14 Cummings et al. Alzheimer's Research & Therapy detect drug-placebo differences or to inform the drug development agenda. have improved the rate of success of drug develop- ment in other settings [15, 21]. Adhering to the “rights of AD drug development” will de-risk many of the challenges of drug development and increase the like- lihood of successful trials of critically needed new treatments for AD. A well-designed phase 3 trial builds on observa- tions made in phase 2. The right trial Drugs have often been ad- vanced to phase 3 based on the interpretation of apparent effects observed in phase 2 unprespecified subgroup analyses that are derived from small non- randomized samples and are rarely if ever repro- duced in phase 3 [22]. Abbreviations A A l id Aβ bb e at o s A: Amyloid; Aβ: Amyloid beta protein; AD: Alzheimer’s disease; ADAS-cog: AD Assessment Scale-Cognitive subscale; ADCOMS: AD Composite Score; ADCS: Alzheimer’s Disease Cooperative Study; ADL: Activities of daily living; ADMET: Absorption, distribution, metabolism, excretion, and toxicity; APOE: Apolipoprotein E; APP: Amyloid precursor protein; ARIA: Amyloid- related imaging abnormalities; ARIA-E: Amyloid-related imaging abnormalities effusion; ARIA-H: Amyloid-related imaging abnormalities hemorrhagic; BACE: Beta-site amyloid precursor protein cleavage enzyme; BBB: Blood-brain barrier; CDR-sb: Clinical Dementia Rating-Sum of Boxes; CIBIC+: Clinical Global Impression of Change with Caregiver Input; CNS: Central nervous system; CSF: Cerebrospinal fluid; DIAN-TU: Dominantly Inherited Alzheimer Network-Treatment Unit; DM: Disease modification; DMN: Default mode network; DMT: Disease-modifying therapy; EEG: Electroencephalography; EMA: European Medicines Agency; E- PAD: European Prevention of Alzheimer’s Disease; fMRI: Functional magnetic resonance imaging; FDA: US Food and Drug Administration; FIH: First-in- human; GAP: Global Alzheimer Platform; IADL: Instrumental Activities of Daily Living Questionnaire; IPSC: Induced pluripotent stem cell; IRB: Institutional review board; MAbs: Monoclonal antibodies; MAD: Multiple ascending dose; MCI: Mild cognitive impairment; MMSE: Mini-Mental State Examination; MOA: Mechanism of action; MTD: Maximum tolerated dose; MRI: Magnetic resonance imaging; MS: Multiple sclerosis; N: Neurodegeneration; NfL: Neurofilament light; NIH: National Institutes of Health; NMDA: N-methyl- D-asparate; NPI: Neuropsychiatric Inventory; NTB: Neuropsychological test battery; PACC: Preclinical Alzheimer Cognitive Composite; pAD: Prodromal Alzheimer’s disease; PD: Pharmacodynamic; PET: Positron emission tomography; PK: Pharmacokinetic; SAD: Single ascending dose; SIB: Severe impairment battery; SILK: Stable isotope-labeled kinetics; SNAP: Suspected non-amyloid pathology; T: Tau; tg: Transgenic; TOMM-40: Translocase of outer mitochondrial membrane 40; TPP: Target product profile; TRC- PAD: Trial-Ready Cohort for Prodromal and Preclinical AD; USD: US dollars Summary and conclusions AD drug development has had a high rate of failure [7]. In many cases, BBB penetration, dose, target engage- ment, or rigorous interrogation of early-stage data has not been adequately pursued. Agents have been ad- vanced to phase 3 with little or no evidence of efficacy in phase 2. Better designed and conducted phase 2 stud- ies will inform further development and enable stopping earlier and preserving resources that can be assigned to testing more drugs in earlier stages (preclinical and FIH), as well as promoting better drugs with a greater chance of success to phase 3 [168]. Deep insight into the biology of AD is currently lacking, and predicting drug success will continue to be challenging; optimizing drug development and clinical trial conduct will reduce this inevitable risk of AD treatment development. Table 3 provides a summary of the integration of the “rights” of AD drug development across the phases of the develop- ment cycle. This “rights” approach to drug development will en- able the precision medicine objective of the right drug, at the right dose, for the right patient, at the right time, tested in the right trial [11–13, 16]. Approaches such as these when used in other therapeutic areas Competing interests 14. Plenge RM. Disciplined approach to drug discovery and early development. Sci Transl Med. 2016;8(349):349ps15. JC has provided consultation to Acadia, Accera, Actinogen, Alkahest, Allergan, Alzheon, Avanir, Axsome, BiOasis Technologies, Biogen, Bracket, Cassava, Denali, Diadem, EIP Pharma, Eisai, Genentech, Green Valley, Grifols, Hisun, Idorsia, Kyowa Kirin, Lilly, Lundbeck, Merck, Otsuka, Proclara, QR, Resverlogix, Roche, Samumed, Samus, Sunovion, Suven, Takeda, Teva, Toyama, and United Neuroscience pharmaceutical and assessment companies. JC is supported by Keep Memory Alive (KMA) and COBRE award from the NIGMC (P20GM109025). 15. Morgan P, Brown DG, Lennard S, Anderton MJ, Barrett JC, Eriksson U, et al. Impact of a five-dimensional framework on R&D productivity at AstraZeneca. Nat Rev Drug Discov. 2018;17(3):167–81. AstraZeneca. Nat Rev Drug Discov. 2018;17(3):167–81. 16. Golde TE, DeKosky ST, Galasko D. Alzheimer’s disease: the right drug, the right time. Science. 2018;362(6420):1250–1. 17. Morgan P, Van Der Graaf PH, Arrowsmith J, Feltner DE, Drummond KS, Wegner CD, et al. Can the flow of medicines be improved? Fundamental pharmacokinetic and pharmacological principles toward improving phase II survival. Drug Discov Today. 2012;17(9–10):419–24. HHF reports research support from the National Institutes of Aging: Alzheimer Disease Cooperative Study U19AG10483-26, Alzheimer's Disease Research Center P50 AG005131, Canadian Institutes of Health Research 201901CNA-417847-CAN-ABPI-32054/#254450, Brain Canada #4669, Biohaven Pharmaceuticals, Toyama Pharmaceuticals, UC Cures for Alzheimer's Disease Initiative BRD-16-501346 and development grant funding from Vivoryon (formerly Probiodrug), service agreements with Eisai Pharmaceuticals, Genen- tech/Roche Pharmaceuticals, Banner Health Institute, Samus Therapeutics, Merck Pharmaceuticals, Tau RX, Arkuda Therapeutics, and Samumed; speaker fees from World Events Forum, Medscape and Optum; and travel expenses from Axon Neurosciences, Alion Pharmaceuticals, Vivoryon (formerly Probio- drug), and Dominantly Inherited Alzheimer's Disease. 18. Milligan PA, Brown MJ, Marchant B, Martin SW, van der Graaf PH, Benson N, et al. Model-based drug development: a rational approach to efficiently accelerate drug development. Clin Pharmacol Ther. 2013;93(6):502–14. 19. Visser SA, Aurell M, Jones RD, Schuck VJ, Egnell AC, Peters SA, et al. Model- based drug discovery: implementation and impact. Drug Discov Today. 2013;18(15–16):764–75. 20. Dolgos H, Trusheim M, Gross D, Halle JP, Ogden J, Osterwalder B, et al. Translational medicine guide transforms drug development processes: the recent Merck experience. Drug Discov Today. 2016;21(3):517–26. Acknowledgements Not applicable. Acknowledgements Not applicable. Table 3 Five “rights” implemented across the spectrum of drug development Right element Target identification Drug candidate optimization Non-clinical assessment Phase 1 Phase 2 Phase 3 Target Druggable target identified in AD biology PD effect supported PD effect may be assessed with biomarkers PD effect supported by biomarkers PD effect supported by biomarkers and clinical outcomes Drug Chemical properties ADME; toxicity; efficacy in animals PK, ADME in healthy volunteers; MTD established; BBB penetration established PK, PD in AD PD in AD Biomarker Development of biomarkers useful in trials Toxicity biomarkers Patient selection; target engagement biomarkers Patient selection; DM; toxicity; predictive biomarkers Patient Healthy volunteers; AD for immuuno-therapy trials Prodromal AD, AD dementia High-risk normal subjects; prodromal AD; AD dementia Trial Single ascending dose; multiple ascending dose Drug-placebo difference at endpoint; adaptive designs Drug-placebo difference at endpoint; adaptive designs; delay to milestone AD Alzheimer’s disease; ADME absorption, distribution, metabolism, excretion; DM disease modification; PK pharmacokinetics; PD pharmacodynamic Table 3 Five “rights” implemented across the spectrum of drug development AD Alzheimer’s disease; ADME absorption, distribution, metabolism, excretion; DM disease modification; PK pharmacokinetics; PD pharmacodynamic Alzheimer’s disease; ADME absorption, distribution, metabolism, excretion; DM disease modification; PK pharmacokinetics; PD pharmac Page 11 of 14 Page 11 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Consent for publication Not applicable. 13. Peng X, Xing P, Li X, Qian Y, Song F, Bai Z, et al. Towards personalized intervention for Alzheimer’s disease. Genomics Proteomics Bioinformatics. 2016;14(5):289–97. Availability of data and materials Not applicable. 11. Hampel H, O’Bryant SE, Castrillo JI, Ritchie C, Rojkova K, Broich K, et al. Precision medicine - the golden gate for detection, treatment and prevention of Alzheimer’s disease. J Prev Alzheimers Dis. 2016;3(4):243–59. Ethics approval and consent to participate Not applicable. Author details 1 f 1Department of Brain Health, School of Integrated Health Sciences, UNLV and Cleveland Clinic Lou Ruvo Center for Brain Health, 888 West Bonneville Ave, Las Vegas, NV 89106, USA. 2Department of Neurosciences, Alzheimer’s Disease Cooperative Study, University of California San Diego, San Diego, CA, USA. 3Alzheimer Center Amsterdam, Amsterdam University Medical Centers, Amsterdam, The Netherlands. 25. Kobayashi H, Ohnishi T, Nakagawa R, Yoshizawa K. The comparative efficacy and safety of cholinesterase inhibitors in patients with mild-to-moderate Alzheimer’s disease: a Bayesian network meta-analysis. Int J Geriatr Psychiatry. 2016;31(8):892–904. 26. Tan CC, Yu JT, Wang HF, Tan MS, Meng XF, Wang C, et al. Efficacy and safety of donepezil, galantamine, rivastigmine, and memantine for the treatment of Alzheimer’s disease: a systematic review and meta-analysis. J Alzheimers Dis. 2014;41(2):615–31. Received: 15 May 2019 Accepted: 13 August 2019 27. Tariot PN, Farlow MR, Grossberg GT, Graham SM, McDonald S, Gergel I. Memantine treatment in patients with moderate to severe Alzheimer disease already receiving donepezil: a randomized controlled trial. JAMA. 2004;291(3):317–24. Funding The authors acknowledge the support of a COBRE grant from the NIH/ NIGMS (P20GM109025) and Keep Memory Alive. 9. Cummings J, Lee G, Ritter A, Zhong K. Alzheimer’s disease drug development pipeline: 2018. Alzheimers Dement. 2018;4:195–214. development pipeline: 2018. Alzheimers Dement. 2018;4:195–214. Moser J, Verdin P. Trial watch: burgeoning oncology pipeline raises 10. Moser J, Verdin P. Trial watch: burgeoning oncology pipeline raises questions about sustainability. Nat Rev Drug Discov. 2018;17(10):698–9. Availability of data and materials Not applicable. Availability of data and materials Not applicable. Competing interests PS has received research support from Merck, GE Healthcare, Piramal, Alzheimer Nederland, Dioraphte, and Stichting VUmc Fonds and Stichting Alzheimer & Neuropsychiatrie; served as a consultant for AbbVie, Avraham, ARC, Janssen Research Foundation, MD Start, Nutricia, Takeda, Probiodrug, and Genentech and EIP Pharma; and received speaker fees from Piramal, Roche Diagnostics and GE Healthcare. He is the co-editor-in-chief of Alzhei- mer’s Research & Therapy and associate editor of Alzheimer’s Disease and Asso- ciated Disorders. He had no role in the peer review process for this article. 21. Cook D, Brown D, Alexander R, March R, Morgan P, Satterthwaite G, et al. Lessons learned from the fate of AstraZeneca’s drug pipeline: a five- dimensional framework. Nat Rev Drug Discov. 2014;13(6):419–31. 22. Cummings J. Lessons learned from Alzheimer disease: clinical trials with negative outcomes. Clin Transl Sci. 2017;11:147–52. 23. Becker RE, Greig NH. Increasing the success rate for Alzheimer’s disease drug discovery and development. Expert Opin Drug Discov. 2012;7(4):367–70. 24. Karran E, Hardy J. A critique of the drug discovery and phase 3 clinical programs targeting the amyloid hypothesis for Alzheimer disease. Ann Neurol. 2014;76(2):185–205. Authors’ contributions 6. Hurd MD, Martorell P, Langa KM. Monetary costs of dementia in the United States. N Engl J Med. 2013;369(5):489–90. All authors contributed in the writing and revising of the manuscript, and all authors read and approved the final manuscript. 7. Cummings JL, Morstorf T, Zhong K. Alzheimer’s disease drug-development pipeline: few candidates, frequent failures. Alzheimers Res Ther. 2014;6(4):37–43. 8. Cummings J, Lee G, Mortsdorf T, Ritter A, Zhong K. Alzheimer’s disease dr development pipeline: 2017. Alzheimers Dement. 2017;3(3):367–84. Ethics approval and consent to participate Not applicable. 12. Hampel H, O’Bryant SE, Durrleman S, Younesi E, Rojkova K, Escott-Price V, et al. A precision medicine initiative for Alzheimer’s disease: the road ahead to biomarker-guided integrative disease modeling. Climacteric. 2017;20(2):107–18. References Alzheimer’s disease biomarkers in animal models: closing the translational gap. Am J Neurodegener Dis. 2013; 2(2):108–20. 41. Schneider JA, Arvanitakis Z, Leurgans SE, Bennett DA. The neuropathology of probable Alzheimer disease and mild cognitive impairment. Ann Neurol. 2009;66(2):200–8. 69. Yang J, Li S, He XB, Cheng C, Le W. Induced pluripotent stem cells in Alzheimer’s disease: applications for disease modeling and cell-replacement therapy. Mol Neurodegener. 2016;11(1):39. 42. Woodward M, Mackenzie IRA, Hsiung GY, Jacova C, Feldman H. Multiple brain pathologies in dementia are common. Eur Ger Med. 2010;1:259–65. 70. Liu Q, Waltz S, Woodruff G, Ouyang J, Israel MA, Herrera C, et al. Effect of potent gamma-secretase modulator in human neurons derived from multiple presenilin 1-induced pluripotent stem cell mutant carriers. JAMA Neurol. 2014;71(12):1481–9. 43. Devi G, Scheltens P. Heterogeneity of Alzheimer’s disease: consequence for drug trials? Alzheimers Res Ther. 2018;10(1):122. 71. Choi SH, Kim YH, Quinti L, Tanzi RE, Kim DY. 3D culture models of Alzheimer’s disease: a road map to a “cure-in-a-dish”. Mol Neurodegener. 2016;11(1):75. 44. Plenge RM, Scolnick EM, Altshuler D. Validating therapeutic targets through human genetics. Nat Rev Drug Discov. 2013;12(8):581–94. 45. Li P, Nie Y, Yu J. An effective method to identify shared pathways and common factors among neurodegenerative diseases. PLoS One. 2015;10(11):e0143045. 72. Pangalos MN, Schechter LE, Hurko O. Drug development for CNS disorders: strategies for balancing risk and reducing attrition. Nat Rev Drug Discov. 2007;6(7):521–32. 46. International Genomics of Alzheimer’s Disease C. Convergent genetic and expression data implicate immunity in Alzheimer’s disease. Alzheimer Dement. 2015;11(6):658–71. 73. Greenberg BD, Carrillo MC, Ryan JM, Gold M, Gallagher K, Grundman M, et al. Improving Alzheimer’s disease phase II clinical trials. Alzheimers Dement. 2013;9(1):39–49. 47. Guerreiro R, Hardy J. Genetics of Alzheimer’s disease. Neurotherapeutics. 2014;11(4):732–7. 74. Gray JA, Fleet D, Winblad B. The need for thorough phase II studies in medicines development for Alzheimer’s disease. Alzheimers Res Ther. 2015;7(1):67. 48. Townsend MJ, Arron JR. Reducing the risk of failure: biomarker-guided trial design. Nat Rev Drug Discov. 2016;15(8):517–8. 49. Kosik KS, Sejnowski TJ, Raichle ME, Ciechanover A, Baltimore D. A path toward understanding neurodegeneration. Science. 2016;353(6302):872–3. 75. Cartwright ME, Cohen S, Fleishaker JC, Madani S, McLeod JF, Musser B, et al. Proof of concept: a PhRMA position paper with recommendations for best practice. Clin Pharmacol Ther. 2010;87(3):278–85. 50. Breder CD, Du W, Tyndall A. What’s the regulatory value of a target product profile? Trends Biotechnol. 2017;35(7):576–9. 76. References Rizk ML, Zou L, Savic RM, Dooley KE. Importance of drug pharmacokinetics at the site of action. Clin Transl Sci. 2017;10(3):133–42. 33. Cummings JL, Fox N. Defining disease modification for Alzheimer’s disease clinical trials. J Prev Alz Dis. 2017;4:109–15. 62. Wan HI, Jacobsen JS, Rutkowski JL, Feuerstein GZ. Translational medicine lessons from flurizan’s failure in Alzheimer’s disease (AD) trial: implication for future drug discovery and development for AD. Clin Transl Sci. 2009;2(3):242–7 34. Fauman EB, Rai BK, Huang ES. Structure-based druggability assessment-- identifying suitable targets for small molecule therapeutics. Curr Opin Chem Biol. 2011;15(4):463–8. 63. Ameen-Ali KE, Wharton SB, Simpson JE, Heath PR, Sharp P, Berwick J. Review: Neuropathology and behavioural features of transgenic murine models of Alzheimer's disease. Neuropathol Appl Neurobiol. 2017;43(7):553–70. 35. Gashaw I, Ellinghaus P, Sommer A, Asadullah K. What makes a good drug target? Drug Discov Today. 2011;16(23–24):1037–43. 36. Gashaw I, Ellinghaus P, Sommer A, Asadullah K. What makes a good drug target? Drug Discov Today. 2012;17(Suppl):S24–30. 64. Sabbagh JJ, Kinney JW, Cummings JL. Animal systems in the development of treatments for Alzheimer’s disease: challenges, methods, and implications. Neurobiol Aging. 2013;34(1):169–83. target? Drug Discov Today. 2012;17(Suppl):S24–30. 37. Lo AW, Ho C, Cummings J, Kosik KS. Parallel discovery of Alzheimer’s therapeutics. Sci Transl Med. 2014;6(241):241cm5. 65. Laurijssens B, Aujard F, Rahman A. Animal models of Alzheimer’s disease and drug development. Drug Discov Today Technol. 2013;10(3):e319–27. 38. Refolo LM, Snyder H, Liggins C, Ryan L, Silverberg N, Petanceska S, et al. Common Alzheimer’s disease research ontology: National Institute on Aging and Alzheimer’s Association collaborative project. Alzheimers Dement. 2012;8(4):372–5. 66. Potter R, Patterson BW, Elbert DL, Ovod V, Kasten T, Sigurdson W, et al. Increased in vivo amyloid-β42 production, exchange, and loss in presenilin mutation carriers. Sci Transl Med. 2013;5(189):189ra77. 39. Tabrizi SJ, Leavitt BR, Landwehrmeyer GB, Wild EJ, Saft C, Barker RA, et al. Targeting huntingtin expression in patients with Huntington’s disease. N Engl J Med. 2019;380(24):2307–16. 67. Xu G, Ran Y, Fromholt SE, Fu C, Yachnis AT, Golde TE, et al. Murine Aβ over- production produces diffuse and compact Alzheimer-type amyloid deposits. Acta Neuropathol Commun. 2015;3:72. 40. Schneider LS, Raman R, Schmitt FA, Doody RS, Insel P, Clark CM, et al. Characteristics and performance of a modified version of the ADCS-CGIC CIBIC+ for mild cognitive impairment clinical trials. Alzheimer Dis Assoc Disord. 2009;23(3):260–7. 68. Sabbagh JJ, Kinney JW, Cummings JL. References References 1. Alzheimer’s Association. 2015 Alzheimer’s disease facts and figures. Alzheimers Dement. 2015;11(3):332–84. 2. Alzheimer's Association. 2016 Alzheimer’s disease facts and figures. Alzheimer Dement. 2016;12(4):459–509. 3. Mar J, Soto-Gordoa M, Arrospide A, Moreno-Izco F, Martinez-Lage P. Fitting the epidemiology and neuropathology of the early stages of Alzheimer’s disease to prevent dementia. Alzheimers Res Ther. 2015;7(1):2. 4. Brookmeyer R, Johnson E, Ziegler-Graham K, Arrighi HM. Forecasting the global burden of Alzheimer’s disease. Alzheimers Dement. 2007;3(3):186–91. 5. Alzheimer’s Association. 2017 Alzheimer’s disease facts and figures. Alzheimers Dement. 2017;13(4):325–73. 28. Doody RS, Tariot PN, Pfeiffer E, Olin JT, Graham SM. Meta-analysis of six- month memantine trials in Alzheimer’s disease. Alzheimers Dement. 2007; 3(1):7–17. 28. Doody RS, Tariot PN, Pfeiffer E, Olin JT, Graham SM. Meta-analysis of six- month memantine trials in Alzheimer’s disease. Alzheimers Dement. 2007; 3(1):7–17. 29. Greig SL. Memantine ER/donepezil: a review in Alzheimer’s disease. CNS Drugs. 2015;29(11):963–70. 29. Greig SL. Memantine ER/donepezil: a review in Alzheimer’s disease. CNS Drugs. 2015;29(11):963–70. 30. Reisberg B, Doody R, Stoffler A, Schmitt F, Ferris S, Mobius HJ, et al. Memantine in moderate-to-severe Alzheimer’s disease. N Engl J Med. 2003; 348(14):1333–41. 30. Reisberg B, Doody R, Stoffler A, Schmitt F, Ferris S, Mobius HJ, et al. Memantine in moderate-to-severe Alzheimer’s disease. N Engl J Med. 2003; 348(14):1333–41. 31. Strohle A, Schmidt DK, Schultz F, Fricke N, Staden T, Hellweg R, et al. Drug and exercise treatment of Alzheimer disease and mild cognitive 31. Strohle A, Schmidt DK, Schultz F, Fricke N, Staden T, Hellweg R, et al. Drug and exercise treatment of Alzheimer disease and mild cognitive Page 12 of 14 Page 12 of 14 Page 12 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 (2019) 11:76 impairment: a systematic review and meta-analysis of effects on cognition in randomized controlled trials. Am J Geriatr Psychiatry. 2015;23(12):1234–49. impairment: a systematic review and meta-analysis of effects on cognition in randomized controlled trials. Am J Geriatr Psychiatry. 2015;23(12):1234–49. mpairment: a systematic review and meta-analysis of effects on cognit 60. Shen DD, Artru AA, Adkison KK. Principles and applicability of CSF sampling for the assessment of CNS drug delivery and pharmacodynamics. Adv Drug Deliv Rev. 2004;56(12):1825–57. 32. Sevigny J, Chiao P, Bussiere T, Weinreb PH, Williams L, Maier M, et al. The antibody aducanumab reduces Aβ plaques in Alzheimer’s disease. Nature. 2016;537(7618):50–6. 61. References Satlin A, Wang J, Logovinsky V, Berry S, Swanson C, Dhadda S, et al. Design of a Bayesian adaptive phase 2 proof-of-concept trial for BAN2401, a putative disease-modifying monoclonal antibody for the treatment of Alzheimer’s disease. Alzheimers Dement (N Y). 2016;2(1):1–12. 51. Wyatt PG, Gilbert IH, Read KD, Fairlamb AH. Target validation: linking target and chemical properties to desired product profile. Curr Top Med Chem. 2011;11(10):1275–83. 52. Cummings J, Ritter A, Zhong K. Clinical trials for disease-modifying therapies in Alzheimer’s disease: a primer, lessons learned, and a blueprint for the future. J Alzheimers Dis. 2017; In press. 77. Dalgaard L. Comparison of minipig, dog, monkey and human drug metabolism and disposition. J Pharmacol Toxicol Methods. 2015;74:80–92. 78. Galijatovic-Idrizbegovic A, Miller JE, Cornell WD, Butler JA, Wollenberg GK, Sistare FD, et al. Role of chronic toxicology studies in revealing new toxicities. Regul Toxicol Pharmacol. 2016;82:94–8. p 53. Banks WA. Developing drugs that can cross the blood-brain barrier: applications to Alzheimer’s disease. BMC Neurosci. 2008;9(Suppl 3):S2. 54. Walters WP. Going further than Lipinski’s rule in drug design. Expert Opin Drug Discov. 2012;7(2):99–107. 79. Coric V, Salloway S, van Dyck CH, Dubois B, Andreasen N, Brody M, et al. Targeting prodromal Alzheimer disease with avagacestat: a randomized clinical trial. JAMA Neurol. 2015;72(11):1324–33. 55. Ticehurst MD, Marziano I. Integration of active pharmaceutical ingredient solid form selection and particle engineering into drug product design. J Pharm Pharmacol. 2015;67(6):782–802. 80. Egan MF, Kost J, Voss T, Mukai Y, Aisen PS, Cummings JL, et al. Randomized trial of verubecestat for prodromal Alzheimer’s disease. N Engl J Med. 2019; 380(15):1408–20. 56. Mignani S, Huber S, Tomas H, Rodrigues J, Majoral JP. Compound high- quality criteria: a new vision to guide the development of drugs, current situation. Drug Discov Today. 2016;21(4):573–84. 81. Doody RS, Raman R, Sperling RA, Seimers E, Sethuraman G, Mohs R, et al. Peripheral and central effects of gamma-secretase inhibition by semagacestat in Alzheimer’s disease. Alzheimers Res Ther. 2015;7(1):36. 57. Steinmetz KL, Spack EG. The basics of preclinical drug development for neurodegenerative disease indications. BMC Neurol. 2009;9(Suppl 1):S2. 82. Hoffman KB, Dimbil M, Tatonetti NP, Kyle RF. A pharmacovigilance signaling system based on FDA regulatory action and post-marketing adverse event reports. Drug Saf. 2016;39(6):561–75. 58. Lacombe O, Videau O, Chevillon D, Guyot AC, Contreras C, Blondel S, et al. In vitro primary human and animal cell-based blood-brain barrier models as a screening tool in drug discovery. References Mol Pharm. 2011;8(3):651–63. In vitro primary human and animal cell-based blood-brain barrier models as a screening tool in drug discovery. Mol Pharm. 2011;8(3):651–63. 83. Hauber AB, Johnson FR, Fillit H, Mohamed AF, Leibman C, Arrighi HM, et al. Older Americans’ risk-benefit preferences for modifying the course of Alzheimer disease. Alzheimer Dis Assoc Disord. 2009;23(1):23–32. 59. Lin JH. CSF as a surrogate for assessing CNS exposure: an industrial perspective. Curr Drug Metab. 2008;9(1):46–59. Page 13 of 14 Page 13 of 14 Page 13 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 (2019) 11:76 t al. Alzheimer's Research & Therapy (2019) 11:76 in Alzheimer’s disease: results of a randomized, double-blind, placebo- controlled phase 2a study. Alzheimers Res Ther. 2018;10(1):107. 84. Wehling M. Assessing the translatability of drug projects: what needs to be scored to predict success? Nat Rev Drug Discov. 2009;8(7):541–6. in Alzheimer’s disease: results of a randomized, double-blind, placebo- controlled phase 2a study. Alzheimers Res Ther. 2018;10(1):107. 109. Portelius E, Zetterberg H, Dean RA, Marcil A, Bourgeois P, Nutu M, et al. Amyloid-beta (1-15/16) as a marker for gamma-secretase inhibition in Alzheimer’s disease. J Alzheimers Dis. 2012;31(2):335–41. 85. Wendler A, Wehling M. Translatability score revisited: differentiation for distinct disease areas. J Transl Med. 2017;15(1):226. 86. Wendler A, Wehling M. Translatability scoring in drug development: eight case studies. J Transl Med. 2012;10:39. 110. Luo D, Smith JA, Meadows NA, Schuh A, Manescu KE, Bure K, et al. A quantitative assessment of factors affecting the technological development and adoption of companion diagnostics. Front Genet. 2015;6:357. 87. Figueroa-Magalhaes MC, Jelovac D, Connolly R, Wolff AC. Treatment of HER2-positive breast cancer. Breast. 2014;23(2):128–36. and adoption of companion diagnostics. Front Genet. 2015;6:357 111. Cummings J. The role of biomarkers in Alzheimer’s disease drug development. Adv Exp Med Biol. In Press. 88. Green DJ, Liu XI, Hua T, Burnham JM, Schuck R, Pacanowski M, et al. Enrichment strategies in pediatric drug development: an analysis of trials submitted to the US Food and Drug Administration. Clin Pharmacol Ther. 2018;104(5):983–8. 112. Cummings J, Zhong K, Cordes D. Drug development in Alzheimer’s disease: the role of default mode network assessment in phase II. US Neurol. 2017; 13(2):67–9. 89. Arneric SP, Kern VD, Stephenson DT. Regulatory-accepted drug development tools are needed to accelerate innovative CNS disease treatments. Biochem Pharmacol. 2018;151:291–306. 113. Schultz AP, Chhatwal JP, Hedden T, Mormino EC, Hanseeuw BJ, Sepulcre J, et al. References Phases of hyperconnectivity and hypoconnectivity in the default mode and salience networks track with amyloid and tau in clinically normal individuals. J Neurosci. 2017;37(16):4323–31. 90. Cummings J. The National Institute on Aging-Alzheimer’s Association framework on Alzheimer’s disease: application to clinical trials. Alzheimers Dement. 2018; Epub ahead of print. https://doi.org/10.1016/j.jalz.2018.05.006 114. Sepulcre J, Sabuncu MR, Li Q, El Fakhri G, Sperling R, Johnson KA. Tau and amyloid beta proteins distinctively associate to functional network changes in the aging brain. Alzheimers Dement. 2017;13(11):1261–9. 91. Jack CR Jr, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA research framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14(4):535–62. 115. Buckley RF, Schultz AP, Hedden T, Papp KV, Hanseeuw BJ, Marshall G, et al. Functional network integrity presages cognitive decline in preclinical Alzheimer disease. Neurology. 2017;89(1):29–37. 92. Kuhle J, Disanto G, Lorscheider J, Stites T, Chen Y, Dahlke F, et al. Fingolimod and CSF neurofilament light chain levels in relapsing-remitting multiple sclerosis. Neurology. 2015;84(16):1639–43. 116. Koch K, Myers NE, Gottler J, Pasquini L, Grimmer T, Forster S, et al. Disrupted intrinsic networks link amyloid-beta pathology and impaired cognition in prodromal Alzheimer’s disease. Cereb Cortex. 2015;25(12):4678–88. 93. Molinuevo JL, Ayton S, Batrla R, Bednar MM, Bittner T, Cummings J, et al. Current state of Alzheimer’s fluid biomarkers. Acta Neuropathol. 2018;136(6):821–53. 117. Weiler M, de Campos BM, Nogueira MH, Pereira Damasceno B, Cendes F, Balthazar ML. Structural connectivity of the default mode network and cognition in Alzheimers disease. Psychiatry Res. 2014;223(1):15–22. 94. Mills SM, Mallmann J, Santacruz AM, Fuqua A, Carril M, Aisen PS, et al. Preclinical trials in autosomal dominant AD: implementation of the DIAN-TU trial. Rev Neurol (Paris). 2013;169(10):737–43. 95. Reiman EM, Langbaum JB, Tariot PN. Alzheimer’s prevention initiative: a proposal to evaluate presymptomatic treatments as quickly as possible. Biomark Med. 2010;4(1):3–14. 118. Wilson FJ, Danjou P. Early decision-making in drug development: the potential role of pharmaco-EEG and pharmaco-sleep. Neuropsychobiology. 2015;72(3–4):188–94. 96. Qian J, Wolters FJ, Beiser A, Haan M, Ikram MA, Karlawish J, et al. APOE- related risk of mild cognitive impairment and dementia for prevention trials: an analysis of four cohorts. PLoS Med. 2017;14(3):e1002254. 119. Wilson FJ, Leiser SC, Ivarsson M, Christensen SR, Bastlund JF. Can pharmaco- electroencephalography help improve survival of central nervous system drugs in early clinical development? Drug Discov Today. 2014;19(3):282–8. 97. References Zeitlow K, Charlambous L, Ng I, Gagrani S, Mihovilovic M, Luo S, et al. The biological foundation of the genetic association of TOMM40 with late-onset Alzheimer’s disease. Biochim Biophys Acta. 2017. 120. Chen MK, Mecca AP, Naganawa M, Finnema SJ, Toyonaga T, Lin SF, et al. Assessing synaptic density in Alzheimer disease with synaptic vesicle glycoprotein 2A positron emission tomographic imaging. JAMA Neurol. 2018;75(10):1215–24. 98. Boluda S, Toledo JB, Irwin DJ, Raible KM, Byrne MD, Lee EB, et al. A comparison of Abeta amyloid pathology staging systems and correlation with clinical diagnosis. Acta Neuropathol. 2014;128(4):543–50. 121. Liu E, Schmidt ME, Margolin R, Sperling R, Koeppe R, Mason NS, et al. Amyloid-beta 11C-PiB-PET imaging results from 2 randomized bapineuzumab phase 3 AD trials. Neurology. 2015;85(8):692–700. 99. de Souza LC, Sarazin M, Teixeira-Junior AL, Caramelli P, Santos AE, Dubois B. Biological markers of Alzheimer’s disease. Arq Neuropsiquiatr. 2014;72(3): 227–31. 122. Panza F, Solfrizzi V, Imbimbo BP, Giannini M, Santamato A, Seripa D, et al. Efficacy and safety studies of gantenerumab in patients with Alzheimer’s disease. Expert Rev Neurother. 2014;14(9):973–86. 100. Scholl M, Lockhart SN, Schonhaut DR, O’Neil JP, Janabi M, Ossenkoppele R, et al. PET imaging of tau deposition in the aging human brain. Neuron. 2016;89(5):971–82. 123. Cummings J, Blennow K, Johnson K, Keeley M, Bateman RJ, Molinuevo JL, et al. Anti-tau trials for Alzheimer’s disease: a report from the EU/US/CTAD Task Force. J Prev Alzheimers Dis. 2019;6(3):157–63. 101. Sevigny J, Suhy J, Chiao P, Chen T, Klein G, Purcell D, et al. Amyloid PET screening for enrichment of early-stage Alzheimer disease clinical trials: experience in a phase 1b clinical trial. Alzheimer Dis Assoc Disord. 2016; 30(1):1–7. 124. Apostolova LG, Zarow C, Biado K, Hurtz S, Boccardi M, Somme J, et al. Relationship between hippocampal atrophy and neuropathology markers: a 7T MRI validation study of the EADC-ADNI Harmonized Hippocampal Segmentation Protocol. Alzheimers Dement. 2015;11(2):139–50. 102. Jack CR Jr, Knopman DS, Chetelat G, Dickson D, Fagan AM, Frisoni GB, et al. Suspected non-Alzheimer disease pathophysiology--concept and controversy. Nat Rev Neurol. 2016;12(2):117–24. 125. Csernansky JG, Hamstra J, Wang L, McKeel D, Price JL, Gado M, et al. Correlations between antemortem hippocampal volume and postmortem neuropathology in AD subjects. Alzheimer Dis Assoc Disord. 2004;18(4):190–5. 103. Ballard C, Atri A, Boneva N, Cummings JL, Frolich L, Molinuevo JL, et al. Enrichment factors for clinical trials in mild-to-moderate Alzheimer’s disease. Alzheimers Dement (N Y). 2019;5:164–74. 126. References Designing drug trials for Alzheimer’s disease: what we have learned from the release of the phase III antibody trials: a report from the EU/US/CTAD Task Force. Alzheimers Dement. 2013;9(4):438–44. 132. Henley D, Raghavan N, Sperling R, Aisen P, Raman R, Romano G. Preliminary results of a trial of atabecestat in preclinical Alzheimer’s disease. N Engl J Med. 2019;380(15):1483–5. 155. Das S, Lo AW. Re-inventing drug development: a case study of the I-SPY 2 breast cancer clinical trials program. Contemp Clin Trials. 2017;62:168–74. 133. Sperling RA, Jack CR Jr, Black SE, Frosch MP, Greenberg SM, Hyman BT, et al. Amyloid-related imaging abnormalities in amyloid-modifying therapeutic trials: recommendations from the Alzheimer’s Association Research Roundtable Workgroup. Alzheimers Dement. 2011;7(4):367–85. 156. Finger EC, Berry S, Cummings J, Coleman K, Hsiung R, Feldman HH, et al. Adaptive cross-over designs for assessment of symptomatic treatments targeting behavior in neurodegenerative disease: a phase 2 clinical trial of intranasal oxytocin for frontotemporal dementia (FOXY). Alzheimers Res Ther. 2018;10:102–9. 134. Ketter N, Brashear HR, Bogert J, Di J, Miaux Y, Gass A, et al. Central review of amyloid-related imaging abnormalities in two phase III clinical trials of bapineuzumab in mild-to-moderate Alzheimer’s disease patients. J Alzheimers Dis. 2017;57(2):557–73. 157. Ritchie CW, Molinuevo JL, Truyen L, Satlin A, Van der Geyten S, Lovestone S. Development of interventions for the secondary prevention of Alzheimer’s dementia: the European Prevention of Alzheimer’s Dementia (EPAD) project. Lancet Psychiatry. 2016;3(2):179–86. 135. Ostrowitzki S, Lasser RA, Dorflinger E, Scheltens P, Barkhof F, Nikolcheva T, et al. A phase III randomized trial of gantenerumab in prodromal Alzheimer’s disease. Alzheimers Res Ther. 2017;9(1):95. 158. Bateman RJ, Benzinger TL, Berry S, Clifford DB, Duggan C, Fagan AM, et al. The DIAN-TU Next Generation Alzheimer’s prevention trial: adaptive design and disease progression model. Alzheimers Dement. 2017;13(1):8–19. 136. Dubois B, Feldman HH, Jacova C, Cummings JL, Dekosky ST, Barberger- Gateau P, et al. Revising the definition of Alzheimer’s disease: a new lexicon. Lancet Neurol. 2010;9(11):1118–27. 159. Grill JD, Raman R, Ernstrom K, Aisen P, Dowsett SA, Chen YF, et al. Comparing recruitment, retention, and safety reporting among geographic regions in multinational Alzheimer’s disease clinical trials. Alzheimers Res Ther. 2015;7(1):39–53. 137. Dubois B, Feldman HH, Jacova C, Hampel H, Molinuevo JL, Blennow K, et al. Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol. 2014;13(6):614–29. 138. Herscovici R, Kutyifa V, Barsheshet A, Solomon S, McNitt S, Polonsky B, et al. References Whitwell JL, Jack CR Jr, Pankratz VS, Parisi JE, Knopman DS, Boeve BF, et al. Rates of brain atrophy over time in autopsy-proven frontotemporal dementia and Alzheimer disease. Neuroimage. 2008;39(3):1034–40. 104. Soares HD. The use of mechanistic biomarkers for evaluating investigational CNS compounds in early drug development. Curr Opin Investig Drugs. 2010;11(7):795–801. 127. Fox NC, Black RS, Gilman S, Rossor MN, Griffith SG, Jenkins L, et al. Effects of Abeta immunization (AN1792) on MRI measures of cerebral volume in Alzheimer disease. Neurology. 2005;64(9):1563–72. 105. Wagner JA. Strategic approach to fit-for-purpose biomarkers in drug development. Annu Rev Pharmacol Toxicol. 2008;48:631–51. 128. Novak G, Fox N, Clegg S, Nielsen C, Einstein S, Lu Y, et al. Changes in brain volume with bapineuzumab in mild to moderate Alzheimer’s disease. J Alzheimers Dis. 2015;49(4):1123–34. 106. Bateman RJ, Munsell LY, Morris JC, Swarm R, Yarasheski KE, Holtzman DM. Human amyloid-beta synthesis and clearance rates as measured in cerebrospinal fluid in vivo. Nat Med. 2006;12:856–61. 107. Kennedy ME, Stamford AW, Chen X, Cox K, Cumming JN, Dockendorf MF, et al. The BACE1 inhibitor verubecestat (MK-8931) reduces CNS beta-amyloid in animal models and in Alzheimer’s disease patients. Sci Transl Med. 2016; 8(363):363ra150. 129. Mattsson N, Andreasson U, Zetterberg H, Blennow K, Alzheimer’s Disease Neuroimaging I. Association of plasma neurofilament light with neurodegeneration in patients with Alzheimer disease. JAMA Neurol. 2017;74(5):557–66. 130. Johnson KA, Schultz A, Betensky RA, Becker JA, Sepulcre J, Rentz D, et al. Tau positron emission tomographic imaging in aging and early Alzheimer disease. Ann Neurol. 2016;79(1):110–9. 108. Scheltens P, Hallikainen M, Grimmer T, Duning T, Gouw AA, Teunissen CE, et al. Safety, tolerability and efficacy of the glutaminyl cyclase inhibitor PQ912 Page 14 of 14 Page 14 of 14 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 Cummings et al. Alzheimer's Research & Therapy (2019) 11:76 dementia: diagnostic value of the Amsterdam Instrumental Activities of Daily Living Questionnaire. J Geriatr Psychiatry Neurol. 2013;26(4):244–50. dementia: diagnostic value of the Amsterdam Instrumental Activities of Daily Living Questionnaire. J Geriatr Psychiatry Neurol. 2013;26(4):244–50. 131. Tarawneh R, Head D, Allison S, Buckles V, Fagan AM, Ladenson JH, et al. Cerebrospinal fluid markers of neurodegeneration and rates of brain atrophy in early Alzheimer disease. JAMA Neurol. 2015;72(6):656–65. 154. Vellas B, Carrillo MC, Sampaio C, Brashear HR, Siemers E, Hampel H, et al. References Online study partner-reported cognitive decline in the Brain Health Registry. Alzheimers Dement (N Y). 2018;4:565–74. 145. Langbaum JB, Hendrix SB, Ayutyanont N, Chen K, Fleisher AS, Shah RC, et al. An empirically derived composite cognitive test score with improved power to track and evaluate treatments for preclinical Alzheimer’s disease. Alzheimers Dement. 2014;10(6):666–74. 166. Zhong K, Cummings J. Healthybrains.org: from registry to randomization. J Prev Alz Dis. 2016;3(3):123–6. 166. Zhong K, Cummings J. Healthybrains.org: from registry to randomization. J Prev Alz Dis. 2016;3(3):123–6. 167. Tan MH, Thomas M, MacEachern MP. Using registries to recruit subjects for clinical trials. Contemp Clin Trials. 2015;41:31–8. 167. Tan MH, Thomas M, MacEachern MP. Using registries to recruit subjects for clinical trials. Contemp Clin Trials. 2015;41:31–8. 146. Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. 1993;43(11):2412–4. 168. Paul SM, Mytelka DS, Dunwiddie CT, Persinger CC, Munos BH, Lindborg SR, et al. How to improve R&D productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov. 2010;9(3):203–14. 168. Paul SM, Mytelka DS, Dunwiddie CT, Persinger CC, Munos BH, Lindborg SR, et al. How to improve R&D productivity: the pharmaceutical industry’s grand challenge. Nat Rev Drug Discov. 2010;9(3):203–14. 147. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry. 1984;141(11):1356–64. 148. Harrison J, Minassian SL, Jenkins L, Black RS, Koller M, Grundman M. A neuropsychological test battery for use in Alzheimer disease clinical trials. Arch Neurol. 2007;64(9):1323–9. References Early intervention and long-term outcome with cardiac resynchronization therapy in patients without a history of advanced heart failure symptoms. Eur J Heart Fail. 2015;17(9):964–70. 160. Henley DB, Dowsett SA, Chen YF, Liu-Seifert H, Grill JD, Doody RS, et al. Alzheimer’s disease progression by geographical region in a clinical trial setting. Alzheimers Res Ther. 2015;7(1):43–52. 161. Cummings JL, Atri A, Ballard C, Boneva N, Frolich L, Molinuevo JL, et al. Insights into globalization: comparison of patient characteristics and disease progression among geographic regions in a multinational Alzheimer’s disease clinical program. Alzheimer Res Ther. 2018;10:116–28. 139. Reiman EM, Langbaum JB, Tariot PN, Lopera F, Bateman RJ, Morris JC, et al. CAP--advancing the evaluation of preclinical Alzheimer disease treatments. Nat Rev Neurol. 2016;12(1):56–61. 140. Sperling RA, Jack CR Jr, Aisen PS. Testing the right target and right drug at the right stage. Sci Transl Med. 2011;3(111):111cm33. 162. Kennedy RE, Cutter GR, Wang G, Schneider LS. Challenging assumptions about African American participation in Alzheimer disease trials. Am J Geriatr Psychiatry. 2017;25(10):1150–9. 141. Sperling R, Mormino E, Johnson K. The evolution of preclinical Alzheimer’s disease: implications for prevention trials. Neuron. 2014;84(3):608–22. 163. Cummings JL, Aisen P, Barton R, Bork J, Doody R, Dwyer J, et al. Re- engineering Alzheimer clinical trials: Global Alzheimer Platform Network. J Prevent Alz Dis. 2016;3:114–20. 142. Cox CG, Ryan BAM, Gillen DL, Grill JD. A preliminary study of clinical trial enrollment decisions among people with mild cognitive impairment and their study partners. Am J Geriatr Psychiatry. 2019;27(3):322–32. 164. Mackin RS, Insel PS, Truran D, Finley S, Flenniken D, Nosheny R, et al. Unsupervised online neuropsychological test performance for individuals with mild cognitive impairment and dementia: results from the Brain Health Registry. Alzheimers Dement (Amst). 2018;10:573–82. 143. Romero K, Ito K, Rogers JA, Polhamus D, Qiu R, Stephenson D, et al. The future is now: model-based clinical trial design for Alzheimer’s disease. Clin Pharmacol Ther. 2015;97(3):210–4. 144. Weintraub S, Carrillo MC, Farias ST, Goldberg TE, Hendrix JA, Jaeger J, et al. Measuring cognition and function in the preclinical stage of Alzheimer’s disease. Alzheimers Dement (N Y). 2018;4:64–75. 165. Nosheny RL, Camacho MR, Insel PS, Flenniken D, Fockler J, Truran D, et al. Online study partner-reported cognitive decline in the Brain Health Registry. Alzheimers Dement (N Y). 2018;4:565–74. 165. Nosheny RL, Camacho MR, Insel PS, Flenniken D, Fockler J, Truran D, et al. 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. 149. Wang J, Logovinsky V, Hendrix SB, Stanworth SH, Perdomo C, Xu L, et al. ADCOMS: a composite clinical outcome for prodromal Alzheimer’s disease trials. J Neurol Neurosurg Psychiatry. 2016;87:993–9. 150. Schmitt FA, Ashford W, Ernesto C, Saxton J, Schneider LS, Clark CM, et al. The severe impairment battery: concurrent validity and the assessment of longitudinal change in Alzheimer’s disease. The 6 Study. Alzheimer Dis Assoc Disord. 1997;11(Suppl 2):S51–S6. 151. Galasko D, Bennett D, Sano M, Ernesto C, Thomas R, Grundman M, et al. An inventory to assess activities of daily living for clinical trials in Alzheimer’s disease. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord. 1997;11(Suppl 2):S33–S9. 152. Jutten RJ, Peeters CFW, Leijdesdorff SMJ, Visser PJ, Maier AB, Terwee CB, et al. Detecting functional decline from normal aging to dementia: development and validation of a short version of the Amsterdam IADL Questionnaire. Alzheimers Dement (Amst). 2017;8:26–35. 153. Sikkes SA, Pijnenburg YA, Knol DL, de Lange-de Klerk ES, Scheltens P, Uitdehaag BM. Assessment of instrumental activities of daily living in
https://openalex.org/W3041065007	https://strathprints.strath.ac.uk/73085/1/Hay_WHR_2020_feminism_and_national_identity_in_the_Scottish_abortion_campaign_c_1975_1990.pdf	English	null	‘More than a defence against bills’: feminism and national identity in the Scottish abortion campaign, c. 1975–1990	Women's history review	2,020	cc-by	11,558	© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group 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 cited. Wellcome Library [hereafter WL] SA/NAC/F/1/3, G. Smyth Reply to Statement about NAC made by NPM (1982). I am grateful to my supervisor Dr Laura Kelly for her constant encouragement and support from the beginning stages of this article to the end. I am also thankful for the activists I interviewed, who gave up their time to share their experi- ences with me and provided the beating heart of this work. I would also like to thank Dr Carmen M. Mangion and the two reviewers for their insightful and developmental comments on an earlier draft of this article. Their feedback was tremen- dously helpful and signiﬁcantly improved this work. Finally, I am indebted to the Wellcome Trust for supporting this research (Ref: 208532/Z/17/Z). CONTACT Kristin Hay kristin.hay@strath.ac.uk Department of History, University of Strathclyde, Level 4, Lord Hope Building, 141 St James Road, Glasgow G4 0LT, United Kingdom WOMEN’S HISTORY REVIEW https://doi.org/10.1080/09612025.2020.1791405 WOMEN’S HISTORY REVIEW https://doi.org/10.1080/09612025.2020.1791405 KEYWORDS Abortion; feminism; Scotland; pro-choice; identity; oral history; activism; reproductive rights Kristin Hay Department of History, University of Strathclyde, Glasgow, United Kingdom KEYWORDS Abortion; feminism; Scotland; pro-choice; identity; oral history; activism; reproductive rights Wellcome Library [hereafter WL] SA/NAC/F/1/3, G. Smyth ‘Reply to Statement about NAC made by NPM’ (1982). ‘More than a defence against bills’: feminism and national identity in the Scottish abortion campaign, c. 1975–1990 Kristin Hay Kristin Hay ABSTRACT The Abortion Act is described by Gayle Davis as ‘one of the most signiﬁcant and conten- tious medico-social legislative developments of the twentieth century.’3 Following the suc- cessful, yet highly controversial, Private Members Bill by David Steel MP and subsequent implementation in 1968, the Abortion Act legalised pregnancy terminations conducted by a doctor between the ﬁrst and twenty-fourth week of pregnancy in England, Wales and Scotland. It also enabled doctors to consider socioeconomic factors when deciding to permit an abortion. Although this had theoretically existed in Scotland prior to 1967— where doctors could legally oﬀer terminations in ‘good faith’—this was the ﬁrst piece of abortion legislation to cover all British nations, excluding Northern Ireland.4 Emerging during a time of socio-political change in Britain—considered by some commentators as a ‘Sexual Revolution’—the Act was seen to both facilitate and demonstrate the relaxing of attitudes towards sex and sexuality within Britain.5 Its apparent success was demonstrated the following year, with an increase across Britain from 2,300 registered terminations a year to 9,700 on the National Health Service (NHS) by the end of the 1960s.6 y However, legislation did not lead to a natural acceptance of pregnancy termination. Indeed, the cultural shift from the prejudiced to the permissive took far longer to arrive and, following the implementation of the Act, the provision of abortion facilities remained inadequate, information was limited, and anti-abortion sentiments dominated political, medical and public spheres. This medico-moral paradox was heightened in Scotland, where the provision of abortion care was sporadic and the hostility of doctors towards the procedure were most pronounced. Additionally, historians Roger Davidson and Gayle Davis show that nine attempts were made to restrict and limit legal access to abor- tion in the ﬁrst six years of its implementation—the most infamous of which were for- warded by Scottish MPs.7 This highlights a uniquely hostile climate towards abortion in Scotland, with anti-abortion activism building momentum following the 1967 Act and inﬂuencing implementation. In response to increasing parliamentary challenges to legal abortion and compounded by the disparities in the provision of abortion healthcare, the SAC formed in 1980. The SAC was a coalition of feminists, trade unionists, politicians and health workers who sought to form a strong pro-choice presence in Scotland and defend the Abortion Act from parliamentary attack. ABSTRACT ABSTRACT In 1967, the Abortion Act was passed which legalised terminations conducted by a doctor between the ﬁrst and twenty-fourth week of pregnancy in Britain. Despite having abortion legalised in law, the cultural shift from the prejudiced to permissive took far longer to arrive, and women continued to face barriers to their reproductive autonomy in the 1970s and 1980s. The inequalities in abortion healthcare were most pronounced in Scotland, prompting the emergence of the Scottish Abortion Campaign in 1980. This article explores the role of the Scottish Abortion Campaign and their eﬀorts to defend and extend the Abortion Act of 1967 in post-legalisation Scotland. Through archival research and oral testimony, this article demonstrates that the SAC operated semi-autonomously from the broader National Abortion Campaign, demonstrating a distinctive, Scottish feminist identity that inﬂuenced their activities during this period. We in Scotland … want NAC to be much more than a defence against bills—we want to ﬁght for right to choose, and that means ﬁghting for more and better facilities; ﬁghting against cuts in the NHS and ﬁghting against the anti-abortionists. These are the ﬁghts and campaigns we are conducting in Scotland. These are what keep our local groups active.1 (Geri Smyth, Scot- tish Abortion Campaign, 1982) In the early 1980s, the National Abortion Campaign (NAC) was at a crossroads. Believing that their ‘focus on abortion’ was ‘ignoring too many other, equally important and totally related, issues,’ some Steering Committee members suggested that the organ- isation diversify the organisation to include all reproductive issues. Whilst the motion was supported by a number of branches in England—including Leeds, Merseyside 2 K. HAY 2 K. HAY 2 and Norwich—all branches in Scotland were, in contrast, fervently against reorganis- ation. In response, members of the Scottish Abortion Campaign (SAC) argued: ‘without the right to control her body, a woman cannot control her destiny … why abandon the original demands of the Campaign, when we have not yet won them?’2 Their comments suggest that in the aftermath of the legalisation of abortion, women in Scotland continued to face distinctive barriers to their reproductive autonomy. As a result, the SAC rejected proposals to expand campaign energies, believing that they had not yet obtained ‘a women’s right to choose’ for all women in Britain. This exchange epitomises the diﬀerences between abortion activism on a national and local level fol- lowing the introduction of the Abortion Act of 1967. ABSTRACT The founders of ALRA included notable feminists such as Stella Browne and Janet Chance, as well as physician Joan Malleson.11 The emergence of women in abortion activism reframed abortion as a feminist cause, which played a signiﬁcant role in its subsequent legalisa- tion. Similarly, after legalisation, the NAC—and latterly, the SAC—was born out of the Women’s Liberation Movement of the 1960s, emphasising the inﬂuence of feminism within the organisation with their slogan ‘a women’s right to choose.’12 However, Lesley The history of abortion activism in Britain remains a limited area of historical inquiry —despite the central role of activists in achieving legal abortion during the 1950s and 1960s, and the consequent signiﬁcance of pro-choice activism in the post-legalisation era.10 However, it is universally acknowledged that the campaign for abortion rights in Britain was facilitated by the growing inﬂuence of feminism during the twentieth century. Historian Barbara Brookes’ signiﬁcant work on the legalisation of abortion in Britain shows that women remained ‘on the periphery of the medical and legal debate on abortion,’ until the formation of the Abortion Law Reform Association (ALRA) in 1936. As a key example of early women’s health activism in Britain, ALRA played a central role in the implementation of the Abortion Act of 1967. The founders of ALRA included notable feminists such as Stella Browne and Janet Chance, as well as physician Joan Malleson.11 The emergence of women in abortion activism reframed abortion as a feminist cause, which played a signiﬁcant role in its subsequent legalisa- tion. Similarly, after legalisation, the NAC—and latterly, the SAC—was born out of the Women’s Liberation Movement of the 1960s, emphasising the inﬂuence of feminism within the organisation with their slogan ‘a women’s right to choose.’12 However, Lesley Hoggart’s examination of the organisation suggests that the feminist ambitions of the NAC were often diluted through their connections to the Labour Party and the trade union movement—which, at the time, was highly male-dominated. As a single-issue campaign, the NAC had to compromise to appease the political alignments of all those who were attracted to the cause, which meant their more feminist-leaning aspira- tions ‘only had minority support,’ amongst aﬃliates. ABSTRACT Having operated as individual branches of the NAC from 1975, historian Sarah Browne argues that the perceived distinctiveness of Scottish women’s experiences of abortion encouraged these groups to amalgamate into a ‘decen- tralised group’ of the wider organisation. Although still part of the NAC, the abortion campaign in Scotland exercised a great deal of autonomy in their campaign strategies.8 This was reﬂected in their Constitution: WOMEN’S HISTORY REVIEW 3 3 (1) To establish the right in law for a woman to choose whether to continue or terminate a pregnancy. (1) To establish the right in law for a woman to choose whether to continue or terminate a pregnancy. (2) To campaign for free abortion on demand available on the NHS without let or hin- drance, whatever the nationality of the woman. (2) To campaign for free abortion on demand available on the NHS without let or hin- drance, whatever the nationality of the woman. y To publicise the inequality of abortion provision throughout Scotland. (3) To publicise the inequality of abortion provision throughout Scotland. (4) To work towards abortion being given a more positive place in existing sex education programmes.9 (4) To work towards abortion being given a more positive place in existing sex education programmes.9 Consequently, the Scottish campaign operated diﬀerently to its British counterpart. Responding directly to the needs of Scottish women and their access to legal abortion, they had their own Steering Committee, aﬃliations and campaign strategies. However, this unique example of women’s health activism during the second half of the twentieth century is absent within the dominant historiography. y g p y The history of abortion activism in Britain remains a limited area of historical inquiry —despite the central role of activists in achieving legal abortion during the 1950s and 1960s, and the consequent signiﬁcance of pro-choice activism in the post-legalisation era.10 However, it is universally acknowledged that the campaign for abortion rights in Britain was facilitated by the growing inﬂuence of feminism during the twentieth century. Historian Barbara Brookes’ signiﬁcant work on the legalisation of abortion in Britain shows that women remained ‘on the periphery of the medical and legal debate on abortion,’ until the formation of the Abortion Law Reform Association (ALRA) in 1936. As a key example of early women’s health activism in Britain, ALRA played a central role in the implementation of the Abortion Act of 1967. ABSTRACT As a result, Hoggart concludes that the post-legal abortion activism was mostly reactionary: ‘without a defensive campaign to unite its supporters, NAC was not such a vibrant national organisation, and that it even had diﬃculty maintaining itself.’13 However, the Scottish experience appears to be notably diﬀerent. Whilst Hoggart asserts that in the national campaign feminist poli- tics heightened tension within the movement, Sarah Browne argues that single-issue campaigning surrounding reproductive health provided an avenue for continued femin- ist activism within Scotland.14 By virtue of the more centralised nature of their cam- paign, the SAC were able to operate more homogenously with trade unionists than within the NAC. This meant that their campaign had the freedom to pursue more radical aims—including free abortion on demand and the removal of two doctors’ approval. This indicates that abortion activism in Scotland had a unique, Scottish 4 K. HAY 4 K. HAY feminist identity which inﬂuenced their ambitions of both defending and extending the Abortion Act of 1967. This article combines underutilised archival research with oral testimony to highlight the distinctiveness of the campaign for legal abortion in Scotland during the late 1970s and 1980s. It will explore the heightened disparities in abortion healthcare provision fol- lowing the passing of the Abortion Act in 1967 which provided an impetus for renewed abortion activism; the ways in which the SAC recruited and motivations behind those who joined the campaign; and how the SAC attempted to not only defend the Abortion Act but extend it to free abortion on demand. In the United States, Jennifer Nelson has highlighted the signiﬁcance of what she terms ‘regional diversity,’ wherein the develop- ment of feminist health activism across America was often instigated by local—as opposed to national—conditions.15 The work of Esther Breitenbach and Sarah Browne, among others, further suggests that in the context of Britain, the diﬀerences between Scot- tish feminist activism and British feminist activism requires greater historical attention.16 This provides a useful methodological framework for understanding the divergences between the NAC and SAC during this period, and oﬀers a fresh lens through which to analyse post-legal abortion activism in Britain during the latter twentieth century. ABSTRACT Using the archives of the NAC and SAC from the Wellcome Library and Glasgow Women’s Library, supplemented by three oral history interviews with prominent SAC members, this article will demonstrate that feminism and the perception of a distinctive national identity strongly inﬂuenced the actions of the campaign: from how individuals decided to join to the ‘positive legislation’ put forward by the SAC and the criticisms they had of the Abortion Act itself.17 These underutilised sources emphasise the duality between feminism and Scottish national identity present within the SAC. Finally, it will show that despite the best intentions of the SAC, they ultimately failed to achieve their aim introducing positive legislation to secure free abortion on demand for Scottish women. Hoggart argues that such eﬀorts ‘seems to be hidden from history, possibly because they were not very successful.’18 Her accurate analysis demonstrates that stories of success are privileged in activist history narratives. This article thus seeks to problema- tise the concept of ‘success/failure’ binary of social movements and advocates for an improved way of evaluating the signiﬁcance of feminist health activism not by their suc- cesses but by their eﬀorts to inﬂuence social change. In 2017, Scotland was considered a pioneer in British abortion healthcare by allowing women to take the abortion pill at home.19 This suggests that the legacy of the SAC lies not in their short-term achievements, but their long-term legacy of curating a pro-choice consciousness in Scotland. Motivations for the cause: abortion in post-legalisation Scotland The emergence of the SAC in 1980 was in response to a growing awareness that the experi- ences of legal abortion in Scotland was distinctive to that of England and Wales. However, across Britain, issues occurred during the implementation of the Abortion Act. Although popular memory heralds the 1967 legislation as a monumental shift in attitudes towards female autonomy during the twentieth century, in reality it did little to change the legal position in Scotland and so many barriers remained in place which limited women’s access to pregnancy termination.20 In part, this was the result of the Abortion Act itself which remains, as Lee states, ‘one of the formally most restrictive abortion laws of WOMEN’S HISTORY REVIEW 5 those countries where abortion has been legalised.’ The act did not require Area Health Authorities (AHAs) in Scotland, England and Wales to provide abortion facilities, creating sporadic provision across the three countries. The law also made ‘no reference to freedoms and rights for women who want[ed] to terminate a pregnancy,’ removing any notion of ‘a women’s right to choose’ from the legislation.21 Instead, the 1967 law enabled doctors to decide between ‘deserving’ and ‘undeserving’ abortion seekers. This created an uneven power balance between ‘patients’ and practitioners, facilitated to a greater extent by the ambiguous language used in the act which legitimised doctors’ own interpretation and conscience. Yet, in Scotland, heightened hostilities towards abortion compounded these broader issues. Davis and Davidson show that a number of medical professionals in Scot- land, including doctors and nurses, were hostile towards their increased role in abortion position, through what they consider to be a ‘reluctant medicalisation’ of reproductive health.22 Furthermore, of the nine attempts to restrict the abortion act made between 1970 and 1990, James White, MP for Glasgow Pollock, and John Corrie, MP for Bute and Ayrshire, were among the most fervent and infamous.23 Doctors, MPs and pro-life pressure groups, including the Society for the Protection of the Unborn Child (SPUC), mounted a signiﬁcant anti-abortion lobby in Scotland in the post-legal period, leading to varied provision from region to region. In this melting pot of inaccessibility and hosti- lity, the conditions in Scotland prompted a centralised pro-choice response to emerge. y p p p p g One of the biggest issues which facilitated the emergence of the SAC in 1980 was the clear disparities in the provision of abortion healthcare in Scotland. Motivations for the cause: abortion in post-legalisation Scotland In 1987, 10,093 abor- tions were carried out in Scotland compared to 156,191 in England and Wales. These stat- istics translated into an abortion rate of 14.1 per 1,000 women in England and Wales and a 9 per 1,000 rate in Scotland.24 Although these statistics could be interpreted as an indi- cation of low demand, the particularly low rates in Scotland suggested that obtaining an abortion was more diﬃcult than elsewhere in Britain. This analysis was strengthened by the fact that women in Scotland continued to travel for their terminations. In 1972, the Family Planning Association (FPA) Clinic in Glasgow followed the experiences of 250 unmarried pregnant women. Of the 88 who obtained terminations, 35 were carried out in Glasgow whilst the other 53 were done in Birmingham.25 In 1980, the Scottish Home and Health Department (SHHD) estimated that of the total number of termin- ations carried out in Scotland, an additional fourteen per cent—around 1,000 Scottish women—annually travelled to England to obtain an abortion.26 This emphasises that the problem of access persisted into the later decades of the twentieth century. Thus, Scot- land’s low abortion rate was considered to be the result of inaccessibility rather than low demand. There were also diﬀerences in how terminations were accessed which further limited abortion services. In the early 1980s, ﬁfty per cent of abortions in England and Wales were conducted on the health service, with the other half performed in the charity or private health sector. However, in Scotland, ninety-nine per cent of all abortions were con- ducted on the NHS.27 Whilst, on the surface, it appeared that the Abortion Act was best implemented in the north, in reality the lack of NHS abortion facilities meant that women were limited in their options for where to obtain a termination. In 1978, 56,000 women obtained their abortions in Britain through charities—none of which had clinics in Scot- land.28 Consequently, Scottish women were reliant on the NHS to access abortions which, despite being free, led to long waiting lists and the risk of being assessed by a pro-life 6 K. HAY K. HAY K. HAY 6 6 doctor. Motivations for the cause: abortion in post-legalisation Scotland These issues were epitomised in 1982 at the Western Inﬁrmary, Glasgow, where a lack of facilities and hostile attitudes towards abortion made it ‘impossible for most women to get a termination there even if … they are largely entitled to one.’29 Another report in 1981 further suggested that abortion services were in ‘crisis.’30 In England and Wales, the British Pregnancy Advisory Service (BPAS), a pro-choice charity, charged a small fee for women to obtain abortions. Yet, when BPAS set up their only Scottish clinic in Glasgow, they too could only provide NHS referrals.31 Therefore, Scotland had no charity or private sector to supplement the increasing numbers of women seeking ter- minations through the NHS in the years following legal abortion, and without the require- ment to fund abortion services, these services remained inadequate. This particularly impacted working-class women and women from remote, rural areas. As the FPA in Glasgow argued: Often the only way in which termination can be obtained [in Scotland] is for the girl to be given the address of a pregnancy advisory centre in England … this costs the patient from £45 upwards depending on the stage of the pregnancy.32 The high cost of travelling and paying for a pregnancy termination in England thus priced out many young, working-class women from accessing abortions. The continuation of the need to travel emphasised the poor state of abortion provision acutely in the north, and thus encouraged a coordinated response to a distinctively Scottish issue. Finally, not only were facilities on a national level inadequate, but also on a regional level. Location had a signiﬁcant impact on one’s ability to access abortion healthcare. Indi- vidual area health boards were responsible for the provision of family planning services, which led to variances in abortion facilities. Ann, a long-time trade union activist and SAC member, recalled the consequences of these regional variations: I know we had many health boards at that time, a huge number of Area Health Boards, and they all delivered, em, [the Abortion Act] in a slightly diﬀerent way. There were hotspots where you knew you would get … good health care.33 Regional provision was particularly sporadic in Scotland, with doctors in ‘hotspots’ such as Edinburgh and Aberdeen performing dramatically more abortions than those in Glasgow or the Borders. Motivations for the cause: abortion in post-legalisation Scotland In 1985, the SHHD found that Greater Glasgow Health Board had among the lowest abortion rates with an average of 6.2, whilst Tayside was the highest with an abortion rate of 11.6.34 This highlights a notable disparity in abortion rates from region-to-region as well as nation-to-nation. As previously shown, a lack of facilities in areas such as Glasgow and Lanarkshire partially contributed to an uneven distribution of services. Yet, equally, this was the result of diﬀerences in the attitudes of doctors and their interpretation of the Abor- tion Act. As Davis notes, even in the pre-1967 period, variations between the East and the West existed, with doctors in Aberdeen being far more amenable towards abortion. This can be largely attributed to the pioneering work of Sir Dugald Baird, who liberally interpreted the existing laws in Scotland regarding abortion and created a proactive abortion policy in Aberdeen that inspired much of the 1967 legislation. Due to the fact that the 1861 Oﬀences Against the Person Act—which criminalised abortion in England and Wales— did not apply to Scotland, Baird took advantage of Scotland’s more ambiguous legal position by oﬀering abortions on the basis of ‘good faith,’ and worked with David Steel to include social factors in the Abortion Act.35 His inﬂuence created a more liberal medical community 7 7 7 WOMEN’S HISTORY REVIEW which recognised the social beneﬁts of accessible abortion services. Contrastingly, in Glasgow, Dr Ian Donald, a founding member of SPUC, was Regius Professor of Obstetrics and Gynaecology at the University of Glasgow. His inﬂuence equally played a role in the heightened diﬃculties in accessing abortions in this region. Religion also impacted services in Glasgow where, as Davis suggests, ‘abortion was a notoriously unpopular procedure … due primarily to the relatively high proportion of Roman Catholics.’36 These regional issues further compounded access for working-class women. Motivations for the cause: abortion in post-legalisation Scotland This was demonstrated in 1977 by one SAC member who stated: [In] a town that has the unenviable reputation of the worst housing, the highest death rates … the worst dietary habits, the worst post-natal mortality … a woman coming from this area has about half the chance of a woman coming from the rest of Scotland of having an NHS abortion.37 This account reﬂects the bitter irony of inadequate abortion provision occurring in areas where women traditionally needed access the most: poorer, densely-populated commu- nities.38 Therefore, religious tensions, variations across regions, and the declining national average in abortion provision prompted the SAC to organise a direct response to the state of abortion healthcare in Scotland. These issues represent some of the key agitating factors which encouraged a centralised branch of the NAC to operate within Scotland. However, the reasons of the individuals who joined the cause reﬂect deeper motivations for the emergence of the SAC. Ann explained that whilst access encouraged her to join the SAC, the attacks made by Scottish MPs further inﬂuenced her decision: It [was] more about “we want to improve legislation, we’re not going to have anybody tell us we can’t get access”. But the nature of the abortion campaign itself, it’s the most attacked piece of public health legislation ever in Britain … and a disproportionately, it felt, number of MPs and Westminster who were bringing forward attacks on the ‘67 Act were from Scotland and Scottish constituencies, so we always felt we had a bit of a responsibility to say everybody in Scotland actually doesn’t think this way.39 Ann’s recollection emphasised the inﬂuence of national identity in the creation of a Scot- tish-centric abortion campaign. Whilst she saw the role of the SAC as a proactive—as opposed to reactive—abortion lobby, she ultimately decided to join in retaliation to what she saw as a misrepresentation of Scottish values. Ann believed it was her personal duty to mount a pro-choice resistance to Scottish anti-abortionists and show the permiss- ive side of Scottish attitudes towards pregnancy termination. Motivations for the cause: abortion in post-legalisation Scotland This was reﬂected in another account made by Carol, an ex-secretary of the SAC: It was the fact that we had Scottish MPs who were really anti-abortion who brought forward anti-abortion bills … I felt we had a responsibility because some of these bloody Scottish MPs were responsible for attempts … It was our responsibility to try and do something. Carol’s memory also highlights the signiﬁcance of Scottish national identity on activist recruitment to the cause. Scottish MPs were considered a Scottish problem which required a Scottish response. However, Carol was also inﬂuenced by what she saw as a threat to women’s rights following John Corrie’s private members bill: I think it was the ten o’clock news and they talked about, oh it was just mentioned this bill that would restrict abortion rights was passed by a large majority, and I just—I burst into 8 K. HAY 8 K. HAY 8 tears. It was just terrible. I was so aﬀected by it … I hadn’t ever needed to have an abortion because I hadn’t become pregnant but it was like—you needed to know that was there.40 tears. It was just terrible. I was so aﬀected by it … I hadn’t ever needed to have an abortion because I hadn’t become pregnant but it was like—you needed to know that was there.40 Carol’s recollection demonstrates the role of emotion in inﬂuencing individuals to join social movements. As Jasper notes, ‘anxiety is useful, perhaps necessary for mobilising protest … long-run anxieties are background emotions: more a part of how we understand the world.’41 Carol felt personally attacked by Scottish MPs attempts to restrict abortion because she saw it as a restriction to her autonomy. Her anger and frustration towards the perceived threat to her rights prompted her subsequent course of action, as she con- tinues: ‘I was just so upset by it and I thought, right I have to do something about this. I have to get actively involved in this.’42 Thus, emotion played a key role in encouraging Carol to get involved. However, as Jasper also notes, her fervent reaction to the bill also reﬂects her background as a self-identiﬁed feminist. This strengthened her emotional response towards the Corrie Bill and her resolve to act. Similarly, Liz, a prominent SAC Steering Committee member and women’s activist, stated: ‘basically people used to say we promoted abortions and we never promoted. Motivations for the cause: abortion in post-legalisation Scotland What we promoted was choice. We said women should have the choice.’43 Her centrality of the issue of choice emphasises her feminist ideology and commitment to women’s liberation. This highlights that the wider inﬂuences of both feminism and Scottish national iden- tity facilitated the emergence of the SAC in 1980. Not only did Scotland have distinctive legislative and structural issues surrounding access to legal abortion in the post-1967 years, but the attacks fronted by Scottish MPs further encouraged the creation of a centralised response to anti-abortion sentiments in the north of Britain. Inﬂuenced to a further extent by the feminist ideal of ‘a women’s right to choose,’ the SAC responded to issues from a distinctively Scottish feminist perspective. Although many other attacks on the Abortion Act came from MPs from England—including William Benyon and Ann Wid- dicombe—the Scottish-led proposals cut most deeply in the hearts of SAC activists. This highlights the role of both Scottish and feminist identities in the formation of the abortion campaign and the de-centralisation of the NAC during this period. Building a strong pro-choice alliance and protecting the Abortion Act As mentioned previously, the SAC sought to not only defend access to legal abortion in law but to mobilise a mass resistance against anti-abortion legislation in Scotland. Conse- quently, the SAC sought to inﬂuence Scottish politics in favour of abortion through seeking the support of trade unions. The Scottish Trade Union Congress (STUC) were seen as a key means of inﬂuencing Scottish MPs, most of whom were from the Labour Party. As Ann explained: The trade unions … had much tighter links with the Labour Party so if you were trying to change the minds of your Labour politicians, adding more and more voices to the move- ment—the pro-choice movement—made it harder for individual Labour MPs—and in our case in Scotland—to argue that their conscience prevented them from doing something like supporting the ‘67.44 The SAC—a product of the Women’s Liberation Movement (WLM)—had little links with political parties. Thus, by obtaining the support of the STUC, the SAC secured an impor- tant ally in defence of legal abortion in Scotland. However, getting ‘a women’s right to 9 WOMEN’S HISTORY REVIEW choose’ into trade union policy was far from simplistic. It required framing abortion as a workplace issue and inﬂuencing a male-dominated environment to vote it into trade union policy. One intersection between the two movements was the issue of class. The SAC argued that poor access to abortion disproportionately aﬀected working-class women. As early as 1977, the SAC sought out trade union support, arguing that: ‘abortion is a class issue and the organisation and strength of the working class in the trade union and Labour movement is needed to win the struggle for abortion rights.’45 This highlights the commitment of the SAC to build an alliance with the STUC. The SAC saw their invol- vement as essential in protecting and extending the Abortion Act. As Reverby suggests, ‘class attitudes … shaped what happened to women who had to rely upon publicly funded care.’33 Women who were more likely to be dependent on NHS-funded abortions struggled to a greater extent than those who could aﬀord to travel and pay for the pro- cedure in England. Thus, ‘abortion is a class issue’ became a signiﬁcant slogan for the SAC, who argued that, as poor access to abortion disproportionately aﬀected working- class women, the STUC had a duty to defend it. Building a strong pro-choice alliance and protecting the Abortion Act HAY 10 K. HAY 10 challenge. The eﬀorts of the SAC in lobbying against restrictions during this period were also undeniably strengthened by the support of the STUC. As Browne states, ‘STUC were extremely important in galvanising support … they ﬁnancially supported NAC’s eﬀorts and organised demonstrations.’50 In Scotland, the involvement of the trade unions during protest marches was perceived as a key means to amplify the SAC’s message. Indeed, Ann said: It’s quite diﬀerent to walk through the streets of Glasgow with, I don’t know, several hundred women with nothing to identify who they are, to walk with [pause]—we all had banners who said who was who—“we are health workers we’re rail workers, we’re community groups … we’re bigger than [pause] this movement is bigger than a few hundred people you see in the street.” This emotive quote shows that trade union support helped to augment the SAC’s socio- political presence. The trade union element transformed individual protesters into repre- sentatives of large organisations, which held great inﬂuence over public attitudes towards legal abortion, as Ann continued: Banners are important for me, even if there are three of you, it will be representing a union branch or an organisation … [the public] will recognise … a logo, maybe, from their own trade union membership card and they’d be like “God, I didn’t even know the union cared about this issue!”51 Trade union visibility thus enabled the SAC to engage with wider audiences hitherto dis- engaged with the politics surrounding abortion. Overall, it is clear that the defensive tactics of the SAC were successful, as no attempts to restrict the Act of 1967 were passed during this period. Faced with the most hostile politicians to the abortion issue, the SAC appealed eﬀectively to the Scottish public and successfully organised a mass opposition to anti-abor- tion legislation in Scotland. Trade unions involvement was essential in increasing their visibility and providing both ﬁnancial support and manpower in defence of the Abortion Act of 1967. The fruitful relationship between the SAC and STUC oﬀers an alternative perspective on the relationship between feminist and left-wing social movements during the second half of the twentieth century. Building a strong pro-choice alliance and protecting the Abortion Act Breitenbach and Abrams have noted that historically, women activists utilised ‘the expression of a shared national identity’ to ‘enlarge their sphere of action,’ in Scotland.46 Scottish class consciousness created a dia- logue through which the SAC could appeal to wider audiences, such as trade unions, and eﬀectively recruit them to the cause. Thus, national identity also played a signiﬁcant role in fostering this relationship. Not only did the SAC campaign for union support, they also operated within the STUC system by encouraging members to take motions to their local branch and reaﬃrming it into STUC policy. As Ann recalls: We would do model motions and then send them out to all our supporters and if they were members of any political [party] we’d say, ‘take it there, any trade union take it there, take it to the trade union council’ … every layer of representative democracy.47 This action not only reﬂected their commitment to securing trade union support, but also reﬂected changing attitudes ushered by second-wave feminism. As Ryan notes, during the 1980s ‘activism was being deﬁned not just as unusual, daring or attention-seeking action; but also as the hard, boring, long-term commitment of eﬀorts to eﬀect change.’48 This suggests that the SAC were additionally inﬂuenced to a notable extent by the changing face of the feminist movement during this period. Motions were forwarded by many within the SAC which, in turn, forced the issue to be raised in STUC conferences. The success of their eﬀorts was exempliﬁed in 1987, when the STUC Women’s Conference ‘recognised the diﬃculties experienced by working-class women when an unplanned preg- nancy occurs,’ and made a commitment to ‘mobilise support against any further restrictive legislation proposed in Parliament and actively work for and support pro-choice groups campaigning to extend present NHS abortion facilities in Scotland.’49 This highlights the eﬀectiveness of the SAC in inﬁltrating the STUC and ensuring their support for future campaigns and demonstrations. Through constructing abortion as a class issue, the SAC reinterpreted reproductive health as a workplace matter and successfully secured a strong ally in their defence of the Abortion Act of 1967. Therefore, the SAC suc- cessfully built a strong pro-choice voice in direct response to heightened anti-abortion sentiments in Scotland. Although the aﬃliation of the STUC was a signiﬁcant gain for the SAC, it is undeniable that the bigger challenge remained protecting the Abortion Act from parliamentary 10 K. Building a strong pro-choice alliance and protecting the Abortion Act Whilst it appears that the SAC and STUC worked homoge- nously in their support of the Abortion Act, the aﬃliation of trade unionists within the NAC has been depicted as a tenuous and fragile alliance.52 However, the success of Scot- tish activists in aﬃliating the STUC reﬂected the distinctive context in which the SAC operated. From the 1960s, Scottish women represented an increasing percentage of the workforce which in turn increased their presence in trade unions. As Breitenbach esti- mates, between 300,000 and 353,000 women were unionised in Scotland by the 1980s: forty per cent of the female workforce, higher than the national average.53 This indicates that SAC members had a proclivity to already be unionised. Indeed, all of those inter- viewed were involved in trade unionism before engaging in abortion activism and Carol explicitly discussed the beneﬁts that oﬀered to the campaign: For people like me—and other’s like me who were in trade unions—we were used to working with men … so we didn’t think it was appropriate to exclude them from the campaign … Now that was diﬃcult with [others] to really sit down and talk to them about it. But they accepted that.54 11 11 WOMEN’S HISTORY REVIEW This suggests that tensions did exist between SAC activists and (male) trade unionists. Issues such as the SAC oﬃces being located in the exclusive Women’s Centre in Millar Street, Glasgow, created friction between the two groups. However, Liz’s testimony further reasserts that these issues were kept to a minimum: Further on when we were fundraising and what not we would have fundraising disco nights in the Communist Party Star Club and who would be doing the DJing? Lesbian Line! [Laugh- ter] and our sisters in London couldn’t understand how we could be so accommodating to each other. Liz’s warm anecdote about SAC fundraising emphasises that identity played a signiﬁcant role in the success of the SAC/STUC alliance, as she continued: ‘Whether it was scale or whatever in Glasgow, or Scotland, we seemed to have a very good [relationship].’55 Her suggestion that the centrality of the SAC allowed for more homogeneity across aﬃliations presents a compelling hypothesis in understanding the distinctive operations of the SAC. By virtue of being a much smaller campaign, SAC could operate more smoothly across the board, whilst the NAC had to contend with a broad spectrum of political alignments. Building a strong pro-choice alliance and protecting the Abortion Act This, combined with the greater participation of women in trade unions prior to the emergence of the SAC, provides some insight into the successful engagement between the SAC and STUC. The eﬀorts of the SAC in defending the Abortion Act of 1967 went beyond traditional modes of political opposition. It was a long process which required building a strong foun- dation of pro-choice support in Scotland in preparation for both parliamentary attack and future pro-choice lobbying. The SAC’s relationship with the trade union movement was a pragmatic and ultimately fruitful aﬃliation. Consequently, the role of the SAC in redeﬁn- ing abortion as a class issue—in a period of heightened political hostility towards legal abortion—must not be understated. Unlike the NAC, the SAC welcomed the STUC almost unanimously and recognised the beneﬁts of their alliance despite the inevitable involvement of men in the campaign. This highlights the inﬂuence of the local context within larger organisations and the politics of national identity in single-issue social movements. Extending the Abortion Act: a failed enterprise? Despite having exerted considerable energy in defending the Abortion Act, the SAC was ultimately unsatisﬁed by the law and sought to extend the rights of women to legal abor- tion. The SAC Constitution reasserted their commitment to both ensuring the full implementation of the abortion act as well as seeking to establish ‘a women’s right to choose’ in law.56 However, like NAC, they were ultimately unsuccessful in fulﬁlling these ‘positive’ aims. The Abortion Act remained virtually unchanged and women contin- ued to experience issues of access throughout the second half of the twentieth century. For this reason, as Hoggart has argued, their eﬀorts in seeking to extend the Abortion Act have received limited historical interest. This has led to an erasure of what could arguably be described as the more feminist aspects of their work. Indeed, much of their attempts to introduce ‘positive legislation’ in regards to abortion reﬂected the ideals of second-wave feminism, including women’s autonomy and the creation of feminist-driven health care. This emphasises the inﬂuence of the WLM on the SAC and its enduring role in 12 K. HAY 12 K. HAY 12 their subsequent activism. In contrast to the dominant narrative which argued that femin- ism was in decline during this period, the SAC’s attempts to revolutionise abortion health- care strengthens Browne’s assertion that ‘far from the [women’s] movement declining, it actually fragmented into a number of diﬀerent campaigns.’57 Thus, the SAC’s proactive activity during this period reﬂects the diversiﬁcation of feminist activism in Scotland. y g p As reﬂected in the ﬁrst aim of the organisation, one of the key ways the SAC sought to positively advocate for improved abortion healthcare was through challenging the power of doctors to delay or refuse to terminate a pregnancy. This contradicts Halfmann’s assess- ment that ‘feminist demands for abortion on request were part of the American reform debate, but not the British one.’58 In fact, both the SAC and NAC maintained the aim of ‘a woman’s right to choose’ without limitations. The SAC sought to redeﬁne the role of the health practitioner in pursuit of greater patient autonomy for women seeking abor- tions in Scotland. Extending the Abortion Act: a failed enterprise? As Davis shows, Scottish doctors were particularly reluctant to embrace the purported new freedoms oﬀered by the 1967 act, seeing ‘a fundamental conﬂict between the established role of doctors to save and preserve life and the more destructive implications of abortion.’59 In response, the SAC sought to reform abortion provision through advocating for doctors to be sympathetic when consulting a woman on her options and to be open with her about their own beliefs. As Carol explained: ‘we were [campaigning for things] like doctors … should have to tell a woman if he was anti-abor- tion instead of just saying ‘I’ll think about it,’ and then delaying things.’60 Her statement reﬂects a number of reported instances where an anti-abortion doctor would delay decision-making until a woman could not legally have an abortion.61 Thus, appealing for doctors to provide honest, impartial information on abortion in a timely manner was seen as a key way to speed up the process and alleviate the confusion surrounding abortion. As O’Neill argues, women went to great lengths to overcome the professional barriers maintained by anti-abortionist doctors ‘creating a situation which some have argued facilitated performance and game playing.’62 The SAC’s campaign to appeal to doctors thus responded directly to these issues. Pamphlets were also created and distrib- uted by the NAC and the SAC throughout the 1980s explaining in detail how to get an abortion, the law, and the diﬀerent types of abortion available. These booklets aimed to inform women about abortion law and practice. This was a further reaction to another misconception perpetuated by some members of the medical profession that they could outright deny women access to an abortion. 63 As Ann recalled: Women were phoning in from parts of Scotland where they’d been told wrongly by their GP that it wasn’t going to happen, and GPs were not supposed to do that. For conscientious reasons they could say they wouldn’t deal with it themselves, the health service is supposed to refer you … You weren’t just supposed to say no.64 The explicit misinformation spread by certain medical professionals in regards to abortion led the SAC to include sections in their pamphlets such as ‘You and Your Doctor,’ which oﬀered practical advice to women on how to manoeuvre around unsympathetic GPs. Extending the Abortion Act: a failed enterprise? This included recommendations such as to ‘register as a temporary patient with a doctor you know is sympathetic using the address of a friend in the area. Say you are on holiday.’65 This advice directly responded to the uneven power balance between (male) doctors and their patients, and empowered women to challenge anti-abortion doctors in Scotland. Moreover, it reﬂected the centrality of the power of knowledge in SAC campaigning, 1 WOMEN’S HISTORY REVIEW 13 reﬂecting the wider inﬂuences of feminism on the movement. As Morgen states, a key issue of second wave feminism was that doctors ‘monopolised knowledge of (thus exerting control over) women’s bodies and reproductive decisions.’66 Therefore, education became an important means for women to take control of their reproductive lives and acted as a form of empowerment. The work of the SAC in circumventing doctor’s control was thus clearly inﬂuenced by their feminist roots. Furthermore, the SAC sought to alleviate the poor provision of services through raising awareness and advocating for increased facilities. As mentioned previously, the vast majority of abortions in Scotland were carried out on the NHS, with the alternative for Scottish women requiring them to travel to England to a charitable abortion service. However, national abortion statistics did not include women who had been refused a ter- mination, nor did they include individual hospital statistics, class, or race. Consequently, in 1983, the SAC embarked on a campaign to improve SHHD statistics. Through lobbying individual health boards, the SAC aimed to collect information regarding regional diver- gences on a hospital-to-hospital basis. This was an attempt to illuminate the individualistic provision of abortion services and the lower abortion rates in Scotland compared to England and Wales. They also sought to diversify statistics to include ethnicity and social class, and to document those who were refused an abortion to highlight the unre- presentative nature of available abortion statistics.67 The limited information available on abortion meant that there was little public discourse concerning women’s experiences. By shedding light on these regional variations, the SAC contributed to emphasising the inade- quacies in NHS abortion healthcare. They also advocated for the creation of out-patient abortion clinics across Scotland, in particular Glasgow, where, as previously shown, issues of access were particularly prevalent. Extending the Abortion Act: a failed enterprise? For this, the SAC held two fundraising discos, sold a newssheet which they advertised directly to the public and collected signa- tures for a petition which they delivered to Glasgow Council. Their eﬀorts to secure more abortion facilities further raised awareness of abortion inequalities and reﬂected the enduring inﬂuence of feminism within the campaign. This was demonstrated by a state- ment made in reference to the proposed outpatient clinic in 1977: ‘unless better NHS facilities appear in Glasgow then phrases “a women’s right to choose”, “abortion on request” and “the sexual revolution” are meaningless.’68 The connection made between abortion and the sexual liberation of women epitomised the politicisation of accessing reproductive health facilities, which in turn demonstrates the commitment of the SAC to feminist ideals. Consequently, the actions of the SAC are distinctive within the broader historiography. As Schoen shows, in response to poor state-coordinated facilities in America, feminist abortion activists set up their own clinics as a means of rejecting capi- talist-driven, male-dominated medical control over women.69 Yet, in Britain the existence of the welfare state made channelling campaign energy into separate feminist-driven healthcare impractical.70 The SAC recognised that they could not radicalise health care in Britain and instead had to work to reform the system from within. Yet, far from making the movement less feminist, the practicalities speciﬁc to Scotland strengthened the SAC’s commitment to the cause and led to an exceptional example of feminist abortion activism. This shows the impact of local surroundings in responses to abortion inequality and provides a diﬀerent example of feminist praxis operating within social healthcare. Finally, the feminist inﬂuence of the SAC was best embodied by the activists them- g Finally, the feminist inﬂuence of the SAC was best embodied by the activists them- selves. The SAC emerged largely from activists who had previously been involved in the 14 K. HAY 14 K. HAY 14 WLM. Indeed, all women commented on their identities as feminists and their engage- ment with the women’s movement. Ann recalled ﬁnding feminism during university which she said ‘opened my eyes,’ to women’s issues.71 Similarly, when asked about whether feminism inﬂuenced her work with the SAC, Carol responded: ‘totally. I mean I read extensively, everything that I could ﬁnd.’72 Both testimonies stressed the importance of their feminist identity. Both also emphasised education in their feminism: Ann through university; Carol through literature. Extending the Abortion Act: a failed enterprise? They also all had a commitment to educating women on their experiences, citing the Boston Women’s Health Collective’s Our Bodies, Ourselves, as a major source of inspiration, and Liz describing it as her ‘bible.’73 The feminist iden- tities of the SAC activists inﬂuenced the spaces they occupied, featuring frequently in fem- inist magazines such as MsPrint and SpareRib, and attending events such as Women’s Health Fairs.74 The ﬁrst Women’s Health Fair in Scotland was set up in 1983 and acted as a medium for women to share their experiences and politicise their engagement with healthcare. The SAC was involved in the planning and organisation of the fair and held a ‘bookstall and exhibition,’ at the event.75 Fairs gave exposure to the SAC and, through their increased visibility, helped to raise awareness of abortion as a women’s rights issue. The SAC’s engagement with Women’s Health Fairs solidiﬁes the fact that, at its roots, the organisation was a continuation of second-wave feminist activism. A ﬁnal echo of the WLM within Scottish abortion activism was the continuing practice of con- sciousness-raising within the SAC. The SAC was more than a political movement; for both activists and the public it became a space through which women could learn about reproductive health and, more broadly, their shared experiences as women. Central to this were so-called Consciousness-Raising ‘CR Groups,’ small groups where activists came together to discuss their daily lives and learn to politicise their shared experiences.76 Whilst Liz and Carol both discussed the ‘support’ they drew from the WLM, Ann was most vocal about its impact on her life: ‘I got a lot of strength from the Women’s Movement … [women would] maybe talk about their own cultural backgrounds … or arguments that had been used against them when they just said something that they thought was completely obvious and, you know, you’d use it to strengthen your own conﬁdence in your argument. And you’d always ﬁnd other women who’d had exactly the same response to something that you thought had only happen to you.’77 From Ann’s testimony, it is clear that holding CR meetings were a signiﬁcant way of sharing experiences and thus improving public engagement strategies. Historicising Scottish feminist abortion activism during the ‘second wave’ In summary, the SAC was a distinctive form of post-legal abortion activism in Britain during the second half of the twentieth century. Prompted by the acute inequalities in abortion provision in Scotland following the passing of the Abortion Act, the SAC sought to not only defend legal abortion, but to establish a strong pro-choice voice in Scotland in the name of women’s reproductive rights. Their inﬂuence on Scottish MPs, coupled with their unique relationship to the STUC, demonstrates the signiﬁcance of Scottish national identity within the campaign which inﬂuenced their strategies to a notable extent. However, feminism also played a key role in the politicisation of abortion healthcare, which inspired many of the SAC’s actions. The legacy of the SAC on Scottish society and culture was demonstrated in 2017 by the decision of the Scottish Parliament to allow women to take the second abortion pill—misoprostol—at home.80 This indi- cates the emergence of a pro-choice consciousness in Scotland: from one of the most hostile countries towards abortion in 1970, to a pioneer in accessible abortion care in Britain today. The shift in sociocultural attitudes towards abortion rights reinforces the importance of abortion activism in the closing decades of the twentieth century, and the work of the SAC in agitating for institutional change in support of ‘a women’s right to choose.’ The role of the respondents’ during the research process for this article must not be understated. Through their own words, they aided in reconstructing the work of the SAC which had been previously underappreciated within the historic literature. Yet, this also revealed key issues in the historicising of feminist abortion activism in twentieth century Britain. Firstly, the erasure of the work of SAC activists to improve access had an impact on respondents’ sense of self, as characterised by Liz’s assertion that: Nobody ever thanks you [by] saying ‘thank you for ﬁghting in the parliament so that I can get access,’ because a lot of younger women, we found—well I still do ﬁnd—think … that its always been there, and abortion has always been there as an option. Extending the Abortion Act: a failed enterprise? Additionally, as Doyal notes, such meetings were ‘concerned primarily … with the bolstering of women’s conﬁdence,’ which further highlights the signiﬁcance of these meetings in empowering women to the abortion cause.78 Furthermore, they acted as means of support and solidarity. This sense of camaraderie went further than abortion activism, for Ann it additionally provided support during her employment in a male-dominated railway industry: From Ann’s testimony, it is clear that holding CR meetings were a signiﬁcant way of sharing experiences and thus improving public engagement strategies. Additionally, as Doyal notes, such meetings were ‘concerned primarily … with the bolstering of women’s conﬁdence,’ which further highlights the signiﬁcance of these meetings in empowering women to the abortion cause.78 Furthermore, they acted as means of support and solidarity. This sense of camaraderie went further than abortion activism, for Ann it additionally provided support during her employment in a male-dominated railway industry: ‘You’d go back into the Women’s Movement meeting and it’s like you’d get like a wee boost … but day-to-day I was in an environment that didn’t talk about it these things and [I] wouldn’t have [otherwise].’79 Therefore, consciousness-raising was also a place of support for women engaging in a new labour market. This highlights that the SAC continued to uphold the values of women’s 1 15 WOMEN’S HISTORY REVIEW liberation. It was an environment where female campaigners could foster relationships with other members and share their experiences both as activists and women. Therefore, the role of feminism within the SAC cannot be understated. Whilst ultimately unsuccessful in their attempts to revolutionise legal abortion in Scot- land, the attempts made by the SAC to improve inequalities in abortion healthcare empha- sise the centrality of feminism in the organisation. This reaﬃrms Browne’s assessment that abortion campaigning acted as a site for continued women’s rights activism during this period and calls into question the decline of the broader WLM in the second half of the twentieth century. Historicising Scottish feminist abortion activism during the ‘second wave’ You know, it didn’t just happen.81 Her feelings of being unacknowledged for her work in the SAC was partly inﬂuenced by the marginalisation of their work in historical literature which, as shown, was considered ineﬀective, thereby insigniﬁcant.82 Yet, as this study highlights, the institutional change that the SAC sought to implement was idealistic and proved to be a greater challenge than previously estimated. As Ryan analysed in her study of feminist campaigners in America: 16 K. HAY 16 K. HAY 16 Activists no longer expected to see the changes they hoped for in their lifetime … these issues were seen as more complex, more deeply rooted, and more diﬃcult to achieve than they had seemed in the earlier period.83 This indicates that deﬁning success within such a limited time frame becomes problematic in the context of the ongoing struggle for women’s rights in the West. As this article proves, through analysing the inﬂuences, motivators and reasoning behind the actions of activist movements, we produce a richer, more comprehensive understanding of their activities during this time. Through avoidance of deﬁning activists by their successes and failures, we can appreciate the movements themselves and their contribution to the continuing pursuit of progress. A second observation was that decentralised experiences have been further obscured from the dominant narrative. This article highlighted that despite the representation of the NAC as a site of continued feminist conﬂict towards the end of the second wave, the SAC operated fairly harmoniously across a broad political spectrum. This trope within feminist histories was observed by interviewees. Indeed, Carol considered it a ‘misrepresentation,’ of experiences, commenting: [Historians] really focused on the split … between socialist feminists and radical feminists … it wasn’t as awful as [history books] have made out. There were diﬃculties, but it wasn’t that bad, at all … that didn’t happen to us in Glasgow.84 Her rejection of the dominant narrative emphasises the importance of including local experi- ences within broader social movements. The actions of the SAC was infuenced by a distinc- tively Scottish feminist identity which requires greater analysis within the historiography. Moreover, the experience of the SAC oﬀers a revision to the assumption that the 1970s and 1980s represented the decline of feminist activity in Britain. Historicising Scottish feminist abortion activism during the ‘second wave’ As Orleck argues, ‘historians of social movements … have been drawn to moments of turbulent unrest,’ which often comes to deﬁne the movement as a whole.85 However, activist movements are inherently sites of conﬂict and tension. Whilst it is important to analyse these conﬂicts, it is equally important to avoid privileging their signiﬁcance in an attempt to ﬁt into the dominant historical narrative. Appreciating the complexity and inherently intense nature of feminist health activism leads to a deeper analysis of social movements and a greater overview of the history of feminism in the second half of the twentieth century. Consequently, this work provides an avenue for continued reassessment and greater his- torical inquiry into the role of feminism in women’s health activism in twentieth century Britain. By diverting focus away from the ‘successes’ and ‘conﬂicts’ within the movement, we can gain a deeper understanding how feminist activism changed throughout the course of the 1970s and 1980s. Through a bottom-up approach, we see how individual, local campaigns contributed to broader organisations diﬀerently due to their distinctive cul- tural identities. It is essential that historians trace this history in order to contextualise current events and to provide a better understanding of the development of abortion rights in the U.K. At a time when women across the world are gaining and losing their right to reproductive autonomy, understanding the controversial history of legal abortion and its signiﬁcance to women’s liberation holds acute relevance to present times. Notes 1. Ibid. Notes 1. Ibid. 17 WOMEN’S HISTORY REVIEW 2. H. Kanter, S. Lefanu, and S. Shah (Ed.) Sweeping Statements: Writings from the Women’s Lib- eration Movement 1981–83 (London: Women’s Press, 1984), 235. 3. G. Davis, ‘The Medical Community and Abortion Law Reform: Scotland in National Context, c 1960–1980’, in Lawyer’s Medicine: The Legislature, the Courts and Medical Prac- tice, 1760–2000, ed. I. Goold and C. Kelly (Oxford: Hart, 2009), 143. 4. G. Davis and R. Davidson, ‘“A Fifth Freedom” or “Hideous Atheistic Expediency”? The Medical Community and Abortion Law Reform in Scotland, c. 1960–1975’, Medical History 50 (2006): 35. 5. For a comparative discussion of the Sexual Revolution, see H. Cook, The Long Sexual Revolu- tion and J. Weeks, Sex, Politics and Society: The regulation of Sexuality since 1800 (Harlow: Pearson, 2012), C.G. Brown, ‘Sex, Religion and the Single Woman c. 1950–75: The Impor- tance of a “Short” Sexual Revolution to the English Religious Crisis of the Sixties’, Twentieth Century British History 22, no. 2 (2011). 6. Weeks, Sex, Politics and Society, 334. It must be noted that these ﬁgures do not include those provided through the charity or private sector. 7. R. Davidson and G. Davis, The Sexual State: Sexuality and Social Governance 1950–80 (Edin- burgh: University of Edinburgh Press, 2012), 116–120. 8. S. Browne, The Women’s Liberation Movement in Scotland (Manchester: Manchester Univer- sity Press, 2014), 127. 9. WL, SA/NAC/F/1/3, Scottish Abortion Campaign, ‘Constitution (Updated)’, (1988). 9. WL, SA/NAC/F/1/3, Scottish Abortion Campaign, ‘Constitution (Updated)’, (1988). 10. There have also been new developments in the history of the pro-life movement in Britain 9. WL, SA/NAC/F/1/3, Scottish Abortion Campaign, ‘Constitution (Updated)’, (1988). 10. There have also been new developments in the history of the pro-life movement in Britai See O. Dee, The Anti-abortion Campaign in England 1966–1989 (New York: Routledg g 10. There have also been new developments in the history of the pro-life movement in Britain. See O. Dee, The Anti-abortion Campaign in England 1966–1989 (New York: Routledge, 2020). 11. B. Brookes, Abortion in England 1900–1967 (London: Croom Helm, 1988), 79–94. . WL, PP/RJH/H.6/8, National Abortion Campaign, Abortion: Our Struggle for Contro (London: National Abortion Campaign, 1980), 5. 13. L. Hoggart, Feminist Campaigns for Birth Control and Abortion Rights in Britain (Lampster: Edwin Mellen, 2003), 186–9. 14. Browne, WLM in Scotland. 15. J. Notes Nelson, More than Medicine: A History of the Feminist Women’s Health Movement (New York: New York University Press, 2015). 16. E. Breitenbach, ‘Sisters are doing it for Themselves—The Women’s Liberation Movement in Scotland’, in The Scottish Government Yearbook, ed. A. Brown and R. Parry (University of Edinburgh, 1990), Browne, The WLM in Scotland. 17. Although this project is among the ﬁrst to interview SAC activists directly about abortion activism, it is preceded by oral history interviews with prominent members of the WLM who were involved in the Campaign. See Browne, The WLM in Scotland. 18. Hoggart, Feminist Campaigns for Birth Control, 181. 19. BBC News, ‘Women in Scotland Allowed to Take Abortion Pill at Home’, 26 October 2 19. BBC News, ‘Women in Scotland Allowed to Take Abortion Pill at Home’, 26 October 2017. 20. J. Brown, ‘Scotland and the Abortion Act 1967: Historic Flaws, Contemporary Problems’, Juridical Review 2 (2015): 9. 19. BBC News, ‘Women in Scotland Allowed to Take Abortion Pill at Home’, 26 October 2017. 20 J Brown ‘Scotland and the Abortion Act 1967: Historic Flaws Contemporary Problems’ 20. J. Brown, ‘Scotland and the Abortion Act 1967: Historic Flaws, Contemporary Proble Juridical Review 2 (2015): 9. 21. E. Lee, ‘Tensions in the Regulation of Abortion in Britain’, Journal of Law and Society 30, no. 4 (2003): 533. 22. Davis and Davidson, ‘“A Fifth Freedom?”’, 31. 23. Davidson and Davis, The Sexual State, 116. 24. Glasgow Women’s Library [hereafter GWL] GWL, SAC 3/1, ‘Abortion Statistics’, Abortion Review 13 (1989): 1. 25. E.M. Briggs and A.E. Mark, ‘Termination of Pregnancy in the Unmarried’, Scottish Medical Journal (1972): 400. 26. Mitchell Library [hereafter ML], HB77/2/17, Scottish Health Education Unit, Themes in Family Planning: Report no 3, ‘Abortion Trends in Scotland’ (1980). 27. WL, SA/NAC/F/1/3, C. Thomson, ‘Submission by the SAC to Co-ordinating Committee in Defense [sic.] of the 1967 Abortion’, (1983). 18 K. HAY 18 K. HAY 28. WL, SA/NAC/B/3/6, BPAS, ‘Abortion Amendment Bill 1979: How the Charitable Services would be Aﬀected’, (1979). 29. Glasgow Herald, ‘Abortion: An Unfair Barrier to Treatment?’ 1 August 1982. 30. Evening Standard ‘Abortion Shuttle that Stops at Edinburgh’, 24 May 1981. g g y 31. WL, SA/NAC/F/1/3, A. McGowan, ‘Trends in Scottish Abortion Figures Show a Marked Diﬀerence from those in England and Wales’, (1980): 68. 31. WL, SA/NAC/F/1/3, A. McGowan, ‘Trends in Scottish Abortion Diﬀerence from those in England and Wales’, (1980): 68. Notes 32. Briggs and Mark, ‘Termination of Pregnancy in the Unmarried,’ 400. 33. Interview with Ann H., 13 July 2018, 21. 34. Mitchell Library [hereafter ML], HB77/2/17, Scottish Health Education Unit, Themes in Family Planning: Report no. 1, ‘Abortion Trends in Scotland’, (1985). 35. Davis and Davidson, ‘“A Fifth Freedom?”’, 35. 36. Davis, ‘The Medical Profession and Abortion Law Reform’, 148. 37. WL, PP/RJH/H.6/8, NAC, ‘Abortion—The Evidence’, 22. 38. Ibid., 22. 38. Ibid., 22. 39. Interview with Ann, 6. 40. Interview with Carol T., 2 August 2018, 4–26. 40. Interview with Carol T., 2 August 2018, 4–26. 41. J.M. Jasper, The Emotions of Protest (Chicago: U 41. J.M. Jasper, The Emotions of Protest (Chicago: University of Chicago Press, 2019), 40–41. 42. Interview with Carol, 4. 42. Interview with Carol, 4. 43. Interview with Liz A., 2 August 2018, 14. 44. Interview with Ann, 9. 45. GWL, SAC 3/1, ‘A women’s right to choose: Newssheet of the West of Scotland National Abortion Campaign’ (1977). 46. E. Breitenbach and L. Abrams, ‘Gender and Scottish Identity’, in Gender in Scottish History Since 1800, ed. L. Abrams, E. Gordon, D. Simonton and E. Janes (Edinburgh: Edinburgh Uni- versity Press, 2006), 35. 47. Interview with Ann, 11. 48. B. Ryan, Feminism and the Women’s Movement: Dynamics of Change in Social Movement Ideology and Activism (London: Routledge, 1992), 87. 49. GWL SAC 3/1, ‘Minutes from the STUC Women’s Conference’ (1987). 50. Browne, The WLM in Scotland, 126. 51. Interview with Ann, 17. 52. Hoggart, Feminist Campaigns for Birth Control, 186. b h ‘ d f h ggart, Feminist Campaigns for Birth Control, 1 53. E. Breitenbach, ‘A Comparative Study of the Women’s Trade Union Conference and the Scottish Women’s Trade Union Conference’, Feminist Review 7 (1981): 69–70. h l 54. Interview with Carol, 5. 55. Interview with Liz, 5. 56. SAC, ‘SAC Constitution (Updated)’. 57. Browne, The WLM in Scotland, 112–3. 58. D. Halfmann, ‘Historical Priorities and the Responses of Doctors’ Associations to Abortion Reform Proposals in Britain and the United States, 1960–1973’, Social Problems 50, no. 4 (2003): 584. 59. Davis, ‘The Medical Profession and Abortion Law Reform’, 153–5. 60 I i i h C l 12 59. Davis, ‘The Medical Profession and Abortion Law Reform’, 153–5. 60. Interview with Carol, 12. 60. Interview with Carol, 12. 61. GWL SAC/2/1, NAC, ‘NAC ﬁghts for NHS facilities for all abortions’, (c. 1977). 62. O’Neill, ‘Abortion Games’, 1. Notes 62. O’Neill, ‘Abortion Games’, 1. 63. Evening Standard, ‘Abortion Shuttle that Stops at Edinburgh’. 63. Evening Standard, ‘Abortion Shuttle that Stops at Edinburgh’. 64. Interview with Ann, 15–6. , 65. Personal Collection, NAC, ‘How to Get an Abortion’ (1986). 65. Personal Collection, NAC, ‘How to Get an Abortion’ (1986). 65. Personal Collection, NAC, ‘How to Get an 66. S. Morgen, Into Our Own Hands: The Women’s Health Movement in the United States (New Brunswick: Rutgers University Press, 2002), 72. 67. GWL, SAC/2/1, ‘NHS Abortion Facilities in Scotland: A Survey by Scottish Abortion Cam- paign, Preliminary Findings and Observations’ (1986). 68. NAC, ‘Abortion—the Evidence’, 22, please note that at this time SAC came under Scottish NAC. WOMEN’S HISTORY REVIEW 19 69. J. Schoen, Abortion after Roe (Chapel Hill: University of North Carolina Press, 2015), 41. 70. See L. Doyal, ‘Women, Health and the Sexual Division of Labor: A Case Study of the Women’s Health Movement in Britain’, in Women’s Health, Politics, and Power: Essays on Sex/Gender, Medicine, and Public Health, ed. E. Fee and N. Krieger (Amityville: Beywood, 1994), 64. 71. Interview with Ann, 3 72. Interview with Carol, 22. 73. Boston Women’s Health Collective, Our Bodies, Ourselves: A Health book by and for women (Harmondsworth: Penguin Books, British Ed., 1978), Interview with Liz, 23. 73. Boston Women s Health Collective, Our Bodies, Ourselves: A Health book by and for women (Harmondsworth: Penguin Books, British Ed., 1978), Interview with Liz, 23. 74. See, for example, SAC, ‘Abortion in Aberdeen’, Spare Rib (1980). (Harmondsworth: Penguin Books, British Ed., 1978), Interview with Liz, 23. 74. See, for example, SAC, ‘Abortion in Aberdeen’, Spare Rib (1980). 74. See, for example, SAC, ‘Abortion in Aberdeen’, Spare Rib (1980). 74. See, for example, SAC, ‘Abortion in Aberdeen’, Spare Rib (1980). 75. GWL SAC 2/3, ‘SAC Annual Conference’ (1984). 76. Browne, WLM in Scotland, 45. 77. Interview with Ann, 16. 78. Doyal, ‘Women, Health and the Sexual Division of Labor’, 628. 78. Doyal, ‘Women, Health and the Sexual Division of Labor’, 628. 79. Interview with Ann, 16. 80. BBC News, ‘Women in Scotland Allowed Abortion Pill at Home 81. Interview with Liz, 20. 83. Ryan, Feminism and the Women’s Movement, 88. 84. Interview with Carol, 23. 84. Interview with Carol, 23. 85. A. Orleck, Rethinking American Women’s Activism (New York: Routledge, 2015), x. Acknowledgements I would like to thank Dr Carmen M. Mangion and the two reviewers for their insightful and devel- opmental comments on an earlier draft of this article. Their feedback was tremendously helpful and signiﬁcantly improved this work. I am also grateful to my supervisor Dr Laura Kelly for her con- stant encouragement and support from the beginning stages of this article to the end. I am also thankful for the activists I interviewed, who gave up their time to share their experiences with me and provided the beating heart of this work. Finally, I am indebted to the Wellcome Trust for supporting this research. Disclosure statement No potential conﬂict of interest was reported by the author(s). Notes on contributor Kristin Hay (BA, MSc) is a PhD student at the Centre for the Social History of Health and Health- care at the University of Strathclyde, Glasgow, U.K. Her doctoral research focuses on the history of birth control in Scotland during the second half of the twentieth century. This article was prepared as part of her MSc project on Abortion Activism in Scotland, supported by the Wellcome Trust under Grant 208532/Z/17/Z. ORCID Kristin Hay http://orcid.org/0000-0002-0109-9167 Kristin Hay http://orcid.org/0000-0002-0109-9167
https://openalex.org/W1991944147	https://www.scielo.br/j/bor/a/dNtGHSP5cX9cvYb9njsCZ9f/?lang=en&format=pdf	English	null	CBCT-based evaluation of integrity of cortical sinus close to periapical lesions	Brazilian Oral Research	2,015	cc-by	4,359	Original Research Periodontics Original Research Radiology Original Research Periodontics Original Research Radiology Original Research Periodontics Original Research Radiology CBCT-based evaluation of integrity of cortical sinus close to periapical lesions Abstract: The purpose of the present study was to test the applicability of cone beam computed tomography (CBCT) to evaluate the integrity of the cortical sinus close to periapical lesions. Two observers analyzed samples of 64 alveoli of first molar roots in macerated swine maxillas prepared with perchloric acid to simulate periapical lesions. The specimens were evaluated using CBCT (55-mm high and 100-mm diameter cylinder at 0.2-mm voxel resolution) for the presence of oroantral communication (OAC) caused by the simulated lesions. Sensitivity, specificity, and accuracy were calculated. Fair values were obtained for accuracy (66%-78%) and good values for specificity (70%-98%), whereas the values for sensibility showed relevant variation (41%-78%). For this reason, the interobserver agreement score was weak. CBCT proved capable of evaluating the integrity of the cortical sinus (absence of oroantral communication) when it lies close to an apical periodontitis lesion. However, the low interobserver agreement reflects the difficulty in performing diagnoses when OAC is adjacent to a periapical lesion, using the acquisition protocol adopted in this research. This could be attributed to the high level of image noise. Universidade de São Paulo – USP, School of Dentistry, Department of Stomatology, São Paulo, SP, Brazil Keywords: Periapical Periodontitis; Models, Animal; Cone-Beam Computed Tomography; Diagnosis; Maxillary Sinus. Declaration of Interests: The authors certify that they have no commercial or associative interest that represents a conflict of interest in connection with the manuscript. Braz Oral Res [online]. 2015;29(1):1-7 Introduction A C D B B A teeth is considerably high, thus giving CBCT a role of clinical relevance in performing radioanatomical and radiodiagnostic observations of the maxillary sinus floor and the alveolar bone and teeth.5 A B The proximity between the root apices of the maxillary teeth and the maxillary sinus may generate an image of overlapping structures in two-dimensional (2D) radiographs, hiding a periapical lesion that compromises the integrity of the cortical sinus and triggers odontogenic sinusitis.3,6,7,8 Several studies have shown better sensitivity and specificity of CT, as compared with 2D radiographs, but there is a lack of studies comparing CBCT with the gold standard.9 The proximity between the root apices of the maxillary teeth and the maxillary sinus may generate an image of overlapping structures in two-dimensional (2D) radiographs, hiding a periapical lesion that compromises the integrity of the cortical sinus and triggers odontogenic sinusitis.3,6,7,8 Several studies have shown better sensitivity and specificity of CT, as compared with 2D radiographs, but there is a lack of studies comparing CBCT with the gold standard.9 For this reason, the objective of the present study was to test CBCT applicability in evaluating the integrity of the cortical sinus close to periapical lesions. The sensitivity, specificity, accuracy and reproducibility of CBCT were tested in evaluating simulated periapical lesions in order to detect the presence or absence of oroantral communication (OAC) in macerated swine maxillas, thereby establishing this set of determinants (simulated periapical lesions/ oroantral communication/ macerated swine maxillas) as the gold standard. D C C D For this reason, the objective of the present study was to test CBCT applicability in evaluating the integrity of the cortical sinus close to periapical lesions. The sensitivity, specificity, accuracy and reproducibility of CBCT were tested in evaluating simulated periapical lesions in order to detect the presence or absence of oroantral communication (OAC) in macerated swine maxillas, thereby establishing this set of determinants (simulated periapical lesions/ oroantral communication/ macerated swine maxillas) as the gold standard. Figure 1. Preparation of samples: A) Pig’s head; B) Macerated swine maxilla; C) Odontosection; D) Application of perchloric acid. Introduction Some sites (n = 32) were exposed to acid for a shorter time, to generate incipient lesions (Time 1: 3 hours), whereas others (n = 32) were exposed for a longer time, to simulate larger lesions (Time 2: 4 hours).11,14 The action of the acid was stopped by rinsing the alveoli in running water after the specified time. The impairment of cortical air spaces evolved in some cases to create the intended disruption, thus generating OAC (n = 22). Proof that there was communication between the alveoli and the air space was verified by visual inspection, i.e., by observing a periodontal probe that was introduced into the alveoli, and that passed through the oral cavity into the air space. This is how the gold standard was established for this study. Sites with OAC were recorded, for comparison with diagnosis by CT scans. Preparation of samples Swine maxillas (n = 20) were macerated by boiling them in water, as the preferred method to preserve bone quality of the sample.10 They were then scanned by CBCT to verify the integrity of the alveolar bone of first molar mesiobuccal and distobuccal roots. Maxillas with alveolar bone defects or third molars near the region of interest (ROI) were excluded from the sample. The eligible maxillas were selected for lesion simulation (n = 16). First molars were extracted after odontosection (n = 32), taking care to maintain the integrity of the alveoli and roots, as verified by visual inspection.11 Introduction Early diagnosis of odontogenic sinusitis is important, because it is a common condition, and differs in pathophysiology and management from other sinus diseases of different etiology. Odontogenic sinusitis occurs when the sinus membrane is perforated, possibly caused by periapical lesions in maxillary teeth.1 Corresponding Author: Marcelo Cavalcanti E-mail: mgpcaval@usp.br Several studies have demonstrated the importance of performing a diagnosis when the source of sinusitis is odontogenic, so that an effective treatment plan can be established for the adequate healing of tooth infection, mainly by endodontic treatment, and for the effective prevention of sinusitis recurrence and evolution.1,2,3 When CBCT imaging is available, these image data may be used in preventing post-treatment complications, in the diagnosis of pathology and in improved visualization of the anatomical structure in the posterior maxilla. Apical periodontitis accessed with conventional images is frequently underestimated; however, the application of a higher radiation dose cautions use of CBCT except when justified or supported by clinical examination.4 Anatomical and pathological maxillary sinus involvement in posterior upper Submitted: Feb 21, 2014 Accepted for publication: Sep 08, 2014 Lat revision: Nov 27, 2014 1 Braz Oral Res [online]. 2015;29(1):1-7 CBCT-based evaluation of integrity of cortical sinus close to periapical lesions Figure 1. Preparation of samples: A) Pig’s head; B) Macerated swine maxilla; C) Odontosection; D) Application of perchloric acid. Image acquisition A water-filled plastic container was used to simulate soft tissue attenuation, such as an in vivo situation.15,16 A CBCT (ProMax 3D Max, Planmeca Oy / Helsinki, Finland) scan was performed for each maxilla, with the following acquisition protocol: the limits of the imaging area consisted of a 55-mm high and 100-mm diameter cylinder; the voxel size was 0.2 mm. These parameters were established according to the selection of a default The same operator, not involved in interpreting the images, performed erosions in all the alveoli, to simulate lesions. This was done by applying 70% perchloric acid (Merck Chemicals, Darmstadt, Germany) in the mesiobuccal and distobuccal root alveoli (Figure 1), thus generating the sites (n = 64) to be analyzed on CT images.12,13 Braz Oral Res [online]. 2015;29(1):1-7 2 Santos Junior O, Pinheiro LR, Umetsubo OS, Cavalcanti MGP acquisition protocol for the maxillary region, made available by this CBCT unit. Figure 3. Sagittal CBCT images: A) Sample with oroantral com- munication (arrow) caused by periapical lesion; B) Sample without disruption of cortical sinus (arrow) close to periapical lesion. B A B A A B Sixteen CBCT scans of maxillas (with periapical lesions simulated in the distobuccal and mesiobuccal roots of first molars on both sides) were obtained using this procedure. There were two double-blinded observers, who were experienced oral and maxillofacial radiologists and also trained and calibrated on the tomographic features of periapical lesions affecting the cortical sinus, with and without OAC. Each observer analyzed 64 sites in the CBCT scans. Figure 3. Sagittal CBCT images: A) Sample with oroantral com- munication (arrow) caused by periapical lesion; B) Sample without disruption of cortical sinus (arrow) close to periapical lesion. After image acquisition, the data were stored in Digital Imaging Communication in Medicine (DICOM) file format and imported by OsiriX version 3.8.1 (OsiriX Foundation, Geneva, Switzerland), an open-source DICOM viewer installed in a iMAC 27 – Mac OS X Snow Leopard workstation (Apple, Cupertino, USA). (true negative – TN); correct identification of OAC (true positive – TP); identification of OAC in a maxillary sinus floor without disruption (false positive – FP); and failure to identify OAC in a maxillary sinus floor with disruption (false negative – FN). Radiographic assessment Statistical analyses were carried out using the Validity and Kappa tests. The k coefficients were calculated to assess the degree of intra- and interobserver agreement, and were scored as weak (0.20-0.39), moderate (0.40-0.59), and relevant (0.60-0.79).17 These coefficients made it possible to check the tomographic diagnosis against the gold standard. The level of significance was set at a = 0.05. The data were analyzed using SPSS software (v 17.0.0; SPSS Inc.; Chicago, USA). The entire volume of each sample was analyzed by the observers, using multiplanar reconstructed axial, coronal (Figure 2), and sagittal (Figure 3) images. The sequence of observations was randomized by the Random.org website (www.random.org). The interpretation time was not restricted, and the same images were analyzed repeatedly after a 2-week interval, so that the observers would not remember the previous image interpretation. Presence of OAC was diagnosed by a dichotomous (yes/no) evaluation, and the scores were labeled as follows: correct identification or absence of OAC Sensitivity, specificity and accuracy of the analyses were calculated as follows: Sensitivity = TP / (TP + FN), Specificity = TN / (TN + FP), Accuracy = (TP + TN)/ (TP + FP + FN + TN). The positive and negative predictive values (PPV and NPV) were also calculated to determine the probability of true diagnosis. Statistical analyses were performed using BioEstat software (v 5.3; Bioestat, Belém, Brazil). Figure 2. Coronal CBCT images: A) Sample with oroantral com- munication (arrow) caused by periapical lesion; B) Sample without disruption of cortical sinus (arrow) close to periapical lesion. B A B A A B Braz Oral Res [online]. 2015;29(1):1-7 Results Value First evaluation Second evaluation Observer 1 x Observer 2 Observer 1 x Observer 2 k 0.3469 0.2120 p 0.00001 0.0373 Table 3. Percentage of True (T) and False (F) diagnosis of oroantral communication caused by periapical lesions. Observer 1 Observer 2 Total T (%) F (%) T (%) F (%) T (%) F (%) Fist Evaluation 78 22 73 27 76 24 Second Evaluation 77 23 66 34 71 29 Table 2. Intraobserver reliability. Value First evaluation x Second evaluation First evaluation x Second evaluation Observer 1 Observer 2 k 0.4928 0.5573 p < 0.00001 < 0.00001 A shared consensus in these studies is that the closer proximity of posterior tooth roots with the maxillary sinus floor increases the risk of odonto- genic sinusopathy, mucositis and sinusitis. Neverthe- less, studies reveal that infection and inﬂammatory mediators can affect the sinus mucosa without per- forating the cortical bone of the sinus floor.18 In this study, we simulated lesions in order to induce OAC as a relevant cause of bacterial colonization of odon- togenic origin, thus triggering complications. Table 2. Intraobserver reliability. Value First evaluation x Second evaluation First evaluation x Second evaluation Observer 1 Observer 2 k 0.4928 0.5573 p < 0.00001 < 0.00001 The proximity between the root apices of the max- illary teeth and the maxillary sinus can generate an image of overlapping structures in 2D radiographic examination, hiding incipient lesions and damage to the integrity of a cortical sinus possibly having OAC.3,6,7,8 Thus, this in vitro study used 70% perchlo- ric acid to mimic periapical lesions (occurring from inflammatory reactions, and therefore having more irregular boundaries than those made by drills). This procedure made simulated initial lesions look like those observed in vivo, thereby enabling simu- lation of incipient lesions, which are more difficult to detect.11,12,13,19 Table 3. Percentage of True (T) and False (F) diagnosis of oroantral communication caused by periapical lesions. Observer 1 Observer 2 Total T (%) F (%) T (%) F (%) T (%) F (%) Fist Evaluation 78 22 73 27 76 24 Second Evaluation 77 23 66 34 71 29 Table 3. Percentage of True (T) and False (F) diagnosis of oroantral communication caused by periapical lesions. Observer 1 Observer 2 Total T (%) F (%) T (%) F (%) T (%) F (%) Fist Evaluation 78 22 73 27 76 24 Second Evaluation 77 23 66 34 71 29 Results The whole sample (n = 64) was submitted to perchloric acid: one part (n = 32) was exposed at Time 1, resulting in 7 sites with OAC, and the other part (n = 32) was exposed at Time 2, resulting in 15 communications between the alveoli and the maxillary sinus, yielding 64 sites with periapical lesions and 22 OACs. Figure 2. Coronal CBCT images: A) Sample with oroantral com- munication (arrow) caused by periapical lesion; B) Sample without disruption of cortical sinus (arrow) close to periapical lesion. 3 Braz Oral Res [online]. 2015;29(1):1-7 CBCT-based evaluation of integrity of cortical sinus close to periapical lesions Table 1 presents the kappa values for interobserver agreement in detecting OAC caused by periapical lesions, and Table 2 presents the kappa values for intraobserver agreement. The intraobserver values (0.49 to 0.56) were considered as moderate levels of agreement, and the interobserver values (0.21 to 0.35), as low levels of agreement. Table 4. Sensitivity, positive predictive values, specificity, negative predictive values and accuracy rates (%) calculated according to diagnosis of oroantral communication caused by periapical lesions. Table 4. Sensitivity, positive predictive values, specificity, negative predictive values and accuracy rates (%) calculated according to diagnosis of oroantral communication caused by periapical lesions. First evaluation Second evaluation Observer 1 Observer 2 Observer 1 Observer 2 Sensitivity (%) 41 78 55 60 PPV (%) 90 60 70.59 50 Specificity (%) 98 71 88 70 NPV (%) 75.93 85.29 78.72 76.32 Accuracy (%) 78 74 77 66 Table 3 shows the overall TP and FP, as well as TN and FN results for the diagnosis of periapical lesions that damage cortical sinus integrity. The results for overall sensitivity, positive predictive values, specificity, negative predictive values and accuracy are shown in Table 4. These data represent all the observations for both observers. The values for sensitivity and specificity varied for each observer. Accuracy rates were roughly similar among the observers, and significant values (p < .05) were obtained (66% to 78%). based mostly on clinical signs, although radiological findings may be very valuable in identifying etiologi- cal factors.5 Many studies assess etiological factors, like periodontal disease, carious cavities, large res- torations, surgical procedures, apical periodontitis and endodontic therapy in the posterior maxilla, as well as mucosal thickening in the maxillary sinus.1,3,4 Table 1. Interobserver reliability. Discussion The true etiology of maxillary sinusitis plays an important role in clinical management, mainly because treatment of odontogenic sinusitis is different from that of rhinogenic sinusitis. Sinusitis diagnosis is The anatomical proximity between the root apex and the cortical air space is more relevant than the size of a periapical lesion in establishing OAC.3,8 Therefore, the longer acid etching time was not a Braz Oral Res [online]. 2015;29(1):1-7 4 Santos Junior O, Pinheiro LR, Umetsubo OS, Cavalcanti MGP molar adjacent to the cortical sinus) of the porcine model for image acquisition. This protocol generated images with a relatively large FOV and small vox- els. As a result, we found image noise, reported by the observers as the main difficulty in performing their analyses. Another possible reason for the lower image quality could be attributed to the water-filled container, used in the present study, in which the samples were totally immersed when the CBCT scan was performed.16,26 The water-bearing container was intended to cause beam attenuation, and leveled the values of our in vitro study to clinic levels. Because the CBCT unit used in this study did not allow free manipulation of the parameters, filters and calibra- tion could have caused the images to have a higher level of noise. This limitation affected our in vitro research, thus hindering the detection of small OACs adjacent to the periapical lesion. determining factor for establishing OAC. Nonethe- less, the analysis of this anatomic relationship (max- illary sinus and apex) was performed by CBCT with good reliability. This corroborates previous studies5 and shows the clinical relevance of the radioanatomi- cal observation allowed by CBCT. It may also be one of the reasons for the high specificity and accuracy of the results. Pigs are considered a good species to represent human beings in studying craniofacial bones, because their anatomy, morphology, and bone density in this region is similar to that of humans. This makes swine maxillary bones a good sample for radiological stud- ies.20,21 Moreover, the current research corroborated that porcine maxillas are considered a good model for studying this anatomical region. However, some difficulties were found in standardizing the posi- tioning of the ROI of the sample in the center of the FOV during the tomographic acquisitions, due to maxilla size variation. This is precisely one of the limitations observed in this study. Discussion That is to say, the use of human skulls would be more suitable for simi- lar studies in CBCT devices with pre-standardized acquisition protocols. The results showed high CBCT specificity (70%-98%) in the evaluation of the integrity of the cortical sinus close to the apical periodontitis lesion, attributed to the good assessment of the relationship between the maxillary sinus and the posterior root apices. How- ever, the sensitivity values were lower (41%-78%), possibly attributed to the difficulty in analyzing the existence of OAC adjacent to the periapical lesion, caused by image noise. The choice of an acquisition protocol with a smaller FOV could attenuate the level of image noise.15,16 Therefore, we believe that the choice of an acquisition protocol with a smaller FOV and the positioning of the ROI at the center of the FOV can improve these rates. Several studies have compared and attested to higher sensitivity and specificity of CBCT scans, as compared with 2D X-rays to diagnose periapi- cal pathologies in maxillary teeth.4,5,22,23 A system- atic review on radiological diagnosis reported that CBCT has not been compared with a gold standard.9 For this reason, the current research tests a possible comparison between CBCT images and a gold stan- dard (simulated lesions in the maxillas generating or failing to generate OAC, attested by visual inspec- tion and periodontal probe). Although the accuracy indices were similar among the observers, the sensitivity values were distinct. This resulted in a low level of interob- server agreement. The most obvious difficulty, and one reported by the observers, was diagnos- ing a cortical sinus rupture in the presence of an artifact. Clinically, the findings could infer that CBCT is a valuable adjunct to anatomical evalu- ation of posterior maxilla and diagnosis of peri- apical lesions in the root apices close to the max- illary sinus floor (all simulated lesions at both times – Time 1 and Time 2 – were easily noticed by observers). However, OAC-related limitation exists in regions where the periapical lesion is The size of the FOV and the size of the recon- structed voxels are parameters that may vary in dif- ferent imaging protocols of the same CBCT unit, and that may influence image quality. A large FOV with a small voxel reconstruction generates images with a relevant level of noise.24,25 The CBCT unit used in this study allows the option of choosing pre-defined imaging protocols. Braz Oral Res [online]. 2015;29(1):1-7 References on nuclear DNA amplification using human bone. J Forensic Sci. 2010 Jul;55(4):1032-8. 1. Legert KG, Zimmerman M, Stierna P. Sinusitis of odonto- genic origin: pathophysiological implications of early treat- ment. Acta Otolaryngol. 2004 Aug;124(6):655-63. 1. Legert KG, Zimmerman M, Stierna P. Sinusitis of odonto- genic origin: pathophysiological implications of early treat- ment. Acta Otolaryngol. 2004 Aug;124(6):655-63. 1. Legert KG, Zimmerman M, Stierna P. Sinusitis of odonto- genic origin: pathophysiological implications of early treat- ment. Acta Otolaryngol. 2004 Aug;124(6):655-63. 11. Sogur E, Grondahl HG, Baksi BG, Mert A. Does a combi- nation of two radiographs increase accuracy in detecting acid-induced periapical lesions and does it approach the accuracy of cone-beam computed tomography scanning? J Endod. 2012 Feb;38(2):131-6. 2. Cymerman JJ, Cymerman DH, O’Dwyer RS. Evaluation of odontogenic maxillary sinusitis using cone-beam computed tomography: three case reports. J Endod. 2011 Oct;37(10):1465-9. 3. Lee KC, Lee SJ. Clinical features and treatments of odonto- genic sinusitis. Yonsei Med J. 2010 Nov;51(6):932-7. 12. Ozen T, Kamburoglu K, Cebeci AR, Yuksel SP, Paksoy CS. In- terpretation of chemically created periapical lesions using 2 different dental cone-beam computerized tomography units, an intraoral digital sensor, and conventional film. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009 Mar;107(3):426-32. 4. Shahbazian M, Vandewoude C, Wyatt J, Jacobs R. Comparative assessment of periapical radiography and CBCT imaging for ra- diodiagnostics in the posterior maxilla. Odontology. 2013 Dec 28. 5. Shahbazian M, Vandewoude C, Wyatt J, Jacobs R. Compara- tive assessment of panoramic radiography and CBCT imag- ing for radiodiagnostics in the posterior maxilla. Clin Oral Investig. 2014 Jan;18(1):293-300. 13. Umetsubo OS, Gaia BF, Costa FF, Cavalcanti MG. De- tection of simulated incipient furcation involvement by CBCT: an in vitro study using pig mandibles. Braz Oral Res. 2012 Jul-Aug;26(4):341-7. 6. Oberli K, Bornstein MM, von Arx T. Periapical surgery and the maxillary sinus: radiographic parameters for clinical outcome. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007 Jun;103(6):848-53. 14. Tsai P, Torabinejad M, Rice D, Azevedo B. Accuracy of cone- beam computed tomography and periapical radiography in detecting small periapical lesions. J Endod. 2012 Jul;38(7):965-70. 7. Kilic C, Kamburoglu K, Yuksel SP, Ozen T. An assessment of the relationship between the maxillary sinus floor and the maxillary posterior teeth root tips using dental cone-beam computerized tomography. Eur J Dent. 2010 Oct;4(4):462-7. 15. Costa FF, Gaia BF, Umetsubo OS, Cavalcanti MG. Discussion Accordingly, we opted for the protocol related to the region of the maxillary teeth, especially because of the difficulty in standardizing the positioning of the ROI (apical region of the first 5 5 CBCT-based evaluation of integrity of cortical sinus close to periapical lesions adjacent to the cortical sinus, leading to incon- clusive diagnosis of this condition. should be investigated to improve the sensitiv- ity values in detecting oroantral communication very close to periapical lesions. Acknowledgments Our study concluded that CBCT has fair accu- racy in evaluating the integrity of the cortical sinus close to a periapical lesion (values ranging from 66% to 78%) and good specificity in detect- ing the absence of oroantral communication (val- ues ranging from 70% to 98%). However, the low interobserver agreement evidenced the difficulty in diagnosing oroantral communication adjacent to periapical lesions, possibly caused by the high level of image noise. Other acquisition protocols Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq (Marcelo Cavalcanti Research Productivity Scholarship), Brasília, Brazil, grant no. 303847/2009-3 (to M.G.P.C.); Universal Research Proj- ect grant no. 472895/2009-5 (to M.G.P.C.); Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – CAPES (PhD scholarship to O.S.U., and Master’s scholarship to O.S.J.); and Fundação de amparo à Pesquisa do Estado de São Paulo – FAPESP (PhD scholarship to L.R.P.). References Detection of horizontal root fracture with small-volume cone-beam com- puted tomography in the presence and absence of intracanal metallic post. J Endod. 2011 Oct;37(10):1456-9. 8. Pagin O, Centurion BS, Rubira-Bullen IR, Alvares Capelozza AL. Maxillary sinus and posterior teeth: accessing close rela- tionship by cone-beam computed tomographic scanning in a Brazilian population. J Endod. 2013 Jun;39(6):748-51. 16. Costa FF, Gaia BF, Umetsubo OS, Pinheiro LR, Tortamano IP, Cavalcanti MG. Use of large-volume cone-beam computed tomography in identification and localization of horizontal root fracture in the presence and absence of intracanal metal- lic post. J Endod. 2012 Jun;38(6):856-9. 9. Petersson A, Axelsson S, Davidson T, Frisk F, Hakeberg M, Kvist T, et al. Radiological diagnosis of periapical bone tis- sue lesions in endodontics: a systematic review. Int Endod J. 2012 Sep;45(9):783-801. 17. Landis JR, Koch GG. The measurement of observer agree- ment for categorical data. Biometrics. 1977 Mar;33(1):159-74. 18. Maillet M, Bowles WR, McClanahan SL, John MT, Ahmad M. Cone-beam computed tomography evaluation of maxillary sinusitis. J Endod. 2011 Jun;37(6):753-7. 10. Lee EJ, Luedtke JG, Allison JL, Arber CE, Merriwether DA, Steadman DW. The effects of different maceration techniques Braz Oral Res [online]. 2015;29(1):1-7 6 Santos Junior O, Pinheiro LR, Umetsubo OS, Cavalcanti MGP puted tomography and intraoral periapical radiographs in diagnosis of lingual-simulated external root resorptions. Dent Traumatol. 2012 Aug;28(4):268-72. puted tomography and intraoral periapical radiographs in diagnosis of lingual-simulated external root resorptions. Dent Traumatol. 2012 Aug;28(4):268-72. 19. Tirrell BC, Miles DA, Brown CE Jr, Legan JJ. Interpretation of chemically created lesions using direct digital imaging. J Endod. 1996 Feb;22(2):74-8. 20. Stembirek J, Kyllar M, Putnova I, Stehlik L, Buchtova M. The pig as an experimental model for clinical craniofacial research. Lab Anim. 2012 Oct;46(4):269-79. 24. Barrett JF, Keat N. Artifacts in CT: recognition and avoidance. Radiographics. 2004 Nov-Dec;24(6):1679-91. 25. Librizzi ZT, Tadinada AS, Valiyaparambil JV, Lurie AG, Mal- lya SM. Cone-beam computed tomography to detect erosions of the temporomandibular joint: Effect of field of view and voxel size on diagnostic efficacy and effective dose. Am J Orthod Dentofacial Orthop. 2011 Jul;140(1):e25-30. 21. Wang S, Liu Y, Fang D, Shi S. The miniature pig: a useful large animal model for dental and orofacial research. Oral Dis. 2007 Nov;13(6):530-7. 22. Lofthag-Hansen S, Huumonen S, Gröndahl K, Gröndahl HG. Limited cone-beam CT and intraoral radiography for the diagnosis of periapical pathology. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007 Jan;103(1):114-9. References 26. Perrella A, Lopes PM, Rocha RG, Fenyo-Pereira M, Cavalcanti MG. Influence of dental metallic artifact from multislice CT in the assessment of simulated mandibular lesions. J Appl Oral Sci. 2010 Mar-Apr;18(2):149-54. 23. Bernardes RA, Paulo RS, Pereira LO, Duarte MA, Ordinola- Zapata R, Azevedo JR. Comparative study of cone beam com- Braz Oral Res [online]. 2015;29(1):1-7 7
https://openalex.org/W92175041	https://rhle.es/index.php/revista/article/download/330/223	es		Historia del léxico en los diccionarios: la deuda del "Diccionario de la lengua castellana" de la Real Academia Española con los diccionarios de M. Núñez de Taboada	Revista de historia de la lengua española/Revista de historia de la lengua española	2,007	cc-by	16,190	HISTORIA DEL LÉXICO EN LOS DICCIONARIOS: LA DEUDA DEL DICCIONARIO DE LA LENGUA CASTELLANA DE LA REAL ACADEMIA ESPAÑOLA CON LOS DICCIONARIOS DE M. NÚÑEZ DE TABOADA1 G l ò r ia C l a v e r ía N a d a l Universidad Autónoma de Barcelona Resu m en A partir de los datos contenidos en las actas de las juntas de la Real Academia de la primera mitad de siglo xix, se estudia la deuda de dos ediciones del Diccionario de la lengua castellana (5- edición, 1817; 7a edición 1832) con la obra lexicográfica de M. Núñez de Taboada, autor de un diccionario bilingüe francés-español, español-francés de gran difusión (París, 1812) durante el siglo xix y de un diccionario monolingüe (Diccionario de la Lengua Castellana, París, 1825). El análisis de las voces seleccionadas de los diccionarios de este autor proporciona una aproximación a las bases teórico-prácticas que aplicó la Academia en la ampliación onoma siològica de su diccionario vulgar en la primera mitad del siglo x ix y a su concepción del modelo léxico de lengua. P a l a b r a s C l a v e s : historia del léxico, historia de la lexicografía. 1. In t r o d u c c i ó n La lexicografía del siglo x ix presenta u n a evolución que resulta funda m ental para c o m p ren d er la historia del léxico del español y su recepción en los diccionarios: al lado de las diez ediciones del Diccionario de la Lengua Castellana de la Real Academia Española, surge un im portante núm ero de diccionarios m onolingües generales no académicos (Seco, 1987; Bueno Morales, 1995; Esparza, 1999; García Platero, 2003) y algunos de ellos se constituyen en hitos im portantes de la lexicografía española. El interés p o r la admisión de las innovaciones léxicas producidas como consecuencia del progreso de la ciencia y de la técnica es u n o de los aspec tos más relevantes en los diccionarios de esta centuria. Los diccionarios no académicos, au n q u e tom an como punto de partida el Diccionario de la Academia, van p o r delante de la C orporación en la admisión de este tipo ' La investigación n ecesaria para desarrollar este trabajo ha sid o p arcialm en te financiada con una ayuda d e la DGICYT para el p royecto “Banco d e datos d iacròn ico e h isp á n ico ” (HUM2005-08149-C02Ol/FILO) y co n u n a ayuda d el Com issionat p er Universitats i Recerca d e la G eneralitat de Catalunya c o n ced id o al G ru p o d e Lexicografía y Diacronia (SGR2005-00568). R H LE, 2 /2 0 0 7 de léxico, pues la A cadem ia aplica d u ran te el siglo x ix los criterios de au toridad y difusión en la selección de esta clase de palabras, por lo que su inclusión en el Diccionario académico n o se produce a m enos que se cum plan las circunstancias anteriorm ente enunciadas. La cuestión, sin em bargo, no está cen trad a únicam ente e n los neologis mos. En la lexicografía del siglo x ix las voces anticuadas tienen una im por tancia nada despreciable pues obtienen carta de naturaleza en las macroestructuras de los diccionarios p o r su presencia en los textos antiguos. Como se tendrá o p ortunidad de comprobar, la am pliación onomasiológica en los diccionarios sigue tam bién este criterio con especial cuidado. La lexicografía no académ ica en el siglo x ix cuenta con el Diccionario de la Lengua Castellana de M anuel M elchor Núñez de Taboada publicado en 1825 fuera de España, en París, com o uno de sus prim eros ejemplos (Bueno Morales, 1995). Se trata de u n diccionario m onolingüe cuyo título com pleto resulta muy revelador: Diccionario de la Lengua Castellana, para cuya composición se han consultado los mejores vocabularios de esta lengua y el de la Real Academia, últimamente publicado en 1822; aumentado con más de 5.000 voces o artículos que no se hallan en ninguno de ellos\ u n a o b ra de la que se sirvieron otros lexicógrafos del siglo x ix hasta llegar a la copia. Es bien co nocida, por ejemplo, la polém ica que suscitó el sospechoso em pleo que de esta realizó Ju an Peñalver en su Panléxicc?, tam bién recu rre a ella Vicente Salvá, aunque en la “In tro d u cc ió n ” de su Diccionario, cuando se refiere a los autores de los q ue se ha auxiliado, incluye alguna referencia a las obras de Núñez de Taboada juicios negativos: “desdicen en gran m anera del traba jo concienzudo de Capm any los retazos zurcidos p o r N uñez de Taboada, repetidos en su Diccionario de la Lengua Castellana”, del que señala que “su m érito es muy escaso” (Salvá, 1846: X X X ; cfr. Azorín, 2000: 235; Álvarez de M iranda, 2003: 103). El Diccionario de N úñez de Taboada parece, pues, más interesante p o r la utilización que de él realizaron otros lexicógrafos que p o r su difusión com ercial (Alvar, 2002 [1995]: 44), pues com o señala R. Baquero (1992: 455): A p e s a r d e su a p a r e n t e e s c a s a d i f u s i ó n , ta n s ó l o d o s e d i c i o n e s , e r a m u y c o n o c i d o e n t r e lo s l e x i c ó g r a f o s d e l s i g l o x i x . Es m e n c i o n a d o c o m o p r e c e d e n t e e n v a r io s d i c c i o n a r i o s y, l o q u e e s m á s i m p o r t a n t e p e r o q u e s ó l o d e s c u b r i m o s al t é r m i n o d e n u e s t r o e s t u d i o , s ir v ió d e f u e n t e a n u m e r o s o s c o n t i n u a d o r e s . Posiblem ente el predicam ento del Diccionario de la Lengua Castellana de Núñez de Taboada podía deberse a que, al publicar esta obra, su autor ya era conocido com o lexicógrafo, pues no debe olvidarse que tam bién fue autor de u n diccionario bilingüe de gran éxito a lo largo del siglo x ix (Bueno Morales, 1995). 2 M uñoz del M an zan o [C o n d e d e la Vinaza] 1978 [18 93]: cois. 1537 y siguientes; véanse tam bién interesantes noticias e n S eco (1987: 132-133); cfr. B aq u ero Mesa (1992: 455) y García Platero (2003: 268-269). La consulta de las actas de las ju n tas celebradas en la Real Academia Española en la prim era mitad del siglo x ix dem uestra que la labor lexico gráfica de Núñez de T aboada fue, conocida y aprovechada p or la C orpora ción cuando se encon traba revisando distintas ediciones de su Diccionario vulgar. Todo ello acrecienta el valor de este lexicógrafo y de su obra, para e n te n d e r y reconstruir la historia de la lexicografía del siglo xix , la rela ción entre la lexicografía bilingüe y la m onolingüe y, en últim a instancia, la historia de la recepción del léxico en los diccionarios a lo largo de esta centuria. 2. M. N ú ñ e z d e T a b o a d a De la seguram ente interesante vida de M. Núñez de Taboada solo se tienen por el m om ento noticias aisladas5. Barbastro Gil (1993: 75) lo ca racteriza com o un “personaje de sumo interés para conocer la trabazón, en general poco estudiada, entre afrancesados y liberales” y C. M orange (1990: 289), en sus investigaciones sobre Luis Gutiérrez, establece u na com paración entre este y nuestro lexicógrafo, quien habría llegado a Ba yona en 1803 como “español, natural de Galicia, que había abandon ado el m onasterio de San Benito de Valladolid. En marzo de 1805, consiguió u n a bula de secularización; se quedó algunos años en Bayona, dedicán dose a tareas de traducción y dan d o clases de castellano; pasó después a París, donde creó u n a oficina de traducción e interpretación y publicó un diccionario bilingüe (castellano-francés y francés-castellano) varias veces reed itad o ”. Los refugiados a m e n u d o se dedicaban a la enseñanza y a labores de traducción e interpretación (M orange, 2002: 348; Llorens, 19682: 72 y ss.). Parece que com o m edio de subsistencia M. Núñez de Taboada creó en Pa rís u na oficina o escuela de traducción que llevó el nom bre de Interprétation Générale et Traduction de toutes les langues, morts et vivants; el propio Núñez d e Taboada figura com o “D irector de la Interpretación G eneral de Len guas” en algunas de sus obras y traducciones, sin ir más lejos, así aparece en la portada del Diccionario de la Lengua Castellana de 18255. 5 Figura c o m o entrada e n Gil N ovales (1991) p ero la inform ación que se aporta n o va más allá d e sus obras y traducciones. 4 Según Barbastro Gil (1993: 7 5 ), “d ic h o estab lecim ien to, que lleg ó a alcanzar u n a fam a im portan te, se aprovecharía para estab lecer u n a red epistolar entre los distintos g ru p os d e liberales esp añ o les refugiados en Francia, Italia, Países Bajos e Inglaterra”. 5 Muy posib lem en te n uestro lexicógrafo es el N ú ñ ez de Taboada que aparece relacion ad o co n L. F ern án dez de Moratín en A n d ioc (19 6 3 ). D e él se d ice q u e era “bien co n n u des services d e p olice, fré q u en tait assidûm ent les loges m aço n n iq u es et vivait ‘m aritalem ent q u oiqu e m o in e , avec u n e fe m m e des environ s d e Bayonne, qui pensait aussi mal q u e lu i’. En 1827 il s’affiche c o m m e royaliste et vit d e son m étier d e traducteur asserm enté du tribunal d e com m erce, d es tribunaux et d e la Cour Royale; mais sa con d u ite est p eu régulière et notre h o m m e sem b le très habilem ent s’adapter à toutes les situations”. En la d ocu m en tación aportada p or A n d io c aparecen Moratín, Taboada (N ú ñez d e T aboada), Silvela y Salazar com p rom etid os con la lucha contra el A n tigu o Régim en (cfr. Barbastro Gil, 1993: 74). Un p eq u eñ o rastreo bibliográfico'1m uestra que Núñez de Taboada fue autor de diversas traducciones de obras de náutica y música (Civil Engineer, Torpedo War, and sumbmarine explosions d e R obert Fulton, publicada en in glés, en Nueva York, en 1810 y trad ucida al francés como De la Machine in fernale maritime; ou de la Tactique offensive et défensive de la Torpille, Paris, 1812; Lengua universal marítima ó sistema completo de señales de dia y de noche, hechos por los medios más sencillos que se hallan ordinariamente á bordo de cualquier bar co; U tilís im o a lo s n a v e g a n te s de to d a s la s n a c io n e s , por Levin-Joergen Rohde, Burdeos: Im prenta de Suwerinck, 18377; traducción del francés al español de la obra de Francesco Molino, Método completo para aprender á tocar la gui tarra... (París, 1827-1828?) y de la de Jo sep h Virués, Eléments d ’harmonie, ou le contre point expliqué en six leçons (París, 1825), en este caso del español al francés. Figura también como tra d u cto r del Dernier rapport du comité na tional de Vaccine de Londres (en 1811), traducido del inglés y publicado en París en 1812. Tradujo, además, textos de carácter institucional y político: Constitution politique de la monarchie espagnole, promulguée à Cadiz le 19 de mars 1792 (sic), préc. du rapport des Cortés chargée de présenter le projet de cosntitution (París, 18148) y Constitution militaire de la monarchie espagnole adoptée par les Cortés nationales en 1821 (Paris, 1821); Derniers efforts de la politique de Buonaparte pour séparer lEspagne de la coalition formée contre lui, ou Manifeste des Cortés de la nation espagnole sur le traité proposé par Buonaparte dans le mois dedécembre 1813, et sur les motifs de la conduite magnanime de la Nation, dans ce memorable événement (París, 18149); Dernier soupir de l ’inquisition en Espagne, ou Conduite sage et mesurée de la régence du royaume d Espagne envers le dernier défenseur de ce tribunal de sang, recueil de pièces officielles traduites de l ’espagnol parE. Nuñez de Taboada (París, 181410). Todos estos textos denotan u n a im p ortante actividad traductora en tre los años 1812-1814 y 1820-1828. Al m argen de las traducciones, N úñez de Taboada es autor de los instru mentos básicos para la enseñanza y aprendizaje del español como lengua extranjera: aparte de un im portante diccionario bilingüe español-francés, francés-español, es autor de la Grammaire de la langue espagnole à l ’usage des français, reduite à les plus simples éléments, publicada por prim era vez en Paris en 1822, u n a gram ática para extranjeros de carácter explicativo (Sánchez Pérez, 1992: 205). Nuestro lexicógrafo figura, además, com o director de la revisión y corrección de la Guide de conversation, ou Vade-mecum du voyageur en sept langues: anglais, allemand, français, italien, espagnol, russe, barbaresque 6 C atalogue Bn-O pale plus en< h tt p : //c a ta lo g u e .b n f.fr ; h ttp ://w w w .w orld cat.org> 7 La ob ra aparece citada en Biblioteca m arítim a española de M. F ern án dez d e Navarrete (1968 [1851]: II, 3 91 ). Cfr. N ú ñ e z d e Arenas (1963: 338). 8 Se trata d e una d e las primeras traducciones eu ro p ea s d e la C onstitución d e 1812 (Fernández Sarasola, 20 0 0 ). 9 Cfr. A. Palau y D u lcet (1958: n® 197.498). 10 Cfr. A. Palau y D u lcet (1958: n B 197.499). ou moresque; contenant des conversations familières sur diverse besoins et circonstan ces ou il peut se trouver. Ovrage mis en ordre, revu et corregé sou la direction de M. Núñez de Taboada. Suivi d ’u n vocabulaire français-arabigo-barbaresque, et d ’un dialogue dans ces deus langues sur des choses de première nécessité par M. Núñez de Taboada publicada p o r prim era vez en Paris en 1833. Estamos, pues, ante u n a persona que, en palabras de Azorin (2000: 235), “vivía de esta faceta profesional q u e hoy calificaríamos com o ‘m ediador lingüístico’” y que, por las traducciones que llevó a cabo, dom inaba perfectam ente el francés y el inglés y, p o r las publicaciones relacionadas con la enseñanza de lenguas (gramática, diccionario y otros materiales), debió estar plenam ente volca d o a este cam po profesional com o m uchos otros exiliados establecidos por aquellos años en Francia o en Inglaterra. Es además, autor, de un Diccionario de la Lengua Castellana publicado tam bién en París en 1825. 3. Los D I C C I O N A R I O S D E M. N Ú Ñ E Z D E T A B O A D A El Dictionnaire français-espagnol et espagnol-français plus complet et plus co rrect que tous ceux qui ont été publiés ju sq u ’a cerjour, y compris celui de Capmany (Paris, 1812) se constituye en un notable éxito editorial (Verdonk, 1991: 2980) si se juzga a través del considerable núm ero de ediciones que la obra experim entó tanto en Francia como en España11. En el amplio panoram a d e la lexicografía bilingüe español-francés que traza Cazorla Vivas (2002: 389) en su tesis doctoral, señala que esta obra “seguram ente fue el rep er torio bilingüe más vendido en el xix , sobre todo en la prim era mitad del siglo”. Com o se reconoce el mismo título de la obra, este diccionario se encuen tra estrecham ente relacionado con el de A. Capm any (1805) del que incor p o ra ín tegram ente el prólogo (García Bascuñana, 1999: 119; G em m ingen, 2001: 236-237). Existe, sin em bargo, u n a diferencia nada desdeñable entre ellos, pues la o bra de Núñez de Taboada tiene las dos partes propias de un diccionario bilingüe (francés-español, español-francés), mientras que Cap m any renu nció expresam ente a elaborar la parte español-francés. El Dictionnaire français-espagnol et espagnol-français plus complet et plus co rrect que tous ceux qui ont été publiés ju sq u ’a cerjour, y compris celui de Capmany está fo rm ado p o r dos tomos y cada u n o de ellos tiene su propio Prólogo: el p rim er tom o contiene la parte francés-español y com ienza con un “Prólo 11 Véase la in form ación sobre las distintas e d ic io n e s en Fabbri (1979: 137, n” 1278-1283) y e n Palau y D u lcet (1948: n “ 197476 y ss.), Q u em ad a (1968: 6 0 2 ), Cazorla Vivas (2002: 379-388). Se publica p o r prim era vez e n España (Madrid) en el a ñ o 1820 (4 S ed ición , Editorial Sancha), au n q u e Q u em ad a (1968: 602) cita u n a ed ición m adrileña d e 1817. En varios diccion arios bilingües del siglo x i x aparece j u n t o a C apm any y otros co m o diccionario fuente, p o r ejem p lo en el d e j . Marchena, Diccionario m a nu al francés-español, compendiado del de Capmany, N úñez de Taboada, etc., Burdeos, 1821 (Fabbri, 1979: 136, n s 1270). go de los editores” redactado en español en el que estos reconocen el valor de la obra de Capmany: La correspondencia y propiedad de las voces, la delicada elección de frases proverbia les, la exactitud de las definiciones, su estilo genuino y castizo, y otros mil rasgos de su bien cortada pluma fixáron la atención de los editores, decidiéndolos desde lue go á proponérsele por norte y guia en la composicion del que ya tenían em prendido (Núñez de Taboada, 1812:1, v). De m anera que, para este volum en, se tom a como base el Diccionario de Capmany con pocos cambios y algunas mejoras: los editores se contentáron con retocarle, y al executarlo se han impuesto la rigurosa obligación de conservar fielmente su fondo y de ceñirse á corregir las definiciones y fal tas de correspondencia de muchas voces facultativas, añadir otras y un crecido núm ero de términos omitidos, aun de los recibidos por la Academia francesa, á mudar muchas palabras gramaticales contrarias al carácter propiedad de ámbas lenguas, á seguir un plan mas constante y uniform e que el del Señor Capmany, enm endar las innumerables inversiones y faltas de ortografía de que el poco cuidado del impresor ha dexado salpi cada esta obra, que por su naturaleza debe servir de pauta y norma no solo para hablar sino para escribir, finalmente á refundir en el cuerpo de la obra los suplementos, aun el geográfico, para mayor comodidad y ah orro de tiempo de los que tengan necesidad de consultarle (Núñez de Taboada, 1812: I, v). El tom o segundo del Diccionario francés-español y español-francés de N úñez de Taboada contiene la parte español-francés, sin correspondencia alguna con la obra de Capmany. En este caso el Préface (Núñez de Taboada, 1812: II, v-xviij) está redactado en francés, tien e básicamente la misma estructura y contenido que el prólogo del volum en francés-español con una página de presentación y el prólogo p ro ced e n te del Diccionario de Capmany, esta vez traducido al francés. El Préface, sin em bargo, está redactado en prim era persona del singular, en él no aparece el térm ino editores y en la parte de la presentación se consignan algunas observaciones sobre los criterios con cretos de elaboración del Diccionario12. En los dos prólogos13 se indica que se incluyen, en el cuerpo del Diccio nario y siguiendo o rd en alfabético, los suplem entos dedicados a las voces científicas y al Dictionnaire géographique français-espagnol del Diccionario de Capmany. También los dos prólogos m uestran u na especial atención a la inclusión de voces nuevas, algo h e red a d o del Diccionario de Capmany (Anglada, 1997-1998; García Bascuñana, 1996, 1999; Checa, 1989; Fernández Díaz, 1989; Lázaro Carreter, 1985 [1949]: 288; 2002: 3-4; N iederehe, 1987: 23-24; Rodríguez Fernández, 2006; Roig, 1995). Así, en el Prólogo en espa ñol se observa que el Diccionario posee “un a grand copia de térm inos de ciencias, artes comercio, m anufacturas, diplomática, jurisprudencia, adm i nistración, econom ía pública, etc, q u e no se encuentran en ningún otro 12 Para la suerte d e este p ró lo g o e n las e d ic io n e s p osteriores del diccionario bilingüe, véase Bruña (1996; 1999: 107; 2001: 249). 13 Así aparece explicad o e n N ú ñ e z d e Taboada (1812: I, v y xi, nota (); 1812: II, v-vj y ix, n ota (**))■ diccionario frances-español. Todas estas voces han sido tom adas de los m e jo re s autores antiguos y m odernos, y de diccionarios de conocido mérito, e n que se halla perfecta y com pleta nom enclatura de las ciencias, artes y oficios” (Núñez de Taboada, 1812: I, v-vi), palabras que se repiten en el prólogo de Capmany (N úñez de Taboada, 1812:1, ix). Sin paralelismo en el Prólogo de la parte francés-español, en el Préface de la parte español-francés se m encionan las voces antiguas: Les auteurs des dictionnaires espagnols-françois connus ont eu un soin particulier de n ’y insérer q u ’un très-petit nom bre de vieux mots castillans, et d ’en rejeter une quan tité infinie, quoiqu’avoués par l’Académie espagnole: c’étoit un moyen sûr de mettre les lecteurs dans l’impossibilité de pouvoir lire les bons auteurs espagnols anciens, pas même les C a l d e r ó n , les L o p e - d e -V e g a , les G á r c i l a s o , les C e r v a n t e s , les M a r i a n a . Tous ces mots ont trouvé place dans mon Dictionnaire. On jugera de l’utilité de mon travail, lorsqu’on sera dans le cas de lire ou d ’expliquer quelqu’un des auteurs q u e je viens de citer (Núñez de Taboada, 1812 : II, v). Resulta tam bién significativa y sin precedente en Capm any la explica ción sobre las voces de g erm anía introducidas en el Diccionario de Núñez d e Taboada con la m arca bohémien (Núñez de Taboada, 1812 : II, v ). Ambas observaciones se constituyen en datos valiosos para en te n d e r la técnica lexicográfica em pleada p o r N úñez de Taboada y su concepción de la re cepción del léxico en los diccionarios. Las notas que el Diccionario de N úñez de Taboada introduce en la parte reproducida del Prólogo de Capm any apuntan, muy probablem ente, a la existencia de diferencias en los criterios lexicográficos aplicados p o r cada u no de estos lexicógrafos en sus respectivas obras. Las palabras de Cap many en el Prólogo dem uestran que su obra surgió de u n a reflexión p ro funda sobre el establecim iento de la nom enclatura del Diccionario y unos criterios claros de selección del léxico14, mientras que las pocas notas que incluyen el Prólogo y el Préface de la obra de Núñez de Taboada reflejan un criterio más amplio y m enos refinado y un trabajo de cotejo con respecto al Diccionario de la A cadem ia15. A pesar de que se carece, de m om ento, de un estudio detallado de la re lación de contenido en tre el Nuevo Diccionario francés-español de Capmany y 14 Por ejem plo, Capm any exp lica d eta llad am en te cuáles son los artículos d e los d iccionarios de C orm on y Gattel q u e n o ha in co rp orad o a su Diccionario por n o considerarlos necesarios (Capmany, 1805: V-VIII). 15 Buen ejem p lo d e ello es la n ota q u e se in trodu ce a las siguientes palabras d el p rólo g o d e Cap many: “se han d ese ch a d o c o m o im p ertin en tes, é in útiles para los q u e solo buscan la version co rre sp on d ien te, todos aqu ellos artículos d e voces, q u e, ad em as de n o ser francesas, p u es p e rten ecen á todas las lenguas, n o m udan d e estructura, ni d e p ron u n cia ción en castellano: c o m o son todas las d e oficios, em p leos, ritos, costum bres, y esta b lecim ien to d e países extraños, y rem o to s” (N ú ñ e z d e Taboada, 1812: I, ix), texto q u e se apostilla co n la sigu ien te nota: “Se verán en este d iccionario m uchas d e estas voces q u e no se han q u erid o om itir p or hallarse e n el d e la Academ ia francesa q u e ha i'-’b id o m uy graves razones para n o dexar d e segu ir”. La parte francesa tiene este m ism o texto y la n ota c o ("»oondiente: “O n a trouvé con ven ab le d ’insérer au m oin s c e u x q u ’on trouve dans le D ictionnaire d e 1 r- M é m ie , q u ’o n a eu d e b o n n e s raisons p o u r suivre”. de la parte equivalente de la obra de N úñez de Taboada, se puede suponer que este últim o revisó la obra de Capm any según los criterios que apare cen en el Prólogo y que de esta revisión resultó u n im portante aum ento de artículos, pues, según los datos del p ro p io Capmany, su Diccionario consta de 30.000 artículos16 (Capmany, 1805: III; N úñez de Taboada 1812:1, vij; II, vij); en cambio, el Prólogo del Diccionario de N úñez de Taboada m enciona la existencia de 37.000 artículos. Muy posiblem ente la prim era parte (fran cés-español) del Diccionario de N úñez d e T aboada surge a partir de la fu sión del cuerp o principal del Diccionario de Capm any con los suplementos de este (el Supplement que recoge voces científicas con “un uso mas com ún y c o rrie n te” y el Dictionnaire geographiqué1) . A ello, debieron añadirse unos cinco mil artículos, un hecho al que parecen hacer referencia las palabras del Prólogo de los editores citadas anteriorm ente. El segundo volum en del Diccionario bilingüe de Núñez de Taboada con tiene la parte español-francés sin co rresp o n d en cia en el Diccionario de Cap many y alcanza los 70.000 artículos (N úñez de Taboada, 1812: II, v18); la diferencia se refleja en u n aum ento del n ú m e ro de páginas (964 para el prim er tom o y 1.392 para el segundo). Q u ed a p o r estudiar la m etodología em pleada en la selección y acopio de estos 70.000 artículos y su d ep en d e n cia con otros diccionarios bilingües anteriores. Un aspecto muy significativo para la investigación que se desarrolla en este artículo reside en el h echo de que en el Diccionario de Núñez de Taboa da, a diferencia del de Capmany, se señalan con un asterisco las entradas de la nom enclatura que no aparecen en el Diccionario de la Academia Es pañola (parte español-francés) o en d e la Academ ia Francesa (parte fran cés-español) . En este aspecto, el Diccionario de N úñez de Taboada sigue la tradición desarrollada en Francia desde el siglo x v m que B. Q uem ada ha d en o m in ad o académisme y que se m anifiesta especialm ente en la lexico grafía bilingüe en “ouvrages qui se ratta c h e n t d ’un m anière déclarée à la lignée académ ique, soit p ar leurs titres, soit p ar leurs préfaces” (Quemada, 1968: 219). En el caso de la obra de N úñez de Taboada, por tanto, se tom an com o p u n to de partida los diccionarios de las Academias Francesa y Espa ñola, y se am plía la nom enclatura de estos con palabras técnicas, científicas y vulgares, tam bién característica p ro p ia de parte de la lexicografía bilin güe francesa (Q uem ada, 1968: 51) y d e la que se constituye en preceden te claro p ara N úñez de Taboada la o b ra de Capm any (García Bascuñana, 1999yR oig, 1995). 16 N o aclara C apm any si esta cifra incluye los su p lem en to s, p o sib lem en te n o los engloba, p u es un p e q u e ñ o cálcu lo sobre el n ú m ero d e entradas por p á g in a sum a u n as 31.700 entradas. Cfr. Cazorla Vivas (2002: 3 6 6 ), q u iep m e n cio n a unas 32.000 entradas. 17 El si" rimento d e voces técnicas co n tien e 1.500 entradas (F ern án d ez Díaz, 1987: 531) y el Diction na ire r rjgiaphique consta d e p o co m ás d e 500 artículos. S egú n Cazorla Vivas (2002: 396) son 64.500 entradas. En 1825 se publica tam bién en París el Diccionario de la Lengua Caste llana de N úñez de Taboada. Se trata de un o de los primeros diccionarios m onolingües no académicos del siglo x ix que nace con las características propias de este tipo de lexicografía19. Ya en el título del mismo se alude a su relación con respecto a la lexicografía institucional y a la m etodología em pleada para su elaboración con las siguientes palabras: Diccionario de la Lengua Castellana para cuya composición se han consultado los mejores vocabula rios, y el de la Real Academia últimamente publicado en 1822; aumentado con más de 5000 voces ó artículos que no se hallan en ninguno de ellos. Com o justificación de su obra, ap u n ta N úñez de Taboada en el Prefacio que todos los diccionarios anteriores tienen defectos: “pobreza ó esca sez de voces, falta de definiciones, ó definiciones inexactas é inadecuadas, om isiones de acepciones generalm ente usadas y recibidas, lenguage trivial, sin corrección y aun chavacano, ortografía varia é incierta; tales son los de fectos com unes á todos ellos”. Ya en la prim era página se refiere a la Real A cadem ia Española y a su “diccionario nacional”, del que observa que le “h a servido de basa y fundam ento al que ofrezco hoy al público, el que por consiguiente contiene todos los vocablos del diccionario de la Real Aca dem ia Española tanto n um éricam ente com o en orden á sus diversas acep ciones”. Y denuncia lo que a sus ojos constituye la principal deficiencia de esta obra: “Pero este diccionario, au nqu e en sí bueno, está falto de un muy crecido nú m ero de voces pertenecientes á nuestra lengua vulgar; suerte com ún á todos los libros de esta naturaleza”. Para resolver el problem a, el Diccionario de Núñez de Taboada “se halla aum entado con cerca de 5.000 voces de que carece el de la Real Academ ia Española, y que he sacado de su p rop io caudal y de los autores nacionales mas clásicos. Todas estas voces van señaladas con un asterisco” (Núñez de Taboada, 1825: II). Si hasta este pu nto la crítica a la A cadem ia ha sido m edida y se ha m ezclado con elogios, en el párrafo siguiente del Prefacio tom a un carácter más intenso cuando el autor pasa a ex poner detalladam ente los errores del Diccionario de la Academia. En p rim er lugar, se hace referencia a la “notable desigualdad en cuanto tiene de bu en o y de m alo” com o “vicio ca pital” que se atribuye al hecho de que intervengan diversas personas en su elaboración. Se com prende que, para N úñez de Taboada (1825: III), “este inconveniente, que solo podrá evitarse, confiando la egecucion de esta es pecie de obras á u na sola persona con sujeción á la censura de hom bres 19 B u e n o M orales (1995) reco ge d o s diccion arios anteriores al d e N ú ñ ez d e Taboada: Cristóbal Pía y Torres, Diccionario de la lengua castellana, por la Academ ia Española, compendiado, París: Lib. d e B. C o rm o n y Blanc, Imp. de R ignoux, 1816; y V icen te G onzález A rnao, Diccionario de la lengua castellana, para cuya composición se han consultado los mejores vocabularios de esta lengua, y el de la R eal Academia Española últim am ente publicado en 1822; aum entado con mas de 5 0 0 0 voces o artículos, que no se hallan en nin gu n o de ellos, Madrid: Im prenta de d. P. Sanz y Sanz, 1823. Sobre la relación entre la obra d e V. G onzález Arnao y E. N ú ñ e z d e Taboada, bien evid en te e n el título, véase B u e n o Morales (19 95), habría q u e estudiar las rela cion es en tre los d o s diccionarios y tam b ién en tre sus autores, ambos refugiados en Francia y afrancesados-liberales (Barbastro Gil, 1993: 21 y ss., 66-67, 72-76). doctos, dotados de luces especiales, n o subsiste en el mió, porque yo solo he trabajado en él”. En el terren o de las críticas concretas, se reprueban las corresponden cias latinas p o r ser “impropias, bárbaras á veces, casi nunca directas, y p or lo com ún perifraseadas con circunloquios de un latín rom anceado q u ejamas fué, no digo el de Cicerón, pero ni aun el del tosco y poco culto E nio”. En opinión de Núñez de Taboada (1825: iii), las equivalencias deberían haberse suprim ido en la últim a edición, la eliminación de este tipo de in form ación, sin embargo, no se prod ujo hasta la edición de 1884 (12a) y le valió a la Academia bastantes censuras d u ran te el siglo xix. Se critican tam bién cuestiones que atañen a la variación gráfica en ciertas palabras en las que el Diccionario de la Academ ia no concuerda con la Ortografía publicada p o r la C orporación y al uso de los clíticos de tercera persona. Se pon en en tela de juicio de form a atinada aspectos particulares de la configuración de los artículos y sus definiciones: term inaciones de los participios pasados, definiciones de los adjetivos, redacción d e definiciones que afectan al régi m en verbal y definiciones m ejorables de las que se proporciona una página de ejemplos. Se hace referencia, adem ás, al problem a de la ordenación de c h y l l que en la obra de N úñez de Taboada (1825: vn) se consideran com o “dos letras diferentes”. Se concluye de todo lo anterior que “seria nun ca acabar si hubiese de hacer la enum eración de las faltas que he nota do en dicho diccionario. La indicación de estas pocas bastará para que el sabio cuerpo de Académicos Españoles piense en enm endarlas todas”. En realidad, el cuerpo principal del Prefacio del Diccionario es una crítica co n tu n d en te al Diccionario de la Academ ia, un texto que, com o se exami nará más adelante, llega a la C o rporación y hace mella en ella. Todas las deficiencias señaladas en la o bra académ ica sirven a Núñez de Taboada (1825: v n ) para concluir que “estos defectos he procurado evitarlos en mi Diccionario. Sin em bargo n o m e atrevo á afirm ar que vea la luz públi ca exento de faltas; aunque debo confesar con franqueza que no dejo de ten er b u en a opinion de él, y q ue me lisonjeo de que le recibirá el público com o el mas correcto en todas sus partes, y com pleto en cuanto á la copia de voces, que se ha publicado hasta a h o ra ”. El objetivo últim o de su o bra aparece claram ente descrito en el últim o párrafo del Prefacio cuando advierte q ue “todo lo daré por bien em pleado si de algún m odo puede con tribu ir á q ue no se descaste y desfigure la incom parable lengua de los Cervantes e n estos tiempos, para ella calami tosos, en que la m anía de traducir del francés cuanto se presenta, bu en o ó malo, h a cundido hasta cierta clase d e hom bres, verdaderos vándalos de la lengua, dispensados p o r estado y condicion de toda especie de luces y conocim ientos, y en que u n a cáfila de traductores á destajo hacen gem ir la prensa con un diluvio de producciones en gerigonza castellana, con que ciertos contrabandistas de la lengua española de esta capital inundan la Península y el Nuevo M undo” (1825: v m ), palabras que resultan muy sig nificativas en u n traductor e intérprete profesional. El Diccionario m onolingüe de Núñez deT aboada, tanto en su título como en el Prefacio, deja claras las diferencias y mejoras que la obra pretende aportar con respecto al Diccionario académico y contiene u n a crítica directa a este último. Puede interpretarse com o u n a simple estrategia comercial o como u n a p ostura lexicográfica de mayor calado teórico-práctico. Sea como sea, la obra tiene una incidencia im portante tanto en la lexicografía no académica subsiguiente como en la misma lexicografía académica. 4. Los D I C C I O N A R I O S D E N Ú Ñ E Z D E T A B O A D A E N L A A C A D E M I A Se registran en las actas de las ju n tas celebradas en la Real Academia Española del p rim er tercio de siglo x ix varias referencias a Núñez de Taboada y sus diccionarios. La prim era pertenece al acta del 26 de noviembre de 1816 (Libro 19) y en ella se recoge u n a de las tareas desarrolladas en la preparación de la q uin ta edición del Diccionario: H abiendo dado cuenta á la Academia de las voces extractadas del Diccionario de Nuñez Taboada, y que se suponia aunque no con verdad que faltaban en el de la Academia, se eligieron aquellas que parecieron de buen uso y se determ inó que se cotejen con la nueva edición las elegidas quedando las que parezcan dignas para el Suplemento. En aquellos m om entos, la Academia se en contrab a ultim ando la quin ta edición del Diccionario y, tal com o se señala en el acta, las voces que se apru eban están destinadas al Suplemento. Nueva alusión a las obras de este autor aparece en las actas académicas unos años más tarde, concretam ente el 19 de febrero de 1829 (Libro 21), sesión en la que se trata de las críticas a la A cadem ia que contiene esta obra20. D urante aquella ju n ta se leyó el Prólogo del Diccionario de Núñez de Taboada “y se reconocieron algunas de las voces o artículos que expresa no hallarse en nuestro Diccionario y que hace llegar hasta el núm ero de 5000”. De form a elegante, la Academ ia acuerda “no mostrarse por ahora parte agraviada reservando para el tiem po en que se publique la edición del Diccionario, en que entiende actualm ente, el descubrir la falsedad é injusticia con que se h a pretendido degradarla”. Unas semanas más tarde, en la ju n ta del 2 de abril de 1829, se lee u n a lista de voces procedentes del Diccionario de N úñez de Taboada, algunas de ellas son aceptadas, otras son rechazadas p o r no estar autorizado su uso p o r textos adecuados. Así, el 20 Este e p iso d io q u ed a reflejado tam bién en el diario d e J. M usso y Valiente (Martínez A rnaldos y Molina Martínez, 2002: 190): “El secretario ha p resen tad o el prim er to m o d el D iccionario de la len gu a dado a luz e n París p o r N u ñ e z Taboada. Se ha leíd o el p rólog o e n q u e zahiere fu ertem en te a la Acade mia, a u n q u e en casi to d o sin fun d am en to. Y se han le íd o tam bién algunas d e las 500 (sic) voces q u e se d ice haber au m e n ta d o , y n o se cree que hay m otivo para haber insertad o casi todas las tales voces en el D iccion ario”. Cfr. Abad M erin o (2006: 199). secretario, Francisco A ntonio González, hace constar en el acta de aquella junta: Leí vina lista de voces entresacadas del D publicado en París por el S“r. Núñez de Taboada y que son del núm ero de las 5000 añadidas, según ha decantado á las que contiene el nuestro de la Ac.; pocas realmente o casi ninguna autorizada por clasicos de nuestra lengua y solo introducidas por el uso que en varias de ellas no es tan constante y culto como presume el referido editor. Se aprobaron algunas de las pertenecientes a mi re partimiento y cuya definición quedo a mi cargo. U no de los aspectos más interesantes de este episodio se halla en que, al analizar la lista de voces que proceden del Diccionario de Núñez de Taboada, se plantea u n a cuestión teórico-metodológica im portante y que per mite observar bastante de cerca los criterios de selección del léxico para el Diccionario vulgar en aquellos m om entos. T. González Carvajal sugiere, a raíz de todo ello, que “convendría fijar ciertas y estables reglas para ad mitirse cualquiera voz en nuestro Diccionario, ó que por lo menos en con sideración al casi com ún uso de varias se extendiere la lista o catálogo de autores que p u ed an autorizar las ultimas hasta la época de 1800 contando solo aquellos cuyos escritos se han m irado p o r el público com o modelos de erudición, de cultura y de buen lenguage en la últim a época del siglo anterior. Mas com o este p u n to necesita de cierto tino, meditación y delica deza se determ inó encargarlo á una com ision para la que el Sor. Director nom bró a los Sres. Navarrete, González Carvajal y Musso”21. E n las sesiones siguientes la comisión no m b rad a para la revisión de los autores que podían considerarse com o clásicos presenta u n a lista de escritores de la que se acla ra que no se constituyen en “modelos de estilo y sí com o texto o p ru eb a de voces y testigos de frases recibidas en el uso com ún y culto” (junta del 30 de abril de 1829). Por lo que, a pesar de que minimizan el núm ero de voces que p u ed en incluir en el Diccionario académico procedentes de las del Diccionario de N úñez de Taboada, se suscitan, com o consecuencia de todo ello, cuestiones de mayor envergadura p o r cuanto revisan y am plían la nóm ina de autores que se p u ed en usar com o autoridades. Nueva referencia al Diccionario de N úñez de Taboada aparece u n año más tarde en el acta de las ju n tas del 18 de marzo y del 1 de abril de 1830 (Libro 21), en las que tam bién se ap ru eb an algunos artículos, (“muy po cos” según el a cta), que proced en de esta obra22. Las alusiones al Diccionario de N úñez de Taboada se producen en dos m om entos distintos. El prim ero corresponde a finales de 1816 y, por tanto, su contenido afecta a la 5a edición del Diccionario de la Academ ia (1817). El segundo pertenece a los años 1829-1830 y los resultados d eberían reflejarse 21 Cfr. sobre ello la reseñ a q u e h ace J. Musso y Valiente e n su Diario (M olina Martínez, 2001: 261). 22 En esta ép oca, a d iferen cia d e lo q u e ocurre m ás ad elan te, n o se con sign an e n las actas las voces admitidas, por tanto, se d e s c o n o c e cu áles son. Se enca rga d e extractarlasJ. Musso, cfr. Martínez Arnald o s y M olina Martínez (2002: 190); Abad M erino (2006: 199); y, sobre todo, M olina Martínez (20 01). en la 7a edición del Diccionario (1832). Además, si el Diccionario de la Len gua Castellana de N úñez de Taboada se publicó p o r prim era vez en 1825, la prim era referencia a este au to r no puede referirse a esta obra, debería correspon der a su Diccionario bilingüe. A pesar de que la Academ ia había acordado no mostrarse agraviada, al poco de la publicación del Diccionario, salió en su defensa Joaq u ín Lorenzo Villanueva (1826), académ ico exiliado desde 1823 en Inglaterra (Zamora Vicente, 1999: 221-222; Llorens, 19682), con un largo artículo en sus Ocios de españoles emigrados en el que analiza punto p o r punto el prólogo del Diccionario de Núñez de Taboada refutando las críticas vertidas en él al Dic cionario de la Academia; en la segunda parte del artículo, J. L. Villanueva la e m prende con el Diccionario de N úñez de Taboada23 criticando algunas d e finiciones de voces en las que este lexicógrafo h a copiado al Diccionario de la Academia y, por tanto, h a com etido el mismo e rro r que critica (Villanue va, 1826: 364-366); agrega, además, otros errores que se hallan en las voces añadidas p o r Núñez de Taboada, denuncia el hecho de que se utilicen en las definiciones palabras que n o aparecen en el Diccionario, el acortam iento de las definiciones del Diccionario académico con lo que estas quedan in completas, el hecho de que presenta com o voces añadidas palabras en las que la única variación con respecto al Diccionario de la Academia se halla en la grafía, la inclusión de nom bres propios, de voces dem asiado especializa das y de voces calificadas de “exóticas y aun ridiculas” (1846: 374). Acaba su artículo J. L. Villanueva com p o n ien d o un pequeño texto con algunas de las cinco mil palabras que introduce Núñez de Taboada en su Diccionario p ara m ostrar su ridiculez; en algunas de las voces elegidas Villanueva lleva toda la razón ( mañanar, innupto, hijezno, faufau, descultizar, etc.). Tal como había acordado la Academ ia en su día, el Prólogo d e la séptima edición (1832) tiene un largo párrafo en el que se exponen los criterios de selección y admisión de voces nuevas que se han aplicado en la elaboración del Diccionario y, de pasada, se hace u n a discreta referencia al Diccionario de N úñez de Taboada24: 23 Llorens (19682: 313) señala q u e e l artículo se publicó con m otivo d e “una nueva ed ició n del d iccion ario francés-español d e N ú ñ e z d e T ab oa d a ”, p ero el título d el artículo se refiere claram en te al Diccionario de la lengua castellana. Cfr. tam b ién M olin a Martínez (2001: 261, n ota 13). 24 C o m o aparece en el Diario y la c o r r e sp o n d en cia d e J. Musso y Valiente, es este aca d ém ico el e n cargad o d e elaborar el p rólo g o d e la séptim a ed ició n del Diccionario. Musso había a b a n d on ad o Madrid y se había establecido e n Lorca e n j u n i o d e 1830 y d esd e allí corrigió parte d e la séptim a ed ició n del Diccionario que se encontraba ya en la fase final d e elaboración. En u n a carta rem itida a Musso p or F. A. G onzález, a la sazón secretario d e la A cad em ia, este le p rop on e q u e se en cargu e d el p rólo g o y M usso acepta; para q u e p u ed a trabajar e n él, F. A. G on zález le envía los siguientes materiales: “remitiré n o sólo la lista d e voces sacadas p or V. d el sin par T ab oad a sino la lista gen eral d e todas las q u e él llama d e a u m e n to que m andaré sacar; irá e n c o p ia del p r ó lo g o del m ism o autor, a q u ien v. dará su m erecid o y to d o lo qu e parezca, p revin ien d o fo rm a lm en te lo q u e se m e alcance sin echar indirectillas al vacío: esto s o lo lo h ag o co n q u ien ten g o seguridad q u e sabe rem ontarse y cogerlas c o m o V. lo ha h e c h o ”. Más ad e lan te le envía to d o el material a n u n cia d o para q u e Musso pueda escribir el p rólo g o y vuelve a aparecer u n a referencia al “varapalo fuerte y co n clu y en te para el dich o Taboada y para cu an tos se han arrojado como no falta quien todavía acuse de escasez al Diccionario académico, echando me nos en él voces que en su concepto debieran incluirse, la Academia dene por conve niente advertir que ahora y siempre ha procedido con suma circunspección en dar lugar á voces que no le han tenido anteriorm ente; porque el verdadero caudal de las lenguas vivas no se compone sino de las palabras y frases usadas generalmente entre las personas cultas. La Academia ha sido severa en esta calificación; y lejos de adoptar fácilmente y sin mucho examen voces y locuciones nuevas, ha desechado todas las que no se hallan autorizadas por el ejemplo de nuestros escritores clásicos ó por el uso cla ramente reconocido como general y constante. Muchas veces ha diferido la admisión, dando tiempo á que se prescriban voces bien formadas, que empiezan á introducirse, y suenan ya con alguna aceptación entre las demás del lenguaje. Así que ha exclui do nombres caprichosos y pasageros de trajes y modas que hoy se emplean y mañana desaparecen para no volverse á oir nunca. Por igual razón ha excluido infinitas voces técnicas de ciencias, artes y oficios que no pertenecen al lenguaje común, único objeto del Diccionario. Por la misma no ha admitido las voces fácilmente formables, propias del estilo familiar que permite el uso m om entáneo de una voz, pero sin darle derecho al uso perpetuo, de lo que hay ejemplos en nuestros buenos escritores. Si se diera en trada á estas y otras clases de voces en el Diccionario, fácil cosa fuera añadirle no solo cinco sino muchos mas millares de artículos. Sirva esto para satisfacer la delicadeza de los que todavía pudieran tachar al Diccionario de escaso y diminuto25. Y prescinde la Academia de los reparos que se han opuesto al suyo por no hallarse en él algunos artículos de voces extrangeras como mutualmente, ó de voces estropeadas y desfiguradas como Barbarescos:, ó escritas con mala ortografía como uraño; ó pronunciadas á la mane ra del ínfimo vulgo como hespital. Adoptar artículos de esta especie sería no enriquecer sino m anchar el Diccionario, y profanar el carácter del noble y magestuoso idioma de Castilla (DRAE, 1832: Prólogo). La referencia a las cinco mil palabras conduce inequívocamente al Dic cionario de N úñez de Taboada; las palabras que se citan com o ejemplos, tam bién. La Academia tom ó en serio las críticas vertidas por Núñez de T aboada en el Prólogo de su Diccionario y, au n q u e no adm ite los principios p o r los que este lexicógrafo am plía su Diccionario con 5.000 voces, todo ello sirve para que la Corporación profundice en sus propios principios de selección del léxico que debe co nform ar su Diccionario, principios que acaban apareciendo a m odo de justificación en el Prólogo de la séptima edi ción. Com o consecuencia de ello, adem ás, se am plía la lista de autoridades q ue venía utilizándose en la Academia. a publicar u n a obra q u e p or n in gún d erech o les c o r r e s p o n d e ”. J. Musso, sin em bargo, fue m u c h o más p ru d en te d e lo q u e se le in d icó y parece que fu e él m ism o q u ien p ro p u so n o nom brar d irectam ente a N ú ñ e z d e Taboada e n el prólogo, lo cual p areció b ien a F. A. G on zález y a toda la Academ ia (véase para t o d o ello M olina Martínez, 2001: 278-283). 25 Los d o s adjetivos resultan curiosos. El p rim ero p u e d e rem itir a la d en u n cia que N úñez d e Ta b o a d a incluía e n su Prefacio de los d efectos d e los D iccionarios: “p obreza ó escasez de voces” (N ú ñ e z d e Taboada, 1825: II). El adjetivo dim inuto había sid o e m p le a d o p o r Mayans (Lázaro Carreter, 1980 [1972]: 113). 4.1. El D ictionnaire français-espagnol et espagnol-français de Núñez de Taboada y la quinta edición del Diccionario de la Lengua Castellana de la Academia A partir de las noticias que contienen las actas de la Academia, se im p o n e u n a com paración del contenido del Dictionnaire français-espagnol et espagnol-français de Núñez de Taboada y el Suplemento de la edición del Dic cionario de la Academ ia correspondiente a 1817, puesto que, según reflejan las actas de la Academia, las voces seleccionadas en traro n en el Suplemento de la quin ta edición. El análisis perm itirá determ in ar en cierta m edida el aprovecham iento del Diccionario de N úñez de Taboada y de ello se despren derán los criterios que la Academia utilizó al acercarse a un diccionario bilingüe, elaborado siguiendo unas pautas de admisión del léxico distintas a las que había utilizado la Academia d u ran te el siglo x v m y principios del siglo X I X . El Suplemento de la quinta edición del Diccionario de la Academia ocupa cinco páginas (pp. 913-918) y contiene 307 artículos de los cuales un buen n ú m e ro (137 artículos) coincide con las entradas del Diccionario bilingüe de N úñez de Taboada26. Dado que este últim o m arca con un asterisco los artículos que no se encuentran en el Diccionario de la Academia, resulta lógico pensar que la selección se hizo sobre este tipo de palabras, pero a lo m ejor no fue así, pues las coincidencias entre el Suplemento y el Diccionario de N úñez de Taboada son mayores en los lemas desprovistos de asterisco (92 artículos) que en los lemas con este distintivo (45 artículos). Evidentem ente, no resulta posible dem ostrar de m anera inequívoca cuáles de las coincidencias se deben realm ente a u n a selección hecha a partir de este últim o, salvo en casos muy concretos. A pesar de ello, es se guro que en tre las coincidencias tienen que estar las palabras tomadas del Diccionario de N úñez de Taboada a las que se refieren las actas. De los artículos portadores de asterisco conviene destacar los lemas ari dez, acidular, el familiar achispar, adornista, agiotador, alegrete, amueblado, amue blar, astillón, badiana, belemnita, bismuto, cabotage, cachalote, calcar, capitalista, ceñiglo, civilización, civilizar, clasificar, cobalto, codeso, congregante, las variantes czar y czarina, chaflán, chaflanar, chivo-chiva, deshilarse, desmán, desnaturali zación, don diego de noche, esfínter, filoseda, gayuba, gaza, haraposo, heptacordo, hexacordo (con los com puestos sintagmáticos hexacordo mayory hexacordo me nor), hipogloso, mierra, pajarilla y prohombre. Entre ellas se cuenta la voz actriz, vocablo que aparece en el Diccionario académico con la anotación de ser “voz nuevam ente introducida”, y capitalista27, ambas con visos de m oderni dad; aparecen tam bién los térm inos civilización y civilizar, típicam ente die ciochescos (Alvarez de Miranda, 1992), que p o r fin entran a form ar parte 26 E m p leo para todas las com p rob a cion es en obras lexicográficas d e la Academ ia y e n N ú ñ ez d e T aboada (1825) el N T L L E . Las voces citadas aparecen co n a cen tu a ción m oderna. 27 Véase Battaner (1977) para datos d e la segun d a mitad d e siglo x i x . del diccionario oficial; hay, además, varios nom bres de animales y plantas (badiana, cachalote, ceñiglo, codeso, chivo, desmán, don diego de noche, gayuba, pajarilla), de m inerales ( bismuto, cobalto) y voces técnicas de la arquitectura de docum entación antigua en los textos ( chaflán, chaflanar8). Ambos dic cionarios com parten la voz clasificar que creo que resulta muy interesante p or cuanto ya se había introducido en la quinta edición del Diccionario en el lugar correspondiente con la definición “ordenar o disponer p or clases algunas cosas” y se vuelve a añadir al Suplemento con la definición “ord en ar p or clases” que parece reflejar la redacción de Núñez de Taboada “ranger par classes”. El grupo más num eroso de térm inos que com parten el Diccionario de Núñez de Taboada y el Suplemento de la quinta edición del Diccionario la Academia son las palabras que en el prim ero no llevan asterisco. Aparecen en este grupo térm inos de m edicina (accesión, ranina, talparia), de música ( adagio, laúd), de aritm ética {divisor), de náutica (macarrón, timonera, vara dero) , minerales (malaquita, piedra nefrítica) o voces m arcadas como irónicas ( buenaalhaja); no faltan plantas y animales ( bengala, calaguala, corona real, chamiza, choto, farolillos, francesilla, gálgulo, perca, pechicolorado, don pedro de noche, peje araña, peje diablo, peralejo, rabihorcado, sauzgatillo, torpedo, verderón, verderol, vilano, yerbo, zabila, zandía) y otras voces de difícil clasificación (afi cionado, atanquía, bufos, cautelarse, chilla, durillo, efémero, las variantes chufeta y chufleta, disparador, dominaciones, granza de la que se indica en el Dicciona rio de la Academia que “es voz tom ada del francés”, huelga, mejilla, petroso, plataforma, pudrir, remozar, retal, rimar, sacratísimo, subjuntivo, suversión y su vertir com o variantes de las form as en sub-). La existencia en el Suplemento de 1817 de voces com partidas con el Dic cionario de Núñez de Taboada que en este n o figuran con asterisco indica que la revisión que se realizó en aquel m om ento tuvo un alcance relati vam ente amplio y que no solo se tuvieron en cuenta las voces que no se encontraban en el Diccionario académico, sino también otras voces que no llevaban el distintivo fácilm ente identificable. La dep endencia entre estas dos obras es innegable en algunos casos. Por ejemplo, en el Diccionario de Núñez de Taboada aparece bajo la voz alhaja el com puesto buena alhaja que es adm itido en el Suplemento y que figura en él con la form a buenaal haja aunque o rdenado p o r la a-. En la selección de palabras sin asterisco se incluyen algunas acepciones a lemas que ya constaban en el Diccionario académico (adagio, cadena, cohechar (la tierra), corona (real), chilla, rimar, etc.) y tam bién com puestos sintagmáticos y locuciones adverbiales; además, al gunas de las palabras incluidas en el Suplemento corresponden a voces que ya se encontraban en el Diccionario académ ico, con lo que se repite la acep ción (accesión, aficionado, bufos) 29. 28 DCECH, S. V. C H A F L Á N . 29 Eso m ism o ocurre e n la ed ició n d e 1837, algo q u e d en u n cia Salvá (Alvarez d e Miranda, 2003: 105). Destacan en este g ru p o tres aspectos que creo que p u ed en probar la relación entre los dos diccionarios: se trata de algunos casos de u n a palabra ya existente en el Diccionario de la Academia a la que se añade en el Suple mento u na acepción que coincide con el Diccionario bilingüe español-francés; por ejemplo, en la palabra cadena se agrega la acepción ‘lo mismo que c u e r d a por el conjunto de galeotes’ que posiblem ente refleja ‘Galériens attachés avec la m êm e ch aîn e’ de Núñez de Taboada (1812). También en el verbo cohechar existe paralelismo entre la acepción ‘Alzar el barbecho o d ar la última vuelta á la tierra’ (DñAfc',1832) y ‘cohechar la tierra : d o n n er le prem ier ou le d e rn ie r labour à la te rre ’ (Núñez de Taboada, 1812). En el caso de verderol y verderón se establece u n a remisión entre ambas y curiosa m ente en el Diccionario de N úñez de Taboada verderoles, ‘sorte de coquilla g e’ y verderón es ‘verdier, oiseau’, al revés del Diccionario de la Academia. Conviene no olvidar los casos en los que los dos diccionarios compar ten un com puesto sintagmático que aparece en el Diccionario bilingüe de Núñez de Taboada bajo el adjetivo {fiel ejecutor, fiel ejecutoría) y que figuran en el Suplemento o rdenad os p o r el sustantivo. D entro de estos, estilo ó voz fa miliar se debió tom ar del artículo fam iliar del Diccionario bilingüe en el que figura la definición tanto de estilo fam iliar como de voz familiar. Com parten ambos diccionarios otras estructuras complejas com o hombre de expectación, jabón de piedra, señora de honor, torre del homenaje. Hay coincidencia, en tercer lugar, en un buen n ú m e ro de locuciones adverbiales: de bien á bien, á ciegas, en fragante; y en expresiones fraseológi cas: juntar diestra con diestra, hacer las entrañas á una criatura, hacer las entrañas á uno, guardar las espaldas, hablar por detrás ó por las espaldas, hacer espaldas, hacer estrados, hacer ver á uno las estrellas, sin estrépito y figura de juicio, exhalar el espíritu, nunca falta un roto para un descosido, traerlo escrito en la frente, hacer de garganta, extraer de la iglesia, meterse en libros de caballería, seguir los pasos á alguno, tener la sartén ó el cucharon por el mango. Estos ejemplos indican que más que las innovaciones interesaba com pletar el Diccionario desde el p un to de vista del patrim onio ya existente en la lengua, posiblem ente lo que Núñez de Taboada e n ten d ía p o r lengua vulgar, el caso de las locuciones adverbiales resulta m uy significativo. Los ejemplos precedentes dem uestran cómo se llevó a cabo la selección de entradas del Dictionnaire français-espagnol et espagnol-français. No se cen tró, com o podría pensarse en u n prim er m om ento, tanto en los neologis mos y palabras m o dernas com o en el aprovecham iento de otros artículos que denotan algunos de los intereses de ampliación de la nom enclatura académica a principios del siglo xix: los nom bres de animales y plantas, p o r un lado, y la fraseología, p o r otro. 4.2. El Diccionario de la Lengua Castellana de M. Núñez de Taboada y la séptima edición (1832) del Diccionario de la Lengua Castellana de la Real Academia Según las actas de la Academia, e n tre 1829 y 1830, vuelve a utilizar la C orporación la ob ra de Núñez de T aboada con el fin de ampliar la no m enclatura de su propio diccionario y esta vez se recurre a su Diccionario m onolingüe. Más difícil resulta establecer el tipo de selección que se lleva a cabo en este m om ento por cuanto las voces procedentes de esta fuente se integraron en el Diccionario en el lugar correspondiente según el orden alfabético y n o en el Suplemento como en la edición de 1817, donde resultan más fácilm ente identificables. N úñez de Taboada se jactaba, ya en el mismo título de su obra, de que esta contenía 5.000 palabras más que el Diccionario de la Academia (6a edi ción, 1822). La cifra debió producir efecto, pues aparece m encionada va rias veces en las actas y tam bién se refiere a ella Salvá (Alvarez de Miranda, 2003: 103), a pesar de la severa crítica que dedica en este Diccionario y de que su propia o b ra au m en ta con creces la de N úñez de Taboada (Azorín, 2003). En realidad, y a pesar del título, las voces que realm ente aparecen mar cadas con asterisco ascienden solo a 2.75030. Las palabras añadidas son de muy variado tipo: en tre ellas destacan p o r su nú m ero derivados que com pletan u n a familia léxica (acordonamiento, acrimonioso, acutangular, blanque ría, calzoncillero, ebanizar, murmureo, etc.); en tre los derivados figuran ad verbios en -mente ( agradablemente, emancipadamente, monacalmente, obrepticia mente, reducidamente, etc.), derivados en -ble ( agotable, aguantable, amortizable, calculable, conciliable, imborrable, inatacable, mezclable, etc.), bastantes deriva dos con el sufijo -ero (alfilerero, baratillero, gazpachero, municionero, pagugero, rastrillero, sermonero, etc.), diminutivos no lexicalizados ( caponcillo, cobanillo, encrespadillo, endechilla, matorralillo e tc.), superlativos (habladorísimo, limpidí simo, obscurísimo), y voces derivadas con los prefijos in- (inaplicable, inaptitud, inatacable, incicatrizable, etc.) y des- (desacoplar, desacuñador, desalfombrar, des almidonar, desalterado, etc.). Algunas de ellas son derivaciones muy forzadas que p ro b ablem ente no se debían apoyar en el uso (anagramatizador defini do com o ‘El que hace anagram as’, anagramatizar ‘H acer anagram as’). Se desdoblan algunas formas que son simples variaciones flexivas (ado ra, junquitos, líquidos) y tam bién verbos reflexivos ( afinarse, afligirse, ahorrar se, arrastrarse) con significado diferenciado. Se incluyen, además, algunas variantes form ales (ahoguido, amaca, amanillar, gngüescos, mormuración, mor murados piritas). Entre las palabras añadidas propias de u n área de especialidad deter 30 de 1822. H abría q u e restar d e esta cifra 119 palabras q u e, p e se a ten er asterisco, con stan en el Diccionario m inada destacan de form a clara las que pertenecen a m undo de la m ar (N áut. y Mar.): aculebrar, afelpar, agalibar, ahorcaperro, aljadrez, andamiada, bucosidad, caperoles, desabrigada, frescacho, gregalizar, etc., lo cual podría tener relación con algunas de las traducciones de Núñez de Taboada; tam bién existe u n notable nú m ero de voces relacionadas con la m edicina (con las marcas Med., Anat., o incluso sin marca): acracia, aductor, artrítico, confluente, discrasia, equinoftalmia, sublingual, tórax, etc.31; algunas, no muchas, tienen asignada la marca Bot. y perten ecen a la term inología descriptiva de esta ciencia (antera, arundináceas, arvense, liliáceo, monopétalo, polipétalo); hay tér m inos propios de la lingüística (anglicanismo, arcaísmo, aticismo, germanismo, neologismo) y vocablos relacionadas con el m undo de la política ( aristócra ta, aristodemocraáa, autocracia, autonomía, autónomo, oclocracia). Se recogen, además, algunas voces de uso poético (alecto, argentino, enteo, exarar) y al gunas más, con la m arca de familiar (acollonar, antana, antiparras, apandor garse, arlequinada, chapadanza, doctorismo, emprimerar, farcinador, paparrucha, e tc.). Se agregan tam bién voces de realidades que pertenecen a la A ntigüe dad (ágapa, agapetas, alcides, ancilias, crucható) y hay algún nom bre propio, curiosam ente Emanuel. U na de las vías más im portantes de ampliación de la lexicografía m od erna se en cu e n tra en las designaciones de animales y plantas que también se hallan bien representadas (amamelis, anhorca, bana no, céreo, dentaria, fava-crasa, etc.; alpaque, alucón, boyo, copetuda) y e n las que destaca u n a descripción muy am bigua (‘Especie de planta’, ‘Cierta p lanta’, ‘Ave n o c tu rn a ’). Para p o d er valorar adecuadam ente la labor lexicográfica de Núñez de Taboada, sería necesario realizar un análisis comparativo de las voces m arcadas con asterisco en las ediciones publicadas del Diccionario bilingüe hasta 1825 y el m onolingüe, pues parecen existir entre ellos diferencias significativas, también sería necesario estudiar la relación con el Diccionario de Capm any (1805). Se p o d ría pensar, a la luz de lo expuesto anteriorm ente, que parte de las entradas añadidas al Diccionario de la Academia p or Núñez de Taboada son m ateriales de “relleno”, pero quizá no es exactam ente así por cuanto es posible distinguir en las palabras portadoras de asterisco elem entos léxicos de procedencias diferentes. En p rim er lugar y tal com o ya señaló Baquero (1992), están las palabras que N úñez de Taboada debió tom ar de los dic cionarios académicos anteriores: voces que habían form ado parte de la n o m enclatura académica en el siglo x v m , ya fuese el Diccionario de Autoridades o las distintas ediciones de la A cadem ia del siglo x v m y aun del siglo x ix 32, pero q u e habían sido eliminadas en alguna edición de m anera que ya no constaban en la que sirvió de base a N úñez de Taboada (6a edición, 1822). El mism o autor lo señala en el Prefacio de su obra cuando aclara que las pretendidas cinco mil voces las “he sacado de su propio caudal” (Núñez de 31 P odrían estas rem ontarse al Suplem ento d e voces técnicas d e Capm any (1805). 32 D e s d e la primera ed ició n en un so lo to m o (1780) hasta la 5 a ed ición (1817). Taboada, 1825: II; Baquero Mesa, 1992) refiriéndose al caudal académico. Com o se refleja en el cuadro siguiente, 483 artículos de los marcados con asterisco se encuentran en el Diccionario de Autoridades. Además, tam bién debieron consultarse otras ediciones de los diccionarios de la Acade mia, pues 121 artículos de los m arcados con asterisco se hallan en u n a o varias de las ediciones del Diccionario de la Academia: Diccionario de Autoridades 483 artículos O tras ediciones del Diccionario d e la Academ ia hasta 1817 121 artículos TO TAL 604 artículos CUADRO 1 Por tanto, u n a parte de las voces agregadas (24%) se encontraba ya en algún diccionario académico anterior. La o tra gran fuente pudo ser el Dic cionario castellano con las voces de ciencias y artes, pues 551 artículos marcados con asterisco se encuentran en este Diccionario (20%), u n a obra que, por su configuración y contenido, debió resultar muy útil tanto para la lexico grafía bilingüe como para la m onolingüe. Llama la atención, den tro de las voces procedentes de las fuentes ante riores, que un pequeño g ru p o de ellas sea p o rtad o r de u na marca diacrò nica. Se trata de 60 artículos q ue llevan la abreviatura ant. y tres más que incluyen la m arca p. us. P ueden relacionarse estos artículos con su adver tencia del Diccionario español-francés en cuyo Préface se refería a la inclusión de “vieux mots castillans”. La mayoría de estas palabras son variantes anti guas sinónimas de otra form a existente tam bién en el Diccionario (hespedar, gómito, estaría, impusición, etc.) y coinciden con palabras que figuran en la 4a edición del Diccionario de la A cadem ia (1803), u na edición en la que se habían introducido bastantes de estos elem entos y que habían vuelto a desaparecer en la edición siguiente en la que se había hecho la revisión lim piando el Diccionario de este tipo d e form as antiguas. El resto de voces (1476 artículos, 54%) n o se en cuentra en ninguno de los diccionarios m encionados an terio rm en te, aquí pudo tener una partici pación im portante la obra de Capm any (1805) y el trasvase de materiales lexicológicos del Diccionario bilingüe al m onolingüe. La cuestión que interesa es qué selección se llevó a cabo al form ar las listas de palabras que se exam inaron en la Academ ia con el fin de elegir aquellas que se juzgaban adecuadas p a ra incluirlas en la siguiente edición del Diccionario (7a edición, 1832). Si en las actas solam ente aparecen las referencias poco precisas que se han expuesto anteriorm ente, el Diario y la correspondencia d e j . Musso dem u estran que este académico se encargó de form ar las listas de palabras pro ceden tes del Diccionario de N úñez de T aboada que la Academia consideró y que, p ara realizar el trabajo, se hizo un a relación con todas las palabras marcadas con asterisco (Molina Martí nez, 2001: 269, 277-278). J. Musso recoge en su Diario el resultado de todo ello: En una (lista) se hallan unas pocas (voces) que juzgué debían omitirse de ellas las que llevan A fueron realmente admitidas por la Academia, las de R reprobadas y las que una cruz están ya en la sexta edición, aunque el autor las da por nuevas. En otra hay una larguísima retahila de términos que Núñez de Taboada prohija a la lengua castellana y que ninguno de la Academia conoce, ni probablemente ningún español (Molina Martínez, 2001: 277). La selección realizada tiene implicaciones im portantes p ara la historia de la lexicografía p o r cuanto puede ayudar a d eterm in ar los criterios de am pliación de la n o m enclatura académica en la prim era m itad del siglo x ix , una época en la que la Academia aplicó un criterio muy restrictivo en la ampliación. El contraste entre la edición de 1832 y las voces marcadas con asterisco en el Diccionario de Núñez de Taboada indica que el núm ero de vocablos tom ados de esta última obra debió ser, tal com o se señala en las actas, muy pequeño, pues parece que solo se eligen 80 palabras para ser introducidas en la séptim a edición del Diccionario de la Academia. Ello da u n a idea de la selección tan m edida que se realizó. Por todo lo expuesto hasta este m om ento, no resulta difícil com prender que casi la m itad de las voces seleccionadas (40 artículos) ya se encontrara en uno de los diccionarios que sirvió com o fuente a N úñez de Taboada (12 artículos se hallan en el Diccionario de Autoridades, otros dos aparecen en u na edición académ ica an terio r a 182234 y 25 voces figuran en el Diccionario castellano con las voces de ciencias y artes de E. de T erreros35); son palabras, p o r tanto, que de u n a fo rm a u otra ya existían en el siglo x v m . El resto (42 artículos36) son vocablos que recoge p o r prim era vez N úñez de Taboada en un diccionario m onolingüe y que la Academia debió elegir porque eran voces “autorizadas” según se expone en el Prólogo de la séptim a edición. Las ediciones siguientes del Diccionario académico, d u ran te todo el si glo x ix y aun el siglo x x , no dejan de adm itir voces que ya figuraban en el Diccionario de N úñez de Taboada. Otras, sin em bargo, nunca llegan a for m ar parte de la nom enclatura académica y unas pocas (160 artículos) no aparecen en ningún otro diccionario, sea académico o no académico. 33 Apoteosis, candente, carbonizar, cebador, chunga, harmónico, harmonioso, hipotenusa, loto, melodrama, prolifico, tendedor. 34 M alaquita, rabilo. 35 A ntediluviano, apaleo, cedilla, centuplicar, cochinada, cochinería, compilador, criticador, cronómetro, de magogo, desempaquetar, diabla, escarlatina, espato, icario, iniciar, litógrafo, manes, odiosamente, pachorrudo, pajaril, redicho, retroactivo, sestercio, tomista. 36 Alinear, anisar, ardoroso, aristócrata, atascamiento, canelo, canfórico, caníbal, canonical, carbónico, carbo no, chapurrar, chiripa, chiripear, chiripero, conclusivo, contrafallar, corola, criticón, cuévano, defensiva, demago gia, despreocupar, dietético, diplom ada, entusiasmar, fantasm agoiía, hebraizante, intuito, megano, mineralógico, mucilaginoso, nítrico, opresivo, pacotilla, paparrucha, pelosilla, pirueta, pizarreño, sacrista, suavecito, zoología. 5. C o n c l u s ió n Los diccionarios de M. Núñez de Taboada constituyen un eslabón im portante en la historia de la lexicografía española y en la historia de la recepción del léxico en los diccionarios. Por ser autor de un diccionario bilingüe y otro m onolingüe, garantiza la existencia de un im portante tras vase entre los dos tipos de lexicografía. Quizá los materiales léxicos reun i dos de su propia cosecha no respondieran a criterios metodológicos bien establecidos y fueran, en buena m edida, lo que Salvá denom inaba “reta zos zurcidos”, pero, seguram ente p o r m últiples motivos (5000 palabras, la fama de su diccionario bilingüe, las palabras del prólogo dirigidas direc tamente a la Academia) su obra fue lo suficientem ente im portante com o para que la A cadem ia le prestase atención y se sirviese de sus obras com o fuente. Del aprovecham iento de los diccionarios de N úñez de Taboada p o r parte de la Corporación, se desprende la concepción del Diccionario vulgar como m odelo de lengua de la prim era m itad del siglo xix: fuertem ente an clada en los autores anteriores al siglo x ix , en el léxico patrimonial y en las estructuras fraseológicas, y fundam entalm en te circunspecta en la admisión de voces nuevas y m odernas. R e f e r e n c ia s b ib l io g r á f ic a s A b a d M e r i n o , M e r c e d e s (2006): “Jo sé Musso Valiente, lexicógrafo dialectal: contribución al vocabulario m urciano del riego tradicional”, en A . Roldán, R. Escavy, E. H ern án d ez, J. M. H ern án d ez y M- I. López (eds.), Caminos actuales de la historiografía lingüística. Actas del V Congreso Internacional de la Sociedad Española de Lingüística, M urcia, 7-11 de noviembre de 2005, Universidad de Murcia, I, 197-208. A l v a r E z q u e r r a , M a n u e l (2002 [1995]): “Los diccionarios del español en su historia”, International Journal o f Lexicography, 8, 3, 1995, 173-201. Reeditado en De antiguos y nuevos diccionarios del español, M adrid: A rco/L ibros, 2002, 15-50. Á l v a r e z d e M i r a n d a , P e d r o ( 1 9 9 2 ) : Palabras e ideas: el léxico de la Ilustración temprana en España (1680-1760), Madrid: Real Academia Española. — (2003): “Vicente Salvá, editor y corrector del Diccionario de la Academia (1838 y 1841) ”, en Lexicografía y lexicología en Europa y América. Homenaje a Günther Haensch en su 80 aniversario, Madrid: Gredos, 99-114. A n d i o c , R e n é (1963): “L eandro F ernández de Moratín: H óte de la F rance”, en Revue de Littérature Comparée, 37, 2, 268-278. A n c l a d a B o i x , E m i l i a (1997-1998): “T raducción y diccionario. Algunos n eo logismos de la quím ica en el nuevo diccionario francés-español (1805) de A. Capm any”, Revista de Lexicografía, IV, 31-47. A z o r í n F e r n á n d e z , D o l o r e s ( 2 0 0 0 ) : Los diccionarios del español en su perspecti va histórica, Alicante: Universidad. — (2003): “U n proyecto original en la lexicografía española del siglo xix: El Nuevo diccionario de la lengua castellana (1846) de Vicente Salvá”, en Lexicografía y lexicología en Europa y América. Homenaje a Günther Haensch en su 80 aniversario, Ma drid: Gredos, 115-131. B a q u e r o M e s a , R o s a r i o (1992): “Notas en contribución a la historia de la lexicografía española m onolingue del siglo x i x ”, en EURALEX’90 Proceedings. Actas del IV Congreso Internacional, Barcelona: Biblograf, 455-461. B a r b a s t r o G i l , L u i s ( 1993) : Los afrancesados.Primera emigración política del siglos X IX español (1813-1820), Madrid: csic-Instituto de Cultura ‘Juan Gil-Albert”. B a t t a n e r A r i a s , M ' P a z (1977): Vocabulario político-social en España (1868-1873), Madrid: Real Academia Española. B r u ñ a C u e v a s , M a n u e l (1996): “L’universalité du français dans les diction naires bilingues français-espagnol (1648-1815)”, en J. F. García Bascuñana, B. Lep in e tte y C . Roig (eds.), L '“universalité”du français et sa présence dans la Péninsule Ibéique. Actes du colloque de la SIH FLES. Tarragone septembre 1995, Paris: s i h f l e s , 51-61. — (1999): “Las mejoras aportadas a la traducción p o r el Diccionario d e Capm any (1805)”, en F. Lafarga (éd.), La traducción en España (1750-1830). Lengua, literatura, cultura, Universitat de LLeida, 99-110. — (2001): “L’universalité de la langue française dans les gram m aires d e fran çais p o u r les espagnols et dans les dictionnaires bilingues antérieurs à 1815”, en E.F.K. K oerner y H.-J. N iederehe, History o f linguistics in Spain/H istoria de la lingüís tica en España, vol. II, A m sterdam /P hiladelphia: Jo h n Benjamins, 229-262. B u e n o M o r a l e s , A n a M a r í a (1995): L a lexicografía monolingüe no académica del Siglo X IX , tesis de doctorado, dirigida p o r el Dr. M. Alvar Ezquerra, Universidad de Málaga. C a p m a n y , A n t o n i o d e (1805): Nuevo diccionario francés-español, en que va n en mendados, corregidos, mejorados, y enriquecidos considerablemente los de Gattel, y Cormon, Madrid: Im prenta de Sancha. C a z o r l a V i v a s , M a r í a d e l C a r m e n ( 2 0 0 2 ) : Lexicografía bilingüe con el español y el francés de los siglos X V III y X IX , Tesis d e doctorado, dirigida p o r el Dr. M . Alvar Ezquerra, Universidad C om plutense. C h e c a B e l t r á n , J o s é (1989): “Elogio d e la lengua española en C apm any”, Revista de Filología Española, 69, 131-151. D CECH : Corominas, J., y j . A. Pascual (1980-1990): Diccionario crítico etimológico castellano e hispánico, 6 vols., Madrid: Gredos. D R A E 1817: Real A cademia Española, Diccionario de la Lengua Castellana, Q uinta edición, Madrid: Im p ren ta Real. D R A E 1832: Real Academia Española, Diccionario de la Lengua Castellana, Sépti m a edición, Madrid: Im p ren ta Real. E s p a r z a , M i g u e l A n g e l (1999): “La lexicografía m onolingüe española del siglo xix: un conflicto de paradigm as”, Rom anistik in Geschichte u n d Gegenwart, §5.1., 49. 65. F a b b r i , M a u r i z i o (1979): A Bibliography o f H ispanic Dictionaries. Catalan, Galician, Spanish, Spanish in L atin America and the Philipines, Imola: Galeati. F e r n á n d e z d e N a v a r r e t e , M a r t í n (1968 [1851] ): Biblioteca m antím a española, 2 vols., reim presión, New York: B urt Franklin. F e r n á n d e z D í a z , M a r í a d e l C a r m e n (1987): “La contribución de A. de Ca pm any a la cración del vocabulario científico-técnico castellano”, Verba, 14, 527534. — (1989): “A ntonio de Capm any y el pro b lem a de la traducción y aprendizaje del francés en la España del siglo x v m ”, en J. C. Santoyo et al. (eds.), Fides interpres. Actas de las Ijornadas nacionales de historia de la traducción, Universidad de León, II, 272-277. F e r n á n d e z S a r a s o l a , I g n a c i o (2000): “La Constitución española de 1812 y su proyección eu ro p ea e ib ero am erican a”, Fundamentos: Cuadernos monográficos de teoría del estado, derecho público e historia constitucional, 2, 359-457. Tam bién en http://w w w .cervantesvirtual.com . G a r c í a B a s c u Ña n a , J u a n F. (1996): “C ontribución al estudio de los dicciona rios bilingües francés-español, español-francés: aproxim ación histórica y m eto d o lógica”, en E. Forgas (coord.), Léxico y diccionarios, Tarragona: Universitat Rovira i Virgili-Departam ent de Filologies R om àniques, 91-103. — (1999): “De Gattel y B. C o rm o n a C apm any y Núñez de Taboada. En to rn o a ciertos aspectos y procedim ientos d e la lexicografía bilingüe francés-español en tre 1790 y 1812”, en F. Lafarga (ed.), La traducción en España (1750-1830). Lengua, literatura, cultura, Universitat de LLeida, 111-120. G a r c í a P l a t e r o , J u a n M a n u e l (2003): “La lexicografía no académica en los siglos x v iii y x i x ”, en A. M .3 M edina G u e rra (coord.), Lexicografía española, Barce lona: Ariel Lingüística, 263-306. G e m m i n g e n , B a r b a r a v o n (2001): “Estudios sobre el Diccionario e spañol-francés, francés-español del abate M arie-Claude Gattel (1790)”, en A. M3 M edina G ue rra (coord.), Estudios de lexicografia diacrònica del español, Universidad de Málaga, 211-242. G i l N o v a l e s , A i .b e r t o (1991): Diccionario biográfico del trienio liberal, Madrid: El Museo Universal. L á z a r o C a r r e t e r , F e r n a n d o (1985 [1949]): Las ideas lingüísticas en España durante el siglo XVIII, Barcelona: Editorial Crítica. — (1980 [1972]): “El prim er diccionario de la Academia”, discurso de ingreso leído con el título de Crónica del Diccionario de Autoridades (1713-1740), el 11 de ju n io de 1972; Madrid: Real Academia Española. R eproducido en Estudios de Lingüística, Barcelona: Crítica, 83-148. — (2002): “El neologismo en el d r a e ”, 15 de febrero de 2002, en http://w w w . ra e .e s/rae/gestores/gespub000001. n sf/ (voAnexos) L l o r e n s , V i c e n t e ( 1 9 6 8 2): Liberales y románticos. Una emigración española en In glaterra (1823-1834), Madrid: Castalia. M a r t í n e z A r n a l d o s , M a n u e l , y J o s é L u i s M o l i n a M a r t í n e z (2002): L a transición socio-literaria del Neoclasicismo al Romanticismo en el Diario (1827-1838) de José M usso Valiente, M u r c i a : Nostrum. M o l i n a M a r t í n e z , J o s é L u i s (2001): ‘José Musso y Valiente en la Real Academia Española según su Diario (1829-1837). Su intervención en el Diccionario y en la Gra mática de la lengua castellana!', Boletín de la Real Academia Española, l x x x i , cuaderno c c l x x x i i i , 255-320. M o r a n g e , C l a u d e (1990): Siete calas en la crisis del antiguo régimen español y u n panfleto clandestino de 1800, Alicante: Instituto de C ultura “Ju an Gil-Albert”-Diputa- ción de Alicante. — (2002): Paleobiografía (1779-1819) del «pobrecito holgazán» Sebastián de M iñano y Bedoya, Salamanca: Universidad de Salam anca. M u ñ o z d e l M a n z a n o , C i p r i a n o , [ C o n d e d e l a V iñ a z a ] (1978 [1893]): Biblio teca histórica de la filología castellana, M a d r i d : M a n u e l T e llo , voi. i l i , 721-1018. E d . f a c s ím il M a d r i d : A tla s. N i e d e r e h e , H a n s -J o s e f (1987): “Les dictionaires franco-espagnols ju s q u ’en 1800”, Histoire Epistemologie Langage, ix-2, 13-26. N T L L E : R e a l A c a d e m i a E s p a ñ o l a (2000) : Nuevo Tesoro Lexicográfico de la Lengua Española, Madrid: Espasa, edición en d v d . N ú ñ e z d e A r e n a s , M a n u e l (1963) : L ’E spagne des Lumières au Romantisme, París: C entre d e R echerches de l’Institut d ’Etudes Hispaniques. N ú ñ e z d e T a b o a d a , M a n u e l (1812): Dictionnaire français-espagnol et espagnolfrançais, plus complet et p lu s correct que tous ceux qui ont été publiés ju sq u ’a ce jour, y compris celui de Capmany p a r Nuñez de Taboada, 2 tomos, Paris: Brunot-Labbe, T. Barrois. — (1825): Diccionario de la Lengua Castellana, Paris: Librería de Seguin. P a l a u y D u l c e t A n t o n i o ( 1 9 4 8 ) : M a n u a l del librero hispano-americano, Barcelo na: Librería anticuaría d e A. Palau. Q u e m a d a , B e r n a r d (1968): Les dictionnaires du français moderne, 1539-1863. Etude sur leur histoire, leur types et leurs méthodes, Paris: Didier. R o d r í g u e z F e r n á n d e z , A n a Ma (2006): “El Diccionario castellano con las voces de las ciencias y artes de E. Terreros y el Nuevo Diccionario francés-español de A. de Capmany: aproxim ación al estudio de sus influencias”, en J. J. de Bustos Tovar y J. L. G irón A lconchel, Actas del VI Congreso Internacional de Historia la Lengua Es pañola, M adrid, 2 9 de septiembre-3 de octubre de 2003, Madrid: A rco/L ibros, vol. II, 1639-1652. R o i g , C a r m e n (1995): “El Nuevo diccionario francés-español de A ntonio de Cap m any”, en F. Lafarga, A. Ribas y M. Tricás (eds.), La traducción: metodología/historia/ literatura, ámbito hispanofrancés, Barcelona: p p u , 75-80. S alv á, V i c e n t e (1846): Nuevo diccionario de la lengua castellana, que comprende la últim a edición íntegra, muy rectificada y mejorada del publicado por la Academia Española, y unas veinte y seis m il voces, acepciones, frases y locuciones, entre ellas muchas americanas [...], París: V icente Salvá. S á n c h e z P é r e z , A q u i l i n o ( 1 9 9 3 ) : Historia de la enseñanza del español como len gua extranjera, Madrid: s g e l . S e c o , M a n u e l (1987): “El nacim iento de la lexicografía m o d e rn a no académi ca”, en Estudios de lexicografía española, Madrid: Paraninfo, 129-151. Ve r d o n k , R o b e r t A. (1991): “La lexicographie bilingue espagnol-français, français-espagnol”, en F. J. H ausm ann et a l, Wörterbücher, Dictionaries, Dictionaires: Encyclopédie internationale de lexicografie, Berlin-New York: Walter de Gruyter, 3, 29762987. V i l l a n u e v a , J o a q u í n L o r e n z o (1826): “Análisis del Diccionario cle la lengua castellana publicado p o r D. M. Taboada en Paris, año de 1825”, en Ocios de españo les emigrados, tom o V, n ú m ero 25, abril 1826, 347-379. Z a m o r a V i c e n t e , A. (1999): Historia de la Real Academia Española, Madrid: Es pasa.
https://openalex.org/W4389111054	https://jurnal.uniraya.ac.id/index.php/Relation/article/download/396/328	English	null	AN ANALYSIS OF CONNOTATIVE MEANING IN LINKIN PARK “A THOUSAND SUNS ALBUM”	Research on English Language Education	2,022	cc-by	4,907	o o E i i Ap i y AN ANALYSIS OF CONNOTATIVE MEANING IN LINKIN PARK “A THOUSAND SUNS ALBUM” Heldestina Laia(¹), Sitasi Zagoto(²), Senadaman Wau(³) (1)Mahasiswa Prodi Pendidikan Bahasa Inggris FKIP Universitas Nias Raya (2),(3)Dosen Universitas Nias Raya (heldestinalaia123@gmail.com(¹), sitasizagoto669@gmail.com(²), swauuniraya@gmail.com(³)) Heldestina Laia(¹), Sitasi Zagoto(²), Senadaman Wau(³) (1)Mahasiswa Prodi Pendidikan Bahasa Inggris FKIP Universitas Nias Raya (2),(3)Dosen Universitas Nias Raya (heldestinalaia123@gmail.com(¹), sitasizagoto669@gmail.com(²), swauuniraya@gmail.com(³)) Abstract This research aimed to analyze the connotative meaning and the type of connotative meaning found in the twelve lyrics of linkin parks’s songs. This research is designed in qualitative research with descriptive approach. The source of this research is the twelve lyrics of linkin park’s songs. The data is analyzed by using the technique of the data analysis: data condensation, data display, drawing and verifying conclusions. The result of data analysis shows that: there are 30 connotative meaning used in the twelve lyrics and the dominant types of connotative meaning is associative meaning. Besides, it is found 16 associative meaning, 4 allusive meaning, 3 collocative meaning, 7 affective meaning, 0 reflective meaning and attitudional meaning. Based on the finding, it is concluded that the connotative meaning found is carried within words, phrase and sentences in twelve lyrics of linkin parks’s song. In this research, the researcher also wants to suggest the lecturer or the English teacher to lead the students’ use the connotative meaning theory as the material of teaching and learning and also can be implemented as material for increasing the students’ ability in mastering the vocabullary and also connotative meaning material can be propossed to extend students ability in recognizing the meaning of words not only which has provided in dictionary but also which has associated in another aspect. Key words: Connotative meaning; song lyrics; linkin park album Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 https://jurnal.uniraya.ac.id/index.php/Relation A. Introduction In communication, people interact one with the others for the sake of understanding the message or ideas presented using language. Kreidler (1998:19) stated that “language is a system of symbols through which people communicate. There are main parts of linguistics, they are phonetic and phonology, morphology, syntax, pragmatics and semantics. In this case, the researcher foccussed on meaning of words and sentences. However, semantic is also dealing with relation human words and word. Furthermore, in linguistics, semantics is the study of the relationship between words and how the process of drawing meaning from text. Yule (1985:320) stated that “semantic is the study of meaning in words, phrase and sentences”. In semantics, meaning is divided into two parts, literal meaning and non-literal meaning. Literal meaning as a dennotative meaning based on actual words or the real meanings and non-literal meaning is the other way around its accurs when the speaker’s meaning different from what the real meaning, we can say that as connotative meaning. The listener may find it difficult to understand what the speaker wants to convey if the speaker speaks non literally. To expressing that ideas from semantics are used some way through denotative and connotative meaning. while, meaning includes both denotation and connotation. Denotative meaning is what the dictionary attempts to provide. It means that denotative meaning is the objective (dictionary) relationship between a word and the reality to which it refers. While connotative meaning is the communicative power of a word by virtue what it refers to (Leech, 1981:12). Every song has many messages and need to analyze the meaning of the song to understand the song. In daily life, many people listen a lot of song. Song lyric is one of literary works and it is has meaning. Every song contains lyric which has certain connotative meaning to be heard. Song lyrics are often used by the creator as an intermediary to convey and to expressing feeling that expose the author’s toughts are made with various types of language style. The one of it Linkin Park. Furthermore, in linkin park a thousand suns album the researcher found some connotative meaning, the connotative meaning that found is negative emotion such as angry, disapointment, selfish creature, and arrogant. Connotative that found produce as follow: (1) Blood in your eyes. Abstrak Literal meaning as a dennotative meaning based on actual words or the real meanings and non-literal meaning is the other way around Connotative meaning is difficult to interpret for general listener. Based on data that researcher collected that some people did not understand about connotative meaning that provides when listened. Some people does not get the point and the meaning what the speaker say, the listener just enjoy to hearing the lyric songs without understanding and knowing the connotative meaning that consist on it. Related to the statement and explanations above, learning about connotative meaning is very important. While, it can help make it easier to find and understand about what is being conveyed or being told. One of example to find the connotative meaning used in a lyrics song. Every song has many messages and need to analyze the meaning of the song to understand the song. In daily life, many people listen a lot of song. Song lyric is one of literary works and it is has meaning. Every song contains lyric which has certain connotative meaning to be heard. Song lyrics are often used by the creator as an intermediary to convey and to expressing P-ISSN: 2715-1638 E-ISSN: 2828-5522 Universitas Nias Raya y Connotative meaning is difficult to interpret for general listener. Based on data that researcher collected that some people did not understand about connotative meaning that provides when listened. Some people does not get the point and the meaning what the speaker say, the listener just enjoy to hearing the lyric songs without understanding and knowing the connotative meaning that consist on it. Related to the statement and explanations above, learning about connotative meaning is very important. While, it can help make it easier to find and understand about what is being conveyed or being told. One of example to find the connotative meaning used in a lyrics song. Abstrak Penelitian ini bertujuan untuk menganalisis makna konotatif dan jenis makna konotatif yang terdapat pada 12 lirik lagu linkin park. Penelitian ini di rancang dalam penelitian kualitatif dengan pendekatan deskriptif. Sumber penelitian ini adalah 12 lirik lagu linkin park. Data di analisis dengan menggunakan teknik analisis data: kondensasi data, penyajian data, penarikan dan ferifikasi kesimpulan. Hasil analisis data menunjukan bahwa ada 30 makna konotatif yang di gunakan dalam 12 lirik lagu dan jenis makna konotatif yang dominan adalah makna asosiatif, 16 makna asosiatif, 4 makna kiasan, 3 makna kolokatif, 7 makna afektif, 0 makna refleksi dan makna sikap. Berdasarkan temuan tersebut, di simpulkan bahwa makna konotatif yang di temukan terkandung dalam kata, frasa dan kalimat dalam 12 lirik lagu linkin park. Dalam penelitian ini, peneliti juga ingin menyarankan dosen atau guru bahasa ingris untuk mengarah kan siswa menggunakan teori makna konotatif sebagai bahan belajar mengajar dan juga dapat di implementasikan sebagai bahan untuk meningkatkan kemampuan siswa dalam menguasai kosa kata. Dan juga materi makna konotatif dapat di usulkan untuk menigkatkan kemampuan siswa dalam mengenali arti kata kata tidak hanya yang telah di sediakan dalam kamus tetapi juga yang terkait dalam aspek lain. Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya Kata kunci: Makna konotatif; lirik lagu; album linkin park Kata kunci: Makna konotatif; lirik lagu; album linkin park https://jurnal.uniraya.ac.id/index.php/Relation Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya A. Introduction In communication, people interact one with the others for the sake of understanding the message or ideas presented using language. Kreidler (1998:19) stated that “language is a system of symbols through which people communicate. There are main parts of linguistics, they are phonetic and phonology, morphology, syntax, pragmatics and semantics. In this case, the researcher foccussed on meaning of words and sentences. However, semantic is also dealing with relation human words and word. Furthermore, in linguistics, semantics is the study of the relationship between words and how the process of drawing meaning from text. Yule (1985:320) stated that “semantic is the study of meaning in words, phrase and sentences”. In semantics, meaning is divided into two parts, literal meaning and non-literal meaning. A. Introduction The lyrics was taken from the song “Blackout”, this lyric used the types of ‘Associative meaning’ because associative meaning has relationship and connects it with refferences to stereotype. “blood in your eyes” that is not the real meaning. While, the word ‘blood’ is expression of someone who is angry, disappointed and https://jurnal.uniraya.ac.id/index.php/Relation P-ISSN: 2715-1638 E-ISSN: 2828-5522 Research on English Language Education (Relation J Vol. 4 No. 1 Edisi April 2022 emotion that looked from the eyes. So, the word blood is associated with anexpression angry, disappointed and emotion. . Then, based on the explanation above, the writer would like do a researcher and intends to study about entitled “An analysis of connotative meaning in Linkin Park a thousand suns album”.The objectives of this research are : (1)To describe the types of connotative meaning which are found in Linkin Park “a Thousand Suns Album” (2) To find out the meaning of connotative words, phrase and sentences that found in Linkin Park “a Thousand Suns song lyrics Album” Universitas Nias Raya This study used qualitative methods with descriptive approach because the data is the form of words or written language. This is due to research it seeks to expose sentences, phrase or words. Qualitative research is a form of inquiry that explores phenomena in their natural settings and uses multi-methods to interpret, understand, explain, and bring meaning to them (Anderson and Arsenault, 2005:126). In addition, another expert was argues “qualitative research is practised in many diciplines, so a range of data collection methods hasbeen devised to cater for the varied requirements of the different subjects, such as : qualitative interviewing, focus group, participant, discourse and conversation analysis and analysis of text ad documents” (Walliman, 2011:131).The data of this research was taken from song lyrics Linkin Park’s album. The data source from words, phrase or sentences of songs linkin park by album a thousand sun that contains connotative meaning. This research expected have significance both theoretically and practically. Theoretically, this research will be expected to give some contribution in linguistics development especially for semantics study and theoretical aspect of connotative meaning. Connotative meaning material can be proposed to extend students ability in recognizing the meaning of words not only which has provided in dictionary but also which has associated in another aspect. A. Introduction Morever, it is possible for the teacher to motivate the students for having motivation in learning English by comprehending the connotative meaning attractively. Practically, the teachers also can use the connotative meaning theory as the material of teaching and learning and also can be implemented as material for increasing the students’ ability in mastering the vocabulary. This research also can be useful as a reference and additional information if you are interested in analyzing similiar research title like An analysis Linkin park a thousand suns album as comparison for the relevant research. In collecting the data in this research, there were the steps the researcher used. The steps were : (1) Searching and downloading the lyrics of Linkin Park’s album “A Thousand Suns and printing out all the song lyrics that consist in link http://myliriklagupopuler.blogspot.com/201 3/03/linkin-park-thousand-suns-aalbum- lyric.html(2) Printing out all the song lyrics that consist “A Thousand Suns” album. (3) Reading the lyrics of Linkin Park’s album A Thousand suns to know whether it contains connotative meaning. (4) Identifying and Selecting the word, phrase or sentences that contains connotative meaning that would be analyzed. (5) Writing and arranging down the data that into several parts based on type. lyric.html(2) Printing out all the song lyrics that consist “A Thousand Suns” album. (3) Reading the lyrics of Linkin Park’s album A Thousand suns to know whether it contains connotative meaning. (4) Identifying and Selecting the word, phrase or sentences that contains connotative meaning that would be analyzed. (5) Writing and arranging down the data that into several parts based on type. https://jurnal.uniraya.ac.id/index.php/Relation B. Research Method 1 Edisi April 2022 Condensation. Refers to the process of selecting, focusing, simplifying abstracting, and/or transforming tha data that appear in the full corpus (body) of written-up field notes, interview transcripts, document and other empirical material. (2) Data Display a display is an orgnized compressed assembly of information that allows conclusion drawing and action. In this case, researcher displayed the data in a table form to make it easy for researcher in drawing conclusion. (3) Drawing and Verifying Conclusion. At this stage, the data has been analyzed recheck to develop the conclusion regarding the study and describe the result of analyzing the data from the data collection. https://jurnal.uniraya.ac.id/index.php/Relation B. Research Method Data analysis is seen as consist of three activities in analyzing data: (1) Data https://jurnal.uniraya.ac.id/index.php/Relation https://jurnal.uniraya.ac.id/index.php/Relation Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya Condensation. Refers to the process of selecting, focusing, simplifying abstracting, and/or transforming tha data that appear in the full corpus (body) of written-up field notes, interview transcripts, document and other empirical material. (2) Data Display a display is an orgnized compressed assembly of information that allows conclusion drawing and action. In this case, researcher displayed the data in a table form to make it easy for researcher in drawing conclusion. (3) Drawing and Verifying Conclusion. At this stage, the data has been analyzed recheck to develop the conclusion regarding the study and describe the result of analyzing the data from the data collection. C. Research Finding and Discussion The data of this research were song lyrics of Linkin Parks’s “a thousand suns) album, which consist of (12) twelve songs, which have 30 connotative meaning. Table 1 Types of connotative meaning N o Song Title Types of Connotative Meaning Tot al A s m R f m A t m A l m C l m Af m end 6 Blackout 2 2 7 Wretches and kings 1 3 4 8 Wisdom justice and love 1 1 2 9 Irisdescen t 2 - 1 3 1 0 The catalyst 1 1 1 1 4 1 1 The messenge r 1 1 2 1 2 Jornal del muerto - - - - - - - Total 30 Source, researcher result, (2022) Note: As m :Asociative meaning Al m :Allusive meaning Rf m :Reflected meaning Cl m :Collocative meaning At m :Attitudional meaning Af m :Affective meaning The types of connotative meaning a. Associative meaning is that part of the overal meaning of an expression that consists of expectation or ournal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Universitas Nias Raya end 6 Blackout 2 2 7 Wretches and kings 1 3 4 8 Wisdom justice and love 1 1 2 9 Irisdescen t 2 - 1 3 1 0 The catalyst 1 1 1 1 4 1 1 The messenge r 1 1 2 1 2 Jornal del muerto - - - - - - - Total 30 Source researcher result (2022) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Research on English Language Education (Relation J Vol. 4 No. C. Research Finding and Discussion 1 Edisi April 2022 isangry,disappointed and emotion that looked from the eyes. So the word ‘blood’ associated with ‘an expression angry, dissapointed and emotion Universitas Nias Raya describe one’s felling, where blac symbolis strength and seriousness while bluesymbolizes trust and loyalty. So, it means people who are full of strength, seriousness, trust and loyality, run and fight in the heat or the hot sun. So, black and blue collocative with callors. describe one’s felling, where blac symbolis strength and seriousness while bluesymbolizes trust and loyalty. So, it means people who are full of strength, seriousness, trust and loyality, run and fight in the heat or the hot sun. So, black and blue collocative with callors. describe one’s felling, where blac symbolis strength and seriousness while bluesymbolizes trust and loyalty. So, it means people who are full of strength, seriousness, trust and loyality, run and fight in the heat or the hot sun. So, black and blue collocative with callors. b. allusive meaning is given linguistics expression, allusive meaning created when an expression evokes a saying or quotation in such a way, so that the meaning of saying or quotation became part of the overal meaning. b. allusive meaning is given linguistics expression, allusive meaning created when an expression evokes a saying or quotation in such a way, so that the meaning of saying or quotation became part of the overal meaning. d. d. Affective meaning is particular linguistic expression rather than others that might have been used to express the same literalmessage. Affective meaning that is an emotive effect that is worked on the reader or the person spoken by the choice of expression, and tha forms parts of its overall meaning. The expression does not merely denote is to referent but also hints at some attitude of the speaker or writer to the addressee.  Will we burn in the fires of a thousand suns (St.the catalyst, P.74, L.20-21) This sentences is type of allusive meaning because in the real meaning doen’t the fires of a thousand suns, but this illustrates the result of sin is a hell of fire full of misery in the afterlife. Its the allusive meaning because the saying or quotation became part of the overalmeaning. https://jurnal.uniraya.ac.id/index.php/Relation C. Research Finding and Discussion The data of this research were song lyrics of Linkin Parks’s “a thousand suns) album, which consist of (12) twelve songs, which have 30 connotative meaning. Source, researcher result, (2022) Note: As m :Asociative meaning Al m :Allusive meaning Rf m :Reflected meaning Cl m :Collocative meaning At m :Attitudional meaning Af m :Affective meaning w ic ave 30 co otative ea i g. Table 1 Types of connotative meaning N o Song Title Types of Connotative Meaning Tot al A s m R f m A t m A l m C l m Af m 1 The radiance 1 1 2 Burning in the skies 5 1 6 3 When they come for me 3 3 4 Robot boy - - - - - - 5 Waiting for the 1 1 1 3 Th g Table 1 Types of connotative meaning N o Song Title Types of Connotative Meaning Tot al A s m R f m A t m A l m C l m Af m 1 The radiance 1 1 2 Burning in the skies 5 1 6 3 When they come for me 3 3 4 Robot boy - - - - - - 5 Waiting for the 1 1 1 3 Th Table 1 Table 1 Types of connotative meaning The types of connotative meaning a. Associative meaning is that part of the overal meaning of an expression that consists of expectation or assumptions widely associated (right ornot) with referent of the expression. Associative meaning is the meaning that a word has in relationship to circumstances outside the language and connects it with refferences to stereotype.  Blood in your eyes (St.blackout, P.67, L.15) This sentences is type of asssociative meaning, it’s not real meaning. while, the word “blood” is expression the of someone who https://jurnal.uniraya.ac.id/index.php/Relation P-ISSN: 2715-1638 E-ISSN: 2828-5522 P-ISSN: 2715-1638 E-ISSN: 2828-5522 Universitas Nias Raya Research on English Language Education (Relation Jou Vol. 4 No. 1 Edisi April 2022 isangry,disappointed and emotion that looked from the eyes. So the word ‘blood’ associated with ‘an expression angry, dissapointed and emotion Research on English Language Education (Relation Journa Vol. 4 No. 1 Edisi April 2022 isangry,disappointed and emotion that looked from the eyes. So the word ‘blood’ associated with ‘an expression angry, dissapointed and emotion Research on English Language Education (Relation Jou Vol. 4 No. f. Attitudional meaning f. Attitudional meaning Attitudional meaning is the part of the overal meaning of an expression that consists of some widespread attitude to the refferent. The expression does not merely denote the different, but also hints at some attitude to it. Attitudional connotations be important even when the referent is not a person but a system or practice or way of thinking. f. Attitudional meaning Attitudional meaning is the part of the overal meaning of an expression that consists of some widespread attitude to the refferent. The expression does not merely denote the different, but also hints at some attitude to it. Attitudional connotations be important even when the referent is not a person but a system or practice or way of thinking. Based on research findings and discussion, it can be concluded that connotative meaning is the one important aspect which has been used in Linkin Parks a Thousand Suns album. It beautifies the songs and give the powerfull impressions and message tothe song. The data used in this research is the lyrics of Linkin parks a thousand suns album, which consist of 12 songs. Based on the research findings, some suggestions addressed teacher, It suggested to the teacher to use this reasearch finding as the example of comprehending the English song and to find the connotative meaning of English songs. As addition, the teacher can used the theory ofconnotative meaning as material to teach and learning in increase their vocabullary The connotative meaning based on types of attitudional meaning doesn’t found in song lyrics Based on the research finding above, this section presented the discussion which has conducted bycomparing the research findings with the theory applied in this research. Specifically, this research was aimed to find out the types and the meaning of connotative meaning which were used in linkin park “a thousand suns album”. Theoretically, On the other hand, Previously, the research that has related to this research which has conducted by Muslimin (2017), entiled The Analysis of Connotative Meaning in Masya and The Bear Cartoon Movie (A SemioticAnalysis).This research found that the connotative meaning in in Masya and The Bear Cartoon Moviebased on theory of Hervey And the most dominant connotative meaning which is used in some some movie is Allusive meaning. In one case, her study was similiar to this research result reffered to thetypes of connotative meaning. C. Research Finding and Discussion Fires of a thousand suns as hell  Filthy animals beat down low (St.wretches and kings, P.70, L.65) This sentence types of affective meaning, becausefilthy animals are beaten down with low referrals to people who make big mistakes are punished and treated harshly and badly. So, filthy animals beatdown low are negative connotation while its affect on the listener or reader. Because the real meaning its talk about human beat down low not animals down low.Affective meaning reffers on the listener when he/she said “fists” its the bad or negative word, he/she may can say another word that contains possitive expression to describe he or she express  Filthy animals beat down low (St.wretches and kings, P.70, L.65)  This sentence types of affective meaning, becausefilthy animals are beaten down with low referrals to people who make big mistakes are punished and treated harshly and badly. So, filthy animals beatdown low are negative connotation while its affect on the listener or reader. c. Collocative meaning is given to an expression over and above its literal meaning by the meaning of some other expression with which typically collocates to form a commonly used phrase. Collocative meaning lent to a linguistic expression by the meaning of some other expression with which it frequently collocates. Collocative meaning consists of the association a word acquires on account of the meanings of words which tend to occur in its environment. A collocation is made up two or more words that are commonly used together in english. c.  Black and blue run at the sunshine. (St.wretches and kings, P.70, L.67) This sentence type of collocative meaning because It not mean black and blue running in the sun but it e. Reflected meaning is the meaning given to an expression over and above the denotative meaning which it has in that context by the fact that https://jurnal.uniraya.ac.id/index.php/Relation P-ISSN: 2715-1638 E-ISSN: 2828-5522 P-ISSN: 2715-1638 E-ISSN: 2828-5522 Universitas Nias Raya Research on English Language Education (Relation J Vol. 4 No. 1 Edisi April 2022 it also calls to mindanother meaning of the some words or phrases. The connotative meaning based on types of reflected meaning doesn’t found in song lyrics Research on English Language Education (Relation J Vol. 4 No. 1 Edisi April 2022 it also calls to mindanother meaning of the some words or phrases. C. Research Finding and Discussion The connotative meaning based on types of reflected meaning doesn’t found in song lyrics Universitas Nias Raya researches were different in dominant types of connotative meaning which was Associative meaning and another different is the object of this research. https://jurnal.uniraya.ac.id/index.php/Relation f. Attitudional meaning New York: Routledge Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 f. Attitudional meaning On the other hand, both of these Students, It is expected the research findings can be used as the example of teaching material in comprehending the connotative meaning of the English songs.Next researcher, It suggested to the next researcher to use the research as a guidance in conducting the relevant research Finally, researcher realizes that this research has many weaknesses. So that, researcher is expected for the next researcher were suggested to find other types of connotative meaning which were not found in the research. Moreover, it will be possible to use other theories in conducting the next research and more literally word in order enchance the wider knowledge and get the deep and more analysis about connotative meaning. Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya E. Bibliography Anderson, G. & Arsenault, N. 2005. Fundamentals of Education Research. (2ndEd). Falmer Press Chandler, Daniel 2007. The Basis Semiotic. (2ndEd). London and New York:Routledge Chaer, Abdul 2012. Linguistik Umum. Edisi Revisi. Jakarta: PT.Rhineka Cipta Dickins, James. 2019. Discourse in Translation. (1st Ed).New York: Routledge Dickins, J. Hervey S. and Higgins I. 2017.Thinking Arabic Translation. (2ndEd).New York: Routledge Hervey S, Loughbridge M and Higgins I. 2006. German Translation A Course in Translation Method: German To English. (2nd Ed). New York: Routledge Hurford J.R, Heasley B and Smith M.B 2007. Semantics a Coursebook. (2nd Ed).New York:Cambridge University Press Brinton L J and Brinton D M 2010 The Lyons, Jhon 1981. Language And Linguistics An Introduction.New York: Cambridge University Press Miles, Huberman and Saldana 2014. Qualitative Data Analysis:A Methods Sourcebook. Third Edition. America: United states USA” SAGE Walliman, N. 2011. Research Methods The Basics. New York: Routledge Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya E. Bibliography Anderson, G. & Arsenault, N. 2005. Fundamentals of Education Research. (2ndEd). Falmer Press Chandler, Daniel 2007. The Basis Semiotic. (2ndEd). London and New York:Routledge Chaer, Abdul 2012. Linguistik Umum. Edisi Revisi. Jakarta: PT.Rhineka Cipta Dickins, James. 2019. Discourse in Translation. (1st Ed).New York: Routledge Dickins, J. Hervey S. and Higgins I. Lyons, Jhon 1981. Language And Linguistics An Introduction.New York: Cambridge University Press Miles, Huberman and Saldana 2014. Qualitative Data Analysis:A Methods Sourcebook. Third Edition. America: United states USA” SAGE Walliman, N. 2011. Research Methods The Basics. https://jurnal.uniraya.ac.id/index.php/Relation E. Bibliography Chandler, Daniel 2007. The Basis Semiotic. (2ndEd). London and New York:Routledge Chandler, Daniel 2007. The Basis Semiotic. (2ndEd). London and New York:Routledge Sourcebook. T United states Walliman, N. 2011. Chaer, Abdul 2012. Linguistik Umum. Edisi Revisi. Jakarta: PT.Rhineka Cipta Basics. Dickins, James. 2019. Discourse in Translation. (1st Ed).New York: Routledge Dickins, J. Hervey S. and Higgins I. 2017.Thinking Arabic Translation. (2ndEd).New York: Routledge Hervey S, Loughbridge M and Higgins I. 2006. German Translation A Course in Translation Method: German To English. (2nd Ed). New York: Routledge Hurford J.R, Heasley B and Smith M.B 2007. Semantics a Coursebook. (2nd Ed).New York:Cambridge University Press Hurford J.R, Heasley B and Smith M.B 2007. Semantics a Coursebook. (2nd Ed).New York:Cambridge University Press Brinton L.J and Brinton D.M 2010. The Linguistic Structure of Modern English. Amsterdam:University of Southern California Kielmann, Cataldo and Seeley. 2012. Introduction to Qualitative Research a Training Manual. Seotland: Institute for International Health & Development Queen Margaret University Kreidler, Charles W 1998. Inroducing English Semantics.(1stEd). London: Routledge London: Leech, Geofrey 1981. Semantics The Study of Meaning. (2nd Ed). New York:Penguin Books Lincoln, Yvonna S and Guba, Egon G. 2013. The Constructivist Credo. California: Left Coast Press https://jurnal.uniraya.ac.id/index.php/Relation Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya Research on English Language Education (Relation Journal) Vol. 4 No. 1 Edisi April 2022 https://jurnal.uniraya.ac.id/index.php/Relation Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya Research on English Language Education (Relation Journal) P-ISSN: 2715-1638 E-ISSN: 2828-5522 Vol. 4 No. 1 Edisi April 2022 Universitas Nias Raya https://jurnal.uniraya.ac.id/index.php/Relation
https://openalex.org/W3043130971	http://era.daf.qld.gov.au/7602/1/genes-11-00806.pdf	English	null	Genetic Diversity of C4 Photosynthesis Pathway Genes in Sorghum bicolor (L.)	Genes	2,020	cc-by	10,772	Received: 24 June 2020; Accepted: 13 July 2020; Published: 16 July 2020 Abstract: C4 photosynthesis has evolved in over 60 diﬀerent plant taxa and is an excellent example of convergent evolution. Plants using the C4 photosynthetic pathway have an eﬃciency advantage, particularly in hot and dry environments. They account for 23% of global primary production and include some of our most productive cereals. While previous genetic studies comparing phylogenetically related C3 and C4 species have elucidated the genetic diversity underpinning the C4 photosynthetic pathway, no previous studies have described the genetic diversity of the genes involved in this pathway within a C4 crop species. Enhanced understanding of the allelic diversity and selection signatures of genes in this pathway may present opportunities to improve photosynthetic eﬃciency, and ultimately yield, by exploiting natural variation. Here, we present the ﬁrst genetic diversity survey of 8 known C4 gene families in an important C4 crop, Sorghum bicolor (L.) Moench, using sequence data of 48 genotypes covering wild and domesticated sorghum accessions. Average nucleotide diversity of C4 gene families varied more than 20-fold from the NADP-malate dehydrogenase (MDH) gene family (θπ = 0.2 × 10−3) to the pyruvate orthophosphate dikinase (PPDK) gene family (θπ = 5.21 × 10−3). Genetic diversity of C4 genes was reduced by 22.43% in cultivated sorghum compared to wild and weedy sorghum, indicating that the group of wild and weedy sorghum may constitute an untapped reservoir for alleles related to the C4 photosynthetic pathway. A SNP-level analysis identiﬁed purifying selection signals on C4 PPDK and carbonic anhydrase (CA) genes, and balancing selection signals on C4 PPDK-regulatory protein (RP) and phosphoenolpyruvate carboxylase (PEPC) genes. Allelic distribution of these C4 genes was consistent with selection signals detected. A better understanding of the genetic diversity of C4 pathway in sorghum paves the way for mining the natural allelic variation for the improvement of photosynthesis. Keywords: sorghum; C4 pathway; genetic diversity; SNPs; domestication Yongfu Tao 1 , Barbara George-Jaeggli 1,2 , Marie Bouteillé-Pallas 1, Shuaishuai Tai 3 , Alan Cruickshank 2, David Jordan 1 and Emma Mace 1,2,* Yongfu Tao 1 , Barbara George-Jaeggli 1,2 , Marie Bouteillé-Pallas 1, Shuaishuai Tai 3 , Alan Cruickshank 2, David Jordan 1 and Emma Mace 1,2,* 1 Queensland Alliance for Agriculture and Food Innovation (QAAFI), The University of Queensland, Hermitage Research Facility, Warwick, QLD 4370, Australia; y.tao1@uq.edu.au (Y.T.); b.georgejaeggli@uq.edu.au (B.G.-J.); mariebouteille0384@gmail.com (M.B.-P.); david.jordan@uq.edu.au (D.J.) 2 Agri-Science Queensland, Department of Agriculture and Fisheries (DAF), Hermitage Research Facility, Warwick, QLD 4370, Australia; Alan.Cruickshank@daf.qld.gov.au 3 BGI Genomics, BGI-Shenzhen, Shenzhen 518083, China; shuaishuai.tai@bgi.com * Correspondence: emma.mace@uq.edu.au Received: 24 June 2020; Accepted: 13 July 2020; Published: 16 July 2020 1 Queensland Alliance for Agriculture and Food Innovation (QAAFI), The University of Queensland, Hermitage Research Facility, Warwick, QLD 4370, Australia; y.tao1@uq.edu.au (Y.T.); b.georgejaeggli@uq.edu.au (B.G.-J.); mariebouteille0384@gmail.com (M.B.-P.); david.jordan@uq.edu.au (D.J.) 2 Agri-Science Queensland, Department of Agriculture and Fisheries (DAF), Hermitage Research Facility, Warwick, QLD 4370, Australia; Alan.Cruickshank@daf.qld.gov.au 3 BGI Genomics, BGI-Shenzhen, Shenzhen 518083, China; shuaishuai.tai@bgi.com * Correspondence: emma.mace@uq.edu.au q g 3 BGI Genomics, BGI-Shenzhen, Shenzhen 518083, China; shuaishuai.tai@bgi.com * Correspondence: emma.mace@uq.edu.au genes G C A T T A C G G C A T genes G C A T T A C G G C A T genes Keywords: sorghum; C4 pathway; genetic diversity; SNPs; domestication Genetic Diversity of C4 Photosynthesis Pathway Genes in Sorghum bicolor (L.) Yongfu Tao 1 , Barbara George-Jaeggli 1,2 , Marie Bouteillé-Pallas 1, Shuaishuai Tai 3 , Alan Cruickshank 2, David Jordan 1 and Emma Mace 1,2,* 1. Introduction C4 photosynthesis has independently evolved in more than 60 diﬀerent plant taxa [1]. The main driver for this convergent evolution is the tendency of Ribulose-1,5-bisphosphate carboxylase (Rubisco), which catalyzes the net ﬁxation of carbon dioxide (CO2) to also catalyze an unfavorable oxygenation reaction. This reaction produces toxic phosphoglycolate which has to be converted to useful metabolites requiring substantial metabolic energy [2,3]. This wasteful use of CO2 is termed photorespiration. Photorespiration becomes a major constraint of photosynthesis in situations where CO2 to O2 ratios Genes 2020, 11, 806; doi:10.3390/genes11070806 www.mdpi.com/journal/genes www.mdpi.com/journal/genes Genes 2020, 11, 806 2 of 15 are low and temperatures are high. The evolution of C4 photosynthesis coincided with declining atmospheric CO2 concentrations [1,4] as a mechanism to avoid photorespiration by concentrating CO2 in the vicinity of ribulose-1,5-bisphosphate carboxylase (Rubisco). In the majority of C4 plants, this is achieved via spatial separation of the initial CO2 ﬁxation and the Calvin–Benson–Bassham cycle in two diﬀerent cell types, most often mesophyll cells and bundle sheath cells [5]. CO2 concentration in C4 bundle sheath cells is up to 10-fold higher than that found in C3 mesophyll cells [6]. At higher temperatures, C4 photosynthesis is not only more eﬃcient compared with C3 photosynthesis in terms of reducing energy losses from photorespiration, but due to the improved eﬃciency of this pathway, it renders plants more nitrogen- and water-use eﬃcient [7,8]. C4 plants are more productive than C3 plants in areas with high light intensities, warm temperatures, and low rainfall, such as the sub-tropical and tropical areas around the globe. Many of the major crops that originated from warm and dry regions of the world, such as maize, sorghum, millet, sugarcane, miscanthus, and switchgrass, use the C4 pathway [9]. C4 crops account for an estimated 23% of global primary production [10]. Improved photosynthetic capacity has been suggested as the next frontier in lifting crop productivity [11]. The C4 photosynthesis pathway is a good starting point to improve photosynthetic capacity and resource eﬃciency in crop plants. Attempts are currently being undertaken to integrate characteristics of the C4 pathway into C3 crops [7,12–14]. However, possibly due to multiple independent evolutions of C4 photosynthesis in diﬀerent plant taxa [1], large variation also exists among C4 species in terms of the biochemical pathway. 1. Introduction It has long been known that three major biochemical subtypes—nicotinamide adenine dinucleotide phosphate-malic enzyme (NADP-ME), nicotinamide adenine dinucleotide-malic enzyme (NAD-ME) and phospho-enol-pyruvate carboxykinase (PCK)—exist among C4 species [15]. More recently, it has been suggested that mixtures among them exist [16] and that the subtypes vary in their performance under diﬀerent environmental conditions, e.g., low light [17]. Especially among the grasses, which all of the C4 cereals belong to, diﬀerences in pathway and performance are likely to exist, as C4 photosynthesis has evolved at least 25 times in this group of plants [18]. Exploring such variation may provide avenues to further improve C4 photosynthetic eﬃciency [9]. Sorghum is an NADP-ME subtype C4 crop well-known for its adaption to drought and high temperatures. It provides staple food for over 500 million people in the semi-arid tropics of Africa and Asia; in addition to being an important source of feed, ﬁber, and biofuel. Due to these characteristics, it is expected to play an increasingly important role in meeting the challenges of feeding the world’s growing population under the threat of global warming. Substantial variation in photosynthesis and related traits has been revealed in sorghum [19–23], indicating existence of genetic variation of underlying genes. However, this variation has not yet been studied. The recent assembly of whole-genome sequences for a wide range of wild and cultivated sorghum species [24–26] provides an excellent opportunity to explore genetic diversity of genes related to the C4 photosynthetic pathway. Several high-throughput comparative transcriptomics and evolutionary studies using C3 and C4 phylogenetically related species and cell-speciﬁc gene expression have elucidated the key genes and regulatory networks that underpin the C4 photosynthetic pathway [5,27–37]. In the present study, we explored the genetic variation in genes that have previously been identiﬁed as core C4 genes, mined their allelic diversity and investigated signatures of selection during domestication in sorghum. Identiﬁcation of C4 Gene Families This study focuses on 8 key proteins in the NADP-ME photosynthetic pathway in sorghum (Figure 1). A total of 9 genes encoding these proteins with expression and evolutionary evidence supporting their involvement in NADP-ME pathway (hereafter, referred as C4 genes), and their non-C4 isoforms in sorghum were extracted from two previous studies [38,39] (Table 1). These non-C4 isoforms 3 of 15 3 of 15 Genes 2020, 11, 806 Genes 2020, 11, x FOR are homologous of C4 genes but there was no evidences supporting their involvement in the NADP-ME photosynthetic pathway. Homology between these sorghum C4 genes and their non-C4 isoforms was further veriﬁed via a local blast strategy. Protein sequences of these 9 core C4 genes were extracted from the sorghum reference genome V3.1 and were blasted against the reference genome. Blast hits of each gene were ﬁltered using the criteria: E-value <−10, sequence identity >60%, and alignment length >80%. All hits of the same gene satisfying the criteria were plotted based on –log (E-value); only hits of top –log (E-value) class were considered if clear diﬀerentiation among them was visualized, otherwise all hits were used. isoforms are homologous of C4 genes but there was no evidences supporting their involvement in the NADP‐ME photosynthetic pathway. Homology between these sorghum C4 genes and their non‐C4 isoforms was further verified via a local blast strategy. Protein sequences of these 9 core C4 genes were extracted from the sorghum reference genome V3.1 and were blasted against the reference genome. Blast hits of each gene were filtered using the criteria: E‐value <−10, sequence identity >60%, and alignment length >80%. All hits of the same gene satisfying the criteria were plotted based on – log (E‐value); only hits of top –log (E‐value) class were considered if clear differentiation among them was visualized, otherwise all hits were used. Figure 1. Diagram of the nicotinamide adenine dinucleotide phosphate‐malic enzyme (NADP‐ME) biosynthetic pathway of C4 photosynthesis (adapted from [40]). In the mesophyll cells, CO2 is converted to HCO3− catalyzed by carbonic anhydrase (CA) and fixed into the four‐carbon acid, oxaloacetate (OAA), by phosphoenolpyruvate carboxylase (PEPC). Phosphorylation of PEPC is carried out by PEPC kinase (PPCK). The OAA generated by PEPC is then reduced to malate by the NADP‐malate dehydrogenase (NADP‐MDH) or trans‐aminated to aspartate. Identiﬁcation of C4 Gene Families The resultant C4 acids, malate and aspartate, are transported to the bundle sheath and then decarboxylated in the vicinity of Rubisco to release CO2 and pyruvate. Pyruvate is transported back to mesophyll cells to regenerate PEP by pyruvate orthophosphate dikinase (PPDK), while CO2 enters the Calvin–Benson–Bassham cycle and is fixed by ribulose‐1,5‐bisphosphate carboxylase (Rubisco). Activation and inactivation of PPDK is catalyzed by PPDK regulatory protein (PPDK‐RP). Figure 1. Diagram of the nicotinamide adenine dinucleotide phosphate-malic enzyme (NADP-ME) biosynthetic pathway of C4 photosynthesis (adapted from [40]). In the mesophyll cells, CO2 is converted to HCO3−catalyzed by carbonic anhydrase (CA) and ﬁxed into the four-carbon acid, oxaloacetate (OAA), by phosphoenolpyruvate carboxylase (PEPC). Phosphorylation of PEPC is carried out by PEPC kinase (PPCK). The OAA generated by PEPC is then reduced to malate by the NADP-malate dehydrogenase (NADP-MDH) or trans-aminated to aspartate. The resultant C4 acids, malate and aspartate, are transported to the bundle sheath and then decarboxylated in the vicinity of Rubisco to release CO2 and pyruvate. Pyruvate is transported back to mesophyll cells to regenerate PEP by pyruvate orthophosphate dikinase (PPDK), while CO2 enters the Calvin–Benson–Bassham cycle and is ﬁxed by ribulose-1,5-bisphosphate carboxylase (Rubisco). Activation and inactivation of PPDK is catalyzed by PPDK regulatory protein (PPDK-RP). Figure 1. Diagram of the nicotinamide adenine dinucleotide phosphate‐malic enzyme (NADP‐ME) biosynthetic pathway of C4 photosynthesis (adapted from [40]). In the mesophyll cells, CO2 is converted to HCO3− catalyzed by carbonic anhydrase (CA) and fixed into the four‐carbon acid, oxaloacetate (OAA), by phosphoenolpyruvate carboxylase (PEPC). Phosphorylation of PEPC is carried out by PEPC kinase (PPCK). The OAA generated by PEPC is then reduced to malate by the NADP‐malate dehydrogenase (NADP‐MDH) or trans‐aminated to aspartate. The resultant C4 acids, malate and aspartate, are transported to the bundle sheath and then decarboxylated in the vicinity of Rubisco to release CO2 and pyruvate. Pyruvate is transported back to mesophyll cells to regenerate PEP by pyruvate orthophosphate dikinase (PPDK), while CO2 enters the Calvin–Benson–Bassham cycle and is fixed by ribulose‐1,5‐bisphosphate carboxylase (Rubisco). Activation and inactivation of PPDK is catalyzed by PPDK regulatory protein (PPDK‐RP) Figure 1. Diagram of the nicotinamide adenine dinucleotide phosphate-malic enzyme (NADP-ME) biosynthetic pathway of C4 photosynthesis (adapted from [40]). In the mesophyll cells, CO2 is converted to HCO3−catalyzed by carbonic anhydrase (CA) and ﬁxed into the four-carbon acid, oxaloacetate (OAA), by phosphoenolpyruvate carboxylase (PEPC). Phosphorylation of PEPC is carried out by PEPC kinase (PPCK). Identiﬁcation of C4 Gene Families The OAA generated by PEPC is then reduced to malate by the NADP-malate dehydrogenase (NADP-MDH) or trans-aminated to aspartate. The resultant C4 acids, malate and aspartate, are transported to the bundle sheath and then decarboxylated in the vicinity of Rubisco to release CO2 and pyruvate. Pyruvate is transported back to mesophyll cells to regenerate PEP by pyruvate orthophosphate dikinase (PPDK), while CO2 enters the Calvin–Benson–Bassham cycle and is ﬁxed by ribulose-1,5-bisphosphate carboxylase (Rubisco). Activation and inactivation of PPDK is catalyzed by PPDK regulatory protein (PPDK-RP). 4 of 15 Genes 2020, 11, 806 Table 1. Single nucleotide polymorphism (SNP) information and selection signals across 27 genes from C4 gene families. Identiﬁcation of C4 Gene Families Gene ID Enzyme GL CDSL NoS NoSiC NoNS NoSS UPSGL UBSGL NoSUPS NoNSUPS NoSUBS NoNSUBS Sobic.002G230100 CA 4823 1014 115 14 4 10 No No 0 0 1 0 Sobic.003G234200 CA 10440 1371 475 33 7 26 No No 1 1 0 0 Sobic.003G234400 CA 4749 615 138 13 3 10 No No 0 0 0 0 Sobic.003G234500 CA 2986 609 173 11 5 6 No No 0 0 0 0 Sobic.003G234600 CA 4750 771 210 18 10 8 No No 0 0 0 0 Sobic.007G166200 NADP-MDH 3354 1308 53 11 6 5 No No 0 0 0 0 Sobic.007G166300 NADP-MDH 3816 1290 108 12 3 9 No No 0 0 0 0 Sobic.003G036000 NADP-ME 6107 1941 111 11 4 7 No No 0 0 0 0 Sobic.003G036200 NADP-ME 5447 1911 141 12 3 9 No No 0 0 0 0 Sobic.003G280900 NADP-ME 5691 1782 175 22 13 9 No No 1 1 0 0 Sobic.003G292400 NADP-ME 4527 1782 95 22 8 14 No No 10 2 0 0 Sobic.009G069600 NADP-ME 3624 1713 118 34 10 24 No No 3 1 0 0 Sobic.002G167000 PEPC 5632 2904 41 11 6 5 No No 0 0 0 0 Sobic.003G100600 PEPC 8881 3117 371 43 9 34 No No 0 0 21 2 Sobic.003G301800 PEPC 7610 2901 138 19 3 17 No No 0 0 0 0 Sobic.004G106900 PEPC 6977 2883 146 34 5 29 No No 0 0 7 0 Sobic.007G106500 PEPC 5616 2895 64 12 8 4 No No 1 1 0 0 Sobic.010G160700 PEPC 6647 3087 193 28 9 19 No No 0 0 2 0 Sobic.004G219900 PPCK 1612 924 40 9 1 8 No No 0 0 2 0 Sobic.004G338000 PPCK 1749 855 37 9 4 4 No No 0 0 0 0 Sobic.006G148300 PPCK 1997 900 64 4 1 3 No No 0 0 0 0 Sobic.001G326900 PPDK 8494 2730 321 46 18 28 No Yes 0 0 24 5 Sobic.009G132900 PPDK 12748 2847 441 16 0 16 No No 3 0 0 0 Sobic.002G324400 PPDK-RP 2507 1290 79 22 8 14 No No 0 0 3 0 Sobic.002G324500 PPDK-RP 3072 1260 69 20 5 15 No No 4 0 0 0 Sobic.002G324700 PPDK-RP 4662 1587 222 28 19 9 No No 1 1 2 2 Sobic.005G042000 RbcS 1556 510 45 7 4 3 No No 0 0 0 0 Gene ID is according to sorghum reference genome V3.1. in bold indicate their C4 genes. Enzyme: Encoded enzyme. GL: Gene length. CDSL: Length of coding sequence (CDS). umber of SNPs identiﬁed in CDS. NoNS: Number of non-synonymous SNPs. NoSS: Number of synonymous SNPs. UBSGL: Under balancing selection based on gene level analysis NoSUPS: Number of SNPs under purifying selection 4. Gene-Level Population Genetic Analyses Population genetic parameters including nucleotide diversity (θπ) [41], Tajima’s D [42], and Watterson’s Estimator (hW) [43] were directly calculated for each of the 27 genes using the Bio::PopGen::Statistics module. FST [44], which measures population diﬀerentiation, was also calculated for each of the 27 genes using the Bio::PopGen::PopStats module [26]. The Bio::PopGen::IO module was used to read input ﬁle, which was prepared using an in-house Perl script for calculation of these population genetic parameters. The criteria used in Mace et al. (2013) were employed to identify genes under purifying selection and balancing selection, respectively. Criteria for purifying selection included: (1) θπ and hW < 5% of the empirical distribution in the cultivated group, (2) FST between the group of cultivated sorghum and the group of wild and weedy sorghum > 95% of the population pairwise distribution, (3) Tajima’s D < 0. Criteria for balancing selection included: (1) θπ and hW > 25% of the empirical distribution in the cultivated group, (2) FST between the group of cultivated sorghum and the group of wild and weedy sorghum < 90% of the population pairwise distribution, (3) Tajima’s D > 5% of the empirical distribution. Identiﬁcation of C4 Gene Families Gene IDs in bold indicate their C4 genes. Enzyme: Encoded enzyme. GL: Gene length. CDSL: Length of coding sequence (CDS). NoS: Total number of SNPs identiﬁed across the gene. NoSiC: Number of SNPs identiﬁed in CDS. NoNS: Number of non-synonymous SNPs. NoSS: Number of synonymous SNPs. UPSGL: Under purifying selection based on gene level analysis. UBSGL: Under balancing selection based on gene level analysis. NoSUPS: Number of SNPs under purifying selection. NoNSUPS: Number of non-synonymous SNPs under purifying selection. NoSUBS: Number of SNPs under balancing selection. NoNSUBS: Number of non-synonymous SNPs under balancing selection. ymorphism (SNP) information and selection signals across 27 genes from C4 gene families. Gene ID is according to sorghum reference genome V3.1. Gene IDs in bold indicate their C4 genes. Enzyme: Encoded enzyme. GL: Gene length. CDSL: Length of coding sequence (CDS). NoS: Total number of SNPs identiﬁed across the gene. NoSiC: Number of SNPs identiﬁed in CDS. NoNS: Number of non-synonymous SNPs. NoSS: Number of synonymous SNPs. UPSGL: Under purifying selection based on gene level analysis. UBSGL: Under balancing selection based on gene level analysis. NoSUPS: Number of SNPs under purifying selection. NoNSUPS: Number of non-synonymous SNPs under purifying selection. NoSUBS: Number of SNPs under balancing selection. NoNSUBS: Number of non-synonymous SNPs under balancing selection. 5 of 15 Genes 2020, 11, 806 3. Plant Material and Genomic Data Sequence data of the identiﬁed C4 genes were extracted from 48 accessions of Sorghum bicolor with high mapping depth (~22× per accession, ranging from 16 to 45×) reported in previous studies [24–26]. These 48 accessions represent all major cultivated sorghum races and some wild progenitors (Table S1). 7. Results Nucleotide Diversity of Core C4 Gene Families in Sorghum 6. Phylogenetic and Haplotype Analysis A phylogenetic tree was constructed based on CDS of all 27 genes from C4 gene families using the neighbor-joining method with default settings (bootstrapped 100 times; support threshold, 50%) in Geneious 8.1.2 (https://www.geneious.com/, Biomatters Ltd., Auckland, New Zealand). Analysis of haplotype network was conducted using a combination of the R package ape [46] and pegas [47]. All 48 sorghum accessions were classiﬁed into four groups: Cultivated, wild and weedy, Guinea margaritiferum and S. propinquum (Table S2). 5. SNP-Level Identiﬁcation of Selection Signature Population genetics parameters including θπ, Tajima’s D, and FST between the group of cultivated sorghum and the group of wild and weedy sorghum were computed for these 27 genes using CDS sequence in PopGenome, a population genomics package implemented in the R environment (http://cran.r-project.org/) [45]. Speciﬁcally, commands diversity.stats, F_ST.stats, and neutrality.stats were called to calculate θπ, FST, and Tajima’s D for each single nucleotide polymorphism (SNP), respectively, with a slide window of 1-bp and 1-bp step size. Functional annotation of each SNP was conducted using get.codons command. Fold decrease of θπ in the cultivated sorghum group compared to the group of wild and weedy sorghum was calculated to represent reduction of diversity (RoD). The following criteria were adopted to identify sites with signature of purifying selection: (1) A RoD greater than the average of neutral genes; (2) FST > 0; (3) Tajima’s D < 0. The following criteria were adopted to identify sites with signature of balancing selection: (1) An increase in diversity (IoD) in the cultivated group and the group of wild and weedy comparison; (2) FST > 0; (3) Tajima’s D > 0. Nucleotide Diversity of Core C4 Gene Families in Sorghum Based on 9 genes corresponding to 8 core C4 enzymes in sorghum, 18 homologous genes were identiﬁed across the sorghum genome. In total, 5 CA genes, 2 NADP-MDH genes, 5 NADP-ME genes, 6 PEPC genes, 3 PPCK genes, 2 PPDK genes, 3 PPDK-RP genes, and 1 rbcS gene were identiﬁed 6 of 15 Genes 2020, 11, 806 (Table 1). Nucleotide diversity (θπ) of these 27 genes was investigated using sequence data of 48 genotypes covering wild and weedy, and cultivated sorghum (Mace et al., 2013). A total number of 4183 single nucleotide polymorphisms (SNPs) were identiﬁed in these 27 genes with 521 SNPs located in coding sequence (CDS) regions (Table 1). These C4 gene families displayed an average overall nucleotide diversity of θπ = 2.09 × 10−3, which is comparable to that of 130 housekeeping genes (θπ = 1.97 × 10−3, Mace et al., 2013) (t-test, p-value > 0.05). Nucleotide diversity varied dramatically among the C4 gene families, with the NADP-MDH genes displaying the lowest levels of diversity across all genotypes (average θπ = 0.25 × 10−3), followed by NADP-ME genes (θπ = 0.93 × 10−3), PPCK genes (θπ = 1.20 × 10−3), PEPC genes (θπ = 2.11 × 10−3), CA (θπ = 2.26 × 10−3), and PPDK-RP (θπ = 2.96 × 10−3), while PPDK genes showed the highest level of diversity (θπ = 5.21 × 10−3) (Table 2, Figure 2A). The only gene encoding ribulose bisphosphate carboxylase/oxygenase small-subunit (rbcS), Sobic.005G042000, had relatively high genetic diversity among C4 gene families with θπ = 4.32 × 10−3 across all 48 genotypes, 5.72 × 10−3 in the wild and weedy group, and 3.03 × 10−3 in the cultivated group. Table 2. Genetic diversity (θπ) and ﬁxation index (FST) of 27 genes from C4 gene families. Nucleotide Diversity of Core C4 Gene Families in Sorghum Nucleotide diversity of C4 genes in the other gene families wa within the range of variation of their non C4 isoforms Mixed trends were found when comparing C4 genes with non-C4 isoforms in each gene family with the average overall genetic diversity of C4 genes being comparable to that of their non-C4 counterpart (Table 2). The C4 PPDK-RP gene (Sobic.007G166300) and C4 NADP-MDH gene (Sobic.002G324400) had an overall θπ which was 161.76% and 79.85% higher than their non-C4 isoforms, respectively, whereas the θπ of the C4 PPDK gene (Sobic.009G132900) was 75.16% lower than that of the non-C4 PPDK isoform. Nucleotide diversity of C4 genes in the other gene families was within the range of variation of their non-C4 isoforms. within the range of variation of their non‐C4 isoforms. Genetic diversity across C4 gene families was significantly reduced during sorghum domestication (paired t‐test, p‐value < 0.05). Averaged across all C4 gene families genetic diversit was reduced by 22.44% in the domesticated compared with the wild and weedy group and when jus the 9 core C4 genes were considered, the reduction was 22.98%. However, the reduction of geneti diversity during domestication in C4 genes was not significantly different from that in housekeepin genes (Table S2) (t‐test, p‐value > 0.05). Among the 27 genes, Sobic.003G292400, a non‐C4 NADP‐ME isoform, exhibited the most severe reduction in genetic diversity, with a reduction of 98.23%. The C Genetic diversity across C4 gene families was signiﬁcantly reduced during sorghum domestication (paired t-test, p-value < 0.05). Averaged across all C4 gene families genetic diversity was reduced by 22.44% in the domesticated compared with the wild and weedy group and when just the 9 core C4 genes were considered, the reduction was 22.98%. However, the reduction of genetic diversity during domestication in C4 genes was not signiﬁcantly diﬀerent from that in housekeeping genes (Table S2) (t-test, p-value > 0.05). Among the 27 genes, Sobic.003G292400, a non-C4 NADP-ME isoform, exhibited the most severe reduction in genetic diversity, with a reduction of 98.23%. The C4 version of that gene, the NADP-ME gene (Sobic.003G036200), showed the greatest loss of genetic diversity (51.89%) among the C4 genes, with an FST between the cultivated and wild and weedy groups of 0.06 (Figure 2B). Nucleotide Diversity of Core C4 Gene Families in Sorghum GeneID Enzyme θπ–All θπ-Cultivated θπ-W&W FST Sobic.002G230100 CA 0.80 0.74 0.90 0.19 Sobic.003G234200 CA 2.65 2.46 2.66 0.16 Sobic.003G234400 CA 1.01 0.91 0.88 0.37 Sobic.003G234500 CA 5.55 5.51 4.56 0.07 Sobic.003G234600 CA 1.27 1.35 0.65 0.06 Sobic.007G166200 NADP-MDH 0.18 0.21 0.13 0.07 Sobic.007G166300 NADP-MDH 0.33 0.33 0.42 0.08 Sobic.003G036000 NADP-ME 0.88 0.65 1.59 0.15 Sobic.003G036200 NADP-ME 0.89 0.67 1.39 0.06 Sobic.003G280900 NADP-ME 0.93 0.85 1.11 0.09 Sobic.003G292400 NADP-ME 1.43 0.08 4.44 0.32 Sobic.009G069600 NADP-ME 0.52 0.49 0.10 0.45 Sobic.002G167000 PEPC 0.58 0.51 0.85 0.04 Sobic.003G100600 PEPC 5.36 5.18 3.56 0.05 Sobic.003G301800 PEPC 0.64 0.22 2.37 0.22 Sobic.004G106900 PEPC 3.18 3.02 2.14 0.07 Sobic.007G106500 PEPC 0.44 0.22 0.47 0.21 Sobic.010G160700 PEPC 2.49 2.25 2.86 0.04 Sobic.004G219900 PPCK 2.08 1.94 2.12 0.12 Sobic.004G338000 PPCK 1.03 0.96 0.91 0.03 Sobic.006G148300 PPCK 0.48 0.39 0.13 0.41 Sobic.001G326900 PPDK 8.34 5.64 5.64 0.40 Sobic.009G132900 PPDK 2.07 1.79 2.19 0.13 Sobic.002G324400 PPDK-RP 5.04 3.82 4.55 0.41 Sobic.002G324500 PPDK-RP 1.27 0.10 3.75 0.24 Sobic.002G324700 PPDK-RP 2.58 2.50 3.51 0.05 Sobic.005G042000 rbcS 4.32 3.41 5.72 0.12 Gene ID is according to sorghum reference genome V3.1. Gene IDs in bold indicate the C4 gene versions. Enzyme: Encoded enzyme. θπ-All: Nucleotide diversity across all 48 genotypes. θπ-Cultivated: Nucleotide diversity across cultivated genotypes. θπ-W&W: Nucleotide diversity across wild and weedy genotypes. All θπ values are in unites of per kb. FST: Fixation index between cultivated genotypes and wild and weedy genotypes. Table 2. Genetic diversity (θπ) and ﬁxation index (FST) of 27 genes from C4 gene families. xation index (FST) of 27 genes from C4 gene families. Gene ID is according to sorghum reference genome V3.1. Gene IDs in bold indicate the C4 gene versions. Enzyme: Encoded enzyme. θπ-All: Nucleotide diversity across all 48 genotypes. θπ-Cultivated: Nucleotide diversity across cultivated genotypes. θπ-W&W: Nucleotide diversity across wild and weedy genotypes. All θπ values are in unites of per kb. FST: Fixation index between cultivated genotypes and wild and weedy genotypes. es 2020, 11, 806 7 of relatively high genetic diversity among C4 gene families with θπ = 4.32 × 10−3 across all 48 genoty 5.72 × 10−3in the wild and weedy group, and 3.03 × 10−3 in the cultivated group. Figure 2. Genetic diversity and fixation index (FST) of C4 gene families between cultivated sorghum and the wild and weedy group. (A) Genetic diversity (pi) for each of the C4 gene families. Gene IDs in red indicate core C4 genes. Nucleotide Diversity of Core C4 Gene Families in Sorghum Red bars represent the pi of cultivated sorghum, while dark blue bars represent the pi of wild and weedy. (B) FST between cultivated and wild and weedy of each of C4 gene families. Gene IDs in red indicate core C4 genes. Figure 2. Genetic diversity and ﬁxation index (FST) of C4 gene families between cultivated sorghum and the wild and weedy group. (A) Genetic diversity (pi) for each of the C4 gene families. Gene IDs in red indicate core C4 genes. Red bars represent the pi of cultivated sorghum, while dark blue bars represent the pi of wild and weedy. (B) FST between cultivated and wild and weedy of each of C4 gene families. Gene IDs in red indicate core C4 genes. 7 of 15 enotypes Genes 2020, 11, 806 y g 5.72 × 10−3in th Figure 2. Genetic diversity and fixation index (FST) of C4 gene families between cultivated sorghum and the wild and weedy group. (A) Genetic diversity (pi) for each of the C4 gene families. Gene ID in red indicate core C4 genes. Red bars represent the pi of cultivated sorghum, while dark blue bar represent the pi of wild and weedy. (B) FST between cultivated and wild and weedy of each of C4 gene families. Gene IDs in red indicate core C4 genes. Figure 2. Genetic diversity and ﬁxation index (FST) of C4 gene families between cultivated sorghum and the wild and weedy group. (A) Genetic diversity (pi) for each of the C4 gene families. Gene IDs in red indicate core C4 genes. Red bars represent the pi of cultivated sorghum, while dark blue bars represent the pi of wild and weedy. (B) FST between cultivated and wild and weedy of each of C4 gene families. Gene IDs in red indicate core C4 genes. Mixed trends were found when comparing C4 genes with non‐C4 isoforms in each gene famil with the average overall genetic diversity of C4 genes being comparable to that of their non‐C counterpart (Table 2). The C4 PPDK‐RP gene (Sobic.007G166300) and C4 NADP‐MDH gen (Sobic.002G324400) had an overall θπ which was 161.76% and 79.85% higher than their non‐C isoforms, respectively, whereas the θπ of the C4 PPDK gene (Sobic.009G132900) was 75.16% lowe than that of the non‐C4 PPDK isoform. Nucleotide Diversity of Core C4 Gene Families in Sorghum In contrast, another non-C4 isoform of NADP-ME (Sobic.009G069600), a non-C4 isoform of PPCK (Sobic.006G148300), and a non-C4 CA isoform (Sobic.003G234600) showed a more than 2-fold increase in genetic diversity in the cultivated group. 8. Identiﬁcation of Selection Signals during Domestication across the 27 Genes The selection signature of these C4 gene families was ﬁrstly investigated at the gene level. Based on thresholds of genome-wide rankings described in Mace et al. (2013), only one gene (Sobic.001G326900, non-C4 PPDK isoform) was identiﬁed as being under balancing selection, which maintains diversity of selected genes, during sorghum domestication, while no gene was identiﬁed as being under purifying selection, which reduces diversity of selected genes (Table 1). Subsequent to this, a higher resolution detection of selection signature was conducted at the SNP level using the CDS of the 27 genes. Among 521 SNPs across 27 CDS, 176 were non-synonymous. 8 of 15 Genes 2020, 11, 806 The number of non-synonymous SNPs within genes varied from 19 in the non-C4 PPDK-RP isoform (Sobic.002G324700) to 0 in the C4 PPDK (Sobic.009G132900). The C4 PEPC gene (Sobic.010G160700) had the highest number of non-synonymous SNPs (9) among the 9 C4 genes (Table 1). In contrast to the gene-level analysis, SNP-level analysis identiﬁed 24 SNPs across 8 genes under purifying selection, including 7 non-synonymous SNPs in 6 genes (Table S3). Genes with SNPs under purifying selection included two C4 isoforms, PPDK (Sobic.009G132900) and CA (Sobic.003G234200), three of 4 non-C4 NADP-ME (Sobic.003G280900, Sobic.003G292400, Sobic.009G069600), both two non-C4 PPDK-RP (Sobic.002G324500, Sobic.002G324700), and a non-C4 PEPC gene (Sobic.007G106500). Among the 2 C4 genes with SNPs under selection, Sobic.009G132900 had 3 synonymous SNPs under purifying selection, while Sobic.003G234200 had a non-synonymous SNP under purifying selection. A total of 60 SNPs across 8 genes were identiﬁed as being under balancing selection, 7 of which were non-synonymous SNPs distributed across 2 genes (Table S4). The non-C4 PPDK (Sobic.001G326900) had 24 SNPs under balancing selection including 5 non-synonymous SNPs, and additionally had an overall gene-level signature of balancing selection based on the previous analysis. Two C4 isoforms, PPDK-RP (Sobic.002G324400) and PEPC (Sobic.010G160700), were identiﬁed with 3 and 2 SNPs under balancing selection, respectively, although none of them were non-synonymous SNPs. Two non-C4 PEPC (Sobic.003G100600, Sobic.004G106900) were identiﬁed with SNPs under balancing selection, with Sobic.003G100600 having 21 SNPs including 2 non-synonymous SNPs exhibiting signatures of balancing selection. The other 2 genes with SNPs under balancing selection were a non-C4 CA isoform, Sobic.002G230100, and a non-C4 PPCK isoform, Sobic.004G219900. 9. Allelic Variation of Core C4 Genes under Selection in Sorghum A phylogenetic tree was constructed using the CDS of these 27 genes to depict the genetic relationship of 48 accessions (Figure S1). The inter-and intra-species distribution of private haplotypes of each gene is detailed in Table S5, with the majority (~90%) of the genes with private inter-species haplotypes from S. propinquum, e.g., 4 unique haplotypes were observed for the C4 isoform of PEPC, with the 2 S. propinquum accessions sharing a single private haplotype. To investigate allelic variation of 4 core C4 genes with SNPs under selection in sorghum, haplotype networks were constructed using CDS SNPs. Based on 16 SNPs within the CDS of the PPDK gene (Sobic.009G132900), 8 haplotypes were identiﬁed. Five haplotypes were identiﬁed in the wild and weedy genotypes, with 3 being private haplotypes and two of them being maintained in cultivated sorghum; two new haplotypes arose in cultivated sorghum after domestication (Figure 3A). Ten haplotypes of one CA gene (Sobic.003G234200) were revealed using 33 SNPs, with 4 distinct haplotypes being characterized by the wild and weedy genotypes. Two of the wild and weedy haplotypes were maintained in cultivated sorghum during domestication, with three new haplotypes arising after domestication (Figure 3B). The loss of wild and weedy haplotypes in cultivated sorghum in these two genes was consistent with the ﬁnding that they were under purifying selection. The PPDK-RP gene (Sobic.002G324400) had 22 SNPs in the CDS, based on which 5 haplotypes were identiﬁed. Two haplotypes were characterized by the wild and weedy genotypes, with the main wild haplotype maintained and further diversifying into two new haplotypes in the cultivated group (Figure 3C). Based on 28 SNPs in the CDS of the C4 PEPC gene (Sobic.010G160700), 4 haplotypes were identiﬁed. Wild and weedy genotypes encompassed 3 haplotypes and all of them were maintained in cultivated sorghum (Figure 3D). S. propinquum had unique haplotypes across all 4 genes, while the Sorghum bicolor race guinea margaritiferum shared haplotypes with the wild and weedy genotypes in most cases, indicating a closer relationship with the wild and weedy group. 9 of 15 9 of 15 Genes 2020, 11, 806 Genes 2020, 11, x FO Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. 10 Di i 10. Discussion 10. Discussion The evolution of C4 photosynthesis has been studied extensively at the cross‐species level with signals of adaptive evolution identified on key genes in the C4 pathway [28,34,48–50]. As the evolution of C4 photosynthesis is driven by environments characterized by low CO2 availability, such as hot and dry environments in which CO2 uptake is limited by stomatal closure, it is likely that within‐species adaptive variation also exists. However, to our knowledge, studies of within‐species The evolution of C4 photosynthesis has been studied extensively at the cross-species level with signals of adaptive evolution identiﬁed on key genes in the C4 pathway [28,34,48–50]. As the evolution of C4 photosynthesis is driven by environments characterized by low CO2 availability, such as hot and dry environments in which CO2 uptake is limited by stomatal closure, it is likely that within-species adaptive variation also exists. However, to our knowledge, studies of within-species allele diversity and signatures of selection on key genes in the C4 pathway have not previously been undertaken. allele diversity and signatures of selection on key genes in the C4 pathway have not previously been undertaken. Knowledge of existing natural variation and levels of genetic diversity is a pre‐requisite for the optimization of C4 photosynthesis. In this study, we performed the first investigation of the genetic diversity of C4 gene families within a C4 species using a collection of 48 sorghum lines. We focused on 9 C4 genes due to their reported key roles in C4 photosynthesis. Our collection of sorghum represents all major cultivated sorghum races, landraces, and wild progenitors, and captures a good proportion of genetic diversity within sorghum. Substantial variation of nucleotide diversity was observed among these 8 C4 gene families in sorghum, with the NADP‐MDH gene family showing the least diversity and the PPDK gene family showing the greatest diversity. Nine core C4 genes also exhibited varying degrees of genetic diversity, ranging from θπ values of 5.04 × 10−3and 4.32 × 10−3 in PPDK‐RP and rbcS to θπ values of 0.33 × 10−3and 0.67 × 10−3in NADP‐MDH and NADP‐ME. However, despite such low levels of diversity, non‐synonymous SNPs were identified in both NADP‐MDH and Knowledge of existing natural variation and levels of genetic diversity is a pre-requisite for the optimization of C4 photosynthesis. 9. Allelic Variation of Core C4 Genes under Selection in Sorghum Group classiﬁcation of sorghum accessions used as detailed in Table S1. Color-coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes. Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK‐RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. Group classification of sorghum accessions used as detailed in Table S1. Color‐coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK-RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. Group classiﬁcation of sorghum accessions used as detailed in Table S1. Color-coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes. 9. Allelic Variation of Core C4 Genes under Selection in Sorghum (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK‐RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. Group classification of sorghum accessions used as detailed in Table S1. Color‐coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK-RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. Group classiﬁcation of sorghum accessions used as detailed in Table S1. Color-coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes. Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK‐RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. Group classification of sorghum accessions used as detailed in Table S1. Color‐coding as follows; cultivated sorghum (red), wild and weedy genotypes (purple), Sorghum propinquum (blue), and Sorghum guinea margaritiferum (green). The size of the circles in the haplotype networks is proportionate to the number of accessions with that haplotype. The branch length represents the genetic distance between two haplotypes Figure 3. Haplotype network of 4 core C4 gene with selection signal based on individual SNP analysis. (A) The PPDK gene (Sobic.009G132900) with signal of purifying selection; (B) one of the CA genes (Sobic.003G234200) with signal of purifying selection; (C) the PPDK-RP gene (Sobic.002G324400) with signal of balancing selection; (D) the PEPC gene (Sobic.010G160700) with signal of balancing selection. 10 Di i 10. Discussion In this study, we performed the ﬁrst investigation of the genetic diversity of C4 gene families within a C4 species using a collection of 48 sorghum lines. We focused on 9 C4 genes due to their reported key roles in C4 photosynthesis. Our collection of sorghum represents all major cultivated sorghum races, landraces, and wild progenitors, and captures a good proportion of genetic diversity within sorghum. Substantial variation of nucleotide diversity was observed among these 8 C4 gene families in sorghum, with the NADP-MDH gene family showing the least diversity and the PPDK gene family showing the greatest diversity. Nine core C4 genes also exhibited varying degrees of genetic diversity, ranging from θπ values of 5.04 × 10−3 and 4.32 × 10−3 in PPDK-RP and rbcS to θπ values of 0.33 × 10−3 and 0.67 × 10−3 in NADP-MDH and NADP-ME. However, despite such low levels of diversity, non-synonymous SNPs were identiﬁed in both NADP-MDH and NADP-ME (Table 1). C4 PPDK was the only gene which did not contain a non-synonymous SNP, despite its fairly large size (gene size, 12748bp; CDS, 2847bp), indicating the function of this gene is highly conserved. NADP‐ME (Table 1). C4 PPDK was the only gene which did not contain a non‐synonymous SNP, despite its fairly large size (gene size, 12748bp; CDS, 2847bp), indicating the function of this gene is highly conserved. Cultivated sorghum was domesticated more than five thousand years ago in Africa [51–53]. This artificial selection process has morphologically and physiologically reshaped sorghum to better suit h d d l l d i b i l d i f i di i id i g g p p g g g y Cultivated sorghum was domesticated more than ﬁve thousand years ago in Africa [51–53]. This artiﬁcial selection process has morphologically and physiologically reshaped sorghum to better suit human needs, and also resulted in substantial reduction of genetic diversity genome wide in cultivated sorghum compared with wild and weedy types [26,54,55]. In this study, reduction of genetic diversity during sorghum domestication was also observed in the C4 gene families, indicating Genes 2020, 11, 806 10 of 15 that wild sorghum, as a repository for genetic diversity, might harbor alleles useful for improving C4 photosynthesis. 10 Di i 10. Discussion However, the overall reduction in diversity of C4 gene families was not signiﬁcantly diﬀerent from the genome-wide average, indicating that this gene family has not been under particularly strong selection pressure. Similarly, none of the 9 core C4 genes showed a domestication signal at the gene level. The absence of large sequence variation at the gene level is also consistent with previous evolutionary studies suggesting that relatively minor changes to pre-existing regulatory networks and the use of pre-existing cis-elements were often suﬃcient to recruit genes into the C4 pathway [56,57]. The C4 isoform of the NADP-ME gene found in maize and sorghum is one such gene that has been found to be activated for C4 photosynthesis via subtle changes to its promoter, while the rest of the gene is highly conserved [33]. This is consistent with the low diversity in this gene family observed in our study. A further high-resolution investigation of domestication signature at the SNP level revealed 2 C4 genes, PPDK (Sobic.009G132900) and CA (Sobic.003G234200), with SNPs under purifying selection, while the other 2 C4 genes, PPDK-RP (Sobic.002G324400) and PEPC (Sobic.010G160700), were identiﬁed with SNPs under balancing selection. Previous studies have demonstrated that SNP-level analysis using less stringent criteria is superior for capturing soft selection signals compared with genome-wide ranking [54,58]. However, the higher sensitivity may come with a cost of a greater chance of false positives, and therefore requires cautious interpretation. The contrasting selection signals on genes from the same pathway within taxa found in this study was also reported previously in signal transduction pathways [59] and the starch biosynthesis pathway [60]. The C4 isoforms of PPDK and PEPC were also found to show signals of positive selection in a previous cross-species evolutionary study using orthologous groups from closely related C3 and C4 grass species including sorghum [28]. PPDK and PPDK-RP regulate the regeneration of PEP and as such have a direct eﬀect on CO2 assimilation rate [61], especially under cool temperatures [62,63]. However, it is thought that only minor changes to the enzyme properties of PPDK were suﬃcient to recruit it into the C4 pathway and its residues and regions involved in catalyzes are highly conserved in C4 species [64], possibly validating the fact that only soft selection signals via SNP-level were found for the C4 isoform of the PPDK gene in our study. 10 Di i 10. Discussion PEPC is also regarded as a potential limiting step in the assimilation of CO2, and variation of its aﬃnity for CO2/HCO3−amongst species has been documented [65–67]. CA is also critical to C4 photosynthesis as it catalyzes the ﬁrst step of the C4 pathway, converting CO2 to HCO3−[68]. It was reported in the C4 dicot Flaveria bidentis, where antisense plants with <10% of wild-type CA activity required high CO2 for growth and showed reduced CO2 assimilation rates [69,70]. Recent experiments showed CA and PEPC will be more limiting when stomates are partially closed, e.g., under water limitation [71]. The signal of soft purifying selection on PPDK and CA may suggest the C4 pathway was indirectly improved during sorghum domestication. Without photosynthetic rate being a direct selection target in breeding programs, a steady increase in leaf photosynthetic rate over time of cultivar release has been shown in other cereals, e.g., in Australian bread wheat [72]. The balancing selection signal on C4 PPDK-RP and PEPC may reﬂect adaptation to diverse environments, as both PPDK-RP and PEPC are associated with abiotic stress [73,74]. Interestingly, within the PPDK-RP and PPDK gene families, the non-C4 genes all showed selection signals contrasting with their C4 counterparts with both two non-C4 PPDK-RP (Sobic.002G324500, Sobic.002G324700) containing SNPs under purifying selection and the non-C4 PPDK (Sobic.001G326900) containing SNPs under balancing selection. After domestication, sorghum was introduced from tropical to temperate areas, and adapted to divergent local environments. New mutations also arose during this diversiﬁcation process, and played an important role in local adaptation. In the haplotype analysis, these haplotypes unique to cultivated sorghum are likely to be young alleles arising after domestication, while haplotypes unique to the wild progenitor indicate that some haplotypes were lost during domestication of sorghum. Nevertheless, the 11 of 15 Genes 2020, 11, 806 loss of wild haplotypes of C4 genes in cultivated sorghum does not mean these haplotypes are inferior in terms of photosynthetic eﬃciency, as photosynthesis was not speciﬁcally targeted during sorghum domestication [11]. On the contrary, bringing these wild haplotypes back to breeding programs after evaluation of their functions may enrich breeders’ toolkits to manipulate photosynthetic eﬃciency, ultimately contributing to yield improvements. C4 photosynthesis has been well studied over the past 50 years and key components of this complex pathway have been identiﬁed following the advent of transgenic and sequencing technologies [9]. Abbreviations CA, carbonic anhydrase; PEPC, phosphoenolpyruvate carboxylase; PPCK, phosphoenolpyruvate carboxylase kinase; NADP-MDH, NADP-malate dehydrogenase; NADP-ME, NADP-malic enzyme; PPDK, pyruvate orthophosphate dikinase; PPDK-RP, PPDK regulatory protein; RbcS, ribulose bisphosphate carboxylase/oxygenase small-subunit. 10 Di i 10. Discussion Understanding the genetic diversity of the key enzymes of the C4 pathway is an important step towards mining the natural allelic variation for the improvement of photosynthesis. Further investigation of these allelic variation to link them with agronomical traits will provide new targets for sorghum improvement [75]. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4425/11/7/806/s1. Table S1: List of re-sequenced sorghum accessions and their racial and geographic origins. Table S2: List of housekeeping genes and their genetic diversity. Table S3 List of SNPs under purifying selection. Table S4: List of SNPs under balancing selection. Table S5: Inter- and intra-species distribution of private alleles across 27 genes from C4 gene families. Figure S1: Phylogenetic tree of 48 sorghum accessions based on CDS of 27 genes from C4 gene families. Author Contributions: D.J. and E.M. conceived the original idea. Y.T., M.B.-P., S.T., A.C., and E.M. analyzed the data. Y.T. and B.G.-J. write the manuscript. All authors discussed the results and contributed to the ﬁnal manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded partially by the Australian Government through the Australian Research Council Centre of Excellence for Translational Photosynthesis (grant number CE140100015) and State Key Laboratory of Agricultural Genomics, China (grant number 2011DQ782025). Acknowledgments: We thank Susanne von Caemmerer and Robert Furbank for their valuable comments and suggestions towards the improvement of this manuscript. Conﬂicts of Interest: The authors declare that there is no conﬂict of interest. References Long, S.P.; Zhu, X.G.; Naidu, S.L.; Ort, D.R. Can improvement in photosynthesis increase crop yields? Plant Cell Environ. 2006, 29, 315–330. [CrossRef] [PubMed] 12. Zhu, X.G.; Shan, L.L.; Wang, Y.; Quick, W.P. C4 Rice—An ideal arena for systems biology research. J. Integr. Plant Biol. 2010, 52, 762–770. [CrossRef] [PubMed] 13. von Caemmerer, S.; Quick, W.P.; Furbank, R.T. The development of C4 rice: Current progress and future challenges. Science 2012, 336, 1671–1672. [CrossRef] [PubMed] 14. Covshoﬀ, S.; Szecowka, M.; Hughes, T.E.; Smith-Unna, R.; Kelly, S.; Bailey, K.J.; Sage, T.L.; Pachebat, J.A.; Leegood, R.; Hibberd, J.M. C4 Photosynthesis in the Rice Paddy: Insights from the Noxious Weed Echinochloa glabrescens. Plant Physiol. 2016, 170, 57–73. [CrossRef] [PubMed] 15. Hatch, M.D.; Kagawa, T.; Craig, S. Subdivision of C4-pathway species based on diﬀering C4 acid decarboxylating systems and ultrastructural features. Funct. Plant Biol. 1975, 2, 111–128. [CrossRef] 16. Bräutigam, A.; Schliesky, S.; Külahoglu, C.; Osborne, C.P.; Weber, A.P.M. Towards an integrative model of C4 photosynthetic subtypes: Insights from comparative transcriptome analysis of NAD-ME, NADP-ME, and PEP-CK C4 species. J. Exp. Bot. 2014, 65, 3579–3593. [CrossRef] 17. Sonawane, B.V.; Sharwood, R.E.; Whitney, S.; Ghannoum, O. Shade compromises the photosynthetic eﬃciency of NADP-ME less than PEP-CK and NAD-ME C4 Grasses. J. Exp. Bot. 2018, 69, 3053–3068. [CrossRef] [PubMed] 18. Grass Phylogeny Working Group II. New grass phylogeny resolves deep evolutionary relationships and discovers C4 origins. New Phytol. 2012, 193, 304–312. [CrossRef] 19. Kidambi, S.P.; Krieg, D.R.; Rosenow, D.T. Genetic variation for gas-exchange rates in grain sorghum. Plant Physiol. 1990, 92, 1211–1214. [CrossRef] 20. Peng, S.B.; Krieg, D.R. Gas-exchange traits and their relationship to water-use eﬃciency of grain sorghum. Crop Sci. 1992, 32, 386–391. [CrossRef] 21. Henderson, S.; von Caemmerer, S.; Farquhar, G.D.; Wade, L.J.; Hammer, G. Correlation between carbon isotope discrimination and transpiration eﬃciency in lines of the C4 species Sorghum bicolor in the glasshouse and the ﬁeld. Aust. J. Plant Physiol. 1998, 25, 111–123. [CrossRef] 22. Balota, M.; Payne, W.A.; Rooney, W.; Rosenow, D. Gas exchange and transpiration ratio in sorghum. Crop Sci. 2008, 48, 2361–2371. [CrossRef] 23. Fernandez, M.G.S.; Strand, K.; Hamblin, M.T.; Westgate, M.; Heaton, E.; Kresovich, S. Genetic analysis and phenotypic characterization of leaf photosynthetic capacity in a sorghum (Sorghum spp.) diversity panel. Genet. Resour. Crop Ev. 2015, 62, 939–950. [CrossRef] 24. References 1. Sage, R.F.; Sage, T.L.; Kocacinar, F. Photorespiration and the evolution of C4 photosynthesis. Annu. Rev. Plant Biol. 2012, 63, 19–47. [CrossRef] [PubMed] 1. Sage, R.F.; Sage, T.L.; Kocacinar, F. Photorespiration and the evolution of C4 photosynthesis. Annu. Rev. Plant Biol. 2012, 63, 19–47. [CrossRef] [PubMed] 2. Sage, R.F. The evolution of C4 photosynthesis. New Phytol. 2004, 161, 341–370. [CrossRef] 2. Sage, R.F. The evolution of C4 photosynthesis. New Phytol. 2004, 161, 341–370. [CrossRef] 3. Zelitch, I.; Schultes, N.P.; Peterson, R.B.; Brown, P.; Brutnell, T.P. High glycolate oxidase activity is required for survival of maize in normal air. Plant Physiol. 2009, 149, 195–204. [CrossRef] [PubMed] 3. Zelitch, I.; Schultes, N.P.; Peterson, R.B.; Brown, P.; Brutnell, T.P. High glycolate oxidase activity is required for survival of maize in normal air. Plant Physiol. 2009, 149, 195–204. [CrossRef] [PubMed] y 4. Edwards, E.J.; Smith, S.A. Phylogenetic analyses reveal the shady history of C4 grasses. Proc. Natl. Acad. Sci. USA 2010, 107, 2532–2537. [CrossRef] 4. Edwards, E.J.; Smith, S.A. Phylogenetic analyses reveal the shady history of C4 grasses. Proc. Natl. Acad. Sci. USA 2010, 107, 2532–2537. [CrossRef] 5. Hibberd, J.M.; Covshoﬀ, S. The regulation of gene expression required for C4 photosynthesis. Annu. Rev. Plant Biol. 2010, 61, 181–207. [CrossRef] 5. Hibberd, J.M.; Covshoﬀ, S. The regulation of gene expression required for C4 photosynthesis. Annu. Rev. Plant Biol. 2010, 61, 181–207. [CrossRef] 6. von Caemmerer, S.; Furbank, R.T. The C4 pathway: An eﬃcient CO2 pump. Photosynth. Res. 2003, 77, 191. [CrossRef] [PubMed] 6. von Caemmerer, S.; Furbank, R.T. The C4 pathway: An eﬃcient CO2 pump. Photosynth. Res. 2003, 77, 191. [CrossRef] [PubMed] 7. Hibberd, J.M.; Sheehy, J.E.; Langdale, J.A. Using C4 photosynthesis to increase the yield of rice—Rationale and feasibility. Curr. Opin. Plant Biol. 2008, 11, 228–231. [CrossRef] [PubMed] 7. Hibberd, J.M.; Sheehy, J.E.; Langdale, J.A. Using C4 photosynthesis to increase the yield of rice—Rationale and feasibility. Curr. Opin. Plant Biol. 2008, 11, 228–231. [CrossRef] [PubMed] 8. Langdale, J.A. C4 cycles: Past, present, and future research on C4 photosynthesis. Plant Cell 2011, 23, 3879–3892. [CrossRef] [PubMed] 9. von Caemmerer, S.; Furbank, R.T. Strategies for improving C4 photosynthesis. Curr. Opin. Plant Biol. 2016, 31, 125–134. [CrossRef] 10. Still, C.J.; Berry, J.A.; Collatz, G.J.; DeFries, R.S. Global distribution of C3 and C4 vegetation: Carbon cycle implications. Glob. Biogeochem. Cy. 2003, 17, 1006. [CrossRef] 12 of 15 Genes 2020, 11, 806 11. References Evolutionary switch and genetic convergence on rbcL following the evolution of C4 Photosynthesis. Mol. Biol. Evol. 2008, 25, 2361–2368. [CrossRef] 36. Gowik, U.; Bräutigam, A.; Weber, K.L.; Weber, A.P.; Westhoﬀ, P. Evolution of C4 photosynthesis in the genus Flaveria: How many and which genes does it take to make C4? Plant Cell 2011, 23, 2087–2105. [CrossRef] 37. Mallmann, J.; Heckmann, D.; Bräutigam, A.; Lercher, M.J.; Weber, A.P.; Westhoﬀ, P.; Gowik, U. The role of photorespiration during the evolution of C4 photosynthesis in the genus Flaveria. eLife 2014, 3, e02478. [CrossRef] 8. Williams, B.P.; Aubry, S.; Hibberd, J.M. Molecular evolution of genes recruited into C4 photosynthesis. Tr Plant Sci. 2012, 17, 213–220. [CrossRef] 9. Wang, X.; Gowik, U.; Tang, H.; Bowers, J.E.; Westhoﬀ, P.; Paterson, A.H. Comparative genomic analysis o photosynthetic pathway evolution in grasses. Genome Biol. 2009, 10, R68. [CrossRef] 40. Ermakova, M.; Danila, F.R.; Furbank, R.T.; von Caemmerer, S. On the road to C4 rice: Advances and perspectives. Plant J. 2020, 101, 940–950. [CrossRef] 41. Nei, M.; Li, W.-H. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proc. Natl. Acad. Sci. USA 1979, 76, 5269–5273. [CrossRef] [PubMed] 42. Tajima, F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 1989, 123, 585–595. [PubMed] 43. Watterson, G.A. On the number of segregating sites in genetical models without recombination. Theor. Popul. Biol. 1975, 7, 256–276. [CrossRef] 44. Hudson, R.; Boos, D.D.; Kaplan, N. A statistical test for detecting geographic subdivision. Mol. Biol. Evol. 1992, 9, 138–151. [PubMed] 45. Pfeifer, B.; Wittelsbürger, U.; Onsins, S.E.R.; Lercher, M.J. PopGenome: An eﬃcient Swiss army knife for population genomic analyses in R. Mol. Biol. Evol. 2014, 31, 1929–1936. [CrossRef] 46. Paradis, E.; Claude, J.; Strimmer, K.J.B. APE: Analyses of phylogenetics and evolution in R language. Bioinformatics 2004, 20, 289–290. [CrossRef] 47. Paradis, E. Pegas: An R package for population genetics with an integrated–modular approach. Bioinformatics 2010, 26, 419–420. [CrossRef] 48. Ehleringer, J.R.; Sage, R.F.; Flanagan, L.B.; Pearcy, R.W. Climate change and the evolution of C4 photosynthesis. Trends Ecol. Evol. 1991, 6, 95–99. [CrossRef] 49. Christin, P.A.; Salamin, N.; Savolainen, V.; Duvall, M.R.; Besnard, G. C4 Photosynthesis evolved in grasses via parallel adaptive genetic changes. Curr. Biol. CB 2007, 17, 1241–1247. [CrossRef] 50. Christin, P.A.; Samaritani, E.; Petitpierre, B.; Salamin, N.; Besnard, G. Evolutionary insights on C4 photosynthetic subtypes in grasses from genomics and phylogenetics. Genome Biol. Evol. References Zheng, L.Y.; Guo, X.S.; He, B.; Sun, L.J.; Peng, Y.; Dong, S.S.; Liu, T.F.; Jiang, S.Y.; Ramachandran, S.; Liu, C.M.; et al. Genome-wide patterns of genetic variation in sweet and grain sorghum (Sorghum bicolor). Genome Biol. 2011, 12, R114. [CrossRef] [PubMed] 25. Paterson, A.H.; Bowers, J.E.; Bruggmann, R.; Dubchak, I.; Grimwood, J.; Gundlach, H.; Haberer, G.; Hellsten, U.; Mitros, T.; Poliakov, A.; et al. The Sorghum bicolor genome and the diversiﬁcation of grasses. Nature 2009, 457, 551–556. [CrossRef] [PubMed] 26. Mace, E.S.; Tai, S.S.; Gilding, E.K.; Li, Y.H.; Prentis, P.J.; Bian, L.L.; Campbell, B.C.; Hu, W.S.; Innes, D.J.; Han, X.L.; et al. Whole-genome sequencing reveals untapped genetic potential in Africa’s indigenous cereal crop sorghum. Nat. Commun. 2013, 4, 2320. [CrossRef] 27. Fankhauser, N.; Aubry, S. Post-transcriptional regulation of photosynthetic genes is a key driver of ontogeny. J. Exp. Bot. 2017, 68, 137–146. [CrossRef] 28. Huang, P.; Studer, A.J.; Schnable, J.C.; Kellogg, E.A.; Brutnell, T.P. Cross species selection scans identify components of C4 photosynthesis in the grasses. J. Exp. Bot. 2017, 68, 127–135. [CrossRef] 29. Burgess, S.J.; Hibberd, J.M. Insights into C4 metabolism from comparative deep sequencing. Curr. Opin. Plant Biol. 2015, 25, 138–144. [CrossRef] 30. Reeves, G.; Grangé-Guermente, M.J.; Hibberd, J.M. Regulatory gateways for cell-speciﬁc gene expression in C4 leaves with Kranz anatomy. J. Exp. Bot. 2017, 68, 107–116. [CrossRef] 31. Christin, P.-A.; Osborne, C.P.; Chatelet, D.S.; Columbus, J.T.; Besnard, G.; Hodkinson, T.R.; Garrison, L.M.; Vorontsova, M.S.; Edwards, E.J. Anatomical enablers and the evolution of C4 photosynthesis in grasses. Proc. Natl. Acad. Sci. USA 2013, 110, 1381–1386. [CrossRef] 32. Moreno-Villena, J.J.; Dunning, L.T.; Osborne, C.P.; Christin, P.-A. Highly expressed genes are preferentially co-opted for C4 photosynthesis. Mol. Biol. Evol. 2018, 35, 94–106. [CrossRef] Genes 2020, 11, 806 13 of 15 13 of 15 33. Borba, A.R.; Serra, T.S.; Gorska, A.; Gouveia, P.; Cordeiro, A.M.; Reyna-Llorens, I.; Knerova, J.; Barros, P.M.; Abreu, I.A.; Oliveira, M.M.O.; et al. Synergistic binding of bHLH transcription factors to the promoter of the maize NADP-ME gene used in C4 photosynthesis is based on an ancient code found in the ancestral C3 state. Mol. Biol. Evol. 2018, 35, 1690–1705. [CrossRef] 34. Christin, P.-A.; Petitpierre, B.; Salamin, N.; Büchi, L.; Besnard, G. Evolution of C4 phosphoenolpyruvate carboxykinase in grasses, from genotype to phenotype. Mol. Biol. Evol. 2008, 26, 357–365. [CrossRef] 35. Christin, P.-A.; Salamin, N.; Muasya, A.M.; Roalson, E.H.; Russier, F.; Besnard, G. References 2009, 1, 221–230. [CrossRef] 1. Clark, J.D.; Stemler, A. Early domesticated sorghum from central Sudan. Nature 1975, 254, 588–591. [Cross 51. Clark, J.D.; Stemler, A. Early domesticated sorghum from central Sudan. Nature 1975, 254, 588–591. [CrossRef] 52. Mann, J.A.; Kimber, C.T.; Miller, F.R. The Origin and Early Cultivation of Sorghums in Africa; Bulletin 1454; Texas Agricultural Experiment Station: College Station, TX, USA, 1983. 2. Mann, J.A.; Kimber, C.T.; Miller, F.R. The Origin and Early Cultivation of Sorghums in Africa; Bulletin 1 Texas Agricultural Experiment Station: College Station, TX, USA, 1983. 53. Wendorf, F.; Close, A.E.; Schild, R.; Wasylikowa, K.; Housley, R.A.; Harlan, J.R.; Królik, H. Saharan exploitation of plants 8,000 years BP. Nature 1992, 359, 721–724. [CrossRef] 54. Tao, Y.; Mace, E.S.; Tai, S.; Cruickshank, A.; Campbell, B.C.; Zhao, X.; Van Oosterom, E.J.; Godwin, I.D.; Botella, J.R.; Jordan, D.R. Whole-genome analysis of candidate genes associated with seed size and weight in Sorghum bicolor reveals signatures of artiﬁcial selection and insights into parallel domestication in cereal crops. Front. Plant Sci. 2017, 8, 1237. [CrossRef] 55. Tao, Y.; Zhao, X.; Mace, E.; Henry, R.; Jordan, D. Exploring and exploiting pan-genomics for crop improvement. Mol. Plant 2019, 12, 156–169. [CrossRef] 14 of 15 Genes 2020, 11, 806 56. Kümpers, B.M.C.; Burgess, S.J.; Reyna-Llorens, I.; Smith-Unna, R.; Boursnell, C.; Hibberd, J.M. Shared characteristics underpinning C4 leaf maturation derived from analysis of multiple C3 and C4 species of Flaveria. J. Exp. Bot. 2017, 68, 177–189. [CrossRef] 57. Külahoglu, C.; Denton, A.K.; Sommer, M.; Maß, J.; Schliesky, S.; Wrobel, T.J.; Berckmans, B.; Gongora-Castillo, E.; Buell, C.R.; Simon, R.; et al. Comparative transcriptome atlases reveal altered gene expression modules between two Cleomaceae C3 and C4 plant species. Plant Cell 2014, 26, 3243–3260. [CrossRef] 58. Massel, K.; Campbell, B.C.; Mace, E.S.; Tai, S.; Tao, Y.; Worland, B.G.; Jordan, D.R.; Botella, J.R.; Godwin, I.D. Whole genome sequencing reveals potential new targets for improving nitrogen uptake and utilization in Sorghum bicolor. Front. Plant Sci. 2016, 7, 1544. [CrossRef] 59. Riley, R.M.; Jin, W.; Gibson, G. Contrasting selection pressures on components of the Ras-mediated signal transduction pathway in Drosophila. Mol. Ecol. 2003, 12, 1315–1323. [CrossRef] 60. Campbell, B.C.; Gilding, E.K.; Mace, E.S.; Tai, S.; Tao, Y.; Prentis, P.J.; Thomelin, P.; Jordan, D.R.; Godwin, I.D. Domestication and the storage starch biosynthesis pathway: Signatures of selection from a whole sorghum genome sequencing strategy. Plant Biotechnol. J. 2016, 14, 2240–2253. [CrossRef] 61. References Wang, Y.-M.; Xu, W.-G.; Hu, L.; Zhang, L.; Li, Y.; Du, X.-H. Expression of maize gene encoding C4-pyruvate orthophosphate dikinase (PPDK) and C4-phosphoenolpyruvate carboxylase (PEPC) in transgenic Arabidopsis. Plant Mol. Biol. Rep. 2012, 30, 1367–1374. [CrossRef] 62. Wang, D.; Portis, A.R.; Moose, S.P.; Long, S.P. Cool C4 photosynthesis: Pyruvate Pi dikinase expression and activity corresponds to the exceptional cold tolerance of carbon assimilation in Miscanthus × giganteus. Plant Physiol. 2008, 148, 557–567. [CrossRef] 63. Naidu, S.L.; Moose, S.P.; AL-Shoaibi, A.K.; Raines, C.A.; Long, S.P. Cold Tolerance of C4 photosynthesis in Miscanthus × giganteus: Adaptation in amounts and sequence of C4 photosynthetic enzymes. Plant Physiol. 2003, 132, 1688–1697. [CrossRef] 64. Chastain, C.J.; Failing, C.J.; Manandhar, L.; Zimmerman, M.A.; Lakner, M.M.; Nguyen, T.H.T. Functional evolution of C4 pyruvate, orthophosphate dikinase. J. Exp. Bot. 2011, 62, 3083–3091. [CrossRef] 65. Bauwe, H.; Chollet, R. Kinetic properties of phosphoenolpyruvate carboxylase from C3, C4, and C3-C4 intermediate species of Flaveria (Asteraceae). Plant Physiol. 1986, 82, 695–699. [CrossRef] 66. von Caemmerer, S.; Edwards, G.E.; Koteyeva, N.; Cousins, A.B. Single cell C4 photosynthesis in aquatic and terrestrial plants: A gas exchange perspective. Aquat. Bot. 2014, 118, 71–80. [CrossRef] 67. Boyd, R.A.; Gandin, A.; Cousins, A.B. Temperature responses of C4 photosynthesis: Biochemical analysis of rubisco, phosphoenolpyruvate carboxylase, and carbonic anhydrase in Setaria viridis. Plant Physiol. 2015, 169, 1850–1861. [CrossRef] [PubMed] 68. Hatch, M.D.; Burnell, J.N. Carbonic anhydrase activity in leaves and its role in the ﬁrst step of C4 photosynthesis. Plant Physiol. 1990, 93, 825–828. [CrossRef] 69. von Caemmerer, S.; Quinn, V.; Hancock, N.; Price, G.D.; Furbank, R.T.; Ludwig, M. Carbonic anhydrase and C4 photosynthesis: A transgenic analysis. Plant Cell Environ. 2004, 27, 697–703. [CrossRef] 70. Cousins, A.B.; Badger, M.R.; von Caemmerer, S. Carbonic anhydrase and its inﬂuence on carbon isotope discrimination during C4 photosynthesis. Insights from antisense RNA in Flaveria bidentis. Plant Physiol. 2006, 141, 232–242. [CrossRef] 71. Osborn, H.L.; Alonso-Cantabrana, H.; Sharwood, R.E.; Covshoﬀ, S.; Evans, J.R.; Furbank, R.T.; von Caemmerer, S. Eﬀects of reduced carbonic anhydrase activity on CO2 assimilation rates in Setaria viridis: A transgenic analysis. J. Exp. Bot. 2016, 68, 299–310. [CrossRef] 72. Watanabe, N.; Evans, J.R.; Chow, W.S. Changes in the Photosynthetic Properties of Australian wheat cultivars over the last century. Aust. J. Plant Physiol. 1994, 21, 169–183. [CrossRef] 73. Liu, X.; Li, X.; Zhang, C.; Dai, C.; Zhou, J.; Ren, C.; Zhang, J. © 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/). 75. Tao, Y.; Zhao, X.; Wang, X.; Hathorn, A.; Hunt, C.; Cruickshank, A.W.; van Oosterom, E.J.; Godwin, I.D.; Mace, E.S.; Jordan, D.R. Large-scale GWAS in sorghum reveals common genetic control of grain size among cereals. Plant Biotechnol. J. 2020, 18, 1093–1105. [CrossRef] 74. Jeanneau, M.; Gerentes, D.; Foueillassar, X.; Zivy, M.; Vidal, J.; Toppan, A.; Perez, P. Improvement of drought tolerance in maize: Towards the functional validation of the Zm-Asr1 gene and increase of water use eﬃciency by over-expressing C4–PEPC. Biochimie 2002, 84, 1127–1135. [CrossRef] References Phosphoenolpyruvate carboxylase regulation in C4-PEPC-expressing transgenic rice during early responses to drought stress. Physiol. Plant. 2017, 159, 178–200. [CrossRef] [PubMed] Genes 2020, 11, 806 15 of 15 74. Jeanneau, M.; Gerentes, D.; Foueillassar, X.; Zivy, M.; Vidal, J.; Toppan, A.; Perez, P. Improvement of drought tolerance in maize: Towards the functional validation of the Zm-Asr1 gene and increase of water use eﬃciency by over-expressing C4–PEPC. Biochimie 2002, 84, 1127–1135. [CrossRef] 75. Tao, Y.; Zhao, X.; Wang, X.; Hathorn, A.; Hunt, C.; Cruickshank, A.W.; van Oosterom, E.J.; Godwin, I.D.; Mace, E.S.; Jordan, D.R. Large-scale GWAS in sorghum reveals common genetic control of grain size among cereals. Plant Biotechnol. J. 2020, 18, 1093–1105. [CrossRef]
https://openalex.org/W2103413680	https://zenodo.org/records/575950/files/BD_article_8925.pdf	English	null	Viability of unstimulated lymphocytes exposed to extremely low frequency electromagnetic fields is dependent on intensity	BioDiscovery	2,012	cc-by	8,383	Abstract The cell viability and DNA damage in unstimulated sheep primary lymphocytes subjected to different extremely low electromagnetic field intensities (5, 50 and 100 µT; 50 Hz) were studied with special emphasis on apoptosis. Sheep primary lymphocytes cultured in RPMI, supplemented with 10% FBS in the absence of mitogens, were exposed till 16 h. The cell viability assessment by 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl tetrazolium bromide (MTT) assay showed a dose dependent enhancement of viability at 16 h. Further, quantitative DNA laddering and flow cytometric analysis showed a significant decrease in apoptosis of the cells subjected to 100 (p<0.01) and 50 µT (p<0.05) for 16 h as compared to control group. There was a statistically significant decrease (p<0.01) in the specific activity of caspase 9 at 100 µT in cells exposed for 16 h. However, no enhancement of DNA damage was observed at 5, 50 and 100 µT as evidenced by comet assay. Comet assay also confirmed the decreased cell death of exposed cells (100 µT). Experimental data suggests decreased apoptosis at 100 µT (50 Hz), possibly by the suppression of caspase 9 activity leading to the enhanced cell viability. Citation: Rajendra P, Sujatha HN, Sashidhar RB, Subramanyam C, Devendranath D, Aradhya RSS. Extremely Low Frequency Electromagnetic Fields is Dependent on Intensity. Biodiscovery 2012; 2: 2; DOI: 10.7750/BioDiscovery.2012.2.2 Copyright: © 2012 : Rajendra et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, provided the original authors and source are credited. Received: 15 June 2012; Accepted: 2 August 2012; Available online /Published: 22 August 2012 Received: 15 June 2012; Accepted: 2 August 2012; Available online /Published: 22 August 2012 Received: 15 June 2012; Accepted: 2 August 2012; Available online /Published: 22 August 2012 Keywords: ELF-EMF, primary lymphocytes, apoptosis, DNA damage, caspase 9 * Corresponding Author: Sashidhar RB; email: sashi_rao@yahoo.com Conflict of Interests: The authors have no competing interests in relation to the work described. Keywords: ELF-EMF, primary lymphocytes, apoptosis, DNA damage, caspase 9 * Corresponding Author: Sashidhar RB; email: sashi_rao@yahoo.com Conflict of Interests: The authors have no competing interests in relation to the work described. RESEARCH ARTICLE RESEARCH ARTICLE RESEARCH ARTICLE RESEARCH ARTICLE BioDiscovery BioDiscovery Rajendra P1, Sujatha HN1, Sashidhar RB1, Subramanyam C1, Devendranath D2, Aradhya RSS2 1 Department of Biochemistry, University College of Science, Osmania University, Hyderabad - 500 007, Andhra Pradesh, India. 2 Central Power Research Institute, Ultra High Voltage Research Laboratory, Hyderabad-500 039, Andhra Pradesh, India. Viability of unstimulated lymphocytes exposed to extremely low frequency electromagnetic fields is dependent on intensity Rajendra P1, Sujatha HN1, Sashidhar RB1, Subramanyam C1, Devendranath D2, Aradhya RSS2 ndra P1, Sujatha HN1, Sashidhar RB1*, Subramanyam C1, Devendranath D2 hya RSS2 Introduction leukemia. The present study on cellular models was mainly aimed to obtain a relationship between ELF-EMF and biological effects. The biological effects caused by Extremely Low Frequency Electromagnetic fields (ELF-EMF) have been undergoing serious debate since the last two decades. Several studies have been conducted to asses the effect of ELF-EMF, during this period using different biological models - animal model [1] cellular model [2], developing embryos [3] and plants [4] so as to explain the possible mechanism [5]. The first pioneering epidemiological study by Werthiemer and Leeper [6] put forward a relationship between EMF exposure and childhood However, the available biological data on the oncogenic potential of ELF-EMF are unable to provide a cause-effect relationship between exposure conditions and induction of tumors. In this context, the question of possible genotoxic effects is particularly relevant. It has been reported that EMF leads to chromosomal breaks, changes in direction, formation and functionality of mitotic spindle and may have genotoxic effects [7] in various cells. There are also other contradictory studies, BioDiscovery \| www.biodiscoveryjournal.co.uk BioDiscovery \| www.biodiscoveryjournal.co.uk 1 August 2012 \| Issue 2 \| 2 Effect of ELF-EMF on primary lymphocytes which shows that EMF may not have any influence on frequencies of sister chromatid exchanges [8, 9], cell cycle progression or chromosomal breakage [10] in lymphocytes. There are several reports indicating possible induction of mutation, following EMF exposure. Single strand DNA breaks have been reported in Molt 4 T-lymphoblastoid cells [11]. A dose dependent increase in DNA single strand breaks was observed in human leukocytes exposed to 50 Hz magnetic fields [12]. The latest published data provided evidence suggesting that ELF-EMF interfere with the DNA repair process leading to accumulation of damaged DNA in cells [13]. It was also stated by Philip et al., 1996 [11], that exposure of Molt 4 T- lymphoblastoid cells to a 100 µT, 60 Hz sinusoidal magnetic fields leads to decreased activity of the DNA repair enzyme, poly (ADP ribose) polymerase, and increased DNA fragmentation in the presence of the DNA damaging agent, etoposide. However, there are negative reports in the literature [14, 15, 16] that indicates no detrimental effect of ELF-EMF on cellular DNA. Influence of magnetic fields on viability of lymphocytes The lymphocyte viability based on MTT assay at different magnetic fields is given in Figure 1. It was observed that upon exposure to 5, 50 and 100 µT for 16 h, there was a dose dependent enhancement in the viability of lymphocytes with reference to control group. Statistically significant (p<0.05) increase in the viability of lymphocytes subjected to 100 µT for 16 h as compared to control group was observed (Figure 1). Many other studies have also shown an increase in proliferation after exposure to EMF in human fibroblasts [19] and lymphocytes [9, 20, 21, 22]. The exposure of 60 Hz magnetic fields (1.0, 1.5 and 2.0 mT) on human lymphocytes was found to modify the cell proliferation and had no influence on the frequency of sister chromatid exchanges [23]. Similarly, ELF-EMF was found to induce partial blockage for the differentiation of Friend erythroleukemia cells, which results in a larger population of cells remaining undifferentiated in the proliferative state suggesting that ELF-EMF may act as a tumor promoter [24, 25]. Figure 1. Effect of magnetic field (5, 50 and 100 µT) on lymphocyte viability. Earlier studies have reported an increase in apoptosis in human myelogenous leukemic cell lines, HL-60 and ML-1 [2, 26], in thymocytes and macrophages of mice exposed to long term ELF-EMF [27, 28]. In macrophages of mice, the increase in apoptosis was accompanied by increased intracellular calcium levels [27]. Similarly, Ca2+ movement was induced in U-373 MG cells by ELF-EMF (50 Hz, 3mT), however there was neither cell proliferation nor cell death [29]. Thymocytes from mice subjected to a long term 0.4 to 1.0 mT, 60 Hz, showed an abnormal response to dexamethazone induced apoptotic stimuli [28]. In contrast, the static and ELF magnetic fields possessed an anticancer activity [30]. Similar effect was observed in three different human tumor cell lines [HTB63 (melanoma), HTB77 IP3 (ovarian carcinoma) and CCL86 (lymphoma)] subjected to prolonged exposure of Figure 1. Effect of magnetic field (5, 50 and 100 µT) on lymphocyte viability. Introduction The studies conducted in vitro and on cell free DNA mismatch repair revealed that the core processes related to the transmission of genetic information (DNA polymerase activity, transcription and repair) are stable under ELF-EMF (60 Hz, at 0.25 to 0.5 T) [17]. Similarly, environmental level magnetic fields (10, 50 and 100 µT) had no effect on in vitro transcription in a cell free system [18]. static magnetic field [31]. The continuous exposure to 1.2 µT, sinusoidal 60 Hz magnetic field was found to influence chemically induced proliferation control processes in breast cancer cells (MCF-7) in vitro [32]. static magnetic field [31]. The continuous exposure to 1.2 µT, sinusoidal 60 Hz magnetic field was found to influence chemically induced proliferation control processes in breast cancer cells (MCF-7) in vitro [32]. In view of the mutually contradictory reports on the oncogenic or proliferative effect of ELF- EMF on cellular models, the effect of ELF-EMF on cell viability and DNA damage in primary lymphocytes (non- transformed) subjected to ELF-EMF with special emphasis on apoptosis was investigated. The magnetic field dosage (5, 50 and 100 µT) was selected based on the national survey of magnetic fields conducted by Central Power Research Institute, Ultra High Voltage Research Laboratory, Hyderabad, India (unpublished data; CPRI report No.11.1.11 2000). In view of the mutually contradictory reports on the oncogenic or proliferative effect of ELF- EMF on cellular models, the effect of ELF-EMF on cell viability and DNA damage in primary lymphocytes (non- transformed) subjected to ELF-EMF with special emphasis on apoptosis was investigated. The magnetic field dosage (5, 50 and 100 µT) was selected based on the national survey of magnetic fields conducted by Central Power Research Institute, Ultra High Voltage Research Laboratory, Hyderabad, India (unpublished data; CPRI report No.11.1.11 2000). DNA damage analysis by comet assay The percent lymphocytes, categorized as "no tail", "nominal tail" and "longer tail" along with positive control (Figure 2a) based on comet assay are given in Figure 2b. There was a statistically significant (p<0.01) decrease in the percent of "nominal tail" lymphocytes subjected to 100 µT for 16 h (Figure 2b). There was also a gradual increase in the percent of "no tail" lymphocytes subjected to 5, 50 and 100 µT for 16 h in comparison with the control group. BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 2 Effect of ELF-EMF on primary lymphocytes A Figure 2. A. Categorization of cells based on comet tail length [a: positive control (H2O2 treated); b: nominal tail; c: no tail; d: longer tail]. B. Depending on the tail length, the cells were categorized into nominal and long tail by scoring 100 cells. Cells, without tail were scored separately. B B B A B Figure 2. A. Categorization of cells based on comet tail length [a: positive control (H2O2 treated); b: nominal tail; c: no tail; d: longer tail]. B. Depending on the tail length, the cells were categorized into nominal and long tail by scoring 100 cells. Cells, without tail were scored separately. B Figure 3. A. DNA laddering analysis (Lane 1 and 2 represents samples in duplicate) showing distinct band of High Molecular Weight (HMW) DNA from viable cells and 200 bp DNA band from dead cells. B. Effect of magnetic fields (5, 50 and 100 µT) on % Fluorescent Intensity (FI) ratio of ‘B’ (200 bp DNA) / ‘A’ (HMW DNA). A statistically significant (p<0.01) decrease in the ratio in cells subjected to 100 µT for 16 h, compared to control group was seen. A B A B A Figure 3. A. DNA laddering analysis (Lane 1 and 2 represents samples in duplicate) showing distinct band of High Molecular Weight (HMW) DNA from viable cells and 200 bp DNA band from dead cells. B. Effect of magnetic fields (5, 50 and 100 µT) on % Fluorescent Intensity (FI) ratio of ‘B’ (200 bp DNA) / ‘A’ (HMW DNA). A statistically significant (p<0.01) decrease in the ratio in cells subjected to 100 µT for 16 h, compared to control group was seen. Influence of magnetic fields on cell death (DNA laddering analysis) A. Flow Cytometric analysis of apoptosis. B. Flow cytometric quantitation of apoptotic cells was carried out using Annexin V-Biotin Apoptosis Detection kit, which uses Annexin V- Biotin conjugate for detection of cell surface changes that occur early in the apoptotic process Figure 5. Spectrofluorometric analysis of caspase 9 activity. The caspase 9 activity was expressed as FI/mg protein by estimating the protein content in the cell lysate using detergent friendly BCA method. Figure 5. Spectrofluorometric analysis of caspase 9 activity. The caspase 9 activity was expressed as FI/mg protein by estimating the protein content in the cell lysate using detergent friendly BCA method. Influence of magnetic fields on apoptosis of lymphocytes (caspase 9 activity) Influence of magnetic fields on cell death (DNA laddering analysis) DNA laddering analysis of lymphocytes showed distinct bands of high molecular weight (HMW) and 200 bp DNA (Figure 3a). The laddering pattern was found to be similar in both control and exposed groups of cells. However, the comparison of percent fluorescent intensity ratio {[FI of 200 bp DNA (B) / FI of HMW DNA (A)] x 100} revealed that there was a statistically significant (p<0.01) decrease in the ratio in cells subjected to 100 µT for 16 h (Figure 3b), whereas, the cells subjected to 5 and 50 µT showed no effect in the ratio as compared to control group. Annexin V stained lymphocytes, subjected to different magnetic fields (5, 50 and 100 µT) for 16 h are represented by histogram obtained by Fluorescence activated cell sorting (FACS) analysis (Figure 4a). The histogram indicates the reduction of peak size corresponding to Annexin V- FITC bound cells in magnetic field exposed groups (50 and 100 µT) as compared to the control group. The quantitative analysis of percent of Annexin V - FITC bound cells (apoptotic cells) showed a statistically significant (p<0.05 and p<0.01) reduction of apoptotic cells at 50 and 100 µT BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 3 Effect of ELF-EMF on primary lymphocytes respectively as compared to control group However Figure 4. A. Flow Cytometric analysis of apoptosis. B. Flow cytometric quantitation of apoptotic cells was carried out using Annexin V-Biotin Apoptosis Detection kit, which uses Annexin V- Biotin conjugate for detection of cell surface changes that occur early in the apoptotic process Figure 5. Spectrofluorometric analysis of caspase 9 activity. The caspase 9 activity was expressed as FI/mg protein by estimating the protein content in the cell lysate using detergent friendly BCA method. Influence of magnetic fields on apoptosis of A B Effect of ELF-EMF on primary lymphocytes Figure 4. A. Flow Cytometric analysis of apoptosis. B. Flow cytometric quantitation of apoptotic cells was carried out using Annexin V-Biotin Apoptosis Detection kit, which uses Annexin V- Biotin conjugate for detection of cell surface changes that occur early in the apoptotic process A B A A A Figure 4. A. Flow Cytometric analysis of apoptosis. B. Flow cytometric quantitation of apoptotic cells was carried out using Annexin V-Biotin Apoptosis Detection kit, which uses Annexin V- Biotin conjugate for detection of cell surface changes that occur early in the apoptotic process B B B Figure 4. Influence of magnetic fields on apoptosis of lymphocytes (caspase 9 activity) respectively as compared to control group. However, there were no changes in the percent of apoptotic cells at 5 µT (Figure 4b). This clearly gives a quantitative picture of reduction of apoptosis in primary lymphocytes subjected to EMF stress in the absence of any phytostimulator. The specific activity of caspase 9 in lymphocytes exposed to different magnetic field intensities (5, 50 and 100 T) for 16 h are given in Figure 5. It is evident that there was a statistically significant decrease (p<0.01) in BioDiscovery \| www.biodiscoveryjournal.co.uk Discussion There are numerous reports, which indicate that power frequency (50 Hz) magnetic fields induce responses at cellular level, particularly in peripheral lymphocytes. Peripheral lymphocytes have been the cellular model of choice because they are well characterized cells as far as culture conditions and growth factor requirements are concerned. Lymphocytes are usually obtained as resting cells from peripheral blood, but can be induced to enter the cell cycle using mitogens, thus mimicking a naturally occurring activation process [22]. The mitogenic response in lymphocytes was reported to be stimulated in the presence of ELF-EMF in vitro [33, 34, 21]. However, there are other reports, which indicate that the mitogenic response in lymphocytes was inhibited [35, 36] or not influenced [10]. There are very few reports in contemporary literature on the effect of ELF-EMF on the unstimulated cells [37, 38]. Vincze et al., 2005 [39] suggested based on theoretical calculations that external low frequency electromagnetic fields (LFEMF) can induce a distribution of field strengths along the cell membrane. This leads to thermal gradients (currents) from the extracellular matrix towards the inside of the cell. This thermal current also carries ions through, leading to thermo-diffusion and thus creating a zero mode electric current, which in turn induces a zero mode electric field in the cell membrane. Therefore, even small fields with zero mode components could elicit biological effects. Further, it has been shown in the previous investigations, that ELF-MF exposure affected the conformation of chromatin in E. coli and human lymphocytes, when used as test systems. ELF-MF, under specific conditions of exposure, acted as a non-toxic but cell-growth stimulating agent, when conformation of chromatin was studied by the method of anomalous viscosity time dependencies (AVTD) method [40, 41]. The present experimental studies were focused mainly to address the biological effects of ELF-EMF (50 Hz) generated during the course of high voltage power transmission. Hence, the study was conducted at very low dosage of magnetic fields (5, 50 and 100 T; 50 Hz). li bli h d h l h l Since MTT and comet assay strongly suggests the possibility of repression of cell death and there by gradual dose dependent enhancement of viability at 5, 50 and 100 T, the present study was mainly focused on apoptosis of lymphocytes under EMF stress. The cellular responses to stress can range from adaptive response to cell death. BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 4 Effect of ELF-EMF on primary lymphocytes have been exposed to magnetic fields for a period of 16 h in the absence of mitogens and a statistically significant increase (p<0.05) in the cell viability was observed at 100 T. Cells subjected to 5 and 50 T magnetic fields also showed a trend in increased cell viability, even though it was statistically not significant (Figure 1). This observation indicates that EMF can enhance the viability of lymphocytes in the absence of mitogens also. As a supporting evidence, comet assay showed a gradual increase of percent of cells with “no tail" (viable cells) at 5, 50 and 100 T, even though, the variation was statistically not significant. It was quite interesting to note that there was a statistically significant decrease (p<0.01) in the percent of lymphocytes having "nominal tail" at 100 T, subjected for 16 h (Figure 2b). This indicates that the percent of cells that undergoes natural death (possibly by apoptosis) was found to be decreased under 100 T. Since there were no mitogens in the medium the chance of cell proliferation under this condition is very low. In this context, the increase in the cell viability at 100 T assessed by MTT assay may be mainly due to the enhancement of life of lymphocytes by suppressing the natural death. The same effect might be the reason for the co-stimulatory activity of EMF on lymphocytes in presence of mitogens [22]. The comet assay analysis also indicates that there are no chances of DNA damage due to magnetic field exposure up to 100 T. This observation is supported by the generally accepted fact that EMF has no direct clastogenic effect on cellular DNA [42]. However, there are several reports indicating possible mutation induction following EMF exposure. Single strand DNA breaks have been reported [11]. A dose dependent increase in DNA single strand breaks was observed in human leukocytes exposed to 50 Hz magnetic fields [12]. It was also stated [11] that exposure of Molt 4 T- lymphoblastoid cells to a 100 T, 60 Hz sinusoidal magnetic fields leads to decreased activity of the DNA repair enzyme, poly (ADP-ribose) polymerase, and increased DNA fragmentation in the presence of the DNA damaging agent, etoposide. BioDiscovery \| www.biodiscoveryjournal.co.uk Possibly, these alterations in cellular DNA under ELF- EMF may be due to higher dosages of magnetic fields than the one that exists in the normal routine life, in the vicinity of biological system Since MTT and comet assay strongly suggests the specific activity of caspase 9 in lymphocytes subjected to 100 T as compared to control group. The specific activity of caspase 9 of lymphocytes subjected to 5 and 50 T did not show any change as compared to control group. BioDiscovery \| www.biodiscoveryjournal.co.uk Discussion One of the classical adaptive responses involves the induction or activation of highly conserved proteins called stress or heat shock proteins (HSPs). The major HSPs of mammalian cells include proteins of 110, 90, 70, 60, 40 and 27 kDa. Some HSP members are It was earlier established that lymphocyte cultures should be exposed for a minimum period of 6 h [22] because all processes related to cell proliferation such as RNA synthesis, gene expression, and DNA synthesis occur within 6 h from the binding of the mitogen to the cell membrane. In the present study the lymphocytes BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 5 Effect of ELF-EMF on primary lymphocytes outer surface of the cells [48]. The primary lymphocytes subjected to 50 and 100 T for 16 h showed a statistically significant reduction (p<0.05 and p<0.01 respectively) in Annexin V bound cell population as compared to control group in a dose dependent manner (Figure 4a and 4b). However, there was no change in lymphocytes subjected to 5 T. This provides a clear quantitative picture of reduction of apoptosis in primary lymphocytes subjected to EMF stress in the absence of any phytostimulator. constitutively expressed whereas others are expressed only after a period of stress (e.g., rise in temperature, exposure to radiation, viral infection etc.). Induction of these proteins in response to stress confers resistance to subsequent stress (thermo tolerance). This resistance is due to the inhibition of apoptosis [42]. The studies conducted on human keratinocytes subjected to 100 T showed that power line electromagnetic fields cannot induce changes in phosphorylation, localization or expression of the HSP 27 [43, 44]. However, the cultures of human endothelial cell lines exposed to mobile phone radiation (RF-EMF) showed an increased expression and phosphorylation of HSP 27. Phosphorylated HSP27 has been shown to inhibit apoptosis by forming complex with apoptosome (complex of Apaf-1 protein, pro-caspase 9 and cytochrome C), or some of its components, and preventing proteolytic activation of pro-caspase 9 into active form of caspase 9. This in turn prevents activation of pro-caspase 3 which is activated by caspase 9, leading to the inhibition of apoptosis [45, 46].Apoptosis is a highly regulated form of cell death. It is a fundamental physiological process crucial to maintaining homeostasis in multicellular organisms acting as a counterbalance to cell division. Discussion All animal cells programmed to undergo apoptosis when they cease to function, are no longer needed or are damaged. Inappropriate apoptosis occurs in certain degenerative diseases, while a reduced level of apoptosis is frequently associated with the development of cancer [47].One of the classic features of apoptosis is the cleavage of genomic DNA into oligo-nucleosomal fragments represented by multiples of 180 - 200 bp. The DNA band corresponding to 200 bp represents the dead cell fraction whereas high molecular weight fraction present on the top of the gel represents live population of cells [48]. The present study showed a statistically significant (p<0.01) reduction in the intensity of 200 bp DNA band in lymphocytes exposed to 100 T as compared to control group (Figure 3a), whereas, the cells subjected to 5 and 50 T remained unaffected. This indicates a reduction in apoptosis in lymphocytes exposed to 100 T for 16 h. Further, this observation supports the data obtained by MTT assay and comet assay since a reduction in the apoptosis under EMF stress in a fixed population of cells will result in enhanced viability as compared to the control group. Apoptosis is characterized by cell shrinkage, nuclear condensation and oligonucleosomal DNA fragmentation. The biochemical basis for these morphological features of apoptosis can also be the action of a family of proteases called caspases. Mitochondria plays a key role in activating caspases by releasing cytochrome C into cytosol where it binds to Apaf-1, facilitating pro-caspase 9 processing, followed by caspase 9 mediated activation of pro-caspase 3 [49]. Active caspase 3 promotes the manifestation of some of the more classical features of apoptosis. Caspase 9 is one of the stress induced caspases involved in the intrinsic mechanism of apoptotic cell death [50]. In this context, the above observation was further subjected to biochemical investigations by assaying the activity of caspase 9 in primary lymphocytic cell lysate after subjecting to different magnetic field intensities (5, 50 and 100 T) for 16 h. The study showed a statistically significant reduction (p<0.01) in the specific activity of caspase 9 in cells exposed to 100 T as compared to the control group (Figure 5). Similarly, there was a minor (non-significant) reduction in the caspase 9 specific activity in lymphocytes subjected to 5 and 50 T. BioDiscovery \| www.biodiscoveryjournal.co.uk Discussion p y y p y j  The reduction in the caspase 9 specific activities under EMF stress further enriches the above observed fact that EMF inhibits apoptosis in primary lymphocytes. The reduction of apoptosis observed in the present study might be a possible reason for the earlier observation where, the proliferation of human peripheral lymphocytes subjected to EMF (50 Hz) was found to be stimulated in the presence of mitogens [33]. Since the molecular mechanism involved in the different type of cellular responses under the influence of ELF-EMF is not yet clear it is difficult to explain the exact mechanism by which apoptosis is regulated under ELF-EMF in the present context. However, the present investigation suggests that the reduction in apoptosis is possibly achieved by inhibiting the activation of caspase 9 by ELF-EMF. The existing literature strongly supports this hypothesis because i) Caspase 9 is one of the stress induced caspases [50] ii) Caspase 9 activity is regulated by phosphorylation. The pro-caspase 9 becomes inhibited upon phosphorylation at Ser 196. The phosphorylation of pro-caspase 9 is mediated by PI3 Kinase/ Akt pathway [51]. Possibly, the biophysical changes of cell membrane in the presence of 50 Hz EMF [52] might increase the Since DNA fragmentation is not a definitive test for apoptosis the observation was further subjected to clarification by conducting the flow cytometric analysis of Annexin V staining of apoptotic cells. Annexin V is a protein that preferentially binds phosphtidyl serine (PS) in a calcium dependent manner. In normal, non-apoptotic cells, PS is segregated to the inner leaflet of the plasma membrane. During early stages of apoptosis, this asymmetry collapses and PS becomes exposed on the BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 6 Effect of ELF-EMF on primary lymphocytes Effect of ELF-EMF on primary lym e 6. Proposed hypothetical mechanism of inhibition of apoptosis by repression of caspase 9 in primary lymphocytes exposed to 50 based on the present experimental investigation and contemporary literature [51, 52, 58]. The biophysical changes of cell me nce of 50 Hz EMF might increase the cytosolic Ca2+ concentration leading to the activation of PI3 kinase leading to the rep otic pathway initiated by pro-caspase 9. Figure 6. Discussion Proposed hypothetical mechanism of inhibition of apoptosis by repression of caspase 9 in primary lymphocytes exposed to 50 Hz ELF- EMF based on the present experimental investigation and contemporary literature [51, 52, 58]. The biophysical changes of cell membrane in presence of 50 Hz EMF might increase the cytosolic Ca2+ concentration leading to the activation of PI3 kinase leading to the repression of apoptotic pathway initiated by pro-caspase 9. BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 7 7 Effect of ELF-EMF on primary lymphocytes cytosolic Ca2+ concentration [53, 54, 55] leading to the activation of PI3 kinase which in turn leads to the repression of apoptotic pathway initiated by pro-caspase 9 as shown in Figure 6. Basically, PI3K/Akt pathways function to promote cellular survival following stress. The present hypothesis is further supported by a recent study on human keratinocytes subjected to UV irradiation, which resulted in the inhibition of caspase 9 by PI3K/Akt pathway [56]. Similarly, exposure of B- lineage lymphoid cells to ELF-EMF stimulated the protein tyrosine kinases (Lyn kinase), that plays pivotal and myriad roles in initiation of signal cascades that affect proliferation and survival of B-lineage lymphoid cells [57]. A possible mechanism of inhibition of apoptosis by repression of caspase 9 in primary lymphocytes exposed to 100 µT (50 Hz) ELF-EMF is given in Figure 6. Figure 7. Magnetic field exposure facility fabricated using Teflon frame wound with Helmholtz coil housed inside the CO2 incubator for studies on cellular models. Figure 7. Magnetic field exposure facility fabricated using Teflon frame wound with Helmholtz coil housed inside the CO2 incubator for studies on cellular models. In summary, the above experimental investigation indicates a reduction in the apoptosis in primary lymphocytes subjected to 100 µT (50 Hz) for 16 h in the absence of mitogens, thereby enhances the viability of the cells. Possibly, the repression of apoptosis in primary lymphocytes is achieved by inhibiting caspases 9, one of the important stress induced initiator caspases. In addition, ELF-EMF (5, 50 and 100 µT; 50 Hz) could not invoke any DNA damage in lymphocytes. It is pertinent to infer that exposure to magnetic fields induces quite a variety of biological effects. Different waveform magnetic field sources results in varied biological effects on living systems, as their frequency spectrums are diverse [39, 58, 59]. Since the magnetic field exposure is a continuous process, an uninterrupted power supply (UPS) was used in order to avoid the supply variation. The output voltage variation of the UPS was limited to less than 1% ensuring stable magnetic field generation. Supply voltage can also produce electric field inside the incubator. If the supply voltage is higher, electric fields produced will also be higher, which is not desirable while studying the biological effects of magnetic fields. BioDiscovery \| www.biodiscoveryjournal.co.uk To reduce the presence of electric field in the exposure set up, the 230 V supply voltage was stepped down to 15V using a step down transformer. The resistance/impedance of the coil being constant at fixed frequencies, the current control was achieved through a variable resistance connected in series with the coil. Heating effect of the coil was reduced by selecting the conductor of higher diameter suitable for carrying the necessary current. The current source (step down transformer, rheostat etc.) was shielded by means of mild steel sheet enclosures. Materials Materials Trypan blue, 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) based in vitro toxicology assay kit and Histopaque-1077 were procured from Sigma (St.Louis, USA). The assay kits for Single cell gel Electrophoresis (comet assay), caspase 9 activity, Annexin V staining were procured from R&D Systems, Inc.(Minniapolis, USA). RPMI 1640, Fetal bovine serum (FBS) and penicillin/streptomycin were obtained from Hyclone (Utah, USA). All other reagents were of analytical grade. The calibration of the exposure facility was carried out to confirm uniform distribution of the magnetic field intensity where the culture flasks would be kept [3]. Isolation of lymphocytes and culturing Peripheral blood was drawn from the jugular vein of healthy male sheep (1 year old), by using heparinized vacutainers. Lymphocytes were separated by Histopaque-1077 density gradient [60]. The cells were re- suspended in RPMI and an aliquot of 0.1 mL cell suspension mixed with 0.1 mL of 0.4% trypan blue and was applied to the edge of the cover slip of hemocytometer. The viable lymphocytes (unstained) were viewed under 40× using a light microscope and were counted in all the four large corner squares. The cells were cultured (1 × 106 cells/mL) in RPMI 1640 medium (Hyclone) supplemented with 10% FBSm (Hyclone) and 1% penicillin/streptomycin (Hyclone) in DNA laddering analysis for cell death DNA laddering analysis was carried out by collecting cell pellet (5  105 cells) in an Eppendorf tube after 16 h of incubation. Cells were lysed by the addition of 20 µL of lysis buffer (TE buffer containing 0.8% SDS) mixed well, and incubated at 37oC for 120 min with RNAse A/T1 cocktail mix. Proteinase K (10 µL) was added and incubated at 50oC overnight. The samples were loaded into dry wells of 1.5% agarose gel containing 0.5 µg/mL of ethidium bromide after mixing with 5 µL of 6X DNA loading buffer. The samples were subjected to electrophoresis at 35V for 4 h [48]. DNA ladders were finally visualized by UV light and the band intensity was quantitated by charge-coupled device (CCD) based digital image analysis system (UVItec, UVIsoft Image acquisition and analysis software, Cambridge, UK). Isolation of lymphocytes and culturing Cell viability by MTT assay The cell viability was monitored by the MTT assay. This assay measures the viable cells based on its mitochondrial dehydrogenase activity by which MTT will be reduced to formazan [61]. MTT was added to cells (4  106) cultured in 96-well plates (in triplicates) at three time points (0 h, 8 h and 16 h). The cultures were incubated for 5 h at 37°C and analyzed in an ELISA reader at 560 nm after solubilization of formazan crystals. The reduced formazan crystals were dissolved in MTT solubilization solution containing 10% Triton X-100 and 0.1 N HCl in anhydrous isopropanol. The number of MTT units in culture samples subjected to EMF was calculated as the ratio (Abs. sample – Abs. blank / Abs. control). Analysis of DNA damage by comet assay y g y y The DNA damage was evaluated by subjecting the lymphocytes for single cell electrophoresis (comet assay) at a concentration of 1 × 105 cells/mL [62, 63] using comet assay kit (R&D Systems, Minniapolis, MN, USA). The principle of the assay is based upon the ability of denatured, cleaved DNA fragments to migrate out of the cell under the influence of an electrical potential, whereas undamaged super coiled DNA remains within the confines of the cell membrane when a DC current is applied. Evaluation of the comet tail shape and migration pattern allows extent and assessment of DNA damage. The assay was carried out by immobilizing the lymphocytes in a bed of low melting agarose on a slide after culturing the lymphocytes for a period of 16 h. The immobilized cells were subjected to gentle lysis using lysis buffer containing 2.5 M NaCl, 0.1M EDTA, 0.01M Tris, 1% sodium-lauryl sarcosinate and 1% Triton X-100. The cells were then subjected to electrophoresis in TBE buffer (0.89 M Tris, 0.88 M boric acid and 0.02 M EDTA; pH 7) in a horizontal electrophoresis apparatus, after treatment with alkali (0.3 M NaOH) to denature the DNA and hydrolyze the sites of damage. The electrophoresis was carried out under potential difference of 1 V/cm for 10 min. The slides were fixed and scored under epi- Magnetic field exposure facility A square-shaped Helmholtz coil, producing ELF- EMFs (50 Hz) proportional to the electric current in the coil, was wound over a Teflon frame of dimension 35 x 35 x 20 cm (l x b x h) suitable for placement inside an air jacketed CO2 incubator (Binder, Tuttlingen, Germany) as shown in Figure 7. Electric current through the coil was regulated as explained below. BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 8 Effect of ELF-EMF on primary lymphocytes the absence of any phytostimulator. Cultures were divided into two groups: fluorescence microscope (Olympus, BX100, New York, USA) using fluorescent dye, SYBR green. Depending on the tail length, the cells were categorized into (i) nominal and (ii) long tail by scoring 100 cells. Cells, without tail were scored separately. A positive control was also separately run by treating the cells with 100 µM H2O2 for 20 min at 4oC. g p a) Group I: Control (n = 4, independent trials) a) Group I: Control (n = 4, independent trials) a) Group I: Control (n = 4, independent trials) b) Group II: ELF-EMF - exposed (n = 4, independent trials). The control group was placed in regular CO2 incubator (Binder, Tuttlingen, Germany) and the experimental group was exposed to varying magnetic field intensities (5 ± 0.25 µT, 50 ± 2.5 µT and 100 ± 5 µT) in the specially designed set up as detailed above. The temperature (37oC) and CO2 (5%) concentrations were monitored continuously in control and ELF-EMF exposed group. Flow Cytometric analysis of apoptosis Flow cytometric quantitation of apoptotic cells was carried out using Annexin V-Biotin Apoptosis Detection kit (R&D Systems, Minniapolis, MN, USA), which uses Annexin V- Biotin conjugate for detection of cell surface changes that occur early in the apoptotic process. Lymphocytes (1 × 106 cells) were collected by centrifugation at 500 g for 5 -10 min at 22oC after 16 h of exposure to magnetic fields. Cells were washed by re- suspending in 500 µL of PBS and pelleted by centrifugation at 500  g. The cells were incubated in 100 µL Annexin V incubation reagent in the dark for 15 min at 22oC followed by incubation with binding buffer containing streptavidin- Fluorescein isothiocyanate (FITC) conjugate in the dark for 15 min at 22oC. Samples were analyzed by flow cytometry (Becton Dickinson, FACS Calibre, USA). BioDiscovery \| www.biodiscoveryjournal.co.uk References in human fibroblasts under extremely low frequency electromagnetic field exposure. Mutat Res 2010; 683: 74-83. in human fibroblasts under extremely low frequency electromagnetic field exposure. Mutat Res 2010; 683: 74-83. 1. McCann J, Robert K, Charles NR. Assessing the potential carcinogenic activity of magnetic fields using animal models. Environ Health Perspect 2000; 108: 79-100. 14. Nordenson I, Mild KH, Anderson G, Sandstrom M. Chromosomal abberations in human amniotic cells after intermittent exposure to fifty hertz magnetic fields. Bioelectromagnetics 1994; 15: 293-301. 2. Narita K, Hanakawa K, Kasahara T, Hisamitsu T, Asano K. Induction of apoptotic cell death in human leukemic cell line, HL- 60, by extremely low frequency electric magnetic fields; analysis of the possible mechanisms in vitro. In vivo 1997; 11: 329 - 335. 15. Hardell L, Sage, C. Biological effects from electromagnetic field exposure and public exposure standards. Biomed Pharmacotherapy 2008; 62: 104-109. 3. Rajendra P, Sujatha HN, Devendranath D, Gunasekaran B, Sashidhar RB, Subramanyam C et al Biological effects of magnetic fields: Neurochemical and toxicological changes in developing chick embryos. BioMagn Res Techno. 2004; 2: 1-9. (URL:http://www.biomagres.com/ content/2/1/1). 16. SSI (Swedish radiation protection authority). SSI independent expert group on electromagnetic fields: Recent research on EMF and health effects. SSI Reports 2008; 12: 1-70 17. Harada S, Yamada S, Kuramata O, Gunji Y, Kawasaki M, Miyakawa T et al. Effects of high ELF magnetic fields on enzyme catalyzed DNA and RNA synthesis in vitro and on a cell– free DNA mismatch repairs. Bioelectromagnetics 2001; 22: 260-266. p g 4. Rajendra P, Sujatha HN, Devendranath D, Gunasekaran B, Subramanyam C, Aradhya, RSS et al. Effects of power frequency electro magnetic fields on growth of germinating Vicia faba. Electromagn Biol Med 2005; 24: 39-54. 18. Miki H, Ohmori M, Hrakawa E, Winters WD. Effects of environmental level magnetic field exposures on transcription of CMV immediate early promoter DNA in a cell free in vitro transcription system. Bioelectromagnetics 1999; 20: 519-521. g 5. Liboff AR. The ion cyclotron resonance hypothesis. In Hand Book of Biological Effects of Electromagnetic Fields. 3rdedition. Edited by Barnes FS, Greenebaum B. Boca Raton FL: Taylor and Francis, 2007, pp 261-292. 19. Kwee S, Rasmark, P. Changes in cell proliferation due to environmental non- ionizing radiation of ELF electromagnetic fields. Bioelectrochem Bioener 1995; 36: 109-114. 6. Wertheimer N, Leeper E. Electric wiring configurations and childhood cancer. Am J Epidemiol 1979; 109: 273-384. 20. References Cossarizza A, Monti D, Bersani F, Cantini M, Cadossi R, Sacchiand A et al. (1989). Extremely low frequency pulsed electromagnetic fields increase cell proliferation in lymphocytes from young and aged subjects. Biochem Biophys Res Commun 1989; 160: 692-698. 7. Koana T, Ikehata M, Nakagawa M. Estimation of genetic effects of static magnetic field by a somatic cell test using mutagen- sensitive mutants of Drosophila melanogaster. Bioelectrochem Bioener 1995; 36: 95-100. 8. Khalil AM, Qassem W. (1991). Cytogenetic effects of pulsing electromagnetic field on human lymphocytes in vitro: Chromosome aberrations, sister chromatid exchanges and cell kinetics. Mutat Res 1991; 247: 141-146. ; 21. Rosenthal M, Obe G. Effects of 50 Hz electromagnetic fields on proliferation and on chromosomal alterations in human peripheral lymphocytes untreated or pretreated with chemical mutagens. Mutat Res 1989; 210: 3329 - 3351. 9. Antonopoulous A, Yang B, Stamm A, Heller WD, Obe G. (1995). Cytological effects of 50 Hz electromagnetic fields on human lymphocytes in vitro. Mutat Res 1995; 346: 151-157. 22. Cadossi R, Bersani R, Cossarizza A, Zucchini P, Emilia G, Torelli G et al. Lymphocytes and low-frequency electromagnetic fields. FASEB J 1992; 6: 2667-2674. 10. Cohen MM, Kunsak A, Astemborski JA, McCulloch D. The effects of low level 60 Hz electromagnetic fields on human lymphoid cells. II. Sister- Chromatid exchanges in peripheral lymphocytes and lymphoblastoid cell lines. Mutat Res 1986; 172: 177-184. 23. Rojas JAH, Rodriguez-De La Fuente AO, del RV Campos M, Leal-Garza CH, Flores LER, de La Fuente-cortez B. Cytological effects of 60Hz magnetic fields on human lymphocytes in vitro: sister chromatid exchanges, cell kinetics and mitotic rate. Bioelectromagnetics 2001; 22: 145-149. 11. Phillips JL, Campbell-Beachler M, Ishida-Jones T, Haggren W, Adey WR. Exposure of Molt 4 T-lymphoblastoid cells to a 1 G sinusoidal magnetic field at 60 Hz: Effects on cellular events related to apoptosis. In: 1997 annual review of research on biological effects of electric and magnetic fields from the generation, delivery and use of electricity, Frederick, MD: W/L associates. Ltd. 1996. 24. Simko M, Kriehuber R, Weiss DG, Luben RA. Effects of 50Hz EMF exposure on micronucleus formation and apoptosis in transformed and non-transformed human cell lines. Bioelectromagnetics 1998; 19: 85-91. 25. Chen G, Upham BL, Sun W, Chang CC, Rothev EJ, Chen KM et al. Effect of EMF exposure on chemically induced differentiation of Friend erythro leukemia cells. Environ Health Perspect 2000; 108: 967-972. 12. Spectrofluorometric analysis of caspase 9 activity The assay of caspase 9 activity was carried out using fluorometric assay kit (R&D Systems, Minniapolis, MN, USA). The lymphocytes that are subjected to magnetic fields for 16 h were first lysed to collect their intracellular contents. The cell lysate was then tested for protease activity by the addition of a caspase - specific peptide (LEHD) that is conjugated to the fluorescent reporter molecule 7-amino-4-trifluoromethyl coumarin (AFC). The cleavage of the peptide by the caspase releases the fluorochrome that when excited by light at 400 nm wavelength, emits fluorescence at 505 nm. The level of caspase enzymatic activity in the cell lysate is directly BioDiscovery \| www.biodiscoveryjournal.co.uk 9 August 2012 \| Issue 2 \| 2 Effect of ELF-EMF on primary lymphocytes Govt. of India is duly acknowledged for providing funds (Contract grant No. 11.1.14) and Research Fellowship to Rajendra at CPRI, Hyderabad and Council for Scientific and Industrial Reasearch (CSIR), Govt. of India, New Delhi for the award of Senior Research Fellowship to continue this study at the Department of Biochemistry, Osmania University, Hyderabad. We acknowledge Mrs. Arunjyothi for her help in establishing the magnetic field exposure set up. proportional to the fluorescence signal detected by a spectrofluorometer (Hitachi F-4500; Hitachi, Tokyo, Japan). The caspase 9 activity was expressed as fluorescence intensity (FI)/mg protein by estimating the protein content in the cell lysate using detergent friendly bicinchoninic acid method [64]. Acknowledgements Central Power Research Institute, Ministry of Power, Effect of ELF-EMF on primary lymphocytes Health Persp 2003; 111: 281-286. et al. Increased apoptosis, changes in intracellular Ca2+, and functional alternations in lymphocytes and macrophages after in vitro exposure to static magnetic field. J Toxicol Environ Health Pt A 1998; 54: 63-76. 45. Pandey P, Farber R, Nakazawa A, Kumar S, Bharti A, Nalini C et al. Hsp 27 functions as a negative regulator of cytochrome c dependent activation of procaspase 3. Oncogene 2000; 19: 1975- 1981. 28. Ismael SJ, Callera F, Gracia AB, Baffa O, Falcao RP. Increased dexamethasone- induced apoptosis of thymocytes from mice exposed to long term extremely low frequency magnetic fields. Bioelectromagnetics 1998; 19: 131-135. 46. Concannon CG, Orrenius S, Samali A. Hsp 27 inhibits cytochrome C mediated caspase activation by sequestering both procaspase 3 and cytochrome C. Gene Expression 2001; 9: 195- 201. 29. Pessina GP, Aldinucci C, Palmi M, Sgaragli G, Benocci A, Meini A et al. Pulsed electromagnetic fields affect the intracellular Ca2+ concentrations in human astrocytoma cells. Bioelectromagnetics 2001; 22: 503-510. 47. Thompson CB. Apoptosis in the pathogenesis and treatment of diseases. Science 1995; 267: 1456 48. Spector DL, Goldman RD, Leinwand IA. Culture and biochemical analysis of cells. In Cells: A laboratory manual. Coldspring, Harbour Press, Vol. I, 1998 pp 11-12. 30. Tofani S, Barone D, Cintorino M, de Santi MM, Ferrara A, Orlassino R et al. (2001). Static and ELF magnetic fields induce tumour growth inhibition and apoptosis. Bioelectromagenetics 2001; 22: 419-428. 49. Li P, Nijhawan D, Budihardio I, Srinivasula SM, Ahmad M, Alnemri ES et al. Cytochrome C and dATP dependent formation of Apaf-1/Caspase 9 complex initiates an apoptotic protease cascade. Cell 1997; 91: 478-489. 31. Raylman RR, Clavo AC, Wahl RL. Exposure to strong static magnetic fields shows the growth of human cancer cells in vitro. Bioelectromagnetics 1996; 17: 358-363 50. Benjamins JA, Neddkoska L, George EB. (2003). Protection of mature oligodendrocytes by inhibitors of caspases and calpains. Neurochem Res 2003; 28: 143-152. 32. Blackman CF, Benane SG, House DE. The influence of 1.2µT, 60 Hz magnetic fields on melatonin and Tamoxifen induced inhibition of MCF-7 cell growth. Bioelectromagnetics 2001; 22: 122-128. 51. Cardone MH, Roy N, Stennicke HR, Salvesen GS, Franke TF, Stanbridge E et al. Regulation of cell death protease caspase 9 by phosphorylation. Science 1998; 282: 1318-1321. 33. Cadossi R, Ceccherelli G, Emilia G, Torelli G, Ruggeri M, Monari P et al. Effect of ELF-EMF on primary lymphocytes Effect of low frequency pulsing electromagnetic fields (PEMFs) on the response of cultured normal and leukemic lymphocytes to the lectins. In Proceedings of the international conference on electric and magnetic fields in medicine and biology, London, 1985, pp 4-8. 52. Paradisi S, Donelli G, Santini MT, Straface E, Malorni WA. A 50Hz magnetic field induces structural and biophysical changes in membranes. Bioelectromagnetics 1993; 14: 247-255. 53. Blackman CF, Benane SG, Elliott DJ, House DE, Pollock MM. Influence of electromagnetic fields on the efflux of calcium ions from brain tissue in vitro: A three model analysis consistent with the frequency response upto 510 Hz. Bioelectromagnetics 1988; 9: 215-227. 34. Cossarizza A, Monti D, Bersani F, Cantini M, Cadossi R, Sacchiand A et al. Extremely low frequency pulsed electromagnetic fields increase cell proliferation in lymphocytes from young and aged subjects. Biochem Biophys Res Commun 1989; 160: 692-698. 54. Blank M. Na, K-ATPase function in alternating electric fields. FASEB J 1992; 6: 2434-2438. ; 35. Conti P, Gigante GE, Cifone MG, Alesse E, Ianni G, Reale M et al. (1983). Reduced mitotic stimulation of human lymphocytes by extremely low frequency electromagnetic fields. FEBS Lett 1983; 162: 156-160. 55. Sancho JG, Montero M, Alrare, J, Fonteriz RI, Sanchez A. Effects of extremely low frequency EMF on ion transport in several mammalian cells. Bioelectromagnetics 1994; 15: 579-588. 56. Wang HQ, Quan T, He T, Franke TF, Voorhees JJ, Fischer GJ. Epidermal growth factor receptor-dependent, NF-kappa B- independent activation of the phosphatidyl inositol 3 –kinase/Akt pathway inhibits ultraviolet irradiation induced caspases-3, -8 and –9 in human keratinocytes. J Biol Chem 2003; 278: 45737-45745. 36. Mooney NA, Smith RE, Watson BW. (1986). Effect of extremely low frequency pulsed magnetic fields on the mitogenic response of peripheral blood mononuclear cells. Bioelectromagnetics 1986; 7: 387-394. 37. Wallezek J. Electromagnetic field effects on cells of the immune system: The role calcium signaling. FASEB J 1992; 6: 3177-3185. 57. Uekun FM, Kurosaki T, Jin J, Jun X, Morgan A, Takata M et al. Exposure of B-lineage lymphoid cells to low energy electromagnetic fields stimulates Lyn Kinase. J Biol Chem 1995; 270: 27666-27670. 38. Azanza M , Moral D. Cell membrane biochemistry andneurobiological approach to biomagnetism. Prog Neurobiol 1994; 44: 517-601. 58. Liboff AR, Macleod BR, Smith, SD. Rotating magnetic fields and ion cyclotron resonance. J Bioelect. 1989; 8: 119-125. 39. Vincze, Gy N. Szasz N, Szasz A. References Ahuja YR, Bhargava A, Sircar S, Rizwani W, Lima S, Devadas AH et al. Comet assay to evaluate DNA damage caused by magnetic fields. In: Proceedings of the International Conference on Electromagnetic Interference and Compatibility, Hyderabad, India, 1997 pp. 273-276. 26. Hisamistu T, Narita K, Kasahara T, Seto A, Yu Y, Asano K. Induction of apoptosis in human leukemic cells by magnetic field. Jpn J Physiol 1997; 47: 307-310. p y ; 27. Flipo D, Fournier M, Benquet C, Roux P, LeBoulaire C, Pinsky C 13. Focke F, Schuermann D, Kuster N, Schar P. DNA fragmentation BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 10 Effect of ELF-EMF on primary lymphocytes Effect of ELF-EMF on primary lymphocytes On the thermal noise limit of cellular membranes. Bioelectromagnetics 2005; 26: 28-35 59. Lednev VV. Possible mechanism for the influence of weak magnetic fields on biological systems, Bioelectromagnetics 1991; 12: 71-75. 40. Belyaev I, Hillert L, Protopopova M, Tamm C, Malmgren LOG, Persson BRR et al. 915 MHz Microwaves and 50 Hz magnetic field affect chromatin conformation and 53BP1Foci in human lymphocytes from hypersensitive and healthy persons. Bioelectromagnetics 2005; 26: 173-184. 60. Boyum A. Separation of leukocytes from blood and bone marrow. Scand J Clin Lab Invest 1968; 97: 77. 61. Mossman T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxic assays. J Immunol Method 1983; 65: 55-63. 41. Belyaev I. Toxicity and SOS-response to ELF magnetic fields and nalidixic acid in E. coli cells. Mutat Res. 2011; 722:56-61 62. Olive PL, Banath JP, Durand RE. Heterogenecity in radiation induced DNA damage and repair in tumour and normal cells measured using the comet assay. Radiat Res 1990; 122: 86-94. 42. Kim JY, Hong SY, Lee YM, Yu SA, Koh WS, Hong JR et al. In vitro assessment of clastogenicity of mobile-phone radiation (835 MHz) using the alkaline comet assay and chromosomal aberration test. Environ Toxicol 2008; 23: 319-327. 63. Singh NP, Danner DB, Tice RR, Brant L, Schnieder EL. DNA damage and repair with age in individual human lymphocytes. Mutat Res 1990; 237: 123-130. 43. Conconnon CG, Gorman AM, Samali A. On the role of Hsp 27 in regulating apoptosis. Apoptosis 2003; 8: 61-70. 44. Shi B, Rarboud B, Nuccitelli R, Isseroff RR. Power line frequency EMF do not induce changes in Phosphorylation, localization or expression of the 27kD hsp in human keratinocytes. Environ 64. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD et al. (1985). Measurement of protein using bicinchonic acid. Anal Biochem 1985; 150: 76-85. BioDiscovery \| www.biodiscoveryjournal.co.uk August 2012 \| Issue 2 \| 2 11
https://openalex.org/W2144901603	https://europepmc.org/articles/pmc3016263?pdf=render	English	null	Obesity related methylation changes in DNA of peripheral blood leukocytes	BMC medicine	2,010	cc-by	7,503	* Correspondence: xwang@mcg.edu 1Georgia Prevention Institute, Department of Pediatrics, Medical College of Georgia, Augusta, GA, USA Full list of author information is available at the end of the article © 2010 Wang 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: Despite evidence linking obesity to impaired immune function, little is known about the specific mechanisms. Because of emerging evidence that immune responses are epigenetically regulated, we hypothesized that DNA methylation changes are involved in obesity induced immune dysfunction and aimed to identify these changes. Method: We conducted a genome wide methylation analysis on seven obese cases and seven lean controls aged 14 to 18 years from extreme ends of the obesity distribution and performed further validation of six CpG sites from six genes in 46 obese cases and 46 lean controls aged 14 to 30 years. Results: In comparison with the lean controls, we observed one CpG site in the UBASH3A gene showing higher methylation levels and one CpG site in the TRIM3 gene showing lower methylation levels in the obese cases in both the genome wide step (P = 5 × 10-6 and P = 2 × 10-5 for the UBASH3A and the TRIM3 gene respectively) and the validation step (P = 0.008 and P = 0.001 for the UBASH3A and the TRIM3 gene respectively). Conclusions: Our results provide evidence that obesity is associated with methylation changes in blood leukocyte DNA. Further studies are warranted to determine the causal direction of this relationship as well as whether such methylation changes can lead to immune dysfunction. See commentary: http://www.biomedcentral.com/1741-7015/8/XX/abstract Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Open Access Obesity related methylation changes in DNA of peripheral blood leukocytes Xiaoling Wang1, Haidong Zhu1, Harold Snieder1,5, Shaoyong Su6, David Munn2,4, Gregory Harshfield1,3, Bernard L Maria2,3,4, Yanbin Dong1, Frank Treiber1, Bernard Gutin1, Huidong Shi4 Background the immune system. Recently, there has been a greater appreciation of the role of epigenetics, meiotically and mitotically heritable changes in gene expression that are not coded in the DNA sequence itself, in the immune and inflammatory responses [7-9]. Therefore, we hypothesize that DNA methylation changes play a role in obesity induced immune dysfunction. The goal of this study was to characterize DNA methylation profile in peripheral blood leukocytes in obese versus lean subjects using a genome wide approach. Identification of methy- lation changes in specific genes will provide important targets for further study into the mechanisms of obe- sity’s effect on the immune system and the potential to develop new therapies to treat multiple obesity comorbidities. Obesity is the epidemic of our time, with sharply and steadily rising rates [1,2]. The major adverse conse- quences of obesity including type 2 diabetes, athero- sclerosis and essential hypertension, when added together, account for a large number of disease related deaths [3,4]. If the obesity-related cancer cases are added to this number, obesity-related mortality by far exceeds that of other common diseases [5]. The latter indicates the urgent need to develop novel efficient ther- apeutic modalities for this condition. The common denominator in the pathogenesis of the co-morbidities of obesity is the presence of an active, low-grade inflammatory process [6]. Despite evidence linking obesity to alterations in inflammatory response, little is known about the specific effects of obesity on Subjects The genome wide methylation analysis was conducted in seven obese and seven age-matched lean controls. Page 2 of 8 Page 2 of 8 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 These 14 subjects were identified from the participants (n = 534) in the Lifestyle, Adiposity, and Cardiovascular Health in Youth (LACHY) study using the following inclusion criteria: (1) African American (AA) ancestry; (2) male; (3) having leukocyte DNA available; (4) obese cases having a body mass index (BMI) ≥99th percentile for age and sex and lean controls having BMI ≤10th percentile for age and sex. The LACHY study consisted of roughly equal numbers of AA and European Ameri- can (EA) adolescents aged 14 to 18 years of both sexes recruited from high schools in the Augusta, Georgia area [10]. stadiometer and a scale, respectively. BMI was calculated as weight/height2. Systolic BP (SBP) and diastolic BP (DBP) were measured with Dinamap monitors, using an appropriately sized BP cuff placed on the subject’s right arm. BP measurements were taken at 11, 13, and 15 minutes, during a 15-minute supine relaxation period. The average of the last two readings was used to repre- sent SBP and DBP values [10-13]. Fasting peripheral blood samples in the LACHY cohort and non-fasting peripheral blood samples in the other three cohorts were collected. The buffy coat and plasma samples were separated and stored at -80°C. DNA was extracted from the buffy coat. In the LACHY cohort, fasting glucose levels were measured using Ekta- chem DT II system (Johnson and Johnson Clinical Diag- nostics, Rochester, NY, USA) and fasting insulin was assayed in duplicate by specific radioimmunoassay (Linco Research, Inc., St Charles, MO, USA) [10]. QUICKI (quantitative insulin-sensitivity check index) was calculated to index insulin sensitivity using the fol- lowing formula: 1/[log(fasting insulin, μU/ml) + log(fast- ing glucose, mg/dl)]. Out of the screening sample which included the seven obese cases and seven lean controls selected from the LACHY cohort, fasting glucose and insulin levels were not available for one case and one control. Genome wide methylation chip Genome wide methylation chip The HumanMethylation27 BeadChip from Illumina (Illumina, San Diego, CA, USA) was used. This chip can quantitatively measure 27,000 CpG sites, covering more than 14,000 well-annotated genes at single-CpG resolution. Each chip can accommodate 12 samples. After bisulfite treatment, 200 ng of the converted DNA was whole genome amplified (WGA) and enzy- matically fragmented. The bisulfite-converted WGA- DNA samples were purified and applied to the Bead- Chips. Image processing and intensity data extraction were performed according to Illumina’s instruction http://www.illumina.com/products/infinium_human- methylation27_beadchip_kits.ilmn. Each methylation data point is represented by fluorescent signals from the methylated and unmethylated alleles. DNA methy- lation beta values are continuous variables between 0 (completely unmethylated) and 1 (completely methy- lated), representing the ratio of the intensity of the methylated bead type to the combined locus intensity. Initial array processing and quality control were per- formed with BeadStudio software. The microarray data discussed in this paper have been deposited in NCBI’s Gene Expression Omnibus and are accessible through GEO Series accession number GSE25301 http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? acc=GSE25301. For all four cohorts self identification by self-reports of each subject or by a parent if the subject was under 18 years of age was used to classify ethnicity according to pre- viously described criteria [14]. Subjects in all the four stu- dies were overtly healthy, free of any acute or chronic illness on the basis of parental reports and were not on anti-hypertensive, lipid lowering, anti-diabetic and anti- inflammatory medications [10-13]. The Institutional Review Board at the Medical College of Georgia approved the stu- dies. Informed consent was obtained from all subjects and by parents if subjects were less than 18 years of age. Subjects The replication cohort included 46 obese (BMI ≥30 kg/ m2 or BMI ≥95th percentile for age and sex if age ≤18) and 46 lean (BMI ≤22 kg/m2 or BMI ≤40th percentile for age and sex if age ≤18) AA males selected from three cohorts, the Blood Pressure (BP) stress study (n = 603) [11], the Georgia Cardiovascular twin study (n = 1,183) [12] and the Prevention of Hypertension in African Amer- ican Teens (PHAT) study (n = 262) [13]. Both the BP Stress study and the twin study are on-going longitudinal studies which have followed the subjects more than 10 years. Both studies included roughly equal numbers of AAs and EAs or males and females. The BP stress study was established in 1989 with subjects aged to 7 to 16 years at baseline and the twin study was established in 1996 with subjects aged 7 to 25 years at baseline [11,12]. The PHAT study was a cross-sectional study and consisted of AA males and females aged 14 to 20 years [13]. Subjects in all the three studies were also recruited from Augusta, GA area. The obese and lean subjects were identified based on the following criteria: (1) having leukocyte DNA available; (2) AA males; (3) only one twin subject from a twin pair was selected if both twins met the criteria; (4) if multiple visits (with multiple leukocyte DNA) were avail- able for a subject, this subject had to be obese or lean on all the visits to be included in the replication sample and the leukocyte DNA collected at the visit when the subject were the most obese or most lean was used. Results Table 1 displays the general characteristics of the screening sample. At this age range, obese subjects already showed significantly higher insulin resistance. Figure 1 is a volcano plot showing the raw P-values for all CpG sites versus mean methylation difference between the obese group and the lean group. We did not observe any CpG sites survive multiple testing cor- rection with the most significant CpG site with a FDR of 0.14 and a raw P-value of 5 × 10-6. Table 2 lists the top 10 most significant CpG sites and one additional CpG site which showed the largest methylation differ- ence between cases and controls (-27.1% for CREB3L3). Since at least two CpG sites were included for the majority of the genes in this genome wide methylation chip, the P-value for the second CpG site in this gene was also listed in Table 2. Out of the 11 CpG sites listed in Table 2 (under the column “more significant CpG site”), we selected six CpG sites for validation in the replication cohort using the following inclusion criteria: (1) the most significant differentially methylated CpG site, which is the CpG site locating at -156 of the UBA- SH3A (ubiquitin-associated and SH3 domain-containing protein A) gene; or (2) the CpG sites in those genes of which the second CpG site also showed a raw P < 0.05. These include the CpG sites from the TRIM3 (Tripartite motif-containing 3), CTSZ (Cathepsin Z preproprotein), HIPK3 (Homeodomain interacting protein kinase 3), CDH5 (Cadherin 5 type 2 preproprotein) and CREB3L3 (cAMP responsive element binding protein 3-like 3) genes. The validation was conducted on the more signif- icant CpG site (with Illumina ID 13517, 17029, 23241, 25599, 8829 and 23739) in each gene by pyrosequen- cing. The general characteristics of the replication cohort are displayed in Table 3. Since the DNA from one obese subject failed on four pyrosequencing assays, the data from this subject was excluded from the analy- sis. Table 4 lists the results in the replication cohort. The findings from the genome wide methylation chip were validated for the CpG sites in the UBASH3A, TRIM3, HIPK3 and CDH5 genes. Measurements For all the four cohorts, height and weight were mea- sured by standard methods using a wall-mounted Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Page 3 of 8 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 least two CpG sites were included for the majority of genes in this genome wide chip, we selected the CpG sites with the lowest p value to represent this gene. Results The methylation sta- tus of the CpG site in the UBASH3A gene was 3.3% higher (P = 0.0019) and the methylation status of the CpG site in the TRIM3, HIPK3 and CDH5 genes was 1.21% (P = 0.0004), 7.31% (P = 0.0135) and 3.09% (P = 0.045) lower in the obese versus the lean group. These significant findings persisted after adjustment of age for the association between obesity and methylation status of the CpG site in the UBASH3A gene (P = 0 008) and Pyrosequencing The methylation levels of the selected six CpG sites from the six genes in the replication cohort were deter- mined by pyrosequencing technology, a rapid and robust method for quantitative methylation analysis. After bisulfite treatment, 10 ng of the converted DNA was used in a PCR reaction to amplify the target region. One of the PCR primers was biotin labeled. Single- stranded biotinylated PCR products were prepared for sequencing by use of the Pyrosequencing Vacuum Prep Tool according to the manufacturer’s instructions. The PCR products (each 10 μl) were sequenced by Pyrose- quencing PSQ96 HS System (Pyrosequencing-Qiagen) following the manufacturer’s instructions. The methyla- tion status of each locus was analyzed individually as a T/C SNP using QCpG software (Biotage, Kungsgatan, Sweden). PCR primers and sequencing primers for these six genes are listed in Additional file 1. All the samples were assayed in the same run and in a random sequence. Statistical analysis For the genome wide methylation analysis, the Limma package [15] was used to analyze each CpG site for dif- ferential methylation between obese and lean subjects. CpG sites on the × and Y chromosomes were excluded from the analysis. Each CpG site was assigned a raw P- value based on a moderated t statistic. To correct for multiple testing, the set of raw P-values were converted to false discovery rates (FDR) according to Benjamini and Hochberg [16]. For the six CpG sites in the replica- tion cohort, a Student’s t-test was used to investigate whether their methylation levels differ between the obese and the lean group. Linear regression was further used to adjust the potential effect of age on methylation levels. We combined the genome wide and the replica- tion steps for these six CpG sites using the weighted z score-based meta-analysis approach implemented in the package METAL [17]. Prior to analysis, methylation levels of the CpG sites in the replication cohort were log-transformed or square root-transformed to obtain a better approximation of the normal distribution. Preli- minary analyses, t-tests and regression analyses were done using STATA 8 (StataCorp, College Station, TX, USA). Gene Ontology analysis was conducted with the FatiGO tool [18]. FatiGO takes two lists of genes and converts them into two lists of GO terms. Then a Fish- er’s exact test for 2 × 2 contingency tables is used to check for significant over-representation of GO terms in one of the sets with respect to the other one. Multiple testing correction (indexed by adjusted P-values) to account for the multiple hypothesis tested (one for each GO term) is applied to reduce false positives. Since at Page 4 of 8 Page 4 of 8 Wang et al. Statistical analysis BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Table 1 General characteristics of the subjects for genome wide methylation analysis Cases Controls N 7 7 Age, years 15.8 ± 1.0 (14.5 to 17.2) 15.9 ± 1.4 (15.1 to 18.1) BMI, kg/m2 39.0 ± 1.7 (37.2 to 41.2) 17.0 ± 0.7 (16.4 to 17.8) BMI percentile 0.996 ± 0.001 (0.995 to 0.997) 0.048 ± 0.022 (0.026 to 0.091) SBP, mmHg 125.8 ± 13.9 (115.3 to 152.7) 112.7 ± 8.9 (100.7 to 124.7) †‡Fasting insulin, uU/ml 35.9 ± 13.6 (19.8 to 58.9) 13.7 ± 3.0 (9.4 to 17.0) ‡Fasting glucose, mg/dl 103.5 ± 14.8 (84 to 126) 88.8 ± 7.6 (80 to 100) ‡QUICKI 0.28 ± 0.01 (0.27 to 0.30) 0.33 ± 0.01 (0.31 to 0.35) Means ± SD (Range) : P < 0.001; †: P < 0.01 ‡: Available for six cases and six controls Table 1 General characteristics of the subjects for genome wide methylation analysis significance for HIPK3 gene (P = 0.05), and disappeared for the CpG site in the CDH5 gene (P = 0.095). The results of the meta-analysis for these six CpG sites on the data from the genome wide step and the replication step are shown in Additional file 2 pilot study in 14 subjects, we did not observe any GO categories survive multiple testing. Table 5 lists the GO categories with raw P-value less than 0.05. Interestingly, we observed enriched functional processes that are potentially relevant for inflammatory response with immune response (GO:0006955), cell activation (GO:0001775), cytokine production (GO:0001816), response to biotic stimulus (GO:0009607) and antigen binding (GO:0003823) among the top GO categories. The results support that obesity might lead to methyla- tion changes in genes involved in inflammatory pathways. Although the pyrosequencing assay was designed to target one specific CpG site for each gene, some of the assays covered several surrounding CpG sites. The results of these CpG sites are also listed in Table 4. The correla- tion within samples among the multiple CpG sites mea- sured within a gene is listed in Additional file 3. Gene Ontology analysis was performed to test whether some common functional trends in molecular functions and biological processes were associated with the genes exhibiting differences between obese cases and lean con- trols in the genome wide chip. Statistical analysis We included those genes with a raw P ≤0.01 to the first list (n = 298) and included all the other genes in the second list. As expected from a Discussion In this study we aimed to identify obesity related methy- lation changes in peripheral blood leukocytes using a genome wide approach in youth and young adults who are free of obesity comorbidities. The primary findings of this study are increased methylation levels at one CpG site in the UBASH3A gene and decreased methyla- tion level at one CpG site in TRIM3 gene in obese sub- jects compared with lean controls. Figure 1 Volcano plot showing raw P-values versus mean methylation difference between obese cases and lean controls. j p The protein encoded by the UBASH3A gene is the “ubiquitin-associated and SH3 domain-containing pro- tein A”, which was previously also known as T-cell ubi- quitin ligand (TULA) and suppressor of T-cell signaling 2 (Sts-2). It is expressed predominantly in T-cells, where it has a suppressing effect on T-cell signaling and activa- tion. It acts in part by inhibiting c-CBL-mediated down- regulation of protein tyrosine kinases (PTKs) that are activated upon T-cell receptor stimulation [19]. UBA- SH3A is also capable of promoting T-cell apoptosis [20]. UBASH3A and UBASH3B (another protein in this family) gene double knock-out mice were hyperrespon- sive to T-cell receptor stimulation with increased cyto- kine secretion, although mice lacking UBASH3A were normal in all respects including T-cell function [21]. However, whether the expression of the UBASH3A gene is regulated by DNA methylation is unknown. No CpG island exists in this gene and the CpG site showing higher methylation levels in obese subjects from this Figure 1 Volcano plot showing raw P-values versus mean methylation difference between obese cases and lean controls. Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Page 5 of 8 Page 5 of 8 Table 2 Top 10 differentially methylated CpG sites and the CpG site showing the largest difference. Gene More significant CpG site Second CpG site ID Dis. to TSS Average Case/Con. (%) Difference (%) P FDR ID Dis. to TSS Average Case/Con. Discussion HIF3A 2879 153 5.35/2.97 2.38 1.08 10 NOTCH4 5969 -51 59.1/68.5 -9.40 1.15 10 CDH5 8829 -243 27.2/36.5 -9.20 1.75 10 CREB3L3 23739 42 9.63/36.7 -27.1 1.78 10 Genes taken forward to the replication stage are shown in bold. ID: ID from Illumina HumanMethylation27 BeadChip; Dis. to TSS: Distance to the T For CTSZ, 7 CpG sites were included in the genome-wide methylation platform Table 3 General characteristics of the subjects of the replication cohort Cases Controls N 45 46 Age, years 20.3 ± 5.0 (14.1 to 29.5) 17.6 ± 3.1 (12.7 to 27.9) †BMI, kg/m2 37.4 ± 6.1 (30.5 to 59.9) 18.9 ± 1.2 (16.4 to 22.0) †SBP, mmHg 125.8 ± 14.2 (97 to 175) 113.5 ± 8.5 (100 to 134) Means ± SD (Range) : P < 0.01; †: P < 0.001 Genes taken forward to the replication stage are shown in bold. ID: ID from Illumina HumanMethylation27 BeadChip; Dis. to TSS: Distance to the Transcription Start Site; FDR: False Discover Rate * For CTSZ, 7 CpG sites were included in the genome-wide methylation platform and the two most significant sites are listed in the table. TRIM3 is one member of the TRIM protein family. These proteins share a conserved tripartite architecture and have a variety of cellular functions including cell proliferation, differentiation, oncogenesis, apoptosis, immune signaling and have been implicated in autoim- mune diseases [24]. In peripheral blood leukocytes, TRIM3 is mainly expressed in macrophages and can be up-regulated by viral infection [25,26]. Similar to UBA- SH3A, the role of methylation in TRIM3 gene expres- sion has not been reported yet. The TRIM3 gene has 1 CpG island located in the promoter and exon 1 (nucleo- tides -446 to 576), one CpG island spanning intron 2 and exon 3 (nucleotides 7759 to 8334), and one CpG island in exon 13 and the downstream region (nucleo- tides 24883 to 25610). The CpG site showing lower methylation levels in obese subjects in both the scanning cohort and the replication cohort locates in the promo- ter CpG island (348 bp upstream to the transcription start site). The methylation status of the other CpG site which showed significant differential methylation levels between obese cases and lean controls in the genome wide methylation analysis also locates in this CpG island (331 bp upstream to the transcription start site). Discussion (%) Difference (%) P UBASH3A 13517 -156 45.6/35.5 10.1 5.00 × 10-6 0.14 146 52 45.6/37.7 7.9 0.145 CTNND1 6160 -431 57.4/70.2 -12.8 2.00 × 10-5 0.20 18016 415 19.7/21.5 -1.8 0.353 TRIM3 17029 -348 6.74/9.28 -2.54 2.40 × 10-5 0.20 19207 -331 12.5/16.2 -3.7 0.035 CTSZ* 23241 -292 13.6/21.0 -7.40 4.40 × 10-5 0.24 1644 -592 17.4/25.6 -8.2 0.001 TRAF5 10097 -319 69.7/58.0 11.7 4.50 × 10-5 0.24 25786 77 1.90/1.79 0.1 0.738 HIPK3 25599 390 51.5/64.9 -13.4 6.50 × 10-5 0.26 5514 314 26.6/35.0 -8.4 0.006 DPCR1 20918 -46 62.5/70.2 -7.70 6.90 × 10-5 0.26 4703 -906 72.4/73.0 -0.6 0.709 HIF3A 2879 153 5.35/2.97 2.38 1.08 × 10-4 0.34 7009 -1445 57.1/63.0 -5.9 0.091 NOTCH4 5969 -51 59.1/68.5 -9.40 1.15 × 10-4 0.34 14700 4 19.3/23.6 -4.3 0.078 CDH5 8829 -243 27.2/36.5 -9.20 1.75 × 10-4 0.44 22286 36 72.5/81.3 -8.8 0.012 CREB3L3 23739 42 9.63/36.7 -27.1 1.78 × 10-3 0.59 13445 -26 5.74/25.8 -20.1 0.004 Genes taken forward to the replication stage are shown in bold. ID: ID from Illumina HumanMethylation27 BeadChip; Dis. to TSS: Distance to the Transcription Start Site; FDR: False Discover Rate * For CTSZ, 7 CpG sites were included in the genome-wide methylation platform and the two most significant sites are listed in the table. Table 2 Top 10 differentially methylated CpG sites and the CpG site showing the largest difference. study locates at the promoter region with a distance of 156 bp to the transcription start site. There is a possibi- lity that methylation of this CpG site or other CpG sites with methylation levels correlated with this CpG site inhibits the interactions between DNA sequence and nuclear proteins, resulting in decreased gene expression. In-silico analysis of the region of this CpG site using TFSEARCH software [22] did not find this CpG site located at any known transcription factor binding sites. However, methylation of this CpG site may suppress gene transcription by recruiting methylcytosine-binding proteins that in turn associate with large protein com- plexes containing corepressors and histone deacetylases. The binding of these complexes to DNA may lead to a change in the chromatin structure from an active to an inactive form [23]. This speculation needs to be confirmed. Discussion BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Page 6 of 8 Page 6 of 8 Table 4 Results from the replication cohort Gene ID Dis. to TSS Average Case/Con. (%) Difference (%) P Age adjusted P UBASH3A 13517 -156 45.1/41.8 3.30 0.0019 0.008 TRIM3 — -357 8.58/8.98 -0.40 0.3128 0.059 17029 -348 8.76/9.97 -1.21 0.0004 0.001 19207 -331 6.91/7.93 -1.02 0.0050 0.022 CTSZ 23241 -292 16.5/17.7 -1.20 0.0953 0.271 — -279 27.8/29.1 -1.30 0.8425 0.798 HIPK3 — 375 48.0/55.3 -7.30 0.0184 0.061 25599 390 49.8/57.1 -7.30 0.0135 0.050 — 422 39.8/45.8 -6.00 0.0164 0.070 — 438 42.5/49.0 -6.50 0.0173 0.061 CDH5 8829 -243 30.3/33.4 -3.10 0.0450 0.092 CREB3L3 — 30 3.23/2.85 0.38 0.7188 0.828 23739 42 18.7/21.6 -2.90 0.3251 0.288 — 48 18.5/22.3 -3.80 0.1549 0.174 ID, ID from Illumina HumanMethylation27 BeadChip Dis. to TSS, Distance to the Transcription Start Site. The targeted CpG sites are highlighted in bold. Table 4 Results from the replication cohort ID, ID from Illumina HumanMethylation27 BeadChip Dis. to TSS, Distance to the Transcription Start Site. The targeted CpG sites are highlighted in bold. change in the epigenetic profile of one specific cell type is related to obesity. In this regard, the actual epigenetic differences might be more substantial than reported here but only present in this specific blood leukocyte cell type. Future studies on epigenetic profiling of var- ious types of cell populations in the leukocytes are war- ranted to gain a greater understanding of the epigenetic dysregulation in obesity. Furthermore, this kind of study will be able to identify the cells specifically reflecting obesity-associated methylation changes, which is of great interest in itself. Transcriptional profiling studies will be very valuable in understanding whether the DNA methylation status differences are associated with differ- ences in gene expression. Unfortunately, cellular RNA is not available for the samples used in the current study, which were selected from several existing cohorts. comparisons with the lean subjects (age adjusted P = 0.022 as shown in Table 4). Similar to the CpG site in the UBASH3A gene, these two CpG sites are not located at any known transcription factor binding sites. Future studies will be needed to test whether the density of the methylation or the methylation of specific CpG sites in this CpG island has effect on TRIM3 gene expression. The observed DNA methylation differences between obese cases and controls were relatively small. Discussion Because the pyrosequencing assay also measured the methylation levels of the -331 CpG site (Illumina ID: 19207), we were able to show that also for this site obese subjects had lower methylation levels in study locates at the promoter region with a distance of 156 bp to the transcription start site. There is a possibi- lity that methylation of this CpG site or other CpG sites with methylation levels correlated with this CpG site inhibits the interactions between DNA sequence and nuclear proteins, resulting in decreased gene expression. In-silico analysis of the region of this CpG site using TFSEARCH software [22] did not find this CpG site located at any known transcription factor binding sites. However, methylation of this CpG site may suppress gene transcription by recruiting methylcytosine-binding proteins that in turn associate with large protein com- plexes containing corepressors and histone deacetylases. The binding of these complexes to DNA may lead to a change in the chromatin structure from an active to an inactive form [23]. This speculation needs to be confirmed. study locates at the promoter region with a distance of 156 bp to the transcription start site. There is a possibi- lity that methylation of this CpG site or other CpG sites with methylation levels correlated with this CpG site inhibits the interactions between DNA sequence and nuclear proteins, resulting in decreased gene expression. In-silico analysis of the region of this CpG site using TFSEARCH software [22] did not find this CpG site located at any known transcription factor binding sites. However, methylation of this CpG site may suppress gene transcription by recruiting methylcytosine-binding proteins that in turn associate with large protein com- plexes containing corepressors and histone deacetylases. The binding of these complexes to DNA may lead to a change in the chromatin structure from an active to an inactive form [23]. This speculation needs to be confirmed. Table 3 General characteristics of the subjects of the replication cohort Cases Controls N 45 46 Age, years 20.3 ± 5.0 (14.1 to 29.5) 17.6 ± 3.1 (12.7 to 27.9) †BMI, kg/m2 37.4 ± 6.1 (30.5 to 59.9) 18.9 ± 1.2 (16.4 to 22.0) †SBP, mmHg 125.8 ± 14.2 (97 to 175) 113.5 ± 8.5 (100 to 134) Means ± SD (Range) : P < 0.01; †: P < 0.001 Table 3 General characteristics of the subjects of the replication cohort Wang et al. Acknowledgements h h The participants in this study were recruited by several NIH funded projects including HL69999, HL56622, HL077230 and HL64157. XW is also funded by the American Heart Association (0730156N) and the NIH (HL086530). Received: 4 November 2010 Accepted: 21 December 2010 Published: 21 December 2010 In this study, we identified several reproducible changes in DNA methylation of peripheral blood leukocytes between obese cases and lean controls. This study pro- vides evidence that obesity is associated with methyla- tion changes in blood leukocyte DNA. Further studies are warranted to determine the causal direction of this relationship as well as whether such methylation changes can lead to immune dysfunction. Such studies will have the ability to identify new insight into disease etiology and provide new targets for prevention of obe- sity related diseases such as cardiovascular diseases and type 2 diabetes. Authors’ contributions XW had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. XW and HS contributed to the study concept and design. XW, SS, HS, HS and DM analyzed and interpreted the data. HZ, YD, BG, FT and GH acquired the data. XW drafted the manuscript. XW, HS, GH and BLM critically revised the manuscript for important intellectual content. XW and SS did the statistical analysis. Author details 1 1Georgia Prevention Institute, Department of Pediatrics, Medical College of Georgia, Augusta, GA, USA. 2Department of Pediatrics, Medical College of Georgia, Augusta, GA, USA. 3The Child Health Discovery Institute, Medical College of Georgia, Augusta, GA, USA. 4The Cancer Research Center, Medical College of Georgia, Augusta, GA, USA. 5Unit of Genetic Epidemiology and Bioinformatics, Department of Epidemiology, University Medical Center Groningen, University of Groningen, The Netherlands. 6Department of Medicine, Division of Cardiology, Emory University School of Medicine, Atlanta, GA, USA. Additional material Additional file 2: Meta-analysis for the six CpG sites. Additional file 3: Correlations among the multiple CpG sites within one gene. Competing interests Th h d l h The authors declare that they have no competing interests. The authors declare that they have no competing interests. Discussion Third, the current study is a small pilot study with the genome wide step conducted only in seven obese and seven lean subjects. Future studies with much larger sample size are war- ranted to discover a more complete profile of obesity related methylation changes. Fourth, although all the obese cases in the genome-wide step were free of clini- cal diseases such as CVD and diabetes, some of them already showed evidence of insulin resistance. It is possi- ble that insulin resistance may be the factor triggering methylation changes and not the obesity per se. Recent literature supports the postulate that glucose restriction alters gene expression through epigenetic mechanisms [27]. We conducted further analysis within the obese cases and did not observe that any of the CpGs showed methylation difference between cases with impaired fast- ing glucose (n = 3) versus those with normal glucose levels (n = 3), with the most significant CpG site having an adjusted P-value of 0.85. However, the limited sample size prevents us from drawing any conclusion based on this analysis. Unfortunately, fasting glucose was not measured in the replication cohort. Future studies with large sample sizes and more detailed phenotypes will be needed to clarify this question. Abbreviations AA: African American; BMI: Body Mass Index; BP: Blood Pressure; CDH5: Cadherin 5 type 2 preproprotein; CREB3L3: cAMP responsive element binding protein 3-like 3; CTSZ: Cathepsin Z preproprotein; DBP: Diastolic Blood Pressure; EA: European American; FDR: False Discovery Rate; HIPK3: Homeodomain interacting protein kinase 3; LACHY: Lifestyle, Adiposity, and Cardiovascular Health in Youth; PHAT: Prevention of Hypertension in African American Teens; QUICKI: Quantitative Insulin-Sensitivity Check Index; SBP: Systolic Blood Pressure; TRIM3: Tripartite motif-containing 3; UBASH3A: Ubiquitin-associated and SH3 domain-containing protein A; WGA: Whole Ubiquitin-associated and SH3 domain-containing protein A; WGA: Whole Genome Amplified. Discussion Second, in the current study, the Infinium HumanMethylation27 Beadchip was selected because of its quantitative mea- sure at each CpG site. However, the limited coverage of this genome wide chip will restrict the findings to cer- tain CpG sites within certain genes. Future studies should use chips with more complete coverage of the genome such as the recently released 450K Infinium Methylation BeadChip from Illumina. Third, the current study is a small pilot study with the genome wide step conducted only in seven obese and seven lean subjects. Future studies with much larger sample size are war- ranted to discover a more complete profile of obesity related methylation changes. Fourth, although all the obese cases in the genome-wide step were free of clini- cal diseases such as CVD and diabetes, some of them already showed evidence of insulin resistance. It is possi- ble that insulin resistance may be the factor triggering methylation changes and not the obesity per se. Recent literature supports the postulate that glucose restriction alters gene expression through epigenetic mechanisms [27]. We conducted further analysis within the obese cases and did not observe that any of the CpGs showed methylation difference between cases with impaired fast- ing glucose (n = 3) versus those with normal glucose levels (n = 3), with the most significant CpG site having an adjusted P-value of 0.85. However, the limited sample size prevents us from drawing any conclusion based on this analysis. Unfortunately, fasting glucose was not measured in the replication cohort. Future studies with large sample sizes and more detailed phenotypes will be needed to clarify this question. cause or the consequence of obesity. Future studies on subjects who changed their body size status will be needed to clarify the causal directions. On the other hand, epigenetic regulation is tissue specific. In this regard, the target tissue to identify epigenetic variations responsible for obesity should be the hypothalamus of the brain rather than peripheral leukocytes. Second, in the current study, the Infinium HumanMethylation27 Beadchip was selected because of its quantitative mea- sure at each CpG site. However, the limited coverage of this genome wide chip will restrict the findings to cer- tain CpG sites within certain genes. Future studies should use chips with more complete coverage of the genome such as the recently released 450K Infinium Methylation BeadChip from Illumina. Discussion They were 10.1% and 2.54% in the genome wide step and 3.3% and 1.2% in the replication step for the UBASH3A gene and the TRIM3 gene, respectively. This modest level of differences raises an important question: what is the biological significance of changes in methylation on this degree? In this study, we used the DNA from leuko- cytes, which represent different cell populations with distinct epigenetic profiles. It is plausible that only the There are several strengths of this study. First, we selected obese cases and controls with extreme pheno- types, which maximizes the power to make discoveries. Second, we focused on youth and young adults with the distinct advantage that the results are not confounded by obesity comorbidities or use of medication, both of which are very common in adult subjects with obesity. Third, a hypothesis-free genome wide approach was used. This approach supersedes the limitations imposed by candidate gene methylation studies and allows searching the whole genome in an unbiased manner. Table 5 Gene-ontology Analysis Biological process GO. ID Term P- value Adjust P Level 3 GO:0006955 Immune response 0.017 0.690 Level 3 GO:0007588 Excretion 0.028 0.690 Level 3 GO:0001775 Cell activation 0.042 0.690 Level 3 GO:0001816 Cytokine production 0.044 0.690 Level 3 GO:0009607 Response to biotic stimulus 0.047 0.690 Molecular function GO. ID Term P value Adjust P Level 3 GO:0019842 Vitamin binding 0.009 0.522 Level 3 GO:0003823 Antigen binding 0.010 0.522 Level 3 GO:0004872 Receptor activity 0.032 0.756 Level 3 GO:0008289 Lipid binding 0.035 0.756 Table 5 Gene-ontology Analysis Table 5 Gene-ontology Analysis Interpretation of these data is also limited by several constraints. First, in this study we hypothesize that obe- sity will lead to methylation changes in the DNA of leu- kocytes, which further lead to obesity related co- morbidities. However, our study design cannot deter- mine whether the identified methylation changes are the Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 Page 7 of 8 Page 7 of 8 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 cause or the consequence of obesity. Future studies on subjects who changed their body size status will be needed to clarify the causal directions. On the other hand, epigenetic regulation is tissue specific. In this regard, the target tissue to identify epigenetic variations responsible for obesity should be the hypothalamus of the brain rather than peripheral leukocytes. Additional file 1: Primer sequences and PCR conditions for bisulfit- pyrosequencing analysis. Additional file 2: Meta-analysis for the six CpG sites. Additional file 3: Correlations among the multiple CpG sites within one gene. Conclusion Received: 4 November 2010 Accepted: 21 December 2010 Published: 21 December 2010 Received: 4 November 2010 Accepted: 21 December 2010 Published: 21 December 2010 References 1. Danaei G, Ding EL, Mozaffarian D, Taylor B, Rehm J, Murray CJ, Ezzati M: The preventable causes of death in the United States: comparative risk assessment of dietary, lifestyle, and metabolic risk factors. PLoS Medicine 2009, 6:e1000058. 1. Danaei G, Ding EL, Mozaffarian D, Taylor B, Rehm J, Murray CJ, Ezzati M: The preventable causes of death in the United States: comparative risk assessment of dietary, lifestyle, and metabolic risk factors. PLoS Medicine 2009, 6:e1000058. 2. Misra A, Khurana L: Obesity and the metabolic syndrome in developing countries. The Journal of Clinical Endocrinology and Metabolism 2008, 93: S9-30. 3. Poirier P, Giles TD, Bray GA, Hong Y, Stern JS, Pi-Sunyer FX, Eckel RH: Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association Scientific Statement on Obesity and Heart Disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation 2006, 113:898-918. Page 8 of 8 Page 8 of 8 Wang et al. BMC Medicine 2010, 8:87 http://www.biomedcentral.com/1741-7015/8/87 4. Wild S, Roglic G, Green A, Sicree R, King H: Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 2004, 27:1047-1053. 24. Ozato K, Shin DM, Chang TH, Morse HC: TRIM family proteins and their emerging roles in innate immunity. Nature Reviews 2008, 8:849-860. 25. Rajsbaum R, Stoye JP, O’Garra A: Type I interferon-dependent and -independent expression of tripartite motif proteins in immune cells. European Journal of Immunology 2008, 38:619-630. 5. Anderson AS, Caswell S: Obesity management–an opportunity for cancer prevention. Surgeon 2009, 7:282-285. 26. Martinez FO, Gordon S, Locati M, Mantovani A: Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression. J Immunol 2006, 177:7303-7311. 6. Karalis KP, Giannogonas P, Kodela E, Koutmani Y, Zoumakis M, Teli T: Mechanisms of obesity and related pathology: linking immune responses to metabolic stress. The FEBS Journal 2009, 276:5747-5754. 7. Wilson AG: Epigenetic regulation of gene expression in the inflammatory response and relevance to common diseases. Journal of Periodontology 2008, 79:1514-1519. 27. Li Y, Liu L, Tollefsbol TO: Glucose restriction can extend normal cell lifespan and impair precancerous cell growth through epigenetic control of hTERT and p16 expression. FASEB J 24:1442-1453. 8. References Nile CJ, Read RC, Akil M, Duff GW, Wilson AG: Methylation status of a single CpG site in the IL6 promoter is related to IL6 messenger RNA levels and rheumatoid arthritis. Arthritis and Rheumatism 2008, 58:2686-2693. Pre-publication history The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1741-7015/8/87/prepub The pre-publication history for this paper can be accessed here: http://www.biomedcentral.com/1741-7015/8/87/prepub 9. Smale ST: The establishment and maintenance of lymphocyte identity through gene silencing. Nature Immunology 2003, 4:607-615. doi:10.1186/1741-7015-8-87 Cite this article as: Wang et al.: Obesity related methylation changes in DNA of peripheral blood leukocytes. BMC Medicine 2010 8:87. 10. Gutin B, Johnson MH, Humphries MC, Hatfield-Laube JL, Kapuku GK, Allison JD, Gower BA, Daniels SR, Barbeau P: Relationship of visceral adiposity to cardiovascular disease risk factors in black and white teens. Obesity (Silver Spring, Md 2007, 15:1029-1035. 11. Wang X, Poole JC, Treiber FA, Harshfield GA, Hanevold CD, Snieder H: Ethnic and gender differences in ambulatory blood pressure trajectories: results from a 15-year longitudinal study in youth and young adults. Circulation 2006, 114:2780-2787. 12. Ge D, Dong Y, Wang X, Treiber FA, Snieder H: The Georgia Cardiovascular Twin Study: influence of genetic predisposition and chronic stress on risk for cardiovascular disease and type 2 diabetes. Twin Res Hum Genet 2006, 9:965-970. 13. Zhu H, Chao J, Guo D, Li K, Huang Y, Hawkins K, Wright N, Stallmann- Jorgensen I, Yan W, Harshfield GA, Dong Y: Urinary prostasin: a possible biomarker for renal pressure natriuresis in black adolescents. Pediatric Research 2009, 65:443-446. 14. Snieder H, Dong Y, Barbeau P, Harshfield GA, Dalageogou C, Zhu H, Carter ND, Treiber FA: Beta2-adrenergic receptor gene and resting hemodynamics in European and African American youth. American Journal of Hypertension 2002, 15:973-979. 15. Smyth GK: Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology 2004, 3:Article3. 15. Smyth GK: Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Statistical Applications in Genetics and Molecular Biology 2004, 3:Article3. 16. Klipper-Aurbach Y, Wasserman M, Braunspiegel-Weintrob N, Borstein D, Peleg S, Assa S, Karp M, Benjamini Y, Hochberg Y, Laron Z: Mathematical formulae for the prediction of the residual beta cell function during the first two years of disease in children and adolescents with insulin- dependent diabetes mellitus. Medical Hypotheses 1995, 45:486-490. 17. Pre-publication history Sanna S, Jackson AU, Nagaraja R, Willer CJ, Chen WM, Bonnycastle LL, Shen H, Timpson N, Lettre G, Usala G, Chines PS, Stringham HM, Scott LJ, Dei M, Lai S, Albai G, Crisponi L, Naitza S, Doheny KF, Pugh EW, Ben- Shlomo Y, Ebrahim S, Lawlor DA, Bergman RN, Watanabe RM, Uda M, Tuomilehto J, Coresh J, Hirschhorn JN, Shuldiner AR, et al: Common variants in the GDF5-UQCC region are associated with variation in human height. Nature Genetics 2008, 40:198-203. 17. Sanna S, Jackson AU, Nagaraja R, Willer CJ, Chen WM, Bonnycastle LL, Shen H, Timpson N, Lettre G, Usala G, Chines PS, Stringham HM, Scott LJ, Dei M, Lai S, Albai G, Crisponi L, Naitza S, Doheny KF, Pugh EW, Ben- Shlomo Y, Ebrahim S, Lawlor DA, Bergman RN, Watanabe RM, Uda M, Tuomilehto J, Coresh J, Hirschhorn JN, Shuldiner AR, et al: Common variants in the GDF5-UQCC region are associated with variation in human height. Nature Genetics 2008, 40:198-203. 18. Al-Shahrour F, Diaz-Uriarte R, Dopazo J: FatiGO: a web tool for finding significant associations of Gene Ontology terms with groups of genes. Bioinformatics (Oxford, England) 2004, 20:578-580. 18. Al-Shahrour F, Diaz-Uriarte R, Dopazo J: FatiGO: a web tool for finding significant associations of Gene Ontology terms with groups of genes. Bioinformatics (Oxford, England) 2004, 20:578-580. 19. Tsygankov AY: Multidomain STS/TULA proteins are novel cellular regulators. IUBMB Life 2008, 60:224-231. 19. Tsygankov AY: Multidomain STS/TULA proteins are novel cellular regulators. IUBMB Life 2008, 60:224-231. 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 20. Collingwood TS, Smirnova EV, Bogush M, Carpino N, Annan RS, Tsygankov AY: T-cell ubiquitin ligand affects cell death through a functional interaction with apoptosis-inducing factor, a key factor of caspase-independent apoptosis. The Journal of Biological Chemistry 2007, 282:30920-30928. 20. Collingwood TS, Smirnova EV, Bogush M, Carpino N, Annan RS, Tsygankov AY: T-cell ubiquitin ligand affects cell death through a functional interaction with apoptosis-inducing factor, a key factor of caspase-independent apoptosis. The Journal of Biological Chemistry 2007, 282:30920-30928. Pre-publication history 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: 21. Mikhailik A, Ford B, Keller J, Chen Y, Nassar N, Carpino N: A phosphatase activity of Sts-1 contributes to the suppression of TCR signaling. Molecular Cell 2007, 27:486-497. 22. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, Kel OV, Ignatieva EV, Ananko EA, Podkolodnaya OA, Kolpakov FA, Podkolodny NL, Kolchanov NA: Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Research 1998, 26:362-367. 23. Attwood JT, Yung RL, Richardson BC: DNA methylation and the regulation of gene transcription. Cell Mol Life Sci 2002, 59:241-257.
https://openalex.org/W3200896096	https://www.zora.uzh.ch/id/eprint/207721/1/Caselli_journal.pone.0255580_2021.pdf	English	null	Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT-qPCR	PloS one	2,021	cc-by	10,258	Zurich Open Repository and Archive University of Zurich University Library Strickhofstrasse 39 CH-8057 Zurich www.zora.uzh.ch Year: 2021 Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT-qPCR Caselli, Emanuele ; Pelliccia, Cristina ; Teti, Valeria ; Bellezza, Guido ; Mandarano, Martina ; Ferri, Ivana ; Hartmann, Kerstin ; Laible, Mark ; Sahin, Ugur ; Varga, Zsuzsanna ; Lupi, Chiara ; Stracci, Fabrizio ; Sidoni, Angelo DOI: https://doi.org/10.1371/journal.pone.0255580 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-207721 Journal Article Published Version The following work is licensed under a Creative Commons: Attribution 4 Posted at the Zurich Open Repository and Archive, University of Zurich ZORA URL: https://doi.org/10.5167/uzh-207721 Journal Article Published Version Originally published at: Caselli, Emanuele; Pelliccia, Cristina; Teti, Valeria; Bellezza, Guido; Mandarano, Martina; Ferri, Ivana; Hartmann, Kerstin; Laible, Mark; Sahin, Ugur; Varga, Zsuzsanna; Lupi, Chiara; Stracci, Fabrizio; Sidoni, Angelo (2021). Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT-qPCR. PLoS ONE, 16(9):e0255580.t DOI: https://doi.org/10.1371/journal.pone.0255580 PLOS ONE RESEARCH ARTICLE Methods Herein, we investigated discrepancies in the assessment of estrogen receptor (ER), proges- terone receptor (PR), human epidermal growth factor receptor 2 (HER2) and marker of pro- liferation Ki67 comparing routinely obtained IHC (and FISH) data (ORI) with the results of manual (REV) and semi-automated (DIA) re-evaluation of the original IHC slides and then with RNA expression data from the same tissue block using the MammaTyper® (MT) gene expression assay. Published: September 23, 2021 Copyright: © 2021 Caselli 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. Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT-qPCR Emanuele CaselliID1☯, Cristina Pelliccia1☯, Valeria Teti1, Guido Bellezza1, Martina MandaranoID1, Ivana Ferri1, Kerstin Hartmann2, Mark Laible2, Ugur Sahin2, Zsuzsanna Varga3, Chiara LupiID4, Fabrizio Stracci4,5, Angelo SidoniID1 Emanuele CaselliID1☯, Cristina Pelliccia1☯, Valeria Teti1, Guido Bellezza1, Martina MandaranoID1, Ivana Ferri1, Kerstin Hartmann2, Mark Laible2, Ugur Sahin2, Zsuzsanna Varga3, Chiara LupiID4, Fabrizio Stracci4,5, Angelo SidoniID1 1 Department of Medicine and Surgery, Section of Anatomic Pathology and Histology, Medical School, University of Perugia, Perugia, Italy, 2 BioNTech Diagnostics GmbH, Mainz, Germany, 3 Institute for Pathology and Molecular Pathology, Universita¨tsspital Zu¨rich, Zu¨rich, Switzerland, 4 Umbria Cancer Registry, Perugia, Italy, 5 Department of Medicine and Surgery, Section of Public Health, University of Perugia, Perugia, Italy a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ☯These authors contributed equally to this work. ☯These authors contributed equally to this work. * caselliemanuele1986@gmail.com ☯These authors contributed equally to this work. * caselliemanuele1986@gmail.com * caselliemanuele1986@gmail.com Purpose Decades of quality control efforts have raised the standards of immunohistochemistry (IHC), the principle method used for biomarker testing in breast cancer; however, computa- tional pathology and reverse transcription quantitative PCR (RT-qPCR) may also hold prom- ise for additional substantial improvements. Editor: Fernando Schmitt, Medical Faculty of Porto University and IPATIMUP, PORTUGAL Received: December 13, 2020 Accepted: July 19, 2021 Published: September 23, 2021 Received: December 13, 2020 Accepted: July 19, 2021 Published: September 23, 2021 Copyright: © 2021 Caselli 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 manuscript and its Supporting Information. Data Availability Statement: All relevant data are within the manuscript and its Supporting Information. OPEN ACCESS Citation: Caselli E, Pelliccia C, Teti V, Bellezza G, Mandarano M, Ferri I, et al. (2021) Looking for more reliable biomarkers in breast cancer: Comparison between routine methods and RT- qPCR. PLoS ONE 16(9): e0255580. https://doi.org/ 10.1371/journal.pone.0255580 Citation: Caselli E, Pelliccia C, Teti V, Bellezza G, Mandarano M, Ferri I, et al. (2021) Looking for more reliable biomarkers in breast cancer: ☯These authors contributed equally to this work. * caselliemanuele1986@gmail.com Introduction Analysis of breast cancer tissues by pathologists generates a wealth of information that can be used to classify disease for prognostic purposes and help guide therapeutic decisions. Recent molecular studies have lent further support to the historical importance of hormone receptors, growth factors and perturbations in proliferation as the key operators of breast cancer initiation and progression [1]. Thus, it is not surprising that assessing the expression of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor two (HER2) and prolifera- tion remains the cornerstone of breast cancer diagnosis and treatment [2], although the well- known problems of accuracy and reproducibility which have necessitated the issuing of various guidelines [3–6], are still persistent [7]. In fact, IHC is methodologically error-prone, as it is a multi-step process, highly exposed to variabilities of various kinds, such as from the use of non- standardized local practices or from different, non-equally performing albeit approved reagents [8, 9]. However, the main source of variation is linked to the intrinsic limits of immunohisto- chemical stains, both for marked influences of the pre-analytical phase [7] and the subjectivity of interpretative criteria [10] as well as extreme intra- and inter-tumoral heterogeneity [11] Some IHC markers are more susceptible to methodological failures owing to these limita- tions. The proliferation marker Ki67 is the most prominent example, because it is expressed heterogeneously in tissues and is continuously distributed, which makes it difficult to identify a single, reproducible cut-off [10]. For HER2, IHC has been greatly improved by FISH. Unfor- tunately, FISH is a method that requires a specific expertise, a fluorescent microscope, and is more time-consuming. The complexities of FISH are exemplified in a recent guideline update, which aims to address an ever-expanding list of methodological intricacies [12]. However, there are other reliable methods of in situ hybridization for the assessment of HER2 amplifica- tion that surpass these drawbacks, such as CISH, SISH and DDISH. Even though modern scientific advances were immensely assisted by corresponding tech- nological breakthroughs, such as gene arrays and reverse transcription quantitative PCR (RT- qPCR), the landscape of breast cancer diagnostics remains dominated by conventional hema- toxylin and eosin (H&E) stains and by immunohistochemistry (IHC) [13]. Conclusions specific roles of these authors are articulated in the ‘author contributions’ section.” The remaining authors received no specific funding for this work. Determination of the breast cancer biomarkers ER, PR, HER2 and Ki67 at the mRNA level shows high degree of correlation with IHC and compares well with correlations between original with subsequent independent manual or semi-automated IHC assessments. The use of methods with wider dynamic range and higher reproducibility such as RT-qPCR may offer more precise assessment of endocrine responsiveness, improve Ki67 standardization and help resolve HER2 cases that remain equivocal or ambiguous by IHC/FISH. In sum- mary, our findings seem to configure RT-qPCR as a complementary method to be used in cases of either equivocal results or presenting, at the traditional determination assays, bio- markers expressions close to the cut-off values. Competing interests: Kerstin Hartmann, Mark Laible, Ugur Sahin: Salary and stock ownership BioNTech Diagnostics GmbH / BioNTech AG. These commercial affiliations do not alter our adherence to PLOS ONE policies on sharing data and materials. Mark Laible: Patent ownership of WO 2015/024942, Commercialized as MammaTyper (TM) Kit. The other authors declare that they have no conflict of interest. Results Correlation for ER and PR was high between ORI IHC and the other three study methods (REV, DIA and RT-qPCR). For HER2, 10 out of 96 discrepant cases can be detected between ORI and REV that involved at least one call in the equivocal category (except for one case). For Ki67, 22 (29.1%) cases were categorized differently by either REV alone (n = 17), DIA alone (n = 15) or both (n = 10) and 28 cases (29.2%) for RT-qPCR. Most of the dis- crepant Ki67 cases changed from low to high between the original and following assess- ment and belonged to the intermediate Ki67 expression range (between 9 and 30%). Funding: KH, ML and US received salary from and have stock ownership of BioNTech Diagnostics GmbH / BioNTech AG. However, BioNTech Diagnostics GmbH / BioNTech AG provided support in the form of salaries for authors KH, ML and US, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 1 / 18 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers specific roles of these authors are articulated in the ‘author contributions’ section.” The remaining authors received no specific funding for this work. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 Patient selection Between March 2010 and February 2012, 288 women underwent surgery for invasive breast carcinoma (Section of Anatomic Pathology and Histology, Perugia, Italy). The IHC results from patient histopathology reports were screened to identify cases with varying levels of diag- nostic difficulty (Fig 1). Available material from 116 cases, including FFPE blocks, H&E slides and IHC slides, were retrieved from the archive. The previously reported results for ER, PR, Ki67 and HER2 were registered as the original data (ORI) and consisted of assessments deliv- ered during diagnostic routine by 6 trained pathologists according to available guidelines at that time [3, 6]. An experienced pathologist and FISH expert (GB) performed the ORI FISH data. Informed consent The current study is a retrospective one, conducted on data and biological material already fully anonymized before the authors accessed them. The patients’ data and corresponding samples were accessed in the period between March 2010 and February 2012. According to the internal rules both of the University of Perugia and of the Hospital of Perugia, we have consid- ered only the biological material in respect of which the patients expressed his/her specific informed consent to use such a material for research purposes; we also have observed all rules concerning confidentiality and protection of person data, in accordance with European Union, International and National rules. Compliance with ethical standards All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Introduction Current consensus does not favor the use of alternative methods such as for example RT-qPCR for assessing the expression of individual markers, and even though reliable and reproducible solutions are commercially available, their role however being mostly complementary [14, 15]. In 2017, the US Food and Drug Administration (FDA) announced the approval of the first whole slide imaging (WSI) system for primary diagnosis in surgical pathology [16]. Aside PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 2 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers from RT-qPCR, automated digital image analysis (DIA) may also pave the way for more pre- cise biomarker assessment and more accurate histopathological breast cancer diagnosis and has the potential to provide the objectivity and reliability required to transform biomarker test- ing in breast cancer. The entire process, however, involves several steps, including image acquisition, storage and management, annotation and viewing or sharing, each one requiring stringent quality control procedures [17]. Therefore, considering the diversity of candidate analytical methods and ongoing chal- lenges with IHC/FISH, the decision for or against alternative methods for biomarker testing in breast cancer remains complex. In this respect, local attempts to benchmark new techniques against testing traditions may help estimate their feasibility and potential for routine use. In this study, we carried out parallel comparisons between 3 independent IHC assessments, including an image analysis solution and comparisons between these and RT-qPCR, using well-established local IHC methods, international scoring guidelines and the MammaTyper1 gene expression assay. The aim of the study was to detect discordance between methods and discuss their clinicopathological implications. Conventional histopathological assessments The re-evaluation dataset (REV) was compiled as follows: An experienced pathologist (EC) that was not involved in the initial screening and was blinded to clinicopathological informa- tion carried out tumor classifications and grading on the available original H&E slides 3 / 18 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Fig 1. Study design. https://doi org/10 1371/journal pone 0255580 g001 Fig 1. Study design. according to international standards in force at the time of the study [18, 19]. Twenty cases were excluded from further analyses due to inadequate fixation. In a second step the same pathologist re-evaluated ER, PR, Ki67 and HER2 on the existing IHC slides, but 2 cases for ER, 3 cases for PR, 2 cases for Ki67 and 12 cases for HER2 were re-stained, owing to missing origi- nal slides or low staining quality including fading. Standard international recommendations— in force when the research project was conducted, particularly according with the 2013 ASCO/ CAP guidelines [4]—were used for the assessment and quantification of staining reactions [4, 6, 20]. Of note, original assessments for Ki67 were based on quick eyeballing, whereas formal counting was used upon re-evaluation. Due to the impossibility of re-evaluation of the original FISH slides, all HER2 equivocal cases by IHC (2+ score) after REV, regardless of ORI FISH data when available, were subjected to FISH analysis (dual-probe, Leica HER2 FISH system). Furthermore, we performed FISH analysis in all discrepant HER2/ERBB2 cases between IHC/ FISH and MammaTyper1. Interpretation was performed by EC and an additional blinded evaluation was performed by GB in selected cases according to the ASCO/CAP 2013 guidelines (in force at the time of the study), either by counting signals in randomly chosen regions of interest with the assistance of a digital tool (Cytovision software, Leica Biosystems) (EC) or according to international standards in force at the time of the study [18, 19]. Twenty cases were excluded from further analyses due to inadequate fixation. In a second step the same pathologist re-evaluated ER, PR, Ki67 and HER2 on the existing IHC slides, but 2 cases for ER, 3 cases for PR, 2 cases for Ki67 and 12 cases for HER2 were re-stained, owing to missing origi- nal slides or low staining quality including fading. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 Conventional histopathological assessments Standard international recommendations— in force when the research project was conducted, particularly according with the 2013 ASCO/ CAP guidelines [4]—were used for the assessment and quantification of staining reactions [4, 6, 20]. Of note, original assessments for Ki67 were based on quick eyeballing, whereas formal counting was used upon re-evaluation. Due to the impossibility of re-evaluation of the original FISH slides, all HER2 equivocal cases by IHC (2+ score) after REV, regardless of ORI FISH data when available, were subjected to FISH analysis (dual-probe, Leica HER2 FISH system). Furthermore, we performed FISH analysis in all discrepant HER2/ERBB2 cases between IHC/ FISH and MammaTyper1. Interpretation was performed by EC and an additional blinded evaluation was performed by GB in selected cases according to the ASCO/CAP 2013 guidelines (in force at the time of the study), either by counting signals in randomly chosen regions of interest with the assistance of a digital tool (Cytovision software, Leica Biosystems) (EC) or PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 4 / 18 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers manually (GB). Of note, there was no disagreement between the two pathologists. Three selected cases with discrepancies to MammaTyper1 ERBB2 determination were sent to an independent pathologist (ZV) for blinded FISH re-analysis (hybridization and evaluation). Computer-assisted IHC assessment The same pathologist (EC) independently analyzed the expression of hormone receptors and Ki67 with a digital semi-automated image analysis (DIA) system (Aperio ImageScope, Leica Biosystems). All slides were scanned on the Aperio AT2 Scanner (Leica Biosystems) with a 40x magnification lens, using single focus layer without Z-stacking and were displayed on a 4K (ultra-high definition) monitor. On the digital slide (40x magnification), the pathologist selected randomly several regions of interest for ER and PR, whereas for Ki67 areas were cho- sen in such a way as to be representative of heterogeneity when present. Digital image analysis for assessing the percentage of positive nuclei was performed according to the Leica nuclear v9 algorithm. As regards HER2, the digital image analysis was not performed because of an opti- mized and reliable algorithm for assessing the membrane immunolabeling was unavailable at the time of the study. Fluorescence in situ hybridization (FISH) FISH was carried out using the automated Leica HER2 FISH System on a BOND-III platform. Importantly, in the original determinations, FISH was performed using the HER-2 DNA Pro- beKit II (PathVysion) with the Vysis Paraffin Pretreatment Reagent Kit. The dual-color probes identified the centromere 17 (Cep 17) as green signal, whereas the HER2/Neu gene/locus as orange one. Positive and negative controls were always available. Immunohistochemistry (IHC) All reactions were carried out on a BOND-III fully automated immunohistochemistry stainer (Leica Biosystems) with antibodies against ER, PR, HER2 and Ki67 according to vendor proto- cols and the use of appropriate positive and negative controls (Table 1). Of note, all 19 stains that had to be repeated were performed on the same block as the origi- nals and without any reagent or protocol deviations compared to those available from the archive. Only for HER2, all slides prepared for re-evaluation were stained on the Autostainer- Link 48(Dako) using Hercept Test™. Dewaxing and rehydration were performed manually. Antigen retrieval (pH = 6) was performed by heating the slides for 40 minutes at 97˚C using the DakoPT-link. RNA isolation and mRNA quantification with RT-qPCR Rules for nominal categories (positive/negative) for all markers are summarized in Table 2. RNA isolation and mRNA quantification with RT-qPCR The MammaTyper1 test was performed on 10 μm sections. In cases with tumor cell content <20%, manual dissection was conducted. MammaTyper1 test was successfully performed in Table 1. Immunhistochemistry staining specifications. Antibody Clone Dilution Manufacturer Platform ER 6F11 Ready to use Leica Biosystems BOND-III fully automated IHC PR 16 Ready to use Leica Biosystems stainer (Leica Biosystems) Ki67 MIB1 1:100 Dako HER2 Rabbit Anti-Human HER2 protein Ready to use Dako BOND-III fully automated IHC stainer (Leica Biosystems) and AutostainerLink 48 Dako (for the 12 restained slides) https://doi org/10 1371/journal pone 0255580 t001 Table 1. Immunhistochemistry staining specifications. Antibody Clone Dilution Manufacturer Platform ER 6F11 Ready to use Leica Biosystems BOND-III fully automated IHC PR 16 Ready to use Leica Biosystems stainer (Leica Biosystems) Ki67 MIB1 1:100 Dako HER2 Rabbit Anti-Human HER2 protein Ready to use Dako BOND-III fully automated IHC stainer (Leica Biosystems) and AutostainerLink 48 Dako (for the 12 restained slides) https://doi.org/10.1371/journal.pone.0255580.t001 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 5 / 18 Table 1. Immunhistochemistry staining specifications. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 5 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers all 96 cases, applying manual dissection in 16 cases. Extraction of total RNA from FFPE sam- ples was performed using a CE-marked paramagnetic bead-based method (RNXtract1, BioN- Tech Diagnostics, Mainz, Germany) according to the manufacturers’ instructions, as previously described in detail [21]. The expression levels of ERBB2, ESR1, PGR and MKI67 were determined by RT-qPCR using the CE-marked MammaTyper1 IVD kit (BioNTech Diagnostics) on the CFX96TM (BIO-RAD1) platform according to the manufacturer’s instructions. Calculations of delta-delta-Cq values were carried out as described previously [21], however data was further transformed to the relative unit scale (RU) to allow easier inter- pretation. The RU values is calculated by centering the delta-delta-Cq around the marker spe- cific cutoff value. In this way, positive values correspond to gene expression higher than the cutoff and negative values represent an expression below the cutoff while the Cq (log2) scale is retained. Relative units (RU) = ((Median Cq target sample–Combined Reference sample)– (Median Cq target Positive Control–Combined Reference Positive Control))–marker specific cutoff. As cutoffs in this study, for ESR1 and PGR, a lower, diagnostic cutoff was applied (RU 0) which was established to fit the 1% IHC staining for ER and PR. For MKI67, the previously clinically validated cutoff was applied [22]. Statistical analysis The median difference, absolute median difference and standard deviation of the absolute median difference between ORI versus REV and ORI versus DIA for ER, PR and Ki67 were assessed. Calculation of intraclass correlation coefficient (ICC) which estimates the impreci- sion in relationship with the intersample variance was calculated using the formula: ICC = intersample variance/total variance [23, 24]. The Pearson correlation coefficient was used to measure the strength of the association between ORI versus REV, ORI versus DIA and ORI versus RT-qPCR for ER, PR and Ki67. Graphs were prepared using GraphPad Prism 6 (GraphPad Software, La Jolla, USA). A two-sided P value <0.05 was considered significant. Concordance between IHC assessments for ER, PR and Ki67 The values for ICC shown in Table 3 indicate the high agreement between the continuous per- centage scores of ORI IHC and the other two study methods (REV and DIA). There was no discordance for ER status between all three IHC assessments, with 10 ER-negative and 86 ER positive cases. Compared with original values, proportion scores differed in 58 cases upon REV (59.4%) and in 80 cases upon DIA (82.3%). In both cases, increase was more common than reduction (35 vs 23 for REV and 45 vs 35 for DIA), but in only 46 cases the change was in the same direction. Among 3 cases with a difference 25%, two belonged to the original group with positivity between 1–49%. Compared to ORI data, ER percentage positivity dif- fered by more than 10% in 17 out of 96 cases (17.7%) for REV and in 22 out of 96 cases (22.9%) for DIA. Variabilities were further graphically analyzed with Bland-Altman plots (Fig Table 2. Instrument-specific relative units (RU) for CFX96TM (BIO-RAD1) and the corresponding IHC equivalents. RU ERBB2 ESR1 PGR MKI67 Negative Equivocal Positive Negative Low Positive High Positive Negative Low Positive High Positive Low High MammaTyper < -0.7 -0.7 < 0 0 < 0 0 < 1.1 1.1 < 0 0 < 0.5 0.5 < 0 0 IHC equivalent 0 / 1+ 2+ 3+ 0 1–49% 50% 0 1–19% 20% <20% 20% https://doi.org/10.1371/journal.pone.0255580.t002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 6 / 18 Table 2. Instrument-specific relative units (RU) for CFX96TM (BIO-RAD1) and the corresponding IHC equivalents. RU ERBB2 ESR1 PGR MKI67 Negative Equivocal Positive Negative Low Positive High Positive Negative Low Positive High Positive Low High MammaTyper < -0.7 -0.7 < 0 0 < 0 0 < 1.1 1.1 < 0 0 < 0.5 0.5 < 0 0 IHC equivalent 0 / 1+ 2+ 3+ 0 1–49% 50% 0 1–19% 20% <20% 20% https://doi.org/10.1371/journal.pone.0255580.t002 Table 2. Instrument-specific relative units (RU) for CFX96TM (BIO-RAD1) and the corresponding IHC equivalents. Concordance between IHC assessments for ER, PR and Ki67 RU ERBB2 ESR1 PGR MKI67 Negative Equivocal Positive Negative Low Positive High Positive Negative Low Positive High Positive Low High MammaTyper < -0.7 -0.7 < 0 0 < 0 0 < 1.1 1.1 < 0 0 < 0.5 0.5 < 0 0 IHC equivalent 0 / 1+ 2+ 3+ 0 1–49% 50% 0 1–19% 20% <20% 20% https://doi.org/10.1371/journal.pone.0255580.t002 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 6 / 18 le 2. Instrument-specific relative units (RU) for CFX96TM (BIO-RAD1) and the corresponding IHC equivalents. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 6 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Table 3. Median proportion scores for IHC. ORI versus REV ORI versus DIA Median of Δ Median of abs. Δ SD of abs. Δ ICC Median of Δ Median of abs. Δ SD of abs. Δ ICC ER 0.0 3.0 8.6 0.949 0.0 4.0 7.4 0.955 PR 0.0 3.5 9.2 0.957 -2.5 5.0 9.3 0.945 Ki67 -3.5 5.0 5.6 0.903 -4.0 5.0 6.3 0.855 For ER, PR and Ki67 median and absolute (Abs.) median of difference between original and re-evaluation, as well as between original and digital image analysis was calculated. ICC: intraclass correlation. Table 3. Median proportion scores for IHC. For ER, PR and Ki67 median and absolute (Abs.) median of difference between original and re-evaluation, as well as between original and digital image analysis was calculated. ICC: intraclass correlation. For ER, PR and Ki67 median and absolute (Abs.) median of difference between original and re-evaluation, as well as betwee calculated. ICC: intraclass correlation. 2A and 2B) and confirmed the trend of larger variations observed for samples with intermedi- ate staining results (30–80% positive cells). Regarding PR, there were originally 18 negative and 78 positive cases with 2 discrepancies, both involving a difference between the original data and consecutive assessments, whereas the latter two were always concordant. Compared to original values, proportion scores differed in 57 cases upon REV (59.4%) and in 76 cases upon DIA (79.2%). In both cases, increase was more common than reduction (41 vs 16 for REV and 51 vs 25 for DIA), but in only 42 cases was the change in the same direction. Among 6 cases with a difference 25%, four belonged to the original group with positivity between 1–49%. Concordance between IHC assessments for ER, PR and Ki67 Compared to ORI data, PR percentage positivity differed by more than 10% in 30 out of 96 cases (31.3%) for REV and in 34 out of 96 cases (35.4%) for DIA (Fig 2C and 2D). As seen for ER, the largest differences in the staining percent assessments are seen in the lower range (15– 80%). There were 58 low and 38 high Ki67 cases, based on the original evaluation, but 22 (29.1%) were categorized differently by either REV alone (n = 17), DIA alone (n = 15) or both (n = 10). Most (twenty out of 22) discrepant cases changed from low to high between the original and following assessments (Fig 2G and 2H). Setting lower and upper Ki-67 cut-offs at 9% and 30% respectively to distinguish unequivocally low from high cases, we found that discrepant cases lied mostly between the two cut-offs, namely in the intermediate Ki67 expression range (median score of discrepant cases = 15%). In stark contrast, cases that were concordant between all 3 assessments, were either clearly low (median Ki67-low score = 5.5%) or clearly high (median Ki67-high score = 33.5%). As expected, discordance rates were significantly higher in the intermediate group (44.2%) compared with the low- and high groups (6.8% and 4.2% respectively). Compared to original values, proportion scores differed in 75 cases upon REV (78.1%) and in 94 cases upon DIA (97.9%). Upgrading was in both cases more common than downgrading (62 vs 21 for REV and 71 vs 23 for DIA) and was in the same direction in 63 cases. Discordant and concordant cases did not differ by any of the available clinic-patho- logical parameters. Compared to ORI data, Ki67 proportion scores differed by more than 10% in 26 out of 96 cases (27.1%) for REV and in 28 out of 96 cases (29.2%) for DIA (Fig 2E and 2F). With regard to absolute differences between the assessments, higher Ki67stainings (>15%) had clearly higher deviations than samples with a low Ki67 staining percentage (<15%). https://doi.org/10.1371/journal.pone.0255580.t003 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 HER2 discrepancies Based on the original combined IHC/FISH data, there were 17 positive cases for HER2. Except for one case, all other discrepancies involved at least one call in the equivocal category (2013 ASCO/CAP guidelines, which were employed at the time of the current study). Here, the REV FISH result was negative (in line with the RT-qPCR result and different than the ORI assess- ment that was positive). As shown in Fig 4A, re-evaluation identified 6 potential false positives. All 13 ORI HER2 equivocal cases remained equivocal upon REV, but REV identified 9 addi- tional equivocal cases, 3 of which were ORI 3+, whereas 6 were ORI 0/1+. These 9 discrepant cases were subjected to FISH—evaluated according to the 2013 ASCO/CAP guidelines, which were in force at the time of the present study—, which agreed with the original IHC assessment in 6 cases. From the 14 cases with an ORI FISH result, FISH analysis was repeated for 13 sam- ples. All 9 negative cases and 1 equivocal case were confirmed by re-evaluation. However, from the 3 ORI FISH positive results only one was confirmed in re-evaluation, while one was identified as negative and one as equivocal during the second round of re-evaluation by ZV. Furthermore, the blinded FISH re-analysis by ZV differed also in the 2 other cases tested: none case has been found amplified, but both were defined equivocal. As shown in Fig 4B, discordance for HER2/ERBB2 was typically associated with cases being equivocal by at least one IHC assessment (ORI or REV). One ERBB2 potential false negative could be rescued by the introduction of an equivocal zone spanning 0.7 RUs to the left of the cut-off (Fig 3G and 3H). Correlation between IHC and RT-qPCR for ER, PR and Ki67 By contrast, discordant cases for progesterone receptor were more often negative by RT-qPCR while being positive by REV or DIA IHC (7/ 9). The RNA expression of discordant cases took values close to the assay cut-off, while corre- sponding protein expression was low (Fig 3C and 3D). Two additional cases were discordant between MammaTyper1 and the original assessment only. The discordance rate between the two methods for the proliferation marker was dependent on the type of IHC assessment being highest for the ORI data (29.2%), followed by comparable rates for DIA and REV (17.7% and 15.6% respectively). Among 32 discordant cases affected by discrepancies between RT-qPCR and any IHC assessment, 23 also displayed discordances between the different IHC data and 29 showed high MKI67 expression (Fig 3E and 3F). Dis- crepant cases had an average median Ki67 score of 16.7% compared to 25% in the concordant group and their RNA expression showed a tendency to cluster around the MKI67cut-off. Dis- cordance was not associated with histology, grade or other available tumor characteristics. Correlation between IHC and RT-qPCR for ER, PR and Ki67 Correlation between original IHC and RT-qPCR was very good for ER, PR and HER2 and good for Ki67 and strikingly comparable to the quality of the correlation between different IHC methods (Table 4). Estrogen receptor status by IHC (all 3 independent assessments) and MammaTyper1 was discordant in 4 cases, with 3 out of 4 being negative at the protein level PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 7 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Fig 2. Correlation between original (ORI) and paired re-evaluated (REV) or digital image analyzed (DIA) samples for IHC. Bland- Altman plots showing differences between original, re-evaluated and digital image analysis for ER (A, B), PR (C, D) and Ki67 (E, F). Cut-off value at ±10% is shown by red dotted lines. Scatterplots showing correlation between original, re-evaluated and digital image analysis for Ki67 (G, H). Cut-off value at 20% for Ki67 is shown by black dotted lines. Red dots highlight discordant cases between original and either re-evaluated or digital image analysis. https://doi.org/10.1371/journal.pone.0255580.g002 Fig 2. Correlation between original (ORI) and paired re-evaluated (REV) or digital image analyzed (DIA) samples for IHC. Bland- Altman plots showing differences between original, re-evaluated and digital image analysis for ER (A, B), PR (C, D) and Ki67 (E, F). Cut-off value at ±10% is shown by red dotted lines. Scatterplots showing correlation between original, re-evaluated and digital image analysis for Ki67 (G, H). Cut-off value at 20% for Ki67 is shown by black dotted lines. Red dots highlight discordant cases between original and either re-evaluated or digital image analysis. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 8 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Table 4. Agreement within IHC (ORI versus REV, ORI versus DIA) and between IHC (ORI) and RT-Qpcr. ER / ESR1 PR /PGR Ki67 / MKI67 IHC cutoff Positive if 1% Positive if 1% Positive if 20% RT-qPCR cutoff Positive if RU 0 Positive if RU 0 Positive if RU 0 ORI vs REV ORI vs DIA ORI vs RT-qPCR ORI vs REV ORI vs DIA ORI vs RT-qPCR ORI vs REV ORI vs DIA ORI vs RT-qPCR Pearson´s coefficient 0.811 0.835 0.826 0.892 0.911 0.912 0.692 0.708 0.709 https://doi.org/10.1371/journal.pone.0255580.t004 and positive at the RNA level (Fig 3A and 3B). Impact of discordance on subtyping Black dotted lines represent the respective cut-off for the RT-qPCR assay (ERBB2: 0.0) and for the positivity of the IHC staining (HER2: 2+). Second dotted line on the x-axis represents a second cut-off for ERBB2 (-0.7) after introduction of an equivocal zone. y-axis and MKI67 RU values on the x-axis. Cut-off values at 20% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots showing correlation between IHC determination (ORI and REV) and RT-qPCR assessment for ERBB2/HER2 (G, H). Black dotted lines represent the respective cut-off for the RT-qPCR assay (ERBB2: 0.0) and for the positivity of the IHC staining (HER2: 2+). Second dotted line on the x-axis represents a second cut-off for ERBB2 (-0.7) after introduction of an equivocal zone. https://doi.org/10.1371/journal.pone.0255580.g003 https://doi.org/10.1371/journal.pone.0255580.g003 Luminal B-like upon re-evaluation (23.7% for manual, 18.4% for digital) or when MammaTy- per1 RNA data were used instead of IHC (34.2%) (Fig 5). Discrepant cases in the assessment of ER led to four luminal to non-luminal switches. As regards HER2, the discrepancies highlighted by Mammatyper1 allowed us to identify and better stratify 3 originally FISH-discrepant cases, overcoming the issues related to both laboratory quality assurance and inter-/intraobserver subjectivity in interpreting its immuno- histochemical and ISH assessment. Impact of discordance on subtyping Each tumor was assigned a molecular subtype according to the surrogate definitions of the St. Gallen expert panel [25, 26]. Not surprisingly considering the classification rules, the effect of discordance in the assessment of Ki67 was expressed as discrepancies mainly in the luminal categories with many originally Luminal A-like cases being subsequently categorized as PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 9 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Fig 3. Comparison between IHC and RT-qPCR (MammaTyper1) for estrogen receptor (A, B), progesterone receptor (C, D), proliferation marker Ki67 (E, F) and human epidermal growth factor receptor 2 (G, H). Scatterplots with ORI IHC ER scores (A) or DIA IHC ER score (B) on the y-axis and ESR1 RU values on the x-axis. Cut-off values at 1% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots with ORI IHC PR scores (C) or DIA IHC PR score (D) on the y-axis and PGR RU values on the x-axis. Cut-off values at 1% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots with ORI IHC Ki67 scores (E) or DIA IHC Ki67 score (F) on th E RT-qPCR assessment of breast cancer biomarker Fig 3. Comparison between IHC and RT-qPCR (MammaTyper1) for estrogen receptor (A, B), progesterone receptor (C, D), proliferation marker Ki67 (E, F) and human epidermal growth factor receptor 2 (G, H). Scatterplots with ORI IHC ER scores (A) or DIA IHC ER score (B) on the y-axis and ESR1 RU values on the x-axis. Cut-off values at 1% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots with ORI IHC PR scores (C) or DIA IHC PR score (D) on the y-axis and PGR RU values on the x-axis. Cut-off values at 1% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots with ORI IHC Ki67 scores (E) or DIA IHC Ki67 score (F) on the PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 10 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers y-axis and MKI67 RU values on the x-axis. Cut-off values at 20% for IHC and 0.0 RU for RT-qPCR are shown by black dotted lines. Scatterplots showing correlation between IHC determination (ORI and REV) and RT-qPCR assessment for ERBB2/HER2 (G, H). PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 Discussion Immunohistochemistry showed a high degree of correlation for both ER and PR, even though the original data were generated by at least 6 different pathologists with variable experi- ence in data interpretation, indicating the critical contribution of the 2010 scoring guidelines, used in both cases, in raising the standards of ER and PR testing by harmonizing interpreta- tion. High correlation was also found between RT-qPCR and IHC; however, some cases were discordant as expected when comparing different methodologies and target molecules [27]. Notably, discrepant tumors expressed lower amounts of hormone receptor protein, while cor- responding RNA levels were also near the cut-off probably the result of the association between weak (but not negative) biological signals and uncertainty of measurements. Discor- dance was overall somewhat higher for progesterone receptor, consistent with this marker’s higher quantitative variability [28]. Obviously, the bimodal frequency distribution of estrogen receptor contributed to the high degree of correlation between RNA and protein data, whereas the more heterogeneous natural distribution of progesterone receptor probably allows for more inter-section variability (which may have contributed to some of the discordance between IHC and RT-qPCR) eventually inflating the dependency of IHC assessments on rather perceptual and cognitive functions [29]. In fact, anecdotal evidence from daily pathol- ogy practice suggests that the concern over missed treatment opportunities prompts many pathologists to categorize breast tumors as ER-positive, even when staining of nuclei does not formally meet the 1% criterion, despite the fact that low ER and/or PR expression has repeat- edly been linked to non-luminal biology and endocrine resistance [30–34]. Interestingly, most discrepancies for progesterone receptor in our study were scored low by IHC and were found negative at the RNA level, which is consistent with the view that, in tumors reacting weakly to antibodies against hormone receptors, gene expression profiles are indicative of more aggres- sive molecular subtypes [35]. These observations seem to confirm that the RNA testing utility better suited to quantitative biological continuity of hormone receptors, as advocated by some authors; however, the results about it were not always consistent [36–38]. Despite slight improvements, reliable assessment of Ki67 remains beyond reach in most parts of the world, owing mainly to the heterogeneity in the expression of this marker on tissue sections, which results in reduced reproducibility particularly in the mid-range of observations [39, 40]. Discussion Increasing the accuracy of breast cancer diagnosis to deliver the most precise and effective therapy is a continuous process involving on one end efforts to improve the performance of conventional methodologies and on the other end attempts to introduce technologically and operationally advanced analytical alternatives. The aim of our study was to explore discrepan- cies in the assessment of ER, PR, HER2 and Ki67 by comparing routinely obtained IHC (and Fig 4. HER2 IHC/FISH discrepancies. Schematic analysis of discrepancies in HER2: a. between the original assessment (ORI) and re-evaluation (REV) and b. between IHC (ORI) and RT-qPCR. Positive calls are indicated by red filled boxes. Negatives calls are presented in green and equivocal calls in grey, according to the 2013 ASCO/CAP guidelines in force at the time of the study. https://doi.org/10.1371/journal.pone.0255580.g004 Fig 4. HER2 IHC/FISH discrepancies. Schematic analysis of discrepancies in HER2: a. between the original assessment (ORI) and re-evaluation (REV) and b. between IHC (ORI) and RT-qPCR. Positive calls are indicated by red filled boxes. Negatives calls are presented in green and equivocal calls in grey, according to the 2013 ASCO/CAP guidelines in force at the time of the study. ncies. Schematic analysis of discrepancies in HER2: a. between the original assessment (ORI) and re-evaluation (REV) and b. R. Positive calls are indicated by red filled boxes. Negatives calls are presented in green and equivocal calls in grey, according s in force at the time of the study. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 11 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Fig 5. Subtype redistribution after re-evaluation (REV) (A), semi-automated analysis (DIA), (B) and RT-qPCR measurement (C). Fig 5. Subtype redistribution after re-evaluation (REV) (A), semi-automated analysis (DIA), (B) and RT-qPCR Fig 5. Subtype redistribution after re-evaluation (REV) (A), semi-automated analysis (DIA), (B) and RT-qPCR measurement (C). https://doi.org/10.1371/journal.pone.0255580.g005 https://doi.org/10.1371/journal.pone.0255580.g005 https://doi.org/10.1371/journal.pone.0255580.g005 12 / 18 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers FISH) data first with the results of manual and semi-automated re-evaluation of the original IHC slides and then with RNA expression data from the same tissue block using MammaTy- per1. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 Discussion Meanwhile, experts remain reluctant to endorse new technologies, as intensive efforts for the standardization of existing methods have only recently started to payoff [8]. In our lab, the quality of HER2 testing improved upon the adoption of another analytical platform, which may explain discrepancies between original assessments and re-evaluation for cases that were re-stained with the current laboratory set-up that produces less background cytoplasmic staining. The use of the 2013 guidelines in the re- evaluation resulted in more equivocal cases, as has been already described [43]. Importantly, 3 FISH-positive cases with discrepant RT-qPCR results were tested independently and found to be equivocal, indicating that RNA may have helped identify potential false-positives by our own in-house FISH methodology. This finding is of importance, as it raises questions regard- ing the use of FISH as the “gold-standard” [44]. Although the present study presented a limita- tion having used outdated guidelines concerning the HER2 ISH interpretation [4] and the current recommendations [5, 12] do not longer provide an equivocal result for the ISH test, the RT-qPCR could be helpful in the interpretation of cases for which the second reader ISH analysis and the activation of the internal procedure to resolve the issue [5, 12] would never- theless not reach a confident result. Another solution for the standardization of breast cancer biomarkers and particularly for Ki67could be represented by automation with the help of image analysis platforms. In our study, performing DIA was labor-intensive, because regions of interest had to be manually selected due to insufficient performance of automatic detection of tumor nuclei; additional sig- nificant delays were caused by several software failures. In addition, we were not able to docu- ment any clear benefit from semi-automated scoring compared with manual re-evaluation. Nonetheless, recent studies have shown highly promising degree of reproducibility for compu- tational scoring of Ki67 on TMAs or biopsies that outperforms conventional manual analyses even when the latter are carried out under highly standardized conditions [45–47]. Therefore, computational techniques may hold real potential for improving the reliability of quantitative determinations compared with the human eye despite shortcomings that affect routine usabil- ity. On the other hand, automated quantification of staining is similarly limited by the short dynamic range of chromogen-based IHC and thus the information that is harvested for prog- nostic or predictive purposes is less than what can be obtained by RT-qPCR [48]. Discussion Our data clearly suggest that guideline-driven formal counting of nuclei, either per- formed manually or with the help of imaging software, may result in considerable differences in Ki67 scores compared to quick “eyeballing”. Upon re-evaluation, the percentage of positive nuclei were assessed across several high-power fields accounting in this way for potential het- erogeneity, especially from the so-called “hot-spots”, which may have been disregarded or preferentially targeted in the original assessment. Similarly, better averaging of RNA expres- sion throughout the whole section may explain why MammaTyper1 resulted in more MKI67-positive cases compared with IHC, a consistent finding in previous studies involving the assay [22]. Discrepancies may also be explained by the fact that the study pathologist was blinded to clinical data, whereas original assessments would be expected to have been far more biased, because, routinely, microscopic evidence is integrated with other available sources of information to reach an appropriate interpretation, including treatment preferences commu- nicated over tumor boards [41, 42]. As expected, differences in Ki67 between original IHC data and MammaTyper1 resulted in discrepancies in the St. Gallen classification, leading to a significant redistribution of the luminal subtypes in favour of the Luminal B-like category. Such redistributions were previously reported in the FinHer trial, in which MKI67 was supe- rior to IHC in predicting patient outcome and detecting interactions between an RNA-based St. Gallen molecular classification and treatment [22]. Owing to the clinical diversity of the PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 13 / 18 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers selected cases in this study, such effects were not investigated, as any associations between bio- marker groups and outcomes would have been extremely difficult to interpret due to multiple confounding factors and limited events. As opposed to controversies surrounding the use of Ki67, IHC for HER2 supplemented by FISH to resolve equivocal cases consistently receive strong votes of confidence across most jurisdictions [13]. HER2 is a critical biomarker, signal- ing for an aggressive form of breast cancer and, when present in the tissues of patients, it allows them to access effective, but also potentially toxic treatment with HER2 blocking agents. Due to its importance, interpretation rules for HER2 IHC and FISH results are periodically revised to address persisting or newly identified challenges [3–5]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 Discussion Another sig- nificant aspect of the routine or investigational use of DIA is the sample drop-outrate. In our study, no test was lost neither by MammaTyper1 nor by DIA; however, folding or twisting of tissue has been reported by others as a reason for considerable sample loss during automated analyses [49]. The decision between DIA and RT-qPCR for biomarker testing in breast cancer is further complicated by deployment conditions and pre-analytical requirements. At a first glance, RT-qPCR appears to represent a more realistic solution, due to savings in diagnostic time, broad accessibility of RT-qPCR instruments across laboratories, strong clinical validity and affordability. However, we are aware that the availability of these instruments is not wide- spread and, mainly for economic reasons, most of the pathology departments will continue to use immunohistochemistry to evaluate the expression of these biopathological parameters, obviously adhering carefully to external quality controls and specific guidelines. On the other PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 14 / 18 PLOS ONE RT-qPCR assessment of breast cancer biomarkers hand, the coarseness of the cutoffs currently used could benefit from more refined methods of analysis, such as gene expression profiles. Our study has some limitations. Although clinic-pathological data including follow-up data were available for all patients, we were unable to show meaningful associations between discor- dance and patient or tumor characteristics probably because of the highly heterogeneous nature of the cohort. For the same reason we reckoned that outcome statistics would be of lim- ited value. In addition, sample size was a limiting factor necessitating random selection of approximately 100 cases among all those that initially fulfilled our selection criteria. Moreover, we did not calculate agreement between methods due to the highly artificial constellation of challenging and/or ambiguous cases, which would prevent any attempt to extend such findings to other clinical/diagnostic settings. Finally, the study used outdated guidelines for the inter- pretation of IHC and FISH results concerning HER2 evaluation, according to the recommen- dations in force at the time of the analysis. In conclusion, the degree of correlation between IHC and RT-qPCR is high and compares well with the correlation between original with subsequent independent manual or semi-auto- mated IHC assessments. Intrinsic marker properties such as the type of protein and RNA fre- quency distributions or spatial heterogeneity in whole sections may interact with interpretation bias to shape the extent of inter-observer or inter-method variability. Acknowledgments We would like to thank Sotirios Lakis for providing medical writing services. S1 Data. (XLSX) S1 Data. (XLSX) S2 Data. (XLSX) Discussion The use of methods with wider dynamic range and higher reproducibility such as RT-qPCR may offer more precise assessment of endocrine responsiveness, improve Ki67 standardization and help resolve HER2 cases that present a difficult interpretation by IHC/FISH. In summary, our results seem to configure RT-qPCR as a complementary method to be used in all cases of equivocal results or close to the cut-off values. References 1. Weigelt B, Baehner FL, Reis-Filho JS. The contribution of gene expression profiling to breast cancer classification, prognostication and prediction: a retrospective of the last decade. J Pathol. 2010; 220 (2):263–80. https://doi.org/10.1002/path.2648 PMID: 19927298 2. Duffy MJ, Harbeck N, Nap M, Molina R, Nicolini A, Senkus E, et al. Clinical use of biomarkers in breast cancer: Updated guidelines from the European Group on Tumor Markers (EGTM). Eur J Cancer. 2017; 75:284–98. https://doi.org/10.1016/j.ejca.2017.01.017 PMID: 28259011 3. Wolff AC, Hammond ME, Schwartz JN, Hagerty KL, Allred DC, Cote RJ, et al. American Society of Clin- ical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007; 25(1):118–45. https://doi.org/10. 1200/JCO.2006.09.2775 PMID: 17159189 4. Wolff AC, Hammond ME, Hicks DG, Dowsett M, McShane LM, Allison KH, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncol- ogy/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013; 31 (31):3997–4013. https://doi.org/10.1200/JCO.2013.50.9984 PMID: 24101045 5. Wolff AC, Hammond MEH, Allison KH, Harvey BE, Mangu PB, Bartlett JMS, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. J Clin Oncol. 2018; 36(20):2105– 22. https://doi.org/10.1200/JCO.2018.77.8738 PMID: 29846122 6. Hammond ME, Hayes DF, Dowsett M, Allred DC, Hagerty KL, Badve S, et al. American Society of Clini- cal Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer (unabridged version). Arch Pathol Lab Med. 2010; 134(7):e48–72. https://doi.org/10.5858/134.7.e48 PMID: 20586616 7. Khoury T. Delay to Formalin Fixation (Cold Ischemia Time) Effect on Breast Cancer Molecules. Am J Clin Pathol. 2018; 149(4):275–92. https://doi.org/10.1093/ajcp/aqx164 PMID: 29471352 8. Pfitzner BM, Lederer B, Lindner J, Solbach C, Engels K, Rezai M, et al. Clinical relevance and concor- dance of HER2 status in local and central testing-an analysis of 1581 HER2-positive breast carcinomas over 12 years. Mod Pathol. 2018; 31(4):607–15. https://doi.org/10.1038/modpathol.2017.171 PMID: 29271415 9. Troxell ML, Long T, Hornick JL, Ambaye AB, Jensen KC. Comparison of Estrogen and Progesterone Receptor Antibody Reagents Using Proficiency Testing Data. Arch Pathol Lab Med. 2017; 141 (10):1402–12. https://doi.org/10.5858/arpa.2016-0497-OA PMID: 28714765 10. Denkert C, Budczies J, von Minckwitz G, Wienert S, Loibl S, Klauschen F. Strategies for developing Ki67 as a useful biomarker in breast cancer. Breast. 2015; 24 Suppl 2:S67–72. https://doi.org/10.1016/ j.breast.2015.07.017 PMID: 26283598 11. Visualization: Emanuele Caselli, Guido Bellezza. Visualization: Emanuele Caselli, Guido Bellezza. Writing – original draft: Emanuele Caselli. Writing – original draft: Emanuele Caselli. Writing – review & editing: Martina Mandarano, Kerstin Hartmann, Mark Laible. Author Contributions Conceptualization: Emanuele Caselli, Cristina Pelliccia, Valeria Teti, Ivana Ferri, Angelo Sidoni. Data curation: Emanuele Caselli, Ivana Ferri, Chiara Lupi, Fabrizio Stracci. Data curation: Emanuele Caselli, Ivana Ferri, Chiara Lupi, Fabrizio Stracci. Formal analysis: Emanuele Caselli, Guido Bellezza, Mark Laible. Funding acquisition: Ugur Sahin. Methodology: Cristina Pelliccia, Valeria Teti, Ivana Ferri. Project administration: Angelo Sidoni. Software: Emanuele Caselli, Martina Mandarano, Mark Laible. Supervision: Angelo Sidoni. Formal analysis: Emanuele Caselli, Guido Bellezza, Mark Laible. Formal analysis: Emanuele Caselli, Guido Bellezza, Mark Laible. Methodology: Cristina Pelliccia, Valeria Teti, Ivana Ferri. Project administration: Angelo Sidoni. Software: Emanuele Caselli, Martina Mandarano, Mark Laible. 15 / 18 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers Validation: Emanuele Caselli, Valeria Teti, Guido Bellezza, Ivana Ferri, Zsuzsanna Varga, Angelo Sidoni. Validation: Emanuele Caselli, Valeria Teti, Guido Bellezza, Ivana Ferri, Zsuzsanna Varga, Angelo Sidoni. References Dowsett M, Nielsen TO, A’Hern R, Bartlett J, Coombes RC, Cuzick J, et al. Assessment of Ki67 in breast cancer: recommendations from the International Ki67 in Breast Cancer working group. J Natl Cancer Inst. 2011; 103(22):1656–64. https://doi.org/10.1093/jnci/djr393 PMID: 21960707 21. Laible M, Schlombs K, Kaiser K, Veltrup E, Herlein S, Lakis S, et al. Technical validation of an RT- qPCR in vitro diagnostic test system for the determination of breast cancer molecular subtypes by quan- tification of ERBB2, ESR1, PGR and MKI67 mRNA levels from formalin-fixed paraffin-embedded breast tumor specimens. BMC cancer. 2016; 16:398. https://doi.org/10.1186/s12885-016-2476-x PMID: 27389414 22. Wirtz RM, Sihto H, Isola J, Heikkila P, Kellokumpu-Lehtinen PL, Auvinen P, et al. Biological subtyping of early breast cancer: a study comparing RT-qPCR with immunohistochemistry. Breast Cancer Res Treat. 2016; 157(3):437–46. https://doi.org/10.1007/s10549-016-3835-7 PMID: 27220750 23. Gisev N, Bell JS, Chen TF. Interrater agreement and interrater reliability: key concepts, approaches, and applications. Res Social Adm Pharm. 2013; 9(3):330–8. https://doi.org/10.1016/j.sapharm.2012. 04.004 PMID: 22695215 24. Eliasziw M, Young SL, Woodbury MG, Fryday-Field K. Statistical methodology for the concurrent assessment of interrater and intrarater reliability: using goniometric measurements as an example. Phys Ther. 1994; 74(8):777–88. https://doi.org/10.1093/ptj/74.8.777 PMID: 8047565 25. Goldhirsch A, Winer EP, Coates AS, Gelber RD, Piccart-Gebhart M, Thurlimann B, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Con- sensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol. 2013; 24(9):2206–23. https:// doi.org/10.1093/annonc/mdt303 PMID: 23917950 26. Curigliano G, Burstein HJ, Winer EP, Gnant M, Dubsky P, Loibl S, et al. De-escalating and escalating treatments for early-stage breast cancer: the St. Gallen International Expert Consensus Conference on the Primary Therapy of Early Breast Cancer 2017. Ann Oncol. 2017; 28(8):1700–12. https://doi.org/10. 1093/annonc/mdx308 PMID: 28838210 27. Liu Y, Beyer A, Aebersold R. On the Dependency of Cellular Protein Levels on mRNA Abundance. Cell. 2016; 165(3):535–50. https://doi.org/10.1016/j.cell.2016.03.014 PMID: 27104977 28. Nadji M, Gomez-Fernandez C, Ganjei-Azar P, Morales AR. Immunohistochemistry of estrogen and pro- gesterone receptors reconsidered: experience with 5,993 breast cancers. Am J Clin Pathol. 2005; 123 (1):21–7. https://doi.org/10.1309/4wv79n2ghj3x1841 PMID: 15762276 29. Muftah AA, Aleskandarany M, Sonbul SN, Nolan CC, Diez Rodriguez M, Caldas C, et al. Further evi- dence to support bimodality of oestrogen receptor expression in breast cancer. Histopathology. 2017; 70(3):456–65. https://doi.org/10.1111/his.13089 PMID: 27648723 30. Yi M, Huo L, Koenig KB, Mittendorf EA, Meric-Bernstam F, Kuerer HM, et al. Which threshold for ER positivity? References Koren S, Bentires-Alj M. Breast Tumor Heterogeneity: Source of Fitness, Hurdle for Therapy. Mol Cell. 2015; 60(4):537–46. https://doi.org/10.1016/j.molcel.2015.10.031 PMID: 26590713 12. Wolff AC, Hammond MEH, Allison KH, Harvey BE, Mangu PB, Bartlett JMS, et al. Human Epidermal Growth Factor Receptor 2 Testing in Breast Cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. Arch Pathol Lab Med. 2018; 142 (11):1364–82. https://doi.org/10.5858/arpa.2018-0902-SA PMID: 29846104 13. Mittendorf EA, Bartlett JMS, Lichtensztajn DL, Chandarlapaty S. Incorporating Biology Into Breast Can- cer Staging: American Joint Committee on Cancer, Eighth Edition, Revisions and Beyond. Am Soc Clin Oncol Educ Book. 2018; 38:38–46. https://doi.org/10.1200/EDBK_200981 PMID: 30231409 14. Varga Z, Lebeau A, Bu H, Hartmann A, Penault-Llorca F, Guerini-Rocco E, et al. An international repro- ducibility study validating quantitative determination of ERBB2, ESR1, PGR, and MKI67 mRNA in breast cancer using MammaTyper(R). Breast Cancer Res. 2017; 19(1):55. https://doi.org/10.1186/ s13058-017-0848-z PMID: 28490348 15. Kos Z, Dabbs DJ. Biomarker assessment and molecular testing for prognostication in breast cancer. Histopathology. 2016; 68(1):70–85. https://doi.org/10.1111/his.12795 PMID: 26768030 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 16 / 18 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers 16. Evans AJ, Bauer TW, Bui MM, Cornish TC, Duncan H, Glassy EF, et al. US Food and Drug Administra- tion Approval of Whole Slide Imaging for Primary Diagnosis: A Key Milestone Is Reached and New Questions Are Raised. Arch Pathol Lab Med. 2018; 142(11):1383–7. https://doi.org/10.5858/arpa. 2017-0496-CP PMID: 29708429 16. Evans AJ, Bauer TW, Bui MM, Cornish TC, Duncan H, Glassy EF, et al. US Food and Drug Administra- tion Approval of Whole Slide Imaging for Primary Diagnosis: A Key Milestone Is Reached and New Questions Are Raised. Arch Pathol Lab Med. 2018; 142(11):1383–7. https://doi.org/10.5858/arpa. 2017-0496-CP PMID: 29708429 17. Robertson S, Azizpour H, Smith K, Hartman J. Digital image analysis in breast pathology-from image processing techniques to artificial intelligence. Transl Res. 2018; 194:19–35. https://doi.org/10.1016/j. trsl.2017.10.010 PMID: 29175265 18. Lakhani SR, Ellis IO, Schnitt SJ, Tan PH, van de Vijver MJ. WHO Classification of Tumours of the Breast. 4th ed. Lyon: International Agency for Research on Cancer; 2012. 19. Lester SC BS, Chen Y-Y, Connolly JL, de Baca ME, Fitzgibbons PL, Hayes DF, et al. Protocol for Examination of Specimens From Patients With Invasive Carcinoma of the Breast. 2016 Jan [Cited 2019 Mar 01]. Available from: https://documents.cap.org/protocols/cp-breast-invasive-2016-v3300.pdf 20. References a retrospective study based on 9639 patients. Ann Oncol. 2014; 25(5):1004–11. https://doi. org/10.1093/annonc/mdu053 PMID: 24562447 31. Balduzzi A, Bagnardi V, Rotmensz N, Dellapasqua S, Montagna E, Cardillo A, et al. Survival outcomes in breast cancer patients with low estrogen/progesterone receptor expression. Clin Breast Cancer. 2014; 14(4):258–64. https://doi.org/10.1016/j.clbc.2013.10.019 PMID: 24325948 32. Sheffield BS, Kos Z, Asleh-Aburaya K, Wang XQ, Leung S, Gao D, et al. Molecular subtype profiling of invasive breast cancers weakly positive for estrogen receptor. Breast Cancer Res Treat. 2016; 155 (3):483–90. https://doi.org/10.1007/s10549-016-3689-z PMID: 26846986 33. Liao GS, Dai MS, Hsu HM, Chu CH, Hong ZJ, Fu CY, et al. Survival outcome of weak estrogen/proges- terone receptor expression in HER2 negative breast cancer is similar to triple negative breast cancer. Eur J Surg Oncol. 2017; 43(10):1855–61. https://doi.org/10.1016/j.ejso.2017.07.001 PMID: 28756019 34. Chen T, Zhang N, Moran MS, Su P, Haffty BG, Yang Q. Borderline ER-Positive Primary Breast Cancer Gains No Significant Survival Benefit From Endocrine Therapy: A Systematic Review and Meta-Analy- sis. Clin Breast Cancer. 2018; 18(1):1–8. https://doi.org/10.1016/j.clbc.2017.06.005 PMID: 28712925 PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 17 / 18 September 23, 2021 PLOS ONE RT-qPCR assessment of breast cancer biomarkers 35. Iwamoto T, Booser D, Valero V, Murray JL, Koenig K, Esteva FJ, et al. Estrogen receptor (ER) mRNA and ER-related gene expression in breast cancers that are 1% to 10% ER-positive by immunohis- tochemistry. J Clin Oncol. 2012; 30(7):729–34. https://doi.org/10.1200/JCO.2011.36.2574 PMID: 22291085 35. Iwamoto T, Booser D, Valero V, Murray JL, Koenig K, Esteva FJ, et al. Estrogen receptor (ER) mRNA and ER-related gene expression in breast cancers that are 1% to 10% ER-positive by immunohis- tochemistry. J Clin Oncol. 2012; 30(7):729–34. https://doi.org/10.1200/JCO.2011.36.2574 PMID: 22291085 36. Mackey JR. Can quantifying hormone receptor levels guide the choice of adjuvant endocrine therapy for breast cancer? J Clin Oncol. 2011; 29(12):1504–6. https://doi.org/10.1200/JCO.2010.34.3202 PMID: 21422423 37. Kim C, Tang G, Pogue-Geile KL, Costantino JP, Baehner FL, Baker J, et al. Estrogen receptor (ESR1) mRNA expression and benefit from tamoxifen in the treatment and prevention of estrogen receptor-pos- itive breast cancer. J Clin Oncol. 2011; 29(31):4160–7. https://doi.org/10.1200/JCO.2010.32.9615 PMID: 21947828 38. Dowsett M, Cuzick J, Wale C, Forbes J, Mallon EA, Salter J, et al. Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study. J Clin Oncol. 2010; 28 (11):1829–34. https://doi.org/10.1200/JCO.2009.24.4798 PMID: 20212256 39. References Laenkholm AV, Grabau D, Moller Talman ML, Balslev E, Bak Jylling AM, Tabor TP, et al. An inter- observer Ki67 reproducibility study applying two different assessment methods: on behalf of the Danish Scientific Committee of Pathology, Danish breast cancer cooperative group (DBCG). Acta Oncol. 2018; 57(1):83–9. https://doi.org/10.1080/0284186X.2017.1404127 PMID: 29202622 40. Varga Z, Cassoly E, Li Q, Oehlschlegel C, Tapia C, Lehr HA, et al. Standardization for Ki-67 assess- ment in moderately differentiated breast cancer. A retrospective analysis of the SAKK 28/12 study. PloS one. 2015; 10(4):e0123435. https://doi.org/10.1371/journal.pone.0123435 PMID: 25885288 41. Dunstan RW, Wharton KA Jr., Quigley C, Lowe A. The use of immunohistochemistry for biomarker assessment—can it compete with other technologies? Toxicol Pathol. 2011; 39(6):988–1002. https:// doi.org/10.1177/0192623311419163 PMID: 21971934 42. Fandel TM, Pfnur M, Schafer SC, Bacchetti P, Mast FW, Corinth C, et al. Do we truly see what we think we see? The role of cognitive bias in pathological interpretation. J Pathol. 2008; 216(2):193–200. https://doi.org/10.1002/path.2395 PMID: 18683853 43. Qian XL, Wen HY, Yang YL, Gu F, Guo XJ, Liu FF, et al. Assessment of dual-probe Her-2 fluorescent in situ hybridization in breast cancer by the 2013 ASCO/CAP guidelines produces more equivocal results than that by the 2007 ASCO/CAP guidelines. Breast Cancer Res Treat. 2016; 159(1):31–9. https://doi. org/10.1007/s10549-016-3917-6 PMID: 27455837 44. Bahreini F, Soltanian AR, Mehdipour P. A meta-analysis on concordance between immunohistochemis- try (IHC) and fluorescence in situ hybridization (FISH) to detect HER2 gene overexpression in breast cancer. Breast Cancer. 2015; 22(6):615–25. https://doi.org/10.1007/s12282-014-0528-0 PMID: 24718809 45. Acs B, Pelekanou V, Bai Y, Martinez-Morilla S, Toki M, Leung SCY, et al. Ki67 reproducibility using digi- tal image analysis: an inter-platform and inter-operator study. Lab Invest. 2019; 99(1):107–17. https:// doi.org/10.1038/s41374-018-0123-7 PMID: 30181553 46. Rimm DL, Leung SCY, McShane LM, Bai Y, Bane AL, Bartlett JMS, et al. An international multicenter study to evaluate reproducibility of automated scoring for assessment of Ki67 in breast cancer. Mod Pathol. 2019; 32(1):59–69. https://doi.org/10.1038/s41379-018-0109-4 PMID: 30143750 47. Klauschen F, Wienert S, Schmitt WD, Loibl S, Gerber B, Blohmer JU, et al. Standardized Ki67 Diagnos- tics Using Automated Scoring—Clinical Validation in the GeparTrio Breast Cancer Study. Clin Cancer Res. 2015; 21(16):3651–7. https://doi.org/10.1158/1078-0432.CCR-14-1283 PMID: 25501130 48. Meyerholz DK, Beck AP. Principles and approaches for reproducible scoring of tissue stains in research. Lab Invest. 2018; 98(7):844–55. https://doi.org/10.1038/s41374-018-0057-0 PMID: 29849125 49. Koopman T, Buikema HJ, Hollema H, de Bock GH, van der Vegt B. PLOS ONE \| https://doi.org/10.1371/journal.pone.0255580 September 23, 2021 49. Koopman T, Buikema HJ, Hollema H, de Bock GH, van der Vegt B. Digital image analysis of Ki67 prolif- eration index in breast cancer using virtual dual staining on whole tissue sections: clinical validation and inter-platform agreement. Breast Cancer Res Treat. 2018; 169(1):33–42. https://doi.org/10.1007/ s10549-018-4669-2 PMID: 29349710 References Digital image analysis of Ki67 prolif- eration index in breast cancer using virtual dual staining on whole tissue sections: clinical validation and inter-platform agreement. Breast Cancer Res Treat. 2018; 169(1):33–42. https://doi.org/10.1007/ s10549-018-4669-2 PMID: 29349710 18 / 18
https://openalex.org/W4392966331	https://journal.umgo.ac.id/index.php/Tulip/article/download/1844/1112	Indonesian	null	Persepsi Wisatawan Domestik Terhadap Kualitas Pelayanan Desa Wisata Kete’kesu Toraja Utara	Tulisan Ilmiah Pariwisata	2,022	cc-by-sa	3,226	Info Artikel This research aims to determine the Perception of Domestic Tourists toward Service Quality. This research used a quantitative descriptive approach. The sample was taken technically with a total sample of 85 respondents. Data collection using observation sheets and questionnaires. Data analysis using the Likert scale on the tourist perception validity test of <0.05=0.05 and Cronbach’s alpha result of 0.708 on service quality and hypothesis test with a value of 0.000 using the multivariate regression method. So, there is an influence on tourist perceptions of the quality of service. TULIP: Tulisan Ilmiah Pariwisata E-ISSN: 2720-9873 Available Online at https://journal.umgo.ac.id/index.php/Tulip/index Vol. 5, No. 1 Juni 2022 DOI: http://dx.doi.org/10.31314/tulip.5.1.1-7.2022 PERSEPSI WISATAWAN DOMESTIK TERHADAP KUALITAS PELAYANAN DESA WISATA KETE’KESU TORAJA UTARA Nur Fajrin Dai, Anggraeni lagalo, Sri Sunarti Program Studi Pariwisata, Universitas Muhammadiyah Gorontalo, Indonesia Email; nurfajrindai@gmail.com, lagaloanggraeni@gmail.com, srisunarti80@gmail.com Abstrak: Penelitian ini bertujuan untuk mengetahui Persepsi Wisatawan Domestik Terhadap Kualitas Pelayanan. Penelitian ini menggunakan desain metode pendekatan deskriptif kuantitatif. Sampel diambil secara tekhnik accidental dengan jumlah sampel 85 responden. Pengumpulan data menggunakan lembar observasi dan kuisioner dan analisis data menggunakan skala Likert pada uji validitas persepsi wisatawan yaitu <0.05=0.05 dan hasil cronbach’s alpha yaitu 0.708 pada kualitas pelayanan dan uji hipotesis dengan nilai 0,000 menggunakan metode regresi multivariat. Sehingga dapat disimpulkan ada pengaruh persepsi wisatawan terhadap kualitas pelayanan. Kata Kunci : Persepsi Wisatawan, Kualitas Pelayanan, Desa Wisata Kete’Kesu PENDAHULUAN peninggalan tersebut salah satunya berupa kuburan batu yang ditafsirkan telah berusia ratusan tahun yang juga merupakan salah satu daya Tarik yang banyak diminati Ketika berkunjung ke Wisata Ke’te Ke’su. Ke’te Ke’su kerap kali menjadi salah satu daya tarik wisata paling favorit di Toraja. Kepopuleran yang disandangnya, tidak jauh dari penawaran panorama indah yang begitu menawan. Masyarakatnya yang masih menjunjung tinggi adat istiadatnya seakan jadi alasan bagi wisatawan untuk mengunjungi daya tarik wisata ini. Adat serta kehidupan tradisionalnya yang masih terasa kental, mengharuskan para tamu untuk mematuhi segala peraturan dan larangan yang telah ditetapkan. Semakin menarik untuk dikunjungi, desa ini menyimpan berbagai peninggalan purbakala. Peninggalan tersebut berupa kuburan batu, yang ditafsirkan telah berumur ratusan tahun. Adat serta kehidupan tradisionalnya yang masih sangat kental, mengharuskan para pendatang wajib mengikuti aturan dan ketetapan yang ada. Desa Wisata ini menyimpan beragam peninggalan purbakala, 1 Vol. 5, No. 1 Juni 2022 Tulisan Ilmiah Pariwisata (TULIP) Dengan adanya peninggalan tersebut, bisa menjadi bukti kehidupan sebelumnya di kawasan tersebut. Adanya peninggalan sejarah, menjadikannya masuk dalam daftar cagar budaya yang perlu dilestarikan. Selain memikat wisatawan dengan temuan peninggalannya, desa ini dikenal sebagai penghasil kerajinan pahat hingga lukis yang telah diakui dunia. Hasil karyanya yang begitu menawan, berhasil memikat hati para wisata asing untuk membelinya. Apabila tertarik untuk membeli salah satu karyanya, silahkan membawa budget lebih. hingga kini masih dijaga dengan baik. Selain rumah adatnya yang menarik, disini pengunjung bisa melihat beberapa makam kuno yang membuat bulu kuduk merinding. Perasaan menyeramkan tersebut semakin terasa, usai mengetahui cerita mistisnya. Bahkan tak jarang para wisatawan yang beruntung, bisa menyaksikan langsung beberapa upacara adatnya. Namun, sebagaimana hasil observasi awal peneliti di Desa Ke’su kelurahan Pantana’kan Lolo bahwa pelayanan terhadap wisatawan masih kurang memuaskan sehingga wisatawan yang datang ditempat itu terbilang kurang ramai pada hari-hari diluar weekend. Pelayanan yang ada di Desa wisata ini masih belum terlalu dinikmati oleh pengunjung, wisatawan yang mengunjungi ¬Ke’se Ke’su hanya monoton dengan aktivitas datang, foto-foto, lalu beranjak pergi, ini membuat interaksi langsung dengan masyarakat di sekitar berkurang. Interaksi yang terjadi hanya seputar tawar menawar harga barang. Tentu ini berpengaruh pada kualitas pelayanan serta kepuasan wisatawan, dengan ini dapat dikatakan minimnya pengetahuan mengenai Daya Tarik Wisata Ke’te Ke’su. Masyarakat yang loyal tentu membuat wisatawan berpikir bahwa adanya inisiatif dari masyarakat setempat dianggap membantu dalam hal pengetahuan seputar Ke’te Ke’su. Sepanjang mata memandang, terdapat rumah adat yang disebut Tongkonan berjejer rapi hampir di setiap jalanan yang ditelusuri. a. Fisiologis Banyak informasi yang masuk melalui panca indera, kemudian informasi yang didapatkan tersebut akan mempengaruhi serta melengkapi kegiatan anda untuk memberikan makna terhadap lingkungan atau feedback. Kemampuan indera masing-masing orang itu tentu berbeda dalam mempersepsikan sesuatu. METODE Skor akhir didapatkan dengan menjumlahkan angka tiap-tiap jawaban. Untuk mengetahui intensitas sikap wisatawan, terlebih dahulu perlu dicari rentang nilainya (interval) dimana dalam menemukan interval, sesuai dengan kategori yang diberikan yaitu satu (1) untuk nilai terendah dan lima (5) untuk nilai tertinggi maka digunakan rumus sebagai berikut: Bagian metode penelitian diuraikan secara ringkas dan jelas. metode penelitian menjelaskan penggunaan metode penelitian, jenis penelitian metode pengumpulan data, analisis data dan sumber data. Metode penelitian bukan dalam bentuk. Bagian ini ditulis maksimal 10% (untuk penelitian kualitatif) atau maksimal 15% (untuk penelitian kuantitatif) dari semua isi artikel. [Cambria 12] Skor tertinggi - Skor terendah = Interval Jumlah Wisatawan Dalam penelitian ini, peneliti menggunakan tekhnik accidental, tekhnik accidental yaitu satu bentuk pengambilan data yang dilakukan dengan memberikan kuisioner kepada responden secara tiba-tiba, Kuisioner tersebut meliputi pernyataan mengenai persepsi wisatawan terhadap pelayanan yang ada di Desa Wisata Ke’te Ke’su. Wisatawan akan mengisi data-data yang tertera dalam kuisioner tersebut demi kelengkapan data bagi peneliti. Dalam penelitian ini peneliti menggunakan 85 wisatawan sebagai responden. PENDAHULUAN Dibuat tanpa menggunakan material paku sedikitpun, rumah adat ini dibangun dengan menumpuk kayunya dengan sedemikian rupa. Jika dilihat sepintas memang serupa dengan rumah panggung yang ada di daerah lainnya, namun ada beberapa perbedaan yang membuatnya terlihat unik nan berbeda. Umumnya hunian hanya akan digunakan sebagai tempat beristirahat para pemiliknya, namun kondisi ini tidak berlaku bagi Tongkonan. Kolong rumahnya yang masih luas, dimanfaatkan sebagai kandang ternak kesayangannya. Bahkan bagian rumahnya pun dibagi menjadi beberapa bagian, dimana salah satu bagiannya digunakan sebagai tempat penyimpanan jenazah dari kerabatnya yang belum bisa dikuburkan. Kualitas pelayanan dapat diliat dari antusias masyarakat atau para pelaku pariwisata di Desa Wisata Ke’te Ke’su ini mencakup pada bagaimana masyarakat menyambut kedatangan para wisatawan sampai dengan bagaimana wisatawan meninggalkan daya Tarik wisata. Hal itu tersebut tidak menonjol, interaksi yang terjalin hanya meliputi jual beli. Ini tentu menjadi tolak ukur yang dapat mempengaruhi kualitas pelayanan. Berdasarkan hasil latar belakang tersebut maka peneliti melakukan penelitian berjudul Persepsi Wisatawan Domestik Terhadap Kualitas Pelayanan Desa Wisata Kete’ Kesu Toraja Utara sehingga apa yang menjadi saran dan masukan dari para wisatawan ini yang akan menjadi bahan pertimbangan bagi para pelaku wisata, pemerintah setempat maupun stakeholder dan kelompok sadar wisata untuk pengembangan Daya Tarik Wisata Kete’Kesu. Ciri khas lainnya dari Tongkonan, yaitu pintu rumahnya yang dibuka ke arah atas. Selain pintunya yang unik, atapnya pun tidak kalah menarik untuk diperhatikan. Bentuknya yang seperti perahu terkelungkup, ternyata hanya dibuat dari buritan yang dilapisi dengan ijuk hitam. Bentuk atapnya yang menarik, tatkala mengibaratkannya seperti tanduk kerbau. Setiap Tongkonan yang dibangun harus menghadap utara, lantaran kepercayaan masyarakatnya. Masyarakat Toraja percaya bahwa para leluhurnya berasal dari utara, sehingga rumah adatnya pun dibangun menghadap ke arah tersebut. Bahkan mereka meyakini jika, setiap orang yang telah meninggal akan berkumpul kembali dengan arwah leluhur yang berada di utara. Kepercayaan tersebut telah dipegangnya sejak beberapat tahun silam, dan 2 Vol. 5, No. 1 Juni 2022 Tulisan Ilmiah Pariwisata (TULIP) HASIL DAN PEMBAHASAN Pembahasan berisi teori berupa hasil telaah kritis peneliti terhadap satu atau beberapa teori yang berhubungan dengan masalah penelitiannya. Teori dapat diambil dari satu teori atau beberapa teori yang dipadukan secara eklektik. Pengambilan harus didasari argumentasi akademis. Teori ini sejalan dengan pokok masalah yang diuraikan. Teori yang dikutip harus menggunakan footnote. Data dalam penelitian ini akan dianalisis menggunakan metode deskriptif kuantitatif. Data numerik yang diperoleh akan diuraikan atau dideskripsikan dalam bentuk narasi untuk memperoleh gambaran dan pemahaman yang jelas mengenai Kualitas Pelayanan di Desa Wisata Ke’te Ke’su, Kecamatan Sanggalangi, di Kawasan Toraja Utara. Untuk menganalisis persepsi wisatawan, data hasil kuesioner diukur menggunakan Skala Likert dengan mengajukan 15 butir pernyataan bersifat positif yang masing-masing diberikan lima alternatif jawaban dengan bobot yang berbeda. Wisatawan yang menjadi responden diminta memilih salah satu jawaban yang mewakili tanggapan mereka terhadap Kualitas pelayanan di Desa wisata Ke’te Ke’su Kecamatan Sanggalangi Kabupaten Toraja Utara. Persepsi merupakan suatu penglihatan atau pemikiran terkait suatu objek atau daya Tarik wisata. Pada penelitian ini berikut peneliti cantumkan beberapa indikator persepsi Wisatawan meliputi: b. Responsiveness e. Tangibles Tangibles atau bentuk fisik bisa berupa fasilitas yang tersedia serta para pemilik jasa. Responsiveness atau ketanggapan ialah satu kemampuan dalam memberikan pelayanan yang sesuai terhadap pelanggan melalui Berdasarkan pengolahan data menggunakan aplikasi SPSS dengan jumlah responden 85 wisatawan dengan kriteria umur yaitu 15-65 tahun, maka berikut peneliti lampirkan hasil dari kuisioner yang telah dibagikan kepada wisatawan informasi yang jelas dan akurat serta tidak membiarkan wisatawan menunggu. d. Empathy h d. Empathy Realibility atau keandalan merupakan satu kemampuan ataupun keahlian dari para pelaku pariwisata atau sebuah instansi maupun perusahaan dalam memberikan pelayanan berdasarkan dengan apa yang telah dijanjikan secara detail dan terperinci. Empathy merupakan salah satu bentuk pemberian perhatian kepada wisatawan dengan perasaan yang sangat tulus dan disertai upaya untuk memahai apa yang menjadi keinginan wisatawan p p b. Perhatian Faktor selanjutnya yang mempengaruhi persepsi ialah Perhatian. Setiap individu yang membutuhkan energi yang dikeluarkan untuk memfokuskan ataupun memperhatikan suatu bentuk fisik serta fasilitas mental yang ada pada suatu objek, perbedaan energi inilah yang membuat fokus perhatian terhadap obyek juga tentunya berbeda dan hal tersebut akan mempengaruhi persepsi atau pandangan terkait satu objek. a. Skor lima diberikan untuk penilaian sangat baik atau Excellent. b. Skor empat diberikan untuk penilaian baik atau Good. c. Skor tiga diberikan untuk penilaian cukup atau Average. kualitas pelayanan merupakan satu pengukuran atau value terhadap pelayanan yang diterima pada waktu tertentu. Adapun d. Skor dua diberikan untuk buruk atau Poor. e. Skor satu untuk penilaian sangat buruk atau Terrible. 3 3 Vol. 5, No. 1 Juni 2022 Tulisan Ilmiah Pariwisata (TULIP) dimensi dari Kualitas Pelayanan adalah sebagai berikut: keamanan, komunikasi agar menimbulkan rasa percaya wisatawan terhadap pemilik jasa. c. Accurance Accurance atau jaminan merupakan tata kramah dan sikap para karyawan maupun masyarakat setempat dalam komponen Tabel 1: Tabel Skala Likert Tabel 1: Tabel Skala Likert Tabel 1: Tabel Skala Likert No Kriteria Penilaian Skor Kategori 1 Sangat baik 5 4,20 – 5,00 2 3 4 5 Baik Cukup Buruk Sangat Buruk 4 3 2 1 3,40 – 4,19 2,60 – 3,39 1,80 – 2,59 1,00 – 1,79 Sumber Data: Hasil Penelitian Sumber Data: Hasil Penelitian Sumber Data: Hasil Penelitian memampudayakan (empowering) SDM pelaksana pariwisata. memampudayakan (empowering) SDM pelaksana pariwisata. Hipotesis dalam penelitian ini adalah “Jika pelayanan terhadap wisatawan lebih di utamakan, maka persepsi wisatawan domestik terhadap Desa Wisata Ke’te Ke’su, Kecamatan Sanggalangi, Kabupaten Toraja Utara akan jauh lebih baik, wisatawan akan semakin nyaman mengunjungi Ke’te Ke’su. Wewenang pemerintah daerah dalam penyelenggaraan penataan ruang, meliputi penyelenggaraan penataan ruang wilayah kabupaten dan penataan ruang kawasan strategis kabupaten, melalui perwujudan organisasi pemerintah yang efisien, tata laksana yang efektif dan sumber daya manusia (SDM) yang mumpuni. Potensi sumber daya alam dan kebudayaan membutuhkan perencanaan wilayah yang baik, penyediaan transformasi struktural antar wilayah, dan penyiapan teknologi, peningkatan kualitas sumber daya manusia menjadi syarat yang penting karena terkait dan berhubungan dengan nilai dan sikap (value andattitude ) masyarakat sebagai mahluk budaya. Pemerintah Kabupaten Toraja Utara menetapkan visi 2010 – 2030 yaitu “Toraja Utara Daerah Wisata Budaya Kaya Pesona Dengan Ragam Kreativitas Dan Kasih Yang Menyejahterakan” dengan misi yakni “Menciptakan Berbagai Kemudahan Yang Memungkinkan Warganya Memiliki Tingkat Kreativitas Yang Dapat Menampilkan HasilHasil Yang Gemilang”. Adapun penjabaran misi mengacu kepada RPJP Nasional di bidang kepariwisataan dan RPJP Provinsi Sulawesi Selatan serta aspirasi masyarakat antara lain membenahi objek-objek wisata multi dimensi, membangun pusat pengembangan budaya dan merevitalisasi nilai-nilai adat dan budaya, serta Pariwisata yang berkembang di Toraja Utara menjadi industri jasa yang memberikan keuntungan secara ekonomi, dan menjadikan sektor pariwisata sebagai sumber devisa, tetapi kurang mempertimbangkan aspek kelestarian lingkungan dan budaya karena 4 4 Tulisan Ilmiah Pariwisata (TULIP) Vol. 5, No. 1 Juni 2022 aka nada penanganan tersendiri bagi siapa saja yang melanggar aturan adat yang berlaku. alasan ekonomi.Paradigma ini menyebabkan pelaku pariwisata cenderung mengembangkan pariwisata dalam skala massive yang berdampak terhadap degradasi lingkungan, baik fisik biotis maupun lingkungan sosial budaya seperti pembangunan tempat-tempat rekreasi/cafe/karaoke yang tidak mempertimbangkan banyak hal akibatnya kerusakan (pencemaran) lingkungan, munculnya kebiasaan yang tidak sesuai budaya, pengaruh modernisasi, prostitusi terselubung dan terpinggirkannya masyarakat di sekitar obyek wisata. c. Accurance Padahal pariwisata seharusnya memberikan pengalaman dan pencerahan (pengetahuan) kepada wisatawan tentang kelestarian alam dan ekosistem, memelihara budaya dan kesenian, menghormati integritas masyarakat lokal sebagai tuan rumah, serta meningkatkan kesejahteraan. Atraksi wisata yang terdapat di Daya Tarik Wisata ini meliputi Wisata Religi, Wisata Budaya. Wisatawan dapat mengunjungi makam Tengkorak yang dapat diakses setelah melewati Rumah adat Tongkonan dan Warung- warung Aksesoris Toraja, wisatawan dapat melihat kearifan lokal masyarakat setempat. y p Akses menuju Ke’te Ke’su terbilang mudah dijangkau, pemerintah Toraja juga sudah menyediakan fasilitas Bandara bagi wisatawan yang mau datang mengunjungi Wisata yang ada di Toraja. Terdapat bis maupun angkutan umum yang bertujuan ke Toraja. Bila sudah sampai di Kete’Kesu, wisatawan lokal akan dikenai harga 15.000 untuk tiket masuk. Harga tersebut sudah dapat digunakan untuk mengunjungi beberapa tempat di dalamnya. Kondisi jalan yang sudah baik memudahkan wisatawan untuk menjangkau desa wisata ini, wisatawan akan merasa nyaman dan bila hujan turun, akan dapat berteduh di lumbung yang telah tersedia. Fasilitas yang memadai berupa toilet, mushola, pusat oleh-oleh dan perbelanjaan membuat wisatawan enggan untuk segera meninggalkan objek wisata ini. Ke‟te‟ Kesu‟ adalah salah satu desa tradisional di daerah pegunung-an dan merupakan desa tertua di distrik singgalangi. Kete Kesu tidak pernah berubah sejak 400 tahun lalu. Ke‟te‟ Kesu‟ adalah suatu desa wisata di kawasan Toraja Utara yang dikenal karena adat dan kehidupan tradisional masyarakat dapat ditemukan di kawasan ini. Dikenal sebagai desa yang menyimpan berbagai cerita sejarah. Ke’te Ke’su kerap kali menjadi salah satu daya tarik wisata paling favorit di Toraja. Kepopuleran yang disandangnya, tidak jauh dari penawaran panorama indah yang begitu menawan. Masyarakatnya yang masih menjunjung tinggi adat istiadatnya seakan jadi alasan bagi wisatawan untuk mengunjungi daya tarik wisata ini. Daya Tarik Wisata Ke’te Ke’su ini meliputi kumpulan rumah adat Tongkonan yang terletak secara berhadapan, pengunjung juga dapat melihat makam-makam kuno yang sangat sakral. Disamping itu di Daya Tarik Wisata ini juga sering menjadi pusat kegiatan ataupun upacara adat Toraja berupa Rambu Solo, Ma’nene dan sebagainya, tidak sedikit pengunjung yang beruntung sehingga dapat menyaksikan upacara adatnya dan beberapa acara sakral di wisata ini. Dilansir dari hal tersebut, wisatawan yang datang mengunjungi wisata ini diberlakukan ketentuan-ketentuan adat berupa larangan untuk merusak benda- benda yang terdapat didalamnya ataupun mengucapkan kata atau kalimat yang tidak pantas selama berada di Daya Tarik Wisata ini, Adat serta kehidupan tradisionalnya yang masih sangat kental, mengharuskan para pendatang wajib mengikuti aturan dan ketetapan yang ada. c. Accurance Desa Wisata ini menyimpan beragam peninggalan purbakala, peninggalan tersebut berupa kuburan batu yang ditafsirkan telah berusia ratusan tahun ANOVA Model Sum of Df Mean Square F Sig Squares Regression 41,545 2 20,773 22,226 ,000b 1 Residual 76,502 82 ,933 Total 118,047 189–210. https://doi.org/10.21043/edukasia.v10 i1.791 189–210. https://doi.org/10.21043/edukasia.v10 i1.791 a. Dependent Variable: total daya tarik b. wisata c. Predictors: (Constant), total kualitas pelayanan, total persepsi wisatawan c. Predictors: (Constant), total kualitas pelayanan, total persepsi wisatawan Andi Sudirman Sulaiman, S. . (2018). Profil Pemerintah Prov. Sulawesi Selatan. Berdasarkan uji hipotesis yang mendapatkan dengan nilai 0,000 menggunakan metode Regresi Multivariat maka dapa disimpulkan adanya pengaruh persepsi wisatawan terhadap kualitas pelayanan di Desa Wisata Ke’te Ke’su Bungaran Antonius Simanjuntak, D. (2017). Sejarah Pariwisata Menuju Perkembangan Pariwisata Indonesia. Yayasan Pustaka Obor Indonesia. p y Berdasarkan dari hasil dari penelitian mengenai Persepsi Wisatawan maka peneliti dapat menyimpulkan bahwa adanya pengaruh persepsi wisatawan terhadap kualitas pelayanan didaya tarik wisata Ke’te Ke’su. Maka dengan hasil tersebut, maka peneliti menyatakan bahwa Persepsi Wisatawan domestik berpengaruh terhadap kualitas pelayanan di Desa Wisata Ke’te Ke’su. Maka Hipotesis statistik yang teruji dalam penelitian ini ialah Hipotesis Alternatif (Ha) Di, C., & Bandung, K. (2020). Persepsi Konsumen Pada Kualitas Produk Pastry Bakery Selama Pandemi Covid 19 Di Kota Bandung. Jurnal Inovasi Penelitian, 1(8), 2315–2320. https://stp-mataram.e- journal.id/JIP/article/view/1167/888 Diarta, P. dan. (2009). Pengantar ilmu pariwisata (FI. Sigit Suyantoro (Ed.); p. 75). Yogyakarta : Andi. Faizi, M. F. (2017). BAB II Tinjauan Pusktaka dan Pengajuan Hipotesis. العدد الحا( 1 ,) 43 . https://doi.org/10.1017/CBO97811074 15324.004 KESIMPULAN Berdasarkan hasil analisis data dari kuisioner yang telah disebar, maka peneliti menyimpulkan bahwa pandangan wisatawan domestik terhadap kualitas pelayanan masyarakat yang ada di Desa Wisata Kete’ Ke’su terbilang sudah baik. Hal itu dikarenakan masih ada sekitar 40 persen wisatawan beranggapan bahwa pelayanan di daya tarik ini masih belum maksimal. Itu disebabkan oleh beberapa faktor pendukung antara lain, fasilitas Toilet yang masih harus dibenah dan diperbaiki juga ditambah, masih banyak sampah yang berserakan dimana-mana, kurangnya interaksi masyarakat dengan para pengunjung, hewan peliharaan yang lalu lalang ditengah-tengah pengunjung yang dapat membuat pengunjung kurang nyaman, akses menuju ke kuburan yang masih sempit, kurangnya pemandu wisata sehingga mempermudah wisatawan dalam mempelajari budaya/aluk Toraja. Maka dari hasil penelitian menggunakan aplikasi SPS menyatakan bahwa Persepsi Wisatawan Domestik Terhadap Kualitas Pelayanan di Desa Wisata Ke’te Ke’su itu berpengaruh. Indrawati, Y. (2013). Kualitas Pelayanan Pada Marina Srikandi Tour & Travel Di Padangbai. 1(1), 20–23. Josephine, A. (2015). Komunikasi Antarbudaya Yang Efektif. Jurnal Ilmu Komunikasi FLOW, 2(8), 1–10. Kanca, I. N., & Wijaya, I. N. (2015). Persepsi Wisatawan Mancanegara Terhadap Kualitas Pelayanan Kantor Depan Di Nusa Dua Beach Hotel & Spa Bali. Sosial Dan Humaniora, 5(1), 42–49. Lagalo Anggraeni., S. P. (2018). Modifikasi Skala LIkert. Kerajinan Sulaman Karawo Sebagai Daya Tarik Wisata Minat Khusus Di Provinsi Gorontalo (Kasus Sentra Kerajinan Sulaman Karawo)., 1, 83. Muharto. (2020). Pariwisata Berkelanjutan. Grup Penerbit CV Budi Utama. Muththalib, A. (2022). Desa Kete Kesu, Destinasi Wisata Sakral di Tana Toraja. Celebes.Com. Noor, A. A., & Karwina, W. (2012). Persepsi Wisatawan Domestik Terhadap Kualitas Pelayanan Museum Geologi Bandung. Industrial Research Workshop and National Seminar, July 2012, 1–8. ANOVA 5 189–210. https://doi.org/10.21043/edukasia.v10 i1.791 REFERENSI Akbar, R. F. (2015). Analisis Persepsi Pelajar Tingkat Menengah Pada Sekolah Tinggi Agama Islam Negeri Kudus. Edukasia : Jurnal Penelitian Pendidikan Islam, 10(1), 6 Tulisan Ilmiah Pariwisata (TULIP) Vol. 5, No. 1 Juni 2022 Perlindungan Hukum Terhadap Wisatawan Menurut Undang-Undang Nomor 10 Tahun 2009 Tentang Kepariwisataan, 2 138 (2018). Talib, D., & Usu, N. F. (2019). Peran Masyarakat Lokal Dalam Pengelolaan Daya Tarik Wisata Alam Arung Jeram Papualangi Kabupaten Gorontalo Utara. Tulisan Ilmiah Pariwisata (TULIP), 2(2), 130. https://doi.org/10.31314/tulip.2.2.130- 148.2019 Pemerintahan Kab Toraja. (2016). Toraja Tourism Map. Peta Administrasi Kab. Toraja. Talib, S. S. (2021). Eksistensi Desa Wisata Religi Bongo di Masa Pandemic Covid19 Menuju Desa Wisata Terbaik Nasional. Tulisan Ilmiah Pariwisata, 4(2), 65–70. Putra, I. B. P. S., Negara, I. M. K., & Wijaya, N. M. S. (2017). Persepsi Wisatawan Terhadap Kualitas Pelayanan Pramuwisata di Bali. Jurnal IPTA, 5(1), 29–34. Tatik, S. (2008). Perilaku Konsumen: Implikasi Pada Strategi Pemasaran. Remaja Posdakarya. Putri, Y. L., & Utomo, H. (2017). Pengaruh Kualitas Pelayanan Terhadap Loyalitas Pelanggan Dengan Kepuasan Sebagai Variabel Intervening (Studi Persepsi Pada Pelanggan Dian Comp Ambarawa). Among Makarti, 10(1), 70–90. https://doi.org/10.52353/ama.v10i1.14 7 Tika, M. P. (2006). Metodologi Riset Bisnis. Jakarta: PT Bumi Aksara. Utama, I. G. B. R. (2016). Pengantar Industri Pariwisata. CV Budi Utama. Variabel, D. A. N. (2017). Bab 3 permasalahan, hipotesis, dan variabel. 27–43. Rakhma, J. (2015). Psikologi Komunikasi. PT Remaja Rosdakarya. Wirawan. (2010). Kepemimpinan: Teori, Psikologi, Aplikasi dan Penelitian. Rajawali Press. Selatan, K. (2010). Persepsi Masyarakat Terhadap Minat Pembelian, Kondisi Pasar. 1. Persepsi Masyarakat Tentang Relokasi Pasar Tradisional Terhadap Minat Membeli DI Desa Poyowa Kecil Kecamatan Kotamubagu Selatan, 5. Zamrodah, Y. (2016). Pengertian Persepsi. 15(2), 1–23. Sutisna. (2001). Perilaku Konsumen dan Komunikasi Pemasaran. PT Remaja Rosdakarya. Swastika P, P., & Mahagangga, I. G. A. O. (2016). Persepsi Wisatawan Terhadap Kualitas Pelayanan Starbucks Coffee Di Bandara Ngurah Rai Bali. Jurnal Destinasi Pariwisata, 4(2), 156. https://doi.org/10.24843/jdepar.2016. v04.i02.p28 Tulisan Ilmiah Pariwisata (TULIP) ------- Vol. 5, No. 1 Juni 2022 7 Tulisan Ilmiah Pariwisata (TULIP)
https://openalex.org/W3032885124	https://www.researchsquare.com/article/rs-14832/v2.pdf	English	null	Identification of Blast Resistance QTLs Based on Two Advanced Backcross Populations in Rice	Rice	2,020	cc-by	9,615	Introduction Rice (Oryza sativa) is a staple food crop for more than 50% of the world’s population. Rice blast, caused by the fungal pathogen Magnaporthe oryzae, is one of the most serious diseases of rice in tropical and temperate areas of the world. Improving disease resistance in crops is crucial for stable food production. Nevertheless, using host resistance gene (R gene) to breed resistant varieties has been proven to be the most effective and economical method to control rice blast, and gene pyramiding is a promising method for providing broad-spectrum and durable resistance (Fukuoka et al. 2009; Liu et al. 2016; Tabien et al. 2002). So far, over 100 blast resistant genes or quantitative trait loci (QTL) have been identified (Su et al. 2015; Vasudevan et al. 2016; Xiao et al. 2017; Zheng et al. 2016). Among them, 37 genes have been cloned (Wang et al. 2017; Wang et al. 2019; Zhao et al. 2018), and most of them belong to the nucleotide-binding site (NBS) leucine-rich repeat (LRR) gene family. Many of these R genes are clustered in the rice genome, especially on chromosomes 6, 11, and 12. Notably, at least 11 R genes have been identified, including Pi2, Pi9, Pi22, Pi25, Pi26, Pi40, Pi42, Pigm, Piz, Pizt, and Pi50, which are concentrated as gene clusters in the short-arm region near the centromere of chromosome 6. Of these, Pi2, Pi9, Pi50, Pigm, and Pizt have been cloned and have shown broad-spectrum resistance (Deng et al. 2017; Qu et al. 2006; Su et al. 2015; Zhou et al. 2006). It had been reported that at least 7 R genes were located in the long-arm of rice chromosome 11, including Pik, Pi- kg(t), Pikm, Pik-h, Pik-p, Pi54 and Pi1 (Ashikawa et al. 2008; Hua et al. 2012; Pan et al. 1998; Sharma et al. 2010; Yuan et al. 2011; Zhai et al. 2011; Zhai et al. 2014). More than 20 R gene were located on rice chromosome 12, most of them were located near the centromere of chromosome 12, including Pi-ta, Pi-ta2, Pi-tan, Pi19, Pi20, Pi30, Pi31 and Ptr (Bryan et al. 2000; Hayashi et al. 1998; Imbe et al. 1997; Sallaud et al. 2003; Zhao et al. 2018). Although the use of race-specific resistance genes is a major strategy for disease control, these genes are vulnerable to counter evolution of pathogens. Original article Keywords: Oryza sativa, Rice blast, Mapping, Quantitative trait loci, durable resistance Posted Date: April 13th, 2020 DOI: https://doi.org/10.21203/rs.2.24273/v2 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Keywords: Oryza sativa, Rice blast, Mapping, Quantitative trait loci, durable resistance Posted Date: April 13th, 2020 DOI: https://doi.org/10.21203/rs.2.24273/v2 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License DOI: https://doi.org/10.21203/rs.2.24273/v2 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 Rice on June 1st, 2020. See the published version at https://doi.org/10.1186/s12284-020-00392-6. Page 1/20 Page 1/20 Abstract Background: Rice blast is an economically important and mutable disease of rice. Using host resistance gene to breed resistant varieties has been proven to be the most effective and economical method to control rice blast and new resistance genes or quantitative trait loci (QTLs) are then needed. Results: In this study, we constructed two advanced backcross population to mapping blast resistance QTLs. CR071 and QingGuAi were as the donor parent to establish two BC3F1 and derived BC3F2 backcross population in the Jin23B background. By challenging the two populations with natural infection in 2011 and 2012, 16 and 13 blast resistance QTLs were identified in Jin23B/CR071 and Jin23B/QingGuAi population, respectively. Among Jin23B/CR071 population, 3 major and 13 minor QTLs have explained the phenotypic variation from 3.50% to 34.08% during six observation times. And, among Jin23B/QingGuAi population, 2 major and 11 minor QTLs have explained the phenotypic variation from 2.42% to 28.95% during six observation times. Conclusions: Sixteen and thirteen blast resistance QTLs were identified in Jin23B/CR071 and Jin23B/QingGuAi population, respectively. QTL effect analyses suggested that major and minor QTLs interaction is the genetic basis for durable blast resistance in rice variety CR071 and QingGuAi. Experimental population The blast resistant indica cultivar CR071 and QingGuAi which provided by the Enshi Academy of Agricultural Science, Hubei, China, were used as the donor parent. The blast susceptible indica cultivar Jin23B, the maintainer line for several elite hybrids in China, was used as the recurrent parent. Two backcross populations derived from the cross between Jin23B and CR071 and between Jin23B and QingGuAi were generated according to the plan outlined in Fig. 1. After the first cross, the F1 generations were backcross to Jin23B, then the BC1F1 seeds were sown in a blast nursery in Xianfeng County, Hubei province, in 2010. Resistance plants identified by their leaf blast reactions were further backcrossed with the recurrent parent Jin23B, and BC2F1 plants were obtained. The BC2F1 seeds were sown in Hainan province in 2011 and were randomly crossed with Jin23B (Fig. 1), two backcross population contained 239 and 237 plants, respectively were obtained. The BC3F1 plants and parents were planted during the normal rice growing seasons (from mid-May to early October) at the experimental field of Xianfeng for phenotypic measurement in 2011 and selfed to produce BC3F2 families. In 2012, the BC3F2 families, each containing 12 plants, were grown in Xianfeng County for blast phenotyping. Selected lines in the BC3F1 population were backcross to Jin23B to obtain BC4F1, and the self-cross seed of these BC4F1 plants were used to develop BC4F2 segregating population of each QTL. The BC4F2 segregating populations of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi were planted in 2013 in Xianfeng. Varieties BL6 and CO39 were used as resistant and susceptible control. Introduction Consequently, most varieties lose their resistance after a few years because of new M. oryzae races. Many studies indicated that the genetic control of blast resistance is complex and involves both major and minor resistance genes with complementary or additive effects, as well as environmental interactions (Bonman 1992; Li et al. 2007; Li et al. 2011; Wang et al. 1994; Wu et al. 2005). New resistance genes are then needed, thus continuing a cycle referred to as an evolutionary ‘‘arms race’’ between crops and pathogens (Jones and Dangl. 2006). Quantitative trait loci, which usually have smaller individual effects than R genes but confer broad-spectrum or non-race-specific resistance, can Page 2/20 Page 2/20 contribute to durable disease resistance (Kou and Wang. 2010). Thus, the discovery and use of novel QTLs and development of broad-spectrum resistant varieties are urgent goals in breeding for blast resistance in rice. Most blast resistance genes confer complete and race-specific resistances that the highly variable fungus can overcome the R gene effects within 2 or 3 years after planting (Wang et al. 2017). The resistance conferred by R genes often do not support sustainable crop production. However, resistance controlled by partially effective resistance genes is often considered to be non race-specific and therefore durable. Durable resistance is the main goal of rice breeding, repeated observations generally suggest that cultivars carrying partial resistance maintain resistance for a long time, possibly because of decreased selection pressure upon the pathogen. CR071 and QingGuAi are indica rice (Oryza sativa L.) cultivar that has provided a high level of durable resistance to blast over past decades, and has been used as a donor of blast resistance in breeding in China. To understand the genetic mechanism of blast resistance in CR071 and QingGuAi, two advanced backcross population BC3F1 and derived BC3F2 population from Jin23B/CR071 and Jin23B/QingGuAi were studied for blast response under conditions of natural infection. The objective was to find blast resistance loci in the donor parents and to explain the underlying mechanism of resistance. Such results should be useful for improving blast resistance in rice breeding. Trait evaluations All plants of the BC3F1 population were scored for leaf blast response at tillering and heading stages and were recorded for neck blast at maturation stages in 2011; the three traits were named 11RT (leaf blast response of plants at tillering, 2011), 11RH (leaf blast response of the plants at heading, 2011) and 11RN (neck blast of the plants at maturation, 2011). The BC3F2 populations were scored on an individual plant basis for leaf blast at tillering heading and for neck blast at the maturation stage, 2012, and results were named 12RT (leaf blast response of plants at tillering, 2012), 12RH (leaf blast response of plants at heading, 2012) and 12RN (neck blast of the plants at maturation, 2012). For the BC3F2 families, 24 plants of each family were planted in two row, and the middle 20 plants were scored for blast response. The mean of each family was used as raw data for QTL analysis. Lines developing asynchronously at the normal tillering or heading stages were excluded so that all data would reflect the same developmental stages of the plants. The BC4F2 segregation Page 3/20 Page 3/20 population were scored for leaf blast response at tillering stage in 2013. The most seriously diseased leaf of the top two or three new leaves was scored for each plant at each stage using the rating scale of Bonman et al. (1986), where 0 = no evidence of infection; 1 = brown specks smaller than 0.5 mm in diameter, no sporulation; 2 = brown specks about 0.5–1.0 mm in diameter, no sporulation; 3 = roundish to elliptical lesions, 1–3 mm in diameter, grey centre surrounded by brown margins, lesions capable of sporulation; 4 = typical spindle-shaped blast lesions capable of sporulation,3 mm or longer; 5 = lesions as in 4 but about half of one to two leaf blades killed by coalescence of lesions. Reaction types 0, 1, 2 and 3 were considered resistant, and 4 and 5 as susceptible (Das et al. 2012). Neck blast severity was recorded as a percentage of infection on the neck of rice panicle at physiological maturity stage. The number of panicles showing symptoms of neck blast was expressed as percent infection. Reaction types 0-25% were considered resistant, and 26%-100% as susceptible (Gao et al. 2008; IRRI 2002). Trait evaluations To induce infection by the pathogen, diseased straw collected during the previous year was evenly dispersed in each plot and the highly susceptible variety; CO39, was planted on both sides of each row and around the experimental population. Field management essentially followed normal agricultural practices with the exception of no use of bactericides. Genetic map construction and QTL analysis A genetic linkage map was constructed using the Kosambi mapping function of MapMaker/Exp3.0 program (Lincoln et al. 1992). QTL analysis was performed by composite interval mapping (CIM) method using Windows QTL Cartographer 2.5 software (Wang et al. 2007) with a logarithm of odds (LOD) threshold of 2.5. Genotypes of the BC3F1 population were determined using SSR markers. The resistance score and genotype of each plant in the BC3F1 population were used for QTL analysis. For QTL detection of the BC3F2 population, the mean of each BC3F2 family was used as the row value and the genotypes of the BC3F1 plants were used as the genotypes of the BC3F2 families. Correlation analysis between six observation times in 2011 and in 2012 were examined by the Pearson correlation coefficient test. The resistance score of each plant in the BC3F1 population were used for correlation analysis. The mean of each BC3F2 family were used as the row value for correlation analysis. Data analyses were performed using Microsoft Excel 2003 or SPSS 17.0. DNA markers Among 1032 simple sequence repeat (SSR) markers, 182 were polymorphic between Jin23B and CR071, and 161 were polymorphic between Jin23B and QingGuAi, the level of polymorphism were 17.63% and 15.60%, respectively. A total of 145 and 113 polymorphic markers covering the whole rice genome were used to develop the genetic linkage map of Jin23B/CR071 and Jin23B/QingGuAi population. The RM marker series were searched in the available rice genomic database (http://www.gramene.org). Insertion/deletion markers were designed based on the references maps of Nipponbare and 9311. Genomic DNA was isolated from leaf tissues using the CTAB method. The SSR assay was performed with 4% urea polyacrylamide gels migration and silver staining as reported by Panaud et al (1996). SSR assay of two backcross population Two backcross populations derived from the cross between Jin23B and CR071 and between Jin23B and QingGuAi contained 239 and 237 plants, respectively. Genomic DNA was isolated from young leaves of seeding from each plant using the CTAB method. A total of 145 polymorphic SSR markers covering the whole rice genome between Jin23B and CR071, and 113 polymorphic SSR markers between Jin23B and QingGuAi were used to detect the genotype of each plant. Theoretically, 87.5% of the markers are Jin23B homozygous genotype of each individual plant in BC3F1 population. In the Jin23B/CR071 background population, the marker ratio of Jin23B homozygous genotype of each plant were from 52.41% to 99.31%, most plants with a marker ratio in 90-100% (Fig. S1). In the Jin23B/QingGuAi background population, the Page 4/20 Page 4/20 marker ratio of Jin23B homozygous genotype of each plant were from 51.33% to 94.69%, most plants with a marker ratio in 80-90% (Fig. S1). marker ratio of Jin23B homozygous genotype of each plant were from 51.33% to 94.69%, most plants with a marker ratio in 80-90% (Fig. S1). Measurements and relationship of the traits The receptor parent Jin23B is an indica variety with susceptible performance to rice blast, and the donor parents CR071 and QingGuAi with resistance performance to rice blast. The leaf blast resistance score of Jin23B were 4.17 and 4.11 at tillering stage, and 4.22 and 4.19 at heading stage in 2011 and 2012 respectively (Table 1). The neck blast resistance of Jin23B were 91.7% and 93.7% in 2011 and 2012 respectively (Table 1). The leaf blast resistance score of CR071 and QingGuAi were 0.83 and 1.06 at tillering stage in 2011, and 0.97 and 1.11 at tillering stage in 2012 (Table 1). The leaf blast resistance score of CR071 and QingGuAi were 1.00 and 1.14 at heading stage in 2011, and 1.00 and 1.19 at heading stage in 2012 (Table 1). For neck blast resistance, CR071 and QingGuAi were 3.33% and 6.67% in 2011, and 4.67% and 8.33% in 2012 (Table 1). The resistance score between CR071 and Jin23B, QingGuAi and Jin23B were significant different at the corresponding measurement stages (Table 1). The resistance control BL6 and susceptible control CO39 had leaf blast scores of 1.20 and 4.60 at tillering stage and 1.30 and 5.00 at heading stage in 2011. And those for neck blast of BL6 and CO39 at maturation stage were 10.11% and 96.53% in 2011, and 12.10% and 100% in 2012. The distributions of lesion scores as measures of blast response at the six observation times for the BC3F1 population in 2011 and BC3F2 population in 2012 are shown in Figure 2. There was transgressive segregation in both directions for all traits. The Pearson correlation coefficients showed significant correlation (p<0.01) between all six traits in both years (Table S1; Table S2). Blast resistance of the BC3F1 population at the tillering and heading stages in 2011 had a remarkable positive relationship with blast resistance of the BC3F2 population at the heading and maturation stages in 2012 (p<0.01). Resistance during different stages also exhibited significant relationships (p<0.01). Leaf blast at the tillering and heading stages and neck blast at the maturation stage were significantly correlated with each other, so we were able to predict the neck blast response level according to the leaf blast response at tillering under natural condition. QTL mapping for blast resistance in Jin23B/CR071 population A total of 16 QTLs for blast resistance were identified on chromosomes 1, 2, 3, 4, 6, 7, 8, 11 and 12 in the Jin23B/CR071 population in two years (Table 2; Fig. 3). The phenotypic variance explained by each QTL ranged from 3.50% to 34.08%. For leaf blast resistance at tillering stage, nine QTLs were detected on chromosome 1, 2, 3, 6, 7, 8 and 12 (Table 2; Fig. 3). Among them, four QTLs, qBR2-1, qBR3-3, qBR6 and qBR12 were detected in both year, five QTLs, qBR1, qBR2-2, qBR2-3, qBR7-1 and qBR8 were detected only in 2012 (Table 2; Fig. 3). The QTLs flanked by SR49 and RM426 on chromosome 3, qBR3-3, was detected in both year and explained 25.42% of the phenotypic variation in 2011 and 34.08% of the phenotypic variation in 2012. A QTLs, qBR6, located between RM539 and R19951 on chromosome 6, was also detected in two years and explained 8.89% and 20.40% of the phenotypic variation, respectively. Two QTLs, qBR2-1 and qBR12, were located flanked by RM236-RM451 on chromosome 2, and RM179-YP6213 on chromosome 12, respectively. qBR2-1 was detected in both years and accounted for 6.64% and 8.17% of the phenotypic variation, respectively. Whereas qBR12 was detected in both year and accounted for 4.25% and 4.48% of the phenotypic variation. For leaf blast resistance at heading stage, eight QTLs were detected on chromosome 1, 2, 3, 6, 8 and 12 (Table 2; Fig. 3), and all the QTLs were detected in both year. These QTLs explained the phenotypic variance ranged from 3.90% to 25.83% in 2011, and from 3.50% to 31.33% in 2012. Among them, the QTL, qBR3-3, flanked by SR49 and RM426 on chromosome 3, have the largest effect which explained 25.83% of the phenotypic variation in 2011 and 31.33% of the phenotypic variation in 2012. For neck blast resistance at maturation stage, eleven QTLs were detected on chromosome 1, 2, 3, 4, 7, 8 and 11 (Table 2; Fig. 3). Among them, seven QTLs, qBR1, qBR3-1, qBR7-1, qBR7-2, qBR11-1, qBR11-2 and qBR11-3 were detected in both Page 5/20 Page 5/20 year, four QTLs, qBR2-3, qBR3-2, qBR4 and qBR8 were detected only in 2011 (Table 2; Fig. 3). These QTLs explained the phenotypic variance ranged from 3.91% to 28.48% in 2011, and from 7.02% to 26.19% in 2012. QTL mapping for blast resistance in Jin23B/CR071 population A QTL, qBR7-1, located between RM501 and RM542 on chromosome 7, explained 28.48% of the phenotypic variation, which have the largest effect in 2011. Whereas the QTL qBR11-1, located between RM181 and RM120 on chromosome 11, accounted for 26.19% of the phenotypic variation, which have the largest effect in 2012. Among the sixteen QTLs in the Jin23B/CR071 population, the gene effect of fifteen QTLs come from donor parent CR071, in contrast, the gene effect of qBR1 come from Jin23B. The QTLs, qBR1, flanked by RM237 and RM486 on chromosome 1, was detected in both year at five observation times and explained phenotypic variations of 6.09% in 12RT, 7.99% in 11RH, 7.25% in 12RH, 8.12% in 11RN and 9.48% in 12RN, respectively (Table 2). Six QTLs, qBR2-1, qBR2-3, qBR3-3, qBR6, qBR8 and qBR12 were detected in both year at four observation times; two QTLs qBR2-2 and qBR7-1 were detected in both year at three observation times; five QTLs qBR3-1, qBR7-2, qBR11-1, qBR11-2 and qBR11-3 were detected in both year at two observation times and these QTLs only have the effect to neck blast in maturation stage; two QTLs, qBR3-2 and qBR4 were only detected in 11RN. QTL mapping for blast resistance in Jin23B/QingGuAi population For neck blast resistance at maturation stage, ten QTLs were detected on chromosome 1, 2, 3, 4, 6, 7 and 12 (Table 3; Fig. 4). Among them, four QTLs, qBR4-1, qBR4-2, qBR6 and qBR7-1 were detected in both year, and explained the phenotypic variance were 12.31%, 8.82%, 31.91% and 27.94%, respectively, in 2011, and were 18.47%, 16.92%, 28.95% and 18.71%, respectively, in 2012. Six QTLs, qBR1-2, qBR2, qBR3, qBR7-2, qBR8 and qBR12 were detected only in one year (Table 3; Fig. 4), and explained the phenotypic variance were 8.55%, 8.92%, 12.72%, 3.43%, 18.89% and 9.58%, respectively. Among the thirteen QTLs in the Jin23B/QingGuAi population, the gene effect of twelve QTLs come from donor parent QingGuAi, in contrast, the gene effect of qBR12 come from Jin23B. The QTLs, qBR12, flanked by RM5927 and RM6296 on chromosome 12, was detected only in 2012 at maturation stage and explained phenotypic variations of 9.58% (Table 3). Two QTLs, qBR6 and qBR7 were detected in both year at six observation times; one QTL, qBR4-2, was detected in both year at three observation times; two QTLs, qBR4-1 and qBR11-2 were detected in both year at two observation times; eight QTLs, qBR1-1, qBR1-2, qBR2, qBR3, qBR7-2, qBR8, qBR11-1 and qBR12 were only detected in one observation time. Page 6/20 Validate the genetic effect of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi QTL mapping for blast resistance in Jin23B/QingGuAi population A total of 13 QTLs for blast resistance were identified on chromosomes 1, 2, 3, 4, 6, 7, 8, 11 and 12 in the Jin23B/QingGuAi population in two years (Table 3; Fig. 4). The phenotypic variance explained by each QTL ranged from 2.42% to 28.95%. For leaf blast resistance at tillering stage, four QTLs were detected on chromosome 1, 6, 7 and 11 (Table 3; Fig. 4). Among them, three QTLs, qBR1-1, qBR6 and qBR7-1 were detected in both year, one QTLs, qBR11-2 was detected only in 2012 (Table 3; Fig. 4). A QTL, qBR1-1, located between RM297 and RM486 on chromosome 1, was detected in two years and explained 24.93% and 3.95% of the phenotypic variation, respectively. The QTL flanked by L6ID3F and ZH6111 on chromosome 6, qBR6, was also detected in both year and explained 8.87% of the phenotypic variation in 2011 and 19.91% of the phenotypic variation in 2012. The QTL, qBR7-1, flanked by RM214 and RM5543 on chromosome 7 was detected in both years and accounted for 6.57% and 4.12% of the phenotypic variation, respectively. Whereas qBR11-2 was detected only in 2012 and accounted for 5.49% of the phenotypic variation. For leaf blast resistance at heading stage, four QTLs were detected on chromosome 4, 6, 7 and 11 (Table 3; Fig. 4). Two QTLs, qBR6 and qBR7-1, were detected in both year and the QTLs explained the phenotypic variance were 36.48% and 16.47% in 2011, respectively, and 25.27% and 26.31 in 2012, respectively. Two QTL, qBR4-2 and qBR11-1, were located flanked by RM241-RM317 on chromosome 4, and RM229-RM547 on chromosome 11, respectively. The QTLs qBR4-2 and qBR11-1 were detected only in 2012 and accounted for 2.51% and 2.42% of the phenotypic variation, respectively. For neck blast resistance at maturation stage, ten QTLs were detected on chromosome For neck blast resistance at maturation stage, ten QTLs were detected on chromosome 1, 2, 3, 4, 6, 7 and 12 (Table 3; Fig. 4). Among them, four QTLs, qBR4-1, qBR4-2, qBR6 and qBR7-1 were detected in both year, and explained the phenotypic variance were 12.31%, 8.82%, 31.91% and 27.94%, respectively, in 2011, and were 18.47%, 16.92%, 28.95% and 18.71%, respectively, in 2012. Six QTLs, qBR1-2, qBR2, qBR3, qBR7-2, qBR8 and qBR12 were detected only in one year (Table 3; Fig. 4), and explained the phenotypic variance were 8.55%, 8.92%, 12.72%, 3.43%, 18.89% and 9.58%, respectively. Discussion QTL mapping using advanced backcross population was proposed as an effective molecular breeding technique for incorporating valuable genes from exotic sources into an adapted background (Eizenga et al. 2013; Tanksley and Nelson 1996). With this method, QTL from the donor parents are detected by backcrossing with the adapted parent to eliminate most of the unwanted genes from the donor parent. In addition, using advanced backcross population can accelerate the crop improvement process because near-isogenic lines containing the desired QTL (genes) from the donor in the background of the recurrent parent can be selected from the advanced backcross population. Several studies have been performed to identify QTLs in advanced backcrossing. For example, Thomson et al. (2003) mapped quantitative trait loci for yield, yield components and morphological traits in rice using a BC2F1 and BC2F2 population by composite interval mapping and Windows QTL Cartographer 1.21. And Eizenga et al. (2013) mapped sheath blight and blast quantitative trait loci in two different advanced backcross populations. In our study, we developed two advanced backcross population with the objective of introgression of useful resistance genes from CR071 and QingGuAi into Jin23B. Using this strategy, we mapped 16 blast resistance QTLs from the Jin23B/CR071 population and 13 blast resistance QTLs from the Jin23B/QingGuAi population. We also obtained maintainer lines resistant to blast in the background of Jin23B, which can be used for rice blast resistance breeding. To date, over 100 blast resistant genes or QTLs have been identified (Su et al. 2015; Vasudevan et al. 2016; Xiao et al. 2017; Zheng et al. 2016). Among them, 37 genes have been cloned (Wang et al. 2017; Wang et al. 2019; Zhao et al. 2018). Many of these resistance genes are clustered on rice chromosomes 6, 11 and 12. Notably, at least 11 resistance genes— including Pi2, Pi9, Piz, Pizt, Pigm, Pi22, Pi25, Pi26, Pi40, Pi42 and Pi50—are concentrated in the short-arm region near the centromere of chromosome 6. In this study, a QTL qBR6 located between RM539 and R19951 in the Jin23B/CR071 population have a significant resistance to rice blast in both two years. When we further compared the position of this region to previous studies, we found that it contained Pi2, Pi9, Pigm, Pizt and Pi50, which are cloned blast resistance genes (Deng et al. 2017; Qu et al. 2006; Su et al. 2015; Zhou et al. 2006). Validate the genetic effect of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi Validate the genetic effect of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi alidate the genetic effect of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from Q Page 6/20 Page 6/20 The BC4F2 segregation population of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi were used to confirm the genetic effect of these QTLs. The qBR3-3 and qBR6 loci from CR071 increased blast resistance by 1.13 and 1.43, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). The qBR6 and qBR7-1 loci from QingGuAi increased blast resistance by 1.70 and 0.83, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). The BC4F2 segregation population of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi were used to confirm the genetic effect of these QTLs. The qBR3-3 and qBR6 loci from CR071 increased blast resistance by 1.13 and 1.43, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). The qBR6 and qBR7-1 loci from QingGuAi increased blast resistance by 1.70 and 0.83, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). The BC4F2 segregation population of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi were used to confirm the genetic effect of these QTLs. The qBR3-3 and qBR6 loci from CR071 increased blast resistance by 1.13 and 1.43, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). The qBR6 and qBR7-1 loci from QingGuAi increased blast resistance by 1.70 and 0.83, respectively, on leaf blast at tillering stage in 2013 (Fig. 5). Abbreviations 11RT: Leaf blast resistance at tillering stage in 2011; 11RH: Leaf blast resistance at heading stage in 2011; 11RN: Neck blast resistance at maturation stage in 2011. 12RT: Leaf blast resistance at tillering stage in 2012; 12RH: Leaf blast resistance at heading stage in 2012; 12RN: Neck blast resistance at maturation stage in 2012. CIM method: Composite interval mapping method; NBS-LRR gene: Nucleotide-binding site leucine-rich repeat gene; QTLs: Quantitative trait loci; R gene: Resistance gene; SSR: Simple sequence repeat. Discussion Many studies have been performed to pyramid resistances gene into rice varieties. Hittalmani et al. (2000) pyramided three blast resistance genes Pi1, Piz-5 and Pita into rice variety CO39 and Jiang et al. (2012) pyramided three blast resistance genes Pi1, Pi2 and D12 into rice variety Jin23B. Their results confirmed that pyramiding of blast resistance genes is an effective way to develop highly resistant varieties. In our study, the donor parent CR071 and QingGuAi have a high level and durable resistance to blast over past decades, and have been used as donor parent for blast resistance in breeding in China. In the study, two advanced backcross population were constructed for analyses the genetic mechanism of blast resistance in CR071 and QingGuAi, and major and minor blast resistance QTLs were identified in the donor parents. QTL effect analyses suggested that major and minor QTLs interaction is the genetic basis for durable blast resistance for CR071 and QingGuAi in the past decade in Wuling mountain area in China. Conclusions Overall, the mapping results showed that sixteen blast resistance QTLs were identified in the Jin23B/CR071 backcross population, in which, one QTLs, qBR1, was detected in both year at five observation times; six QTLs, qBR2-1, qBR2-3, qBR3- 3, qBR6, qBR8 and qBR12 were detected in both year at four observation times; nine QTLs were detected at one to three observation times. Thirteen blast resistance QTLs were identified in Jin23B/QingGuAi population, in which, two major QTLs, qBR6 and qBR7 were detected in both year at six observation times and eleven QTLs were detected at one to three observation times. Major and minor effect QTLs were identified in the two populations, which conferring resistance to blast together at different measuring stage. Discussion QTL qBR6 located between L6ID3F and ZH6111 in the Jin23B/QingGuAi population was also overlapped the Pi2/Pi9 gene cluster on chromosome 6. Thus, qBR6 identified in two population may be the allele of Pi2, Pi9, Pigm, Pizt or Pi50. QTLs qBR1 was located between RM237 and RM486 on chromosome 1 in the Jin23B/CR071 population, and the QTL qBR1-2 was located between RM297 and RM486 in the Jin23B/QingGuAi population, in this region, blast resistance gene Pi37, Pish and Pi35(t) had previously been reported (Lin et al. 2007; Nguyen et al. 2006; Takahashi et al. 2010). The blast resistance gene Pi37 and Pish have been cloned, which encode a NBS-LRR protein. Pi35(t) was identified in a QTL analysis of a population derived from the Japonica rice cultivar Hokkai 188 and the Indica rice cultivar Danghang-Shali. The resistance conferred by Pi35(t) to M. oryzae is classified as partial resistance (quantitative) rather than true resistance (qualitative). It is possible that qBR1 from Jin23B/CR071 population and qBR1-2 from Jin23B/QingGuAi population are allelic to either Pi37, Pish or Pi35(t). QTLs qBR2-3 located between RM530 and RM213 on chromosome 2 in the Jin23B/CR071 population, was close to the cloned gene Pib (Wang et al. 1999). The neck blast resistance QTLs, qBR11-1, was located between RM21 and RM590 on chromosome 11 in the Jin23B/CR071 population, in this region, blast resistance gene Pikm, Pik-h and Pik-p had previously been reported (Ashikawa et al. 2008; Yuan et al. 2011; Zhai et al. 2014). It is possible that qBR11-1 is allelic to either Pikm, Pik-h or Pik-p. In the Jin23B/QingGuAi population, QTL qBR7-1 located between RM214 and RM5543 on chromosome 7 have a significant resistance to rice blast, in this region few gene have been reported, thus qBR7-1 may be a new gene. Page 7/20 Page 7/20 Rice cultivars with durable blast resistance have been recognized in several production systems. The durable resistance of these cultivars is associated with polygenic partial resistance that shows no evidence of race specificity. This partial resistance is expressed as fewer and smaller lesions on the leaf blade but latent period does not appear to be an important component. Many blast resistant varieties with single resistance genes lose resistance after a few years; that is, they have or had nondurable resistance (Babujee and Gnanamanickam 2000). Varieties with durable resistance may contain more than one resistance gene (Zhu et al. 2012). Declarations Ethics approval and consent to participate Authors’ contributions Authors’ contributions YH designed the experiments. HJ, YF, and LQ performed the experiments. GG and QZ helped with field management. HJ, YF and LQ analyzed the data. HJ and YH wrote the manuscript. All authors approved the manuscript. Acknowledgements We are very grateful to Professor Taipin Duan in Enshi Academy of Agricultural Sciences for providing seeds for the donor parents of the blast resistance cultivars. Funding This work was supported by grants from the National Program on R&D of Transgenic Plants (2016ZX08001002–002), the National Natural Science Foundation (31801438) and the earmarked fund for the China Agriculture Research System (CARS-01-03) of China. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Availability of data and material The data sets supporting the results of this article are included within the article and its supporting files. Competing interests Page 8/20 The authors declare that they have no competing interests. References Ashikawa I, Hayashi N, Yamane H, Kanamori H, Wu JZ, Matsumoto T, Ono K, Yano M (2008) Two adjacent nucleotide- binding site-leucine-rich repeat class genes are required to confer Pikm-specific rice blast resistance. Genetics 180(4):2267 Babujee L, Gnanamanickam S (2000) Molecular tools for characterization of rice blast pathogen, Magnaporthe grisea, population and molecular marker-assisted breeding for disease resistance. Curr Sci 78:24-257 Babujee L, Gnanamanickam S (2000) Molecular tools for characterization of rice blast pathogen, Magnaporthe grisea, population and molecular marker-assisted breeding for disease resistance. Curr Sci 78:24-257 Babujee L, Gnanamanickam S (2000) Molecular tools for characterization of rice blast pathogen, Magnaporthe grisea, population and molecular marker-assisted breeding for disease resistance. Curr Sci 78:24-257 Bonman JM (1992) Durable resistance to rice blast disease-environmental influences. Euphytica 63:115-123 Bonman JM (1992) Durable resistance to rice blast disease-environmental influences. Euphytica 63:115-123 Bonman JM, De Dios TIV, Khin MM (1986) Physiologic specialization of Pyricularia oryzae in the Philippines. Plant Dis 70:767-769 Bonman JM, De Dios TIV, Khin MM (1986) Physiologic specialization of Pyricularia oryzae in the Philippines. Plant Dis 70:767-769 Bryan GT, Wu KS, Farrall L, Jia YL, Hershey HP, McAdams SA, Faulk KN, Donaldson GK, Tarchini R, Valent B (2000) A single amino acid difference distinguishes resistant and susceptible alleles of the rice blast resistance gene Pi-ta. Plant Cell 12:2033-2045 Page 9/20 amino acid difference distinguishes resistant and susceptible alleles of the rice blast resistance gene Pi-ta. Plant Cell 2:2033-2045 Das A, Soubam D, Singh PK, Thakur S, Singh NK, Sharma TR (2012) A novel blast resistance gene, Pi54rh cloned from wild species of rice, Oryza rhizomatis confers broad spectrum resistance to Magnaporthe oryzae. Funct Integr Genomics 2:215-228 Deng Y, Zhai K, Xie Z, Yang D, Zhu X, Liu J, Wang X, Qin P, Yang Y, Zhang G, Li Q, Zhang J, Wu S, Milazzo J, Mao B, Wang E, Xie H, Tharreau D, He Z (2017) Epigenetic regulation of antagonistic receptors confers rice blast resistance with yield balance. Science 355:962-965 Fukuoka S, Saka N, Koga H, Ono K, Shimizu T, Ebana K, Hayashi N, Takahashi A, Hirochika H, Okuno K, Yano M (2009) Loss of function of a proline-containing protein confers durable disease resistance in rice. Science 325:998-1001 Gao GJ, Li GJ, Bao L, He YQ (2008) Genetic analysis of rice blast resistance and identification of resistance genes hroughout all stages in rice. References Mol Plant Breed 6(5):825-829 (In Chinese with English abstract) Hayashi N, Ando I, Imbe T (1998) Identification of a new resistance gene to a Chinese blast fungus isolate in the Japanese ice cultivar Aichi Asahi. Phytopathology 88(8):822-827 Das A, Soubam D, Singh PK, Thakur S, Singh NK, Sharma TR (2012) A novel blast resistance gene, Pi54rh cloned from wild species of rice, Oryza rhizomatis confers broad spectrum resistance to Magnaporthe oryzae. Funct Integr Genomics 12:215-228 Das A, Soubam D, Singh PK, Thakur S, Singh NK, Sharma TR (2012) A novel blast resistance gene, Pi54rh cloned from wild species of rice, Oryza rhizomatis confers broad spectrum resistance to Magnaporthe oryzae. Funct Integr Genomics 12:215-228 Deng Y, Zhai K, Xie Z, Yang D, Zhu X, Liu J, Wang X, Qin P, Yang Y, Zhang G, Li Q, Zhang J, Wu S, Milazzo J, Mao B, Wang E, Xie H, Tharreau D, He Z (2017) Epigenetic regulation of antagonistic receptors confers rice blast resistance with yield balance. Science 355:962-965 Fukuoka S, Saka N, Koga H, Ono K, Shimizu T, Ebana K, Hayashi N, Takahashi A, Hirochika H, Okuno K, Yano M (2009) Loss of function of a proline-containing protein confers durable disease resistance in rice. Science 325:998-1001 Gao GJ, Li GJ, Bao L, He YQ (2008) Genetic analysis of rice blast resistance and identification of resistance genes throughout all stages in rice. Mol Plant Breed 6(5):825-829 (In Chinese with English abstract) Hayashi N, Ando I, Imbe T (1998) Identification of a new resistance gene to a Chinese blast fungus isolate in the Japanese rice cultivar Aichi Asahi. Phytopathology 88(8):822-827 Page 9/20 Hittalmani S, Parco A, Mew TV, Zeigler RS, Huang N (2000) Fine mapping and DNA marker-assisted pyramiding of the three major genes for blast resistance in rice. Theor Appl Genet 100:1121-1128 Hua LX, Wu JZ, Chen CX, Wu WH, He XY, Lin F, Wang L, Ashikawa I, Matsumoto T, Wang L, Pan QH (2012) The isolation of Pi1, an allele at the Pik locus which confers broad spectrum resistance to rice blast. Theor Appl Genet 125(5):1047-1055 Imbe T, Ora S, Yanoria MJT, Tsunematsu H (1997) A new gene for blast resistance in rice cultivar, IR24. Rice Genet Newsl 14:60-62 IRRI (2002) Standard evaluation system for rice (SES). References International Rice Research Institute, Los Banos Jiang HC, Feng YT, Bao L, Li X, Gao GJ, Zhang QL, Xiao JH, Xu CG, He YQ (2012) Improving blast resistance of Jin23B and its hybrid rice by marker-assisted gene pyramiding. Mol Breed 30:1679-1688 Jones JD, Dangl JL (2006) The plant immune system. Nature 444:323-239 Kou Y, Wang S (2010) Broad-spectrum and durability: understanding of quantitative disease resistance. Curr Opin Plant Biol 13:181-185 Li YB, Wu CJ, Jiang GH, Wang LQ, He YQ (2007) Dynamic analyses of rice blast resistance for the assessment of genetic and environmental effects. Plant Breed 126:541-547 Li YB, Wu CJ, Xing YZ, Chen HL, He YQ (2011) Dynamic QTL analysis for rice blast resistance under natural infection conditions. Aust J Crop Sci 2:73-82 Lin F, Chen S, Que ZQ, Wang L, Liu XQ, Pan QH (2007) The blast resistance gene Pi37 encodes a nucleotide binding site leucine-rich repeat protein and is a member of a resistance gene cluster on rice chromosome 1. Genetics 177(3):1871 Lincoln SE, Daly MJ, Lander ES (1992) Constructing Genetic Maps with MapMaker/EXP3.0 Liu WD, Wang GL (2016) Plant innate immunity in rice: a defense against pathogen infection. Natl Sci Rev 3:295-308 Nguyen TTT, Koizumi S, La TN, Zenbayashi KS, Ashizawa T, Yasuda N, Imazaki I, Miyasaka A (2006) Pi35(t), a new gene conferring partial resistance to leaf blast in the rice cultivar Hokkai 188. Theor Appl Genet 113:697-704 Pan QH, Wang L, Tanisaka T, Ikehashi H (1998) Allelism of rice blast resistance genes in two Chinese rice cultivars, and identification of two new resistance genes. Plant Pathol 47(2):165-170 Panaud O, Chen X, Mccouch SR (1996) Development of microsatellite markers and characterization of simple sequence length polymorphism (SSLP) in rice (Oryza sativa L.). Mol Gen Genet 252(5):597-607 Qu SH, Liu GF, Zhou B, Bellizzi M, Zeng LR, Dai LY, Han B, Wang GL (2006) The broad-spectrum blast resistance gene Pi9 encodes a nucleotide-binding site-leucine-rich repeat protein and is a member of a multigene family in rice. Genetics 172:1901-1914 Sallaud C, Lorieux M, Roumen E, Tharreau D, Berruyer R, Svestasrani P, Garsmeur O, Ghesquiere A, Notteghem JL (2003) Identification of five new blast resistance genes in the highly blast-resistant rice variety IR64 using a QTL mapping strategy. Theor Appl Genet 106:794-803 Sharma TR, Rai AK, Gupta SK, Singh NK (2010) Broad-spectrum blast resistance gene Pi-kh cloned from rice line Tetep designated Pi54. References Science 355(6328):906-907 Wang J, Liu X, Zhang A, Yulong Ren, Wu FQ, Wang G, Xu Y, Lei CL, Zhu SS, Pan T, Wang YF, Zhang H, Wang F, Tan YQ, Wang YP, Jin X, Luo S, Zhou CL, Zhang X, Liu JL, Wang S, Meng LZ, Wang YH, Chen X, Lin QB, Zhang X, Guo XP, Cheng ZJ, Wang JL, Tian YL, Liu SJ, Jiang L, Wu CY, Wang ET, Zhou JM, Wang YF, Wang HY, Wan JM (2019) A cyclic nucleotide- gated channel mediates cytoplasmic calcium elevation and disease resistance in rice. Cell Res 29:820-831 Wang J, Liu X, Zhang A, Yulong Ren, Wu FQ, Wang G, Xu Y, Lei CL, Zhu SS, Pan T, Wang YF, Zhang H, Wang F, Tan YQ, Wang YP, Jin X, Luo S, Zhou CL, Zhang X, Liu JL, Wang S, Meng LZ, Wang YH, Chen X, Lin QB, Zhang X, Guo XP, Cheng ZJ, Wang JL Tian YL Liu SJ Jiang L Wu CY Wang ET Zhou JM Wang YF Wang HY Wan JM (2019) A cyclic nucleotide- g g g g g g g Wang JL, Tian YL, Liu SJ, Jiang L, Wu CY, Wang ET, Zhou JM, Wang YF, Wang HY, Wan JM (2019) A cyclic nucleotide- gated channel mediates cytoplasmic calcium elevation and disease resistance in rice. Cell Res 29:820-831 Wang S, Basten CJ, Zeng ZB (2007) Windows QTL cartographer 2.5. Department of Statistics, North Carolina State University, Raleigh, NC Wang ZX, Yano M, Yamanouchi U, Iwamoto M, Monna L, Hayasaka H, Katayose Y, Sasaki T (1999) The Pib gene for rice blast resistance belongs to the nucleotide binding and leucine-rich repeat class of plant disease resistance genes. Plant J 19(1):55-64 Wu JL, Fan YY, Li DB, Zheng KL, Leung H, Zhuang JY (2005) Genetic control of rice blast resistance in the durably resistant cultivar Gumei 2 against multiple isolates. Theor Appl Genet 111:50-56 Xiao N, Wu YY, Pan CH, Yu L, Chen Y, Liu GQ, Li YH, Zhang XX, Wang ZP, Dai ZY, Liang CZ, Li AH (2017) Improving of rice blast resistances in Japonica by pyramiding major R Genes. Front Plant Sci 7:1918 Yuan B, Zhai C, Wang WJ, Zeng XS, Xu XK, Hu HQ, Lin F, Wang L, Pan QH (2011) The Pik-p resistance to Magnaporthe oryzae in rice is mediated by a pair of closely linked CC-NBS-LRR genes. References J Plant Biochem Biotechnol 19(1):87-89 Page 10/20 Su J, Wang WJ, Han JL, Chen S,Wang CY, Zeng LX, Feng AQ, Yang JY, Zhou B, Zhu XY (2015) Functional divergence of duplicated genes results in a novel blast resistance gene Pi50 at the Pi2/9 locus. Theor Appl Genet 128(11):2213-2225 Tabien E, Li Z, Patterson AH, Marchetti A, Stansel W, Pinson M (2002) Mapping QTLs for field resistance to the rice blast pathogen and evaluating their individual and combined utility in improved varieties. Theor Appl Genet 105:313-324 Takahashi A, Hayashi N, Miyao A, Hirochika H (2010) Unique features of the rice blast resistance Pish locus revealed by large scale retrotransposon-tagging. BMC Plant Biol 10(1):175-0 Tanksley SD, Nelson JC (1996) Advanced backcross QTL analysis: a method for the simultaneous discovery and transfer of valuable QTLs from unadapted germplasm into elite breeding lines. Theor Appl Genet 92:191-203 Thomson MJ, Tai TH, McClung AM, Lai XH, Hinga ME, Lobos KB, Xu Y, Martinez CP, McCouch SR (2003) Mapping quantitative trait loci for yield, yield components and morphological traits in an advanced backcross population between O fi d th O ti lti J ff Th A l G t 107 479 493 Thomson MJ, Tai TH, McClung AM, Lai XH, Hinga ME, Lobos KB, Xu Y, Martinez CP, McCouch SR (2003) Mapping quantitative trait loci for yield, yield components and morphological traits in an advanced backcross population between Oryza rufipogon and the Oryza sativa cultivar Jefferson. Theor Appl Genet 107:479-493 Oryza rufipogon and the Oryza sativa cultivar Jefferson. Theor Appl Genet 107:479-49 Oryza rufipogon and the Oryza sativa cultivar Jefferson. Theor Appl Genet 107:479-493 Vasudevan K, Gruissem W, Bhullar NK (2016) Corrigendum: Identification of novel alleles of the rice blast resistance gene Pi54. Sci Rep 6(15678):17920 Wang GL, Mackill DJ, Bonman JM, McCouch SR, Champoux MC, Nelson RJ (1994) RFLP mapping of genes conferring complete and partial resistance to blast in a durably resistant rice cultivar. Genetics 136:1421-1434 Wang GL, Valent B (2017) Durable resistance to rice blast. Science 355(6328):906-907 Wang GL, Valent B (2017) Durable resistance to rice blast. References Theor Appl Genet 122(5), 1017-1028 Zhai C, Lin F, Dong Z, Dong ZQ, He XY, Yuan B, Zeng XS, Wang L, Pan QH (2011) The isolation and characterization of Pik, a rice blast resistance gene which emerged after rice domestication. New Phytol 189(1):321-334 Zhai C, Zhang Y, Yao N, Lin F, Liu Z, Dong ZQ, Wang L, Pan QH (2014) Function and interaction of the coupled genes responsible for Pik-h encoded rice blast resistance. PLoS ONE 9(6):e98067 Page 11/20 Zhao HJ, Wang XY, Jia YL, Minkenberg B, Wheatley M, Fan J, Jia M, Famoso A, Edwards J, Wamishe Y, Valent B, Wang GL, Yang Y (2018) The rice blast resistance gene Ptr encodes an atypical protein required for broad-spectrum disease resistance. Nat Commun 9:2039 Zhao HJ, Wang XY, Jia YL, Minkenberg B, Wheatley M, Fan J, Jia M, Famoso A, Edwards J, Wamishe Y, Valent B, Wang GL, Yang Y (2018) The rice blast resistance gene Ptr encodes an atypical protein required for broad-spectrum disease resistance. Nat Commun 9:2039 Zheng WJ, Wang Y, Wang LL, Ma ZB, Zhao JM, Wang P, Zhang LX, Liu ZH, Lu XC (2016) Genetic mapping and molecular marker development for Pi65(t), a novel broad-spectrum resistance gene to rice blast using next-generation sequencing. Theor Appl Genet 129(5):1035-1044 Zheng WJ, Wang Y, Wang LL, Ma ZB, Zhao JM, Wang P, Zhang LX, Liu ZH, Lu XC (2016) Genetic mapping and molecular marker development for Pi65(t), a novel broad-spectrum resistance gene to rice blast using next-generation sequencing. Theor Appl Genet 129(5):1035-1044 Zhou B, Qu SH, Liu GF, Dolan M, Sakai H, Lu GD, Bellizzi M, Wang GL (2006) The eight amino-acid differences within three leucine-rice repeats between Pi2 and Piz-t resistance proteins determine the resistance specificity to Magnaporthe grisea. Mol Plant Microbe In 11:1216-1228 Zhu XY, Chen S, Yang JY, Zhou SC, Zeng LX, Han JL, Su J, Wang L, Pan QH (2012) The identification of Pi50(t), a new member of the rice blast resistance Pi2/Pi9 multigene family. Theor Appl Genet 124:1295-1304 Zhu XY, Chen S, Yang JY, Zhou SC, Zeng LX, Han JL, Su J, Wang L, Pan QH (2012) The identification of Pi50(t), a new member of the rice blast resistance Pi2/Pi9 multigene family. Theor Appl Genet 124:1295-1304 Note: RT, leaf blast resistance at tillering stage. RH, leaf blast resistance at heading stage. RN, neck blast resistance at maturation stage. Table 2 QTL mapping results from Jin23B/CR071 background population in two years Tables Table 1 Statistical description of the parents and the backcross population in two years Trait Year Parent Jin23B/CR071 backcross population Jin23B/QingGuAi backcross population Jin23B CR071 QingGuAi Mean±SD Range Mean±SD Range RT 2011 4.17±0.38 0.83±0.38 1.06±0.47 2.59±1.12 0-5 2.30±1.16 0-5 2012 4.11±0.32 0.97±0.16 1.11±0.40 2.58±1.14 0.17-4.5 2.05±1.05 0-4.75 RH 2011 4.22±0.42 1.00±0.23 1.14±0.35 2.48±1.25 0-5 2.37±1.30 0-5 2012 4.19±0.40 1.00±0.33 1.19±0.40 2.59±1.16 0-4.5 2.42±1.08 0-4.58 RN 2011 91.7% ±2.08% 3.33% ±1.15 6.67% ±0.58% 62.7%±40.7% 0-100% 71.3%±36.91% 0-100% 2012 93.7% ±2.31% 4.67% ±1.52 8.33% ±0.58% 54.8%±33.6% 0-100% 78.58±32.23% 0-100% Note: RT, leaf blast resistance at tillering stage. RH, leaf blast resistance at heading stage. RN, neck blast resistance at maturation stage. Note: RT, leaf blast resistance at tillering stage. RH, leaf blast resistance at heading stage. RN, neck blast resistance at maturation stage. Tables Table 2 QTL mapping results from Jin23B/CR071 background population in two years Page 12/20 Page 12/20 Trait Chr QTL Position LOD Add R2 (%) Trait Chr QTL Position LOD Add R2 (%) 11RT 12RT 1 qBR1 a RM237- RM486 7.39 -0.73a 6.09 2 qBR2-1 RM236- RM452 5.32 1.42 6.64 2 qBR2-1 RM236- RM452 6.17 1.63 8.17 2 qBR2-2 RM324- RM341 4.73 1.20 5.77 2 qBR2-3 RM530- RM213 4.13 0.92 9.15 3 qBR3-3 SR49- RM426 20.15 1.58 25.42 3 qBR3-3 SR49- RM426 30.08 1.81 34.08 6 qBR6 RM539- R19951 4.05 0.92 8.89 6 qBR6 RM539- R19951 10.02 1.92 20.40 7 qBR7-1 RM501- RM542 3.47 1.51 10.09 8 qBR8 RM72- RM404 4.04 1.31 6.29 12 qBR12 RM179- YP6213 2.80 0.57 4.25 12 qBR12 RM179- YP6213 3.57 0.58 4.48 11RH 1 qBR1 a RM237- RM486 9.34 -0.93 7.99 12RH 1 qBR1 a RM237- RM486 8.35 -0.80 7.25 2 qBR2-1 RM236- RM452 3.90 1.41 4.51 2 qBR2-1 RM236- RM452 5.75 1.41 6.38 2 qBR2-2 RM324- RM341 4.03 1.12 4.89 2 qBR2-2 RM324- RM341 3.91 1.13 5.21 2 qBR2-3 RM530- RM213 4.51 0.96 8.68 2 qBR2-3 RM530- RM213 3.23 0.84 7.47 3 qBR3-3 SR49- RM426 23.67 1.76 25.83 3 qBR3-3 SR49- RM426 28.23 1.78 31.33 6 qBR6 RM539- R19951 8.94 0.85 7.79 6 qBR6 RM539- R19951 7.80 0.95 10.40 8 qBR8 RM72- RM404 6.78 1.71 7.00 8 qBR8 RM72- RM404 4.20 1.30 4.83 12 qBR12 RM179- YP6213 3.28 0.59 3.90 12 qBR12 RM179- YP6213 2.93 0.53 3.50 11RN 1 qBR1 a RM237- RM486 10.70 -0.36 8.12 12RN 1 qBR1 a RM237- RM486 10.55 -0.35 9.48 2 qBR2-3 RM530- RM213 3.61 0.28 7.10 3 qBR3-1 RM282- RM411 4.74 0.31 7.19 3 qBR3-1 RM282- RM411 7.50 0.23 7.02 3 qBR3-2 RM411- RM487 4.12 0.23 3.91 4 qBR4 RM252- RM470 3.04 0.22 4.16 7 qBR7-1 RM501- RM542 28.26 0.54 28.48 7 qBR7-1 RM501- RM542 20.27 0.36 18.94 7 qBR7-2 RM214- M5543 25.26 0.48 22.00 7 qBR7-2 RM214- M5543 17.44 0.37 18.94 11 qBR11- 1 RM181- RM120 27.25 0.58 28.16 11 qBR11- 1 RM181- RM120 21.57 0.47 26.19 11 qBR11- 2 RM120- RM536 24.09 0.44 20.26 11 qBR11- 2 RM120- RM536 18.69 0.34 17.93 11 qBR11- 3 RM21- RM590 8.02 0.53 17.48 11 qBR11- 3 RM21- RM590 5.07 0.41 15.94 Note: Chr, Chromosome. LOD, logarithm of odds. Add, the additive effect of each QTL. R2, Goodness of fit, represent the phenotypic variance explained by each QTL. Tables a, the resistance effect of QTL come from Jin23B. 11RT and 12RT, leaf blast esistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 1RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Note: Chr, Chromosome. LOD, logarithm of odds. Add, the additive effect of each QTL. R2, Goodness of fit, represent the phenotypic variance explained by each QTL. a, the resistance effect of QTL come from Jin23B. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Note: Chr, Chromosome. LOD, logarithm of odds. Add, the additive effect of each QTL. R2, Goodness of fit, represent the phenotypic variance explained by each QTL. a, the resistance effect of QTL come from Jin23B. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Tables able 3 QTL mapping results from Jin23B/QingGuAi background population in two years Page 14/20 Trait Chr QTL Position LOD Add R2 (%) Trait Chr QTL Position LOD Add R2 (%) 11RT 12RT 1 qBR1-1 RM297- RM486 2.98 0.54 3.95 6 qBR6 L6ID3F- ZH6111 19.39 1.40 24.93 6 qBR6 L6ID3F- ZH6111 13.73 1.14 19.91 7 qBR7-1 RM214- RM5543 7.83 0.89 8.87 7 qBR7-1 RM214- RM5543 3.18 0.55 4.12 11 qBR11- 1 RM229- RM457 5.17 1.08 6.57 11 qBR11- 2 RM457- YH43 6.94 34.7 5.49 11RH 12RH 4 qBR4-2 RM241- RM317 3.44 0.49 2.51 6 qBR6 ZH6111- RM20069 28.39 1.90 36.48 6 qBR6 L6ID3F- ZH6111 28.35 1.31 25.27 7 qBR7-1 RM214- RM5543 14.98 1.36 16.47 7 qBR7-1 RM214- RM5543 29.33 1.43 26.31 11 qBR11- 1 RM229- RM457 3.2 0.61 2.42 11RN 12RN 1 qBR1-2 RM5- RM488 10.9 24.76 8.55 2 qBR2 ZH282- RM71 2.59 44.12 8.92 3 qBR3 RM3441- RM232 4.46 50.77 12.72 4 qBR4-1 RM471- RM241 3.69 42.68 12.31 4 qBR4-1 RM471- RM241 7.64 40.01 18.47 4 qBR4-2 RM241- RM317 2.53 45.94 8.82 4 qBR4-2 RM241- RM317 6.76 46.65 16.92 6 qBR6 L6ID3F- ZH6111 33.94 50.47 31.91 6 qBR6 L6ID3F- ZH6111 29.23 42.24 28.95 7 qBR7-1 RM214- RM5543 30.17 50.02 27.94 7 qBR7-1 RM214- RM5543 20.75 35.94 18.71 7 qBR7-2 RM432- RM21691 4.61 17.06 3.43 8 qBR8 RM404- RM210 7.80 38.08 18.89 12 qBR12 RM5927- RM6296 8.02 -33.18 a 9.58 Note: Chr, Chromosome. LOD, logarithm of odds. Add, the additive effect of each QTL. R2, Goodness of fit, represent the phenotypic variance explained by each QTL. a, the resistance effect of QTL come from Jin23B. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. 11RN Note: Chr, Chromosome. LOD, logarithm of odds. Add, the additive effect of each QTL. R2, Goodness of fit, represent the phenotypic variance explained by each QTL. a, the resistance effect of QTL come from Jin23B. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Supplementary Figure Legend Supplementary Figure Legend Page 15/20 Fig. S1 Marker ratio of Jin23B homozygous genotype of each plant in two BC3F1 background population. a, marker ratio of Jin23B homozygous genotype of each plant in Jin23B/CR071 background population. b, marker ratio of Jin23B homozygous genotype of each plant in Jin23B/QingGuAi background population. Fig. S1 Marker ratio of Jin23B homozygous genotype of each plant in two BC3F1 background population. a, marker ratio of Jin23B homozygous genotype of each plant in Jin23B/CR071 background population. b, marker ratio of Jin23B homozygous genotype of each plant in Jin23B/QingGuAi background population. homozygous genotype of each plant in Jin23B/QingGuAi background population. Figures Figure 1 Strategy to develop mapping populations for blast resistance QTLs Figures Figure 1 Strategy to develop mapping populations for blast resistance QTLs Figure 1 Figure 1 Strategy to develop mapping populations for blast resistance QTLs Strategy to develop mapping populations for blast resistance QTLs Page 16/20 Figure 2 Frequency distribution of blast resistance of the Jin23B/CR071 (a, b, c) and Jin23B/QingGuAi (d, e f) population in 2011 and 2012. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Blue pillar and green pillar indicate the frequency of BC3F1 and BC3F2 population respectively Figure 2 Frequency distribution of blast resistance of the Jin23B/CR071 (a, b, c) and Jin23B/QingGuAi (d, e f) population in 2011 and 2012. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Blue pillar and green pillar indicate the frequency of BC3F1 and BC3F2 population, respectively. Page 17/20 Figure 3 Distribution of QTLs for blast resistance in the Jin23B/CR071 population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Figure 3 Distribution of QTLs for blast resistance in the Jin23B/CR071 population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012 11RN and 12RN neck blast resistance at maturation stage in 2011 and 2012 Figure 3 Distribution of QTLs for blast resistance in the Jin23B/CR071 population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Distribution of QTLs for blast resistance in the Jin23B/CR071 population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Page 18/20 Figure 4 Distribution of QTLs for blast resistance in the Jin23B/QingGuAi population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Figure 4 Figure 5 Genetic effects of qBR3-3 and qBR6 from CR071 and qBR6 and qBR7-1 from QingGuAi. Blue bar represent alleles from Jin23B, Green bar represent alleles from donor parent. P-value based on two-way t-test. Error bars are based on standard deviation of each genotype. Figure 4 Distribution of QTLs for blast resistance in the Jin23B/QingGuAi population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Distribution of QTLs for blast resistance in the Jin23B/QingGuAi population on the genetic linkage map. 11RT and 12RT, leaf blast resistance at tillering stage in 2011 and 2012. 11RH and 12RH, leaf blast resistance at heading stage in 2011 and 2012. 11RN and 12RN, neck blast resistance at maturation stage in 2011 and 2012. Page 19/20 Page 19/20 Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. SupplementaryTables.pdf Page 20/20
https://openalex.org/W4315853529	https://www.mdpi.com/1648-9144/59/1/157/pdf?version=1673925593	English	null	Non-Hodgkin Lymphoma Developed Shortly after mRNA COVID-19 Vaccination: Report of a Case and Review of the Literature	Medicina	2,023	cc-by	6,790	Citation: Cavanna, L.; Grassi, S.O.; Rufﬁni, L.; Michieletti, E.; Carella, E.; Palli, D.; Zangrandi, A.; Inzerilli, N.; Bernuzzi, P.; Di Nunzio, C.; et al. Non-Hodgkin Lymphoma Developed Shortly after mRNA COVID-19 Vaccination: Report of a Case and Review of the Literature. Medicina 2023, 59, 157. https://doi.org/ 10.3390/medicina59010157 Bernuzzi, P.; Di Nunzio, C.; et al. Non-Hodgkin Lymphoma Developed Shortly after mRNA COVID-19 Vaccination: Report of a Case and Review of the Literature. Medicina 2023, 59, 157. https://doi.org/ 10.3390/medicina59010157 Academic Editors: Mauro Salvatore Alessandro Alaibac and Domenico Albano Received: 15 December 2022 Revised: 3 January 2023 Accepted: 9 January 2023 Published: 12 January 2023 Academic Editors: Mauro Salvatore Alessandro Alaibac and Domenico Albano Received: 15 December 2022 Revised: 3 January 2023 Accepted: 9 January 2023 Published: 12 January 2023 Keywords: COVID-19; vaccination; lymphoma Keywords: COVID-19; vaccination; lymphoma Case Report Non-Hodgkin Lymphoma Developed Shortly after mRNA COVID-19 Vaccination: Report of a Case and Review of the Literature Luigi Cavanna 1,* Sergio Ottavio Grassi 2 Livia Rufﬁni 3 Emanuele Michieletti 4 Egidio Carella 4 Luigi Cavanna 1,* , Sergio Ottavio Grassi 2, Livia Rufﬁni 3 , Emanuele Michieletti 4, Egidio Carella 4, Dante Palli 5, Adriano Zangrandi 6, Nicola Inzerilli 1, Patrizia Bernuzzi 1, Camilla Di Nunzio 1 and Chiara Citterio 1 Luigi Cavanna 1,* , Sergio Ottavio Grassi 2, Livia Rufﬁni 3 , Emanuele Michieletti 4, Egidio Carella 4, Dante Palli 5, Adriano Zangrandi 6, Nicola Inzerilli 1, Patrizia Bernuzzi 1, Camilla Di Nunzio 1 and Chiara Citterio 1 Abstract: We report on a 66-year-old man who presented with a right axillary lymphadenopathy approximately 10 days after receiving the third dose of the BNT162b2 vaccine. The lymphadenopathy gradually enlarged, and physical examination and ultrasound (US) revealed one right axillary 6.99 cm and one right supraclavicular 2.36 cm lymphadenopathy. Histologic examination of the right axillary nodule revealed anaplastic large-cell lymphoma that was ALK negative and CD30 positive. A total body computerized tomography (CT) scan, positron emission tomography (PET) and bone-marrow biopsy showed a stage-II non-Hodgkin lymphoma (NHL). The patient was treated with chemotherapy and a scheme of Brentuximab Vedotin, Cyclophosphamide, Doxorubicin and Prednisone (BV-CHP) for six cycles and is now well and in complete remission. The revision of the literature revealed eight additional cases of NHL developed shortly after COVID-vaccination. There were four cases of diffuse large-B-cell lymphoma (DLBCL) (one in a patient who was a heart transplant recipient and developed an Epstein–Bar-virus-positive DLBCL), one case of extranodal NK/T-cell lymphoma, one patient with subcutaneous panniculitis-like T-cell lymphoma, one case of marginal zone B-cell lymphoma and one primary cutaneous anaplastic large-cell lymphoma (PC-ALCL). In ﬁve cases, the lymphoma developed after BNT162b2 mRNA vaccination, including one case after ChAdOx1 nCOV-19, one case after the adenovirus type 26 (Ad26) vaccine and one after mRNA-1273/Spikevax (ModernaTX). We are aware that the link between COVID-19 vaccination and lymphoma most likely is a chance phenomenon, and that COVID-19 vaccines represent very efﬁcient products for many people around the world. However, we believe that clinical events, even if only temporally associated with novel treatments or novel vaccines, should be reported for the beneﬁt of the patients and the scientiﬁc community. medicina medicina medicina 1 Department of Oncology and Hematology, ASL Piacenza, Via Taverna 49, 29121 Piacenza, Italy 2 ASST Lodi, 26900 Lodi, Italy 3 Nuclear Medicine Division, Azienda Ospedaliero-Universitaria di Parma, 43126 Parma, Italy 4 Department of Radiology, ASL Piacenza, Via Taverna 49, 29121 Piacenza, Italy 5 Surgey and Breast Unit, Guglielmo da Saliceto Hospital, Via Taverna 49, 29121 Piacenza, Italy 6 Pathology Unit, ASL Piacenza, Via Taverna 49, 29121 Piacenza, Italy * Correspondence: luigicavanna53@gmail.com; Tel.: +39-0523302697; Fax: +39-0523302141 Axillary and cervical lymphadenopathy related to recent vaccination is a well-known event following vaccines such as those for smallpox, inﬂuenza, human papilloma virus and other vaccines [1,2]. More recently, COVID-19 vaccination was identiﬁed as a cause of inﬂammatory unilateral axillary and cervical adenopathy [3,4]. The median duration of adenopathy reported after mRna-1273 (Moderna) and BNT162b2 (Pﬁzer) vaccine adminis- tration ranges from 1–2 days to approximately 10 days in duration; however, in some cases, the lymphadenopathies persisted for more than 1 month after vaccination [3,4]. The differ- ential diagnosis of lymphadenopathies reactive with other types of lymphadenopathies, 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/). https://www.mdpi.com/journal/medicina Medicina 2023, 59, 157. https://doi.org/10.3390/medicina59010157 Medicina 2023, 59, 157 2 of 8 active ccinat such as lymphoma, is very important. Generally, post-vaccine-reactive lymphadenopathy shows a progressive decrease within a number of days after vaccination [5]. lowing the third dose of BNT162b2 vaccination, which was suspected to be a re phadenopathy. However, the mass gradually enlarged, and a biopsy of p g y Herein, we discuss a case of right axillary lymphadenopathy developed 10 days following the third dose of BNT162b2 vaccination, which was suspected to be a reactive lymphadenopathy. However, the mass gradually enlarged, and a biopsy of the nodule demonstrated non-Hodgkin lymphoma (NHL). In addition, we performed a review of the literature, and here, we report the results of this review. Written informed consent for the publication of details and images of the case report was obtained from the patient and is available for veriﬁcation by the handling editor if needed. p p y g y g p y demonstrated non-Hodgkin lymphoma (NHL). In addition, we performed a re literature, and here, we report the results of this review. Written informed con publication of details and images of the case report was obtained from the pat available for verification by the handling editor if needed. 2. Case Report 2. Case Report The pati disease In Jan The patient reported on here is a 66-year-old white man with no previous history of disease. In January 2021, he received the ﬁrst dose of the BNT162b2 (Pﬁzer) vaccine, and after one month, he received a second dose of the BNT162b2 vaccine. In October 2021, he received a booster dose of the same vaccine in the right arm, and subsequently, he showed a mobile and painful right axillary lymphadenopathy, which arose approximately 10 days after the administration of the vaccine booster, with no other clinical history of interest. disease. In January 2021, he received the first dose of the BNT162b2 (Pfizer) v after one month, he received a second dose of the BNT162b2 vaccine. In Octob received a booster dose of the same vaccine in the right arm, and subsequently, a mobile and painful right axillary lymphadenopathy, which arose approximat after the administration of the vaccine booster, with no other clinical history o The patient presented to our outpatient clinic 3 months later with enlarge y The patient presented to our outpatient clinic 3 months later with enlarged right axillary adenopathy that was conﬁrmed upon physical examination, which also revealed an ipsilateral supraclavicular lymphadenopathy (Figure 1). The patient presented to our outpatient clinic 3 months later with enlarge illary adenopathy that was confirmed upon physical examination, which also r ipsilateral supraclavicular lymphadenopathy (Figure 1). Figure 1. Physical examination revealed a large right axillary adenopa Figure 1. Physical examination revealed a large right axillary adenopathy. Figure 1. Physical examination revealed a large right axillary ade Figure 1. Physical examination revealed a large right axillary adenopathy. The lymphadenopathies were painful and mobile. A total body computed tomography (CT) examination conﬁrmed the two lymphadenopathies, measuring 6.99 cm and 2.36 cm, respectively (Figure 2). Medicina 2023, 59, 157 3 of 8 gra- 2 36 Figure 2. CT scan revealed two lymphadenopathies: right axillary 6.99 cm (A) and supraclavicular 2.36 cm (B), and multiple adenopathies in the coronal section (C). Figure 2. CT scan revealed two lymphadenopathies: right axillary 6.99 cm (A) and supraclavicular 2.36 cm (B), and multiple adenopathies in the coronal section (C). Figure 2. CT scan revealed two lymphadenopathies: right axillary 6.99 cm (A) and sup 2.36 cm (B), and multiple adenopathies in the coronal section (C). 2. Case Report The pati disease In Jan A US id d fi dl i i bi f h ill l h d thies: right axillary 6 99 cm (A) and su Figure 2. CT scan revealed two lymphadenopathies: right axillary 6.99 cm (A) and supraclavicular 2.36 cm (B), and multiple adenopathies in the coronal section (C). Figure 2. CT scan revealed two lymphadenopathies: right axillary 6.99 cm (A) and supraclavicular 2.36 cm (B), and multiple adenopathies in the coronal section (C). 2.36 cm (B), and multiple adenopathies in the coronal section (C). A US-guided, fine-needle aspiration biopsy of the axillary lymphadenopathy was carried out, and suspicion of NHL was suggested. A surgical biopsy of the right axillary adenopathy revealed anaplastic large-cell lymphoma that was ALK negative and CD30 positive, with the presence of elevated Ki-67 proliferation antigen (90%) (Figures 3 and 4). A US-guided, ﬁne-needle aspiration biopsy of the axillary lymphadenopathy was carried out, and suspicion of NHL was suggested. A surgical biopsy of the right axillary adenopathy revealed anaplastic large-cell lymphoma that was ALK negative and CD30 positive, with the presence of elevated Ki-67 proliferation antigen (90%) (Figures 3 and 4). A US guided, fine needle aspiration biopsy of the axillary lymphadeno carried out, and suspicion of NHL was suggested. A surgical biopsy of the rig adenopathy revealed anaplastic large-cell lymphoma that was ALK negative positive, with the presence of elevated Ki-67 proliferation antigen (90%) (Figur Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large (compared with the normal lymphocytes on the right) with highly atypical nuclei and macronucleoli. Apoptotic bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large with the normal lymphocytes on the right) with highly atypical nuclei and macronucleol bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large (compared with the normal lymphocytes on the right) with highly atypical nuclei and macronucleoli. Apoptotic bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large (co with the normal lymphocytes on the right) with highly atypical nuclei and macronucleoli. Ap bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. 2. Case Report The pati disease In Jan Histologic sample from axillary lymphadenectomy: tumor cells are very larg with the normal lymphocytes on the right) with highly atypical nuclei and macronucleo bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large (compared with the normal lymphocytes on the right) with highly atypical nuclei and macronucleoli. Apoptotic bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large with the normal lymphocytes on the right) with highly atypical nuclei and macronucleo bodies and mitoses are frequent, including atypical mitoses (red arrow). Figure 3. Histologic sample from axillary lymphadenectomy: tumor cells are very large (compared with the normal lymphocytes on the right) with highly atypical nuclei and macronucleoli. Apoptotic bodies and mitoses are frequent, including atypical mitoses (red arrow). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavicular adenopathy and in multiple axillary adenopathies (Figure 5). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavicular adenopathy and in multiple axillary adenopathies (Figure 5). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavicular adenopathy and in multiple axillary adenopathies (Figure 5). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavicular adenopathy and in multiple axillary adenopathies (Figure 5). A bone marrow biopsy was performed, a diagnosis of stage-II NHL was made, and the patient underwent monoclonal antibody anti-CD30 Brentuximab Vedotin in association with CHP chemotherapy (Cyclophosfamide, Adriblastine, Prednisone) for six cycles from March to July 2022, with complete remission after three cycles (early PET). No particular signs of toxicity were observed during treatment, which was well-tolerated. The patient is on follow-up, with a good health condition and persistent complete remission, as conﬁrmed by PET/CT restaging at the end of the treatment (Figure 6). Medicina 2023, 59, 157 Medicina 2023, 59, x FO 4 of 8 4 of Figure 4. Lymphoma cells are positive for T-cell markers: CD3 (A), CD2 (B) and CD4 (C). They a also diffusely and strongly positive for CD30, with a membranous and dot-like pattern (D). AL protein was negative (not shown). Figure 4. Lymphoma cells are positive for T-cell markers: CD3 (A), CD2 (B) and CD4 (C). 2. Case Report The pati disease In Jan 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). Medicina 2023, 59, x FOR PEER REVIEW 5 of 9 18F-Fluorodeoxyglucose po owed increased tracer uptake in ar adenopathy and in multiple a sh ul po ke in grap path sh sh Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). VIEW 5 of 9 Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). VIEW 5 of 9 creased tracer uptake in clavicular adenopathy (S ge-II NHL was made, a ximab Vedotin in asso Prednisone) for six cyc I NHL was made, and ab Vedotin in associa- dnisone) for six cycles es (early PET). No par- was well-tolerated. The nt complete remission, ure 6). k i and i Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in th right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SU max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). A bone marrow biopsy was performed, a diagnosis of stage-II NHL was made, an the patient underwent monoclonal antibody anti-CD30 Brentuximab Vedotin in associ tion with CHP chemotherapy (Cyclophosfamide, Adriblastine, Prednisone) for six cycle from March to July 2022, with complete remission after three cycles (early PET). No pa ticular signs of toxicity were observed during treatment, which was well-tolerated. patient is on follow up, with as confirmed by PET/CT rest 3. Revision of the Literature 3. Revision of the Literature patient is on follow up, with as confirmed by PET/CT rest 3. Revision of the Literature Methods 3. Revision of the Literature Methods 2. Case Report The pati disease In Jan They are also diffusely and strongly positive for CD30, with a membranous and dot-like pattern (D). ALK protein was negative (not shown). Figure 4. Lymphoma cells are positive for T-cell markers: CD3 (A), CD2 (B) and CD4 (C). They are also diffusely and strongly positive for CD30, with a membranous and dot-like pattern (D). ALK protein was negative (not shown). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavic- ular adenopathy and in multiple axillary adenopathies (Figure 5). Figure 4. Lymphoma cells are positive for T-cell markers: CD3 (A), CD2 (B) and CD4 (C). They a also diffusely and strongly positive for CD30, with a membranous and dot-like pattern (D). AL protein was negative (not shown). Figure 4. Lymphoma cells are positive for T-cell markers: CD3 (A), CD2 (B) and CD4 (C). They are also diffusely and strongly positive for CD30, with a membranous and dot-like pattern (D). ALK protein was negative (not shown). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/CT) showed increased tracer uptake in the right axillary adenopathy, in the right supraclavic- ular adenopathy and in multiple axillary adenopathies (Figure 5). 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/C showed increased tracer uptake in the right axillary adenopathy, in the right supraclav ular adenopathy and in multiple axillary adenopathies (Figure 5). Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). cina 2023, 59, x FOR PEER REVIEW 5 of 9 18F-Fluorodeoxyglucose positron emission tomography/CT (18-FDG-PET/C showed increased tracer uptake in the right axillary adenopathy, in the right supraclav ular adenopathy and in multiple axillary adenopathies (Figure 5). Figure 5. 18-FDG-PET/CT at baseline (axial and coronal sections), increased tracer uptake in the right axillary adenopathy mass (SUV max 9.7) (A), in the right supraclavicular adenopathy (SUV max 7.6) (B) and in multiple axillary adenopathies (level I–III) (C,D). Figure 5. 2. Case Report The pati disease In Jan Th patient is on follow up with a good health condition and persistent complete remissio A bone marrow biopsy was performed, a diagnosis of stage-II NHL was made, and the patient underwent monoclonal antibody anti-CD30 Brentuximab Vedotin in associa- tion with CHP chemotherapy (Cyclophosfamide, Adriblastine, Prednisone) for six cycles from March to July 2022, with complete remission after three cycles (early PET). No par- ticular signs of toxicity were observed during treatment, which was well-tolerated. The patient is on follow-up, with a good health condition and persistent complete remission, as confirmed by PET/CT restaging at the end of the treatment (Figure 6). Figure 6. Restaging 18-FDG-PET/CT after chemotherapy showed the complete metabolic response of all the involved sites (A–D). Figure 6. Restaging 18-FDG-PET/CT after chemotherapy showed the complete metabolic response of all the involved sites (A–D). (axial and coronal sections), inc max 9.7) (A), in the right suprac denopathies (level I–III) (C,D). erformed, a diagnosis of stag l antibody anti-CD30 Brentux rformed, a diagnosis of stage II antibody anti-CD30 Brentuxim ophosfamide, Adriblastine, Pre lete remission after three cycle ed during treatment, which w health condition and persisten t the end of the treatment (Figu Fi rig m th th tio fr tic pa as F i th i ), in uxim P eline SUV nal C l from March to July 2022, with complete remission after three cycles (early PET). No pa ticular signs of toxicity were observed during treatment, which was well-tolerated. Th atie t i o follo u ith a ood health o ditio a d e i te t o lete e i io Figure 6. Restaging 18-FDG-PET/CT after chemotherapy showed the complete metabolic response of all the involved sites (A–D). Figure 6. Restaging 18-FDG-PET/CT after chemotherapy showed the complete metabolic response of all the involved sites (A–D). 4. Results of the Literature Revision Eight patients who developed NHL after COVID-19 vaccination were identiﬁed, in- cluding four males and four women. Five patients were vaccinated with the BNT162b2 vaccine (Pﬁzer), one with the ChAdOx1 nCOV-19 vaccine (AstraZeneca, Cambridge, UK), one with mRNA-1273/Spikevax (ModernaTX) and one patient with the recombinant replication-incompetent adenovirus type 26 (Ad26) viral-vector-based COVID-19 vaccine (Janssen Pharmaceuticals, Beerse, Belgium) (Table 1) [6–11]. Five patients had B-cell lym- phoma and three had T-cell lymphoma, while four patients were diagnosed with diffuse large-B-cell lymphoma (DLBCL). One patient had extranodal NK/T-cell lymphoma [7], and one patient, a heart transplant recipient, developed a rapidly growing Epstein–Barr-virus (EBV)-positive diffuse large-B-cell lymphoma 7 days after receiving the ﬁrst dose of the ChAdOx1 nCOV-19 vaccine [8]. One patient had a subcutaneous panniculitis-like T-cell lymphoma (SPTCL) [9], one had an extranodal marginal zone lymphoma (EMZL) [10] and one primary cutaneous anaplastic large-cell lymphoma (PC-ALCL) [11]. The lymphopro- liferative disorder developed shortly after the vaccination, within 1 to 10 days following vaccination (Table 1). Four of these eight patients were treated with systemic treatment, one was treated with radical surgery and radiation therapy, one was treated with systemic treatment and radiation therapy, one was treated through the watchful waiting program for comorbidities and the last one showed spontaneous, almost total remission without treatment after 6 weeks of follow-up. Table 1. Review of the literature: malignant lymphoma diagnosed in the context of the mRNA COVID-19 vaccination (Ad26: adenovirus type 26, ALCL: anaplastic large-cell lymphoma, DLBCL: diffuse large-B-cell lymphoma, EBV: Epstein–Barr virus, EMZL: extranodal marginal zone lymphoma, PC-ALCL: primary cutaneous anaplastic large-cell lymphoma, SPTCL: subcutaneous panniculitis-like T-cell lymphoma). * The two previous vaccination doses were BNT162b2. Case N. patient is on follow up, with as confirmed by PET/CT rest 3. Revision of the Literature Methods 3. Revision of the Literature Methods Gender/Age (Year) (Reference) Time from Vaccination to Onset of Lymphoproliferative Disorder Histopathological Examination Type of COVID-19 Vaccine Site and Diameter of Lymphadenopathy Treatment of Lymphoma 1 M/67 [6] 1 day after 1 dose DLBCL BNT162b2 Left axilla 6.0 cm Chemotherapy plus rituximab 2 F/80 [6] 2 days after 1 dose DLBCL BNT162b2 Left axilla 4.1 cm Chemotherapy plus rituximab 3 F/58 [7] 7 days after 2 dose DLBCL BNT162b2 Left cervical area 4 cm Radical surgery plus radiotherapy 4 M/53 [7] 3 days after 1 dose Extranodal NK/T-cell lymphoma BNT162b2 Erosive lesions upper lip up to 5 mm Chemotherapy plus radiotherapy 5 M/51 [8] 7 days after 1 dose EBV-positive DLBCL ChAdox1 nCOV-19 Mediastinal mass 5 cm Rituximab 6 F/28 [9] “A few days after 1 dose” SPTCL Ad26 viral-vector-based Injection site, upper arm Cyclosporine plus prednisone 7 F/80 [10] 1 day after 1 dose EMZL BNT162b2 Right temporal mass No treatment 8 M/76 [11] 10 days after the booster dose PC-ALCL mRNA-1273 * Right arm upper-external surface 6 cm No treatment 5 of 8 Medicina 2023, 59, 157 The general research terms used included the following: “COVID-19, SARS-CoV-2, COVID-19 vaccine, mRNA vaccine, malignancy, neoplasm, cancer, lymphoma, lymphopro- liferative disease”. The research was carried out on 15 November 2022 and was limited to literature in English. The research was conducted independently by two investigators (LC and CC). patient is on follow up, with as confirmed by PET/CT rest 3. Revision of the Literature Methods 3. Revision of the Literature Methods Case reports of lymphoproliferative disorders related to the COVID-19 vaccine were systematically researched using databases including PUBMED, SCOPUS, Science Direct and EBSCO Host. Case reports of lymphoproliferative disorders related to the COVID-19 vaccine were systematically researched using databases including PUBMED, SCOPUS, Science Direct and EBSCO Host. Medicina 2023, 59, 157 5 of The general research terms used included the following: “COVID-19, SARS-CoV- COVID-19 vaccine, mRNA vaccine, malignancy, neoplasm, cancer, lymphoma, lymphopro liferative disease”. The research was carried out on 15 November 2022 and was limited to literature i English. The research was conducted independently by two investigators (LC and CC). 4. Results of the Literature Revision Eight patients who developed NHL after COVID-19 vaccination were identiﬁed, in cluding four males and four women. Five patients were vaccinated with the BNT162b vaccine (Pﬁzer), one with the ChAdOx1 nCOV-19 vaccine (AstraZeneca, Cambridge, UK one with mRNA-1273/Spikevax (ModernaTX) and one patient with the recombinan replication-incompetent adenovirus type 26 (Ad26) viral-vector-based COVID-19 vaccin (Janssen Pharmaceuticals, Beerse, Belgium) (Table 1) [6–11]. Five patients had B-cell lym phoma and three had T-cell lymphoma, while four patients were diagnosed with diffus large-B-cell lymphoma (DLBCL). One patient had extranodal NK/T-cell lymphoma [7], an one patient, a heart transplant recipient, developed a rapidly growing Epstein–Barr-viru (EBV)-positive diffuse large-B-cell lymphoma 7 days after receiving the ﬁrst dose of th ChAdOx1 nCOV-19 vaccine [8]. One patient had a subcutaneous panniculitis-like T-ce lymphoma (SPTCL) [9], one had an extranodal marginal zone lymphoma (EMZL) [10] an one primary cutaneous anaplastic large-cell lymphoma (PC-ALCL) [11]. The lymphopro liferative disorder developed shortly after the vaccination, within 1 to 10 days followin vaccination (Table 1). Four of these eight patients were treated with systemic treatmen one was treated with radical surgery and radiation therapy, one was treated with systemi treatment and radiation therapy, one was treated through the watchful waiting program for comorbidities and the last one showed spontaneous, almost total remission withou treatment after 6 weeks of follow-up. Table 1. Review of the literature: malignant lymphoma diagnosed in the context of the mRNA COVID-19 vaccination (Ad26: adenovirus type 26, ALCL: anaplastic large-cell lymphoma, DLBCL diffuse large-B-cell lymphoma, EBV: Epstein–Barr virus, EMZL: extranodal marginal zone lymphom PC-ALCL: primary cutaneous anaplastic large-cell lymphoma, SPTCL: subcutaneous panniculitis-lik T-cell lymphoma). * The two previous vaccination doses were BNT162b2. Case N. 5. Discussion Reactive post-vaccination lymphadenopathy has been described in relation to mul- tiple vaccine types [1,2] in most cases. Lymphadenopathies secondary to vaccination are ipsilateral axillary or cervical, corresponding to the site of vaccination. First unilateral axillary lymphadenopathy post-COVID-19 vaccination was reported based on imaging scans in January 2021 among women subjected to ultrasound examination of the breast [12]. Subsequently, several cases of benign unilateral axillary lymphadenopathy after COVID-19 vaccination were reported [13–15], with a transient character [12–15]. p It must be emphasized that unilateral axillary lymphadenopathy is common after both mRNA-1273 (Moderna) and BNT 162b2 (Pﬁzer) [3,4]. ( ) ( ) Mizutani M. and colleagues [6] reported two cases of diffuse large-B-cell lymphoma that developed shortly after BNT162b2 COVID-19 vaccination. The axillary lymphadenopathies developed 1 day after the ﬁrst dose of BNT162b2 vaccination in a 67-year-old Japanese man and 1 day after the second dose of BNT162b2 vaccination in an 80-year-old Japanese woman. The patients were treated with chemotherapy combined with rituximab, showing response to the treatment. More recently, two additional cases of malignant lymphoma were reported [7]. In one case, a 58-year-old female patient presented with a left cervical mass of 4 cm in diameter one week after the second dose of BNT162b2. A histopatholog- ical examination revealed DLBCL, and the patient was treated with radical surgery and radiation therapy, with response to the treatment. In the second case, a 53-year-old male developed a painless ulcer on the mucosal surface of the upper lip three days after the ﬁrst dose of the BNT162b2 COVID-19 vaccine. Subsequently, the ulcer worsened, and a biopsy revealed extranodal NK/T-cell lymphoma that was treated with chemotherapy and radiotherapy, with response to the treatment. A 51-year-old man who previously received heart transplantation developed a mediastinal mass approximately 52 × 50 × 42 mm one week after the ﬁrst dose of the ChAdOx1 nCOV-19 vaccine (Astrazeneca), administrated to the right deltoid muscle. The pathologic examination revealed EBV-positive DLBCL. The patient was in treatment with an immunosuppressant regimen for transplantation, and the ﬁnal diagnosis was consistent with post-transplant lymphoproliferative disorder (PTLD). He underwent debulking surgery combined with rituximab and a 50% reduction in the immunosuppressant dosage and showed response to the treatment [8]. 4. Results of the Literature Revision Gender/Age (Year) (Reference) Time from Vaccination to Onset of Lymphoproliferative Disorder Histopathological Examination Type of COVID-19 Vaccine Site and Diameter of Lymphadenopathy Treatment of Lymphoma 1 M/67 [6] 1 day after 1 dose DLBCL BNT162b2 Left axilla 6.0 cm Chemotherapy plus rituximab 2 F/80 [6] 2 days after 1 dose DLBCL BNT162b2 Left axilla 4.1 cm Chemotherapy plus rituximab 3 F/58 [7] 7 days after 2 dose DLBCL BNT162b2 Left cervical area 4 cm Radical surgery plus radiotherapy 4 M/53 [7] 3 days after 1 dose Extranodal NK/T-cell lymphoma BNT162b2 Erosive lesions upper lip up to 5 mm Chemotherapy plus radiotherapy 5 M/51 [8] 7 days after 1 dose EBV-positive DLBCL ChAdox1 nCOV-19 Mediastinal mass 5 cm Rituximab 6 F/28 [9] “A few days after 1 dose” SPTCL Ad26 viral-vector-based Injection site, upper arm Cyclosporine plus prednisone 7 F/80 [10] 1 day after 1 dose EMZL BNT162b2 Right temporal mass No treatment 8 M/76 [11] 10 days after the booster dose PC-ALCL mRNA-1273 * Right arm upper-external surface 6 cm No treatment Medicina 2023, 59, 157 6 of 8 5. Discussion A case of a very rare SPTCL lymphoma was recently described in a previously healthy 28-year-old woman, and it developed a few days after receiving the primary dose of the Ad26 viral-vector-based COVID-19 vaccine (Janssen Pharmaceuticals). The patient was treated with cyclosporine combined with prednisone, with complete remission [9]. An 80-year-old Japanese woman developed a right temporal mass that suddenly appeared on the morning after she received her ﬁrst COVID-19 vaccine (BNT162b2) in her left deltoid muscle. Since the temporal mass increased to 68.3 × 17.1 mm 16 weeks after the ﬁrst vaccination, a biopsy was performed, and a diagnosis of EMZL was made. Chemotherapy was suggested to the patient; however, she preferred careful monitoring [10]. A 76-year-old man presented with a large erythe- matous rounded tumor of 6 cm in diameter on his right arm ten days after the application of the booster dose of the mRNA-1273 SARS-CoV2 vaccine, with a diagnosis of PC-ALCL. After 6 weeks of follow-up, the lesion showed spontaneous, almost total remission without treatment [11]. The majority of the cases described above and the case of our patient reported here showed similar features, including the short time between vaccination and the development of lymphoma, ranging from 1 to 10 days; the pathologic feature, being B-cell lymphoma in 5/8 cases; and the large dimensions of the lymphadenopathy, ranging from 4 to 6 cm. To the best of our knowledge, this is the ﬁrst review of malignant lymphoma diagnosed shortly after COVID-19 vaccination. Previous reports showed a progression or pseudo- progression of the disease in patients with a pre-existing lymphoma [16–20]. On the other hand, it must be emphasized that spontaneous tumor regression after COVID-19 vaccination was reported in a patient with metastatic myoepithelial carcinoma of the left parotid [21] and in a patient with cutaneous anaplastic large-cell lymphoma [22]. It is Medicina 2023, 59, 157 7 of 8 most likely that lymphoma is a chance phenomenon in these cases and that there is an absence of any evidence to support the notion that COVID-19 vaccination is linked in any way to an increased risk of lymphoma incidence. This study is a case report and is not designed to address the question of whether COVID-19 vaccination is linked to the risk of lymphoma. However, as reported by Zamﬁr MA. et al. [7] and Goldman S. et al. 5. Discussion [16], COVID-19 vaccines showed a potent stimulation of the immune cells, especially the T follicular helper (TFH) cells, resulting in persistent germinal center B-cell responses. For these reasons, our case report and the literature review suggest that the appearance of post-vaccine lymphadenopathy should be followed as recommended [4,12,13], and in cases of suspicion of lymphoproliferative disorder, a histopathological examination should be carried out to obtain a deﬁnite diagnosis. In fact, the clever physician should not take any lymphadenopathy lightly and perhaps keep in mind the possibility of a more sinister process, such as lymphoma. Patients should thus always be followed-up to ensure that any suspected reactive lymphadenopathy resolves over time. Otherwise, it is necessary to consider further work-up, including an excision biopsy. In conclusion, according to the scientiﬁc knowledge, we are aware that the COVID-19 vaccines represent very efﬁcient products with a favorable beneﬁt–risk ratio. However, we believe that it is mandatory to communicate all events to the scientiﬁc community, even those only temporally linked with novel treatments or vaccines such as the COVID-19 vaccines. Author Contributions: Conceptualization, L.C. and S.O.G., investigation, L.R., E.M., E.C., D.P., A.Z., N.I. and P.B.; writing—original draft preparation, L.C., C.D.N., C.C. and N.I.; writing—review and editing, L.C, S.O.G., L.R., E.M., E.C., D.P., A.Z., N.I., P.B., C.D.N. and C.C. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Funding: This research received no external funding. Institutional Review Board Statement: Ethical review and approval were waived for this study due to nature of the study (case report and review). Institutional Review Board Statement: Ethical review and approval were waived for this study due to nature of the study (case report and review). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Conﬂicts of Interest: The authors declare no conﬂict of interest. Conﬂicts of Interest: The authors declare no conﬂict of interest. 8. Tang, W.R.; Hsu, C.W.; Lee, C.C.; Huang, W.L.; Lin, C.Y.; Hsu, Y.T.; Chang, C.; Tsai, M.T.; Hu, Y.N.; Hsu, C.H.; et al. A Case Report of Posttransplant Lymphoproliferative Disorder after AstraZeneca Coronavirus Disease 2019 Vaccine in a Heart Transplant Recipient. Transplant. Proc. 2022, 54, 1575–1578. [CrossRef] References 1. Hartsock, R.J. Postvaccinial lymphadenitis. Hyperplasia of lymphoid tissue that simulates malignant lymphomas. Cancer 1968, 21, 632–649. [CrossRef] [PubMed] 21, 632–649. [CrossRef] [PubMed] 2. Studdiford, J.; Lamb, K.; Horvath, K.; Altshuler, M.; Stonehouse, A. Development of unilateral c lymphadenopathy after human papillomavirus vaccination. Pharmacotherapy 2008, 28, 1194–1197. [ 2. Studdiford, J.; Lamb, K.; Horvath, K.; Altshuler, M.; Stonehouse, A. Development of unilateral cervical and supraclavicular lymphadenopathy after human papillomavirus vaccination. Pharmacotherapy 2008, 28, 1194–1197. [CrossRef] [PubMed] C f C l d l S Ad d S Ad 2. Studdiford, J.; Lamb, K.; Horvath, K.; Altshuler, M.; Stonehouse, A. Development of unilateral cervical and supraclavicular lymphadenopathy after human papillomavirus vaccination. Pharmacotherapy 2008, 28, 1194–1197. [CrossRef] [PubMed] 3. Centers for Disease Control and Prevention. Local Reactions, Systemic Reactions, Adverse Events, and Serious Adverse Events: 3. Centers for Disease Control and Prevention. Local Reactions, Systemic Reactions, Adverse Events, and Serious Adverse Events: Pﬁzer-BioNTech COVID-19 Vaccine. Available online: https://www.cdc.gov/vaccines/covid-19/info-by-product/pﬁzer/ reactogenicity.html (accessed on 10 December 2022). y Pﬁzer-BioNTech COVID-19 Vaccine. Available online: https://www.cdc.gov/vaccines/covid-19/info-by-product/pﬁzer/ reactogenicity.html (accessed on 10 December 2022). g y 4. Becker, A.S.; Perez-Johnston, R.; Chikarmane, S.A.; Chen, M.M.; El Homsi, M.; Feigin, K.N.; Gallagher, K.M.; Hanna, E.Y.; Hicks, M.; Ilica, A.T.; et al. Multidisciplinary recommendations regarding post-vaccine adenopathy and radiologic imaging: Radiology scientiﬁc expert panel. Radiology 2021, 300, E323–E327. [CrossRef] [PubMed] g y 4. Becker, A.S.; Perez-Johnston, R.; Chikarmane, S.A.; Chen, M.M.; El Homsi, M.; Feigin, K.N.; Gallagher, K.M.; Hanna, E.Y.; Hicks, M.; Ilica, A.T.; et al. Multidisciplinary recommendations regarding post-vaccine adenopathy and radiologic imaging: Radiology scientiﬁc expert panel. Radiology 2021, 300, E323–E327. [CrossRef] [PubMed] gy p p gy 5. Garcia-Molina, F.; Cegarra-Navarro, M.F.; Javier Andrade-Gonzales, R.; Martinez-Diaz, F. Cytologic and histologic features of COVID-19 post-vaccination lymphadenopathy. Cytojournal 2021, 18, 34. [CrossRef] [PubMed] gy p p gy 5. Garcia-Molina, F.; Cegarra-Navarro, M.F.; Javier Andrade-Gonzales, R.; Martinez-Diaz, F. Cytologic and histologic features of COVID-19 post-vaccination lymphadenopathy. Cytojournal 2021, 18, 34. [CrossRef] [PubMed] p y p p y y j 6. Mizutani, M.; Mitsui, H.; Amano, T.; Ogawa, Y.; Deguchi, N.; Shimada, S.; Miwa, A.; Kawamura, T.; Ogido, Y. Two cases of axillary lymphadenopathy diagnosed as diffuse large B-cell lymphoma developed shortly after BNT162b2 COVID-19 vaccination. J. Eur. Acad. Dermatol. Venereol. 2022, 36, e613–e615. [CrossRef] p y p p y y j 6. Mizutani, M.; Mitsui, H.; Amano, T.; Ogawa, Y.; Deguchi, N.; Shimada, S.; Miwa, A.; Kawamura, T.; Ogido, Y. References Two cases of axillary lymphadenopathy diagnosed as diffuse large B-cell lymphoma developed shortly after BNT162b2 COVID-19 vaccination. J. Eur. Acad. Dermatol. Venereol. 2022, 36, e613–e615. [CrossRef] 7. Zamﬁr, M.A.; Moraru, L.; Dobrea, C.; Scheau, A.E.; Iacob, S.; Moldovan, C.; Cristian Scheau, C.; Caruntu, C.; Caruntu, A. Hematologic Malignancies Diagnosed in the Context of the mRNA COVID-19 Vaccination Campaign: A Report of Two Cases. Medicina 2022, 58, 874. [CrossRef] 8. Tang, W.R.; Hsu, C.W.; Lee, C.C.; Huang, W.L.; Lin, C.Y.; Hsu, Y.T.; Chang, C.; Tsai, M.T.; Hu, Y.N.; Hsu, C.H.; et al. A Case Report of Posttransplant Lymphoproliferative Disorder after AstraZeneca Coronavirus Disease 2019 Vaccine in a Heart Transplant Recipient. Transplant. Proc. 2022, 54, 1575–1578. [CrossRef] 8 of 8 Medicina 2023, 59, 157 9. Kreher, M.A.; Ahn, J.; Werbel, T.; Motaparthi, K. Subcutaneous panniculitis-like T-cell lymphoma after COVID-19 vaccination. JAAD Case Rep. 2022, 28, 18–20. [CrossRef] p 10. Sekizawa, A.; Hashimoto, K.; Kobayashi, S.; Kozono, S.; Kobayashi, T.; Kawamura, Y.; Kimata, M.; Fujita, N.; Ono, Y.; Obuchi, Y.; et al. Rapid progression of marginal zone B-cell lymphoma after COVID-19 vaccination (BNT162b2): A case re- port. Front. Med. 2022, 9, 963393. [CrossRef] p [ ] 11. Revenga-Porcel, L.; Peñate, Y.; Granados-Pacheco, F. Anaplastic large cell lymphoma at the SARS-CoV2 vaccine injection site. J. Eur. Acad. Dermatol. Venereol. 2023, 37, e32–e34. [CrossRef] 12. Mehta, N.; Sales, R.M.; Babagbemi, K.; Levy, A.D.; McGrath, A.L.; Drotman, M.; Dodelzon, K. Unilateral axillary lymphadenopathy in the setting of COVID-19 vaccine. Clin. Imaging 2021, 75, 12–15. [CrossRef] 13. Lane, D.L.; Neelapu, S.S.; Xu, G.; Weaver, O. COVID-19 Vaccine-Related Axillary and Cervical Lymphadenopathy in Patients with Current or Prior Breast Cancer and Other Malignancies: Cross-Sectional Imaging Findings on MRI, CT, and PET-CT. Korean J. Radiol. 2021, 22, 1938–1945. [CrossRef] [PubMed] J [ ] [ ] 14. Washington, T.; Bryan, R.; Clemow, C. Adenopathy following COVID19 vaccination. Radiology 2021, 299, E280–E281. [CrossRef] 14. Washington, T.; Bryan, R.; Clemow, C. Adenopathy following COVID19 vaccination. Radiology 2021, 299, E280–E281. [CrossRef] 15. Lehman, C.D.; D’Alessandro, H.A.; Mendoza, D.P.; Succi, M.D.; Kambadakone, A.; Lamb, L.R. Unilateral Lymphadenopathy after COVID-19 Vaccination: A Practical Management Plan for Radiologists Across Specialties. J. Am. Coll. Radiol. 2021, 18, 843–852. 14. Washington, T.; Bryan, R.; Clemow, C. Adenopathy following COVID19 vaccination. Radiology 2021, 299, E280–E281. [CrossRef] 15. Lehman, C.D.; D’Alessandro, H.A.; Mendoza, D.P.; Succi, M.D.; Kambadakone, A.; Lamb, L.R. References Unilateral Lymphadenopathy after COVID-19 Vaccination: A Practical Management Plan for Radiologists Across Specialties. J. Am. Coll. Radiol. 2021, 18, 843–852. [CrossRef] [PubMed] , ; , ; , ; , ; , ; , y COVID-19 Vaccination: A Practical Management Plan for Radiologists Across Specialties. J. Am. Coll. Rad [CrossRef] [PubMed] 16. Goldman, S.; Bron, D.; Tousseyn, T.; Vierasu, I.; Dewispelaere, L.; Heimann, P.; Cogan, E.; Goldman, M. Rapid Progression of Angioimmunoblastic T Cell Lymphoma Following BNT162b2 mRNA Vaccine Booster Shot: A Case Report. Front. Med. 2021, 8, 798095. [CrossRef] [PubMed] 17. Suleman, A.; Bilbily, A.; Cheung, M.; Chodirker, L. Hypermetabolic lymphadenopathy on positron emission tomography scan following COVID-19 vaccination: A mimicker of disease progression in Hodgkin lymphoma. eJHaem 2021, 2, 678–679. [CrossRef] 17. Suleman, A.; Bilbily, A.; Cheung, M.; Chodirker, L. Hypermetabolic lymphadenopathy on positron emission tomography scan following COVID-19 vaccination: A mimicker of disease progression in Hodgkin lymphoma. eJHaem 2021, 2, 678–679. [CrossRef] 18. Arens, A.I.J.; Hebeda, K.M.; Hutchings, M.; Plattel, W.J.; Stevens, W.B.C. Axillary lymph nodes on PET in Hodgkin lymphoma after COVID-19 vaccination. eJHaem 2021, 2, 885–886. [CrossRef] 18. Arens, A.I.J.; Hebeda, K.M.; Hutchings, M.; Plattel, W.J.; Stevens, W.B.C. Axillary lymph nodes on PET in Hodgkin lymphoma after COVID-19 vaccination. eJHaem 2021, 2, 885–886. [CrossRef] 19. Xu, G.; Lu, Y. COVID-19 mRNA Vaccination-Induced Lymphadenopathy Mimics Lymphoma Progression on FDG PET/CT. Clin. Nucl. Med. 2021, 46, 353–354. [CrossRef] 20. Landete, E.; Gómez-Fernández, I.; González-Gascón-y-Marín, I.; Durán-Barquero, C.; Churruca, J.; Infante, M.S.; Muñoz-Novas, C.; Foncillas, M.Á.; Marín, K.; Ramos-de-Ascanio, V.; et al. Hypermetabolic abdominal and cervical lymph nodes mimicking Hodgkin lymphoma relapse on FDG PET/CT after adenovirus-vectored COVID-19 vaccine. Hum. Vaccines Immunother. 2021, 17, 5129–5132. [CrossRef] 21. Sousa, L.G.; McGrail, D.J.; Li, K.; Marques-Piubelli, M.L.; Gonzalez, C.; Dai, H.; Ferri-Borgogno, S.; Godoy, M.; Burks, J.; Lin, S.Y.; et al. Spontaneous tumor regression following COVID-19 vaccination. J. Immunother. Cancer 2022, 10, e004371. [CrossRef] 21. Sousa, L.G.; McGrail, D.J.; Li, K.; Marques-Piubelli, M.L.; Gonzalez, C.; Dai, H.; Ferri-Borgogno, S.; Godoy, M.; Burks, J.; Lin, S.Y.; et al. Spontaneous tumor regression following COVID-19 vaccination. J. Immunother. Cancer 2022, 10, e004371. [CrossRef] 22. Gambichler, T.; Boms, S.; Hessam, S.; Tischoff, I.; Tannapfel, A.; Lüttringhaus, T.; Beckman, J.; Stranzenbach, R. Primary cutaneous anaplastic large-cell lymphoma with marked spontaneous regression of organ manifestation after SARS-CoV-2 vaccination. Br. J. Dermatol. 2021, 185, 1259–1262. [CrossRef] 22. References Gambichler, T.; Boms, S.; Hessam, S.; Tischoff, I.; Tannapfel, A.; Lüttringhaus, T.; Beckman, J.; Stranzenbach, R. Primary cutaneous anaplastic large-cell lymphoma with marked spontaneous regression of organ manifestation after SARS-CoV-2 vaccination. Br. J. Dermatol. 2021, 185, 1259–1262. [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/W1581872043	http://juser.fz-juelich.de/record/186493/files/FZJ-2015-00567.pdf	English	null	Production of 2‐ketoisocaproate with <scp><i>C</i></scp><i>orynebacterium glutamicum</i> strains devoid of plasmids and heterologous genes	Microbial Biotechnology	2,014	cc-by	7,409	Received 14 July, 2014; revised 7 October, 2014; accepted 7 October, 2014. For correspondence. E-mail j.van.ooyen@fz-juelich .de; Tel. +49 2461 616205; Fax +49 2461 612710. E-mail m.bott @fz-juelich.de; Tel. +49 2461 613294; Fax +49 2461 612710. Michael Vogt and Sabine Haas contributed equally to this work. doi:10.1111/1751-7915.12237 Funding Information No funding information provided. Introduction Michael Vogt, Sabine Haas, Tino Polen, Jan van Ooyen and Michael Bott** Institute of Bio- and Geosciences, IBG-1: Biotechnology, Forschungszentrum Jülich, D-52425 Jülich, Germany. Corynebacterium glutamicum is the major host for biotechnological production of amino acids, the most important ones being the ﬂavor enhancer L-glutamate and the feed additive L-lysine. In the past decades, C. glutamicum strains have been developed for the pro- duction of various other commercially interesting com- pounds (Becker and Wittmann, 2012), including organic acids (Okino et al., 2008; Litsanov et al., 2012a,b; Wieschalka et al., 2013), diamines (Mimitsuka et al., 2007; Kind and Wittmann, 2011; Schneider and Wendisch, 2011) or alcohols (Inui et al., 2004; Smith et al., 2010; Blombach et al., 2011). Besides small mol- ecules, also heterologous proteins can be efficiently pro- duced with this Gram-positive bacterium (Scheele et al., 2013, and references therein). Thus, C. glutamicum has become a production platform in white biotechnology. Three monographs (Eggeling and Bott, 2005; Burkovski, 2008; Yukawa and Inui, 2013) document the rapidly increasing knowledge on this species, which is based on the genome sequence (Ikeda and Nakagawa, 2003; Kalinowski et al., 2003) and efficient techniques for its genetic engineering (Kirchner and Tauch, 2003). © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Summary 2-Ketoisocaproate (KIC), the last intermediate in L-leucine biosynthesis, has various medical and industrial applications. After deletion of the ilvE gene for transaminase B in L-leucine production strains of Corynebacterium glutamicum, KIC became the major product, however, the strains were auxotrophic for L-isoleucine. To avoid auxotrophy, reduction of IlvE activity by exchanging the ATG start codon of ilvE by GTG was tested instead of an ilvE deletion. The result- ing strains were indeed able to grow in glucose minimal medium without amino acid supplementa- tion, but at the cost of lowered growth rates and KIC production parameters. The best production perfor- mance was obtained with strain MV-KICF1, which carried besides the ilvE start codon exchange three copies of a gene for a feedback-resistant 2- isopropylmalate synthase, one copy of a gene for a feedback-resistant acetohydroxyacid synthase and deletions of ltbR and iolR encoding transcriptional regulators. In the presence of 1 mM L-isoleucine, MV-KICF1 accumulated 47 mM KIC (6.1 g l−1) with a yield of 0.20 mol/mol glucose and a volumetric prod- uctivity of 1.41 mmol KIC l−1 h−1. Since MV-KICF1 is plasmid free and lacks heterologous genes, it is an interesting strain for industrial application and as platform for the production of KIC-derived com- pounds, such as 3-methyl-1-butanol. The spectrum of amino acids produced with C. glutamicum includes the essential branched-chain amino acids (BCAAs) L-valine, L-isoleucine and L-leucine, which are produced in quantities of up to 5000 tons per year in a steadily growing market (Becker and Wittmann, 2012). They have different applications in the food, feed and pharmaceutical industry (Park and Lee, 2010). The biosynthesis pathways of the BCAAs in C. glutamicum are overlapping and partly share the same precursors and enzymes (Fig. 1). The direct precursors of L-valine, L-isoleucine and L-leucine are 2-ketoisovalerate (KIV), 2-keto-3-methylvalerate (KMV), and 2-ketoisocaproate (KIC) respectively. These keto acids are predominantly transaminated to the respective amino acids by the transaminase IlvE (Radmacher et al., 2002; Marienhagen et al., 2005). Similar to their corresponding amino acids, KIV, KMV and KIC have a variety of applications in the medical, biological and food area, since they play an 2 M. Vogt et al. Fig. 1. Biosynthesis pathways and their control by various regulatory mechanisms of the three branched-chain amino acids and keto acids in C. glutamicum. Enzymes and their corresponding genes are shown in boxes. e Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology Summary Lines with ‘+’ indicate activation of g sion; ‘-’ indicates repression of gene expression (solid lines) or transcription attenuation or feedback inhibition (dashed lines). ‘L ‘Ile’ indicate the presence of L-leucine, L-valine and L-isoleucine respectively. Not shown is the avtA gene encoding the branche acid transaminase AvtA, which predominantly transaminates 2-ketoisovalerate to L-valine. Abbreviations: AHAIR, acetohydroxya isomeroreductase; AHAS, acetohydroxyacid synthase; BCAA-E, branched-chain amino acid exporter (BrnFE); BCAA-T, branch acid transaminase IlvE; DHAD, dihydroxyacid dehydratase; IPMD, 3-isopropylmalate dehydratase; IPMDH, 3-isopropylmalate d IPMS, 2-isopropylmalate synthase; Lrp, leucine-responsive regulatory protein; LtbR, leucine and tryptophane biosynthesis regul threonine dehydratase (threonine ammonia-lyase). 2 M. Vogt et al. Fig. 1. Biosynthesis pathways and their control by various regulatory mechanisms of the three branched-chain amino acids and the respective keto acids in C. glutamicum. Enzymes and their corresponding genes are shown in boxes. Lines with ‘+’ indicate activation of gene expres- sion; ‘-’ indicates repression of gene expression (solid lines) or transcription attenuation or feedback inhibition (dashed lines). ‘Leu’, ‘Val’ and ‘Ile’ indicate the presence of L-leucine, L-valine and L-isoleucine respectively. Not shown is the avtA gene encoding the branched-chain amino acid transaminase AvtA, which predominantly transaminates 2-ketoisovalerate to L-valine. Abbreviations: AHAIR, acetohydroxyacid isomeroreductase; AHAS, acetohydroxyacid synthase; BCAA-E, branched-chain amino acid exporter (BrnFE); BCAA-T, branched-chain amino acid transaminase IlvE; DHAD, dihydroxyacid dehydratase; IPMD, 3-isopropylmalate dehydratase; IPMDH, 3-isopropylmalate dehydrogenase; IPMS, 2-isopropylmalate synthase; Lrp, leucine-responsive regulatory protein; LtbR, leucine and tryptophane biosynthesis regulator; TD, threonine dehydratase (threonine ammonia-lyase). Fig. 1. Biosynthesis pathways and their control by various regulatory mechanisms of the three branched-chain amino acids and the respective keto acids in C. glutamicum. Enzymes and their corresponding genes are shown in boxes. Lines with ‘+’ indicate activation of gene expres- sion; ‘-’ indicates repression of gene expression (solid lines) or transcription attenuation or feedback inhibition (dashed lines). ‘Leu’, ‘Val’ and ‘Ile’ indicate the presence of L-leucine, L-valine and L-isoleucine respectively. Not shown is the avtA gene encoding the branched-chain amino acid transaminase AvtA, which predominantly transaminates 2-ketoisovalerate to L-valine. Abbreviations: AHAIR, acetohydroxyacid isomeroreductase; AHAS, acetohydroxyacid synthase; BCAA-E, branched-chain amino acid exporter (BrnFE); BCAA-T, branched-chain amino acid transaminase IlvE; DHAD, dihydroxyacid dehydratase; IPMD, 3-isopropylmalate dehydratase; IPMDH, 3-isopropylmalate dehydrogenase; IPMS, 2-isopropylmalate synthase; Lrp, leucine-responsive regulatory protein; LtbR, leucine and tryptophane biosynthesis regulator; TD, threonine dehydratase (threonine ammonia-lyase). Fig. 1. Biosynthesis pathways and their control by various regulatory mechanisms of the three branched-chain amino acids and the respective keto acids in C. Summary glutamicum. Enzymes and their corresponding genes are shown in boxes. Lines with ‘+’ indicate activation of gene expres- sion; ‘-’ indicates repression of gene expression (solid lines) or transcription attenuation or feedback inhibition (dashed lines). ‘Leu’, ‘Val’ and ‘Ile’ indicate the presence of L-leucine, L-valine and L-isoleucine respectively. Not shown is the avtA gene encoding the branched-chain amino acid transaminase AvtA, which predominantly transaminates 2-ketoisovalerate to L-valine. Abbreviations: AHAIR, acetohydroxyacid isomeroreductase; AHAS, acetohydroxyacid synthase; BCAA-E, branched-chain amino acid exporter (BrnFE); BCAA-T, branched-chain amino acid transaminase IlvE; DHAD, dihydroxyacid dehydratase; IPMD, 3-isopropylmalate dehydratase; IPMDH, 3-isopropylmalate dehydrogenase; IPMS, 2-isopropylmalate synthase; Lrp, leucine-responsive regulatory protein; LtbR, leucine and tryptophane biosynthesis regulator; TD, threonine dehydratase (threonine ammonia-lyase). enhancement of protein synthesis, especially in the skel- etal muscle (Escobar et al., 2010; Zanchi et al., 2011). Additionally, an insulin-releasing action of KIC (Heissig et al., 2005) and an inhibitory effect on glucagon release (Leclercq-Meyer et al., 1979) were discussed. It has been shown that KIC can also serve as a basis for the produc- tion of the biofuel isopentanol (Cann and Liao, 2010). important role in living organisms as regulatory factors in metabolism and key intermediates in biosynthesis (Krause et al., 2010; Zhu et al., 2011; Bückle-Vallant et al., 2014). They are used, for example, in the therapy of chronic kidney disease patients (Aparicio et al., 2012). Similar to L-leucine, KIC has anti-catabolic properties through inhibition of muscle proteolysis and provokes 2-Ketoisocaproate production with C. glutamicum 3 KIV, KMV and KIC are mainly produced by chemical synthesis using harsh reaction conditions and multiple puriﬁcation steps resulting in plenty of waste (Cooper et al., 1983). The biotechnological production of these keto acids is thus an interesting alternative. Besides a biotransformation process with Rhodococcus opacus using L-leucine as substrate for KIC formation (Zhu et al., 2011), fermentative processes with glucose as substrate have recently been described for the production of KIV (Krause et al., 2010) and KIC (Bückle-Vallant et al., 2014), showing that deletion of ilvE in certain engineered C. glutamicum strains results in KIC formation. Whereas these strains contained plasmids and in part heterologous genes, the C. glutamicum KIC production strains devel- oped in our work are plasmid free and lack heterologous genes. step homologous recombination using the vector pK19mobsacB (Schäfer et al., 1994) and a method described previously (Niebisch and Bott, 2001). Initial studies on KIC production using plasmid-containing strains of C. glutamicum Based on recently developed efficient production strains of C. glutamicum ATCC 13032 (Abe et al., 1967) for L-leucine (Vogt et al., 2014), we intended to modify these strains for the production of KIC. The conversion of KIC to L-leucine is catalysed by the transaminase IlvE, which also converts KIV to L-valine and KMV to L-isoleucine using L-glutamate as amino donor (Radmacher et al., 2002; Marienhagen et al., 2005). An ilvE deletion has been reported to cause auxotrophy for L-leucine and L-isoleucine, but not for L-valine, since the transaminase AvtA also effectively converts KIV to L-valine using L-alanine as amino donor (Marienhagen et al., 2005). According to this knowledge, deletion of ilvE in L-leucine production strains should lead to the accumulation of KIC and potentially also KMV. In a ﬁrst series of experiments, we deleted ilvE in the wild-type C. glutamicum ATCC 13032 and transformed the ΔilvE mutant with plasmid pAN6-leuA_B018, carrying an IPTG- inducible leuA allele encoding a feedback-resistant 2-isopropylmalate synthase (IPMS) (Vogt et al., 2014). 2-Isopropylmalate synthase of C. glutamicum is strongly inhibited by L-leucine with a Ki of 0.4 mM (Pátek et al., 1994) and the presence of a feedback-resistant variant is the key for L-leucine overproduction (Vogt et al., 2014). The ΔilvE mutant and the ΔilvE strain with plasmid pAN6- leuA_B018 were cultivated in 500 ml baffled Erlenmeyer ﬂasks with 50 ml CGXII minimal medium (Keilhauer et al., 1993) with 4% (w/v) glucose, 1 mM L-leucine and 1 mM L-isoleucine at 30°C and 120 rpm on a rotary shaker. Keto acids and amino acids were quantiﬁed by high- performance liquid chromatography as described (Vogt et al., 2014). Chromosomal in-frame deletions and integrations of DNA fragments were performed by two- Summary Corynebacterium glutamicum ΔilvE exhibited a growth rate of 0.38 ± 0.01 h−1 and excreted up to 5 mM KIV but no detectable concentrations of KIC (detection limit < 0.1 mM), whereas C. glutamicum ΔilvE pAN6- leuA_B018 showed a growth rate of 0.30 ± 0.01 h−1 and accumulated 37 ± 0.7 mM KIC in the supernatant when induced with 0.1 mM IPTG, conﬁrming that over- expression of the leuA allele encoding the feedback- resistant IPMS increased metabolic ﬂux into the leucine pathway (Fig. 1). Surprisingly, C. glutamicum ΔilvE carry- ing pAN6-leuA_B018 also accumulated L-leucine (12.3 ± 0.4 mM) and in fact was only auxotrophic for L-isoleucine, but not for L-leucine. A possible limitation of L-valine due to a high metabolic ﬂux from KIV towards KIC was excluded for this strain since additional supplementation of L-valine did not improve growth (data not shown). Accumulation of L-leucine was also reported for other KIC-producing ΔilvE strains and explained by the activity of unspeciﬁc transaminases (e.g. AlaT or AvtA) using KIC as substrate when it is present in high concentrations (Bückle-Vallant et al., 2014). Consequently, supplementation of the medium with L-leucine was omitted in the following culti- vations. The formation of L-leucine as by-product addi- tionally necessitates the presence of feedback-resistant IPMS for KIC overproduction. The results described above demonstrated that our previously described L-leucine producers (Vogt et al., 2014) can serve as basis for the construction of KIC production strains. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology Results and discussion Initial studies on KIC production using plasmid-containing strains of C. glutamicum Deletion of ilvE in plasmid-free L-leucine production strains coli strains DH5α F−Φ80lacZΔM15 Δ(lacZYA-argF)U169 recA1 endA1 hsdR17 (rK−, mK+) phoA supE44 λ– thi-1 gyrA96 relA1 Invitrogen (Karlsruhe, Germany) Plasmids pAN6 Kanr; E. coli/C. glutamicum shuttle vector for inducible gene expression (Ptac, lacI q, pBL1 pUC18 oriVE.coli, pBL1 oriVC.glutamicum) Frunzke and colleagues (2008) pAN6-leuA_B018 Kanr; pAN6 derivative containing leuA allele coding for feedback-resistant 2-isopropylmalate synthase under control of the tac promoter Vogt and colleagues (2014) pAN6-leuA_B018-cg1121 Kanr; pAN6-leuA_B018 derivative carrying additional gene coding for cg1121 along with its upstream (94 bp) and downstream (305 bp) regions This study pK19mobsacB Kanr; vector for allelic exchange in C. glutamicum (pK18 oriVE.coli sacB lacZα) Schäfer and colleagues (1994) pK19mobsacB-ΔilvE Kanr, pK19mobsacB derivative for in-frame deletion of gene ilvE Marienhagen and colleagues (2005) pK19mobsacB-GTG-ilvE Kanr, pK19mobsacB derivative for replacement of ATG start codon of ilvE by GTG This study pK19mobsacB-Δcg0018 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg0018 This study pK19mobsacB-Δcg1121 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1121 This study pK19mobsacB-Δcg1219 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1219 This study pK19mobsacB-Δcg1419 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1419 This study pK19mobsacB-Δcg1658 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1658 This study pK19mobsacB-Δcg2557 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg2557 This study pK19mobsacB-Δcg2676 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg2676 This study pK19mobsacB-Δcg3334 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg3334 This study pK19mobsacB-cg1121 Kanr, pK19mobsacB derivative for re-integration of gene coding for Cg1121 into its wild-type locus This study a. All constructed plasmids as well as chromosomal deletions and integrations in engineered strains were veriﬁed by DNA sequencing. repressor of the L-leucine biosynthesis genes, and a replacement of the wild-type leuA gene by the feedback- resistant variant leuA_B018 under control of the strong tuf promoter (Vogt et al., 2014). In shake ﬂask cultivations with CGXII medium containing 4% (w/v) glucose, MV-Leu20 accumulated about 20 mM L-leucine. When cultivated in the same medium supplemented with 1 mM L-isoleucine, the strain MV-Leu20 ΔilvE accumulated 18.0 ± 1.6 mM KIC in the supernatant and formed as by-products 5.6 ± 0.3 mM L-leucine, 2.1 ± 0.5 mM KIV and 7.3 ± 1.7 mM KMV. Deletion of ilvE in plasmid-free L-leucine production strains Analogous to our strategy used for L-leucine strain devel- opment (Vogt et al., 2014), we intended to construct KIC production strains devoid of plasmids, heterologous genes and auxotrophies. Depending on the composition of the medium used in the fermentation process, auxotrophies can necessitate the addition of supplements, increasing the costs of the fermentation process. Plasmids usually necessitate the addition of antibiotics to the medium, which is undesirable for production strains applied in the food and feed industry and can be prohibited by regulatory author- ities (Tauch et al., 2002). Moreover, the absence of plasmids, antibiotic resistance markers and heterologous genes often results in more stable producer strains (Pátek, 2007). The use of heterologous genes is also an undesired trait for strains used in the food and feed industry. In a ﬁrst attempt to construct a plasmid-free KIC producer, we deleted the ilvE gene in the previously constructed L-leucine producer MV-Leu20 (Table 1; Vogt et al., 2014), which contains a deletion of the ltbR gene, encoding a 4 M. Vogt et al. Table 1. Strains and plasmids used in this studya,b. Table 1. Strains and plasmids used in this studya,b. Strain or plasmid Relevant characteristicsc Source or reference C. glutamicum strains Wild type ATCC 13032, biotin-auxotrophic Abe and colleagues (1967) ΔilvE ATCC 13032 derivative with in-frame deletion of ilvE Marienhagen and colleagues (2005) MV-Leu20 Rationally designed C. glutamicum L-leucine producer (ΔltbR ΔleuA::Ptuf-leuA_B018) Vogt and colleagues (2014) MV-Leu20 ΔilvE MV-Leu20 derivative with in-frame deletion of ilvE This study SH-KIC20 MV-Leu20 derivative with chromosomal replacement of ATG start codon of ilvE by GTG start codon This study MV-LeuF1 Rationally designed C. glutamicum L-leucine producer (ΔltbR::Ptuf-leuA_B018 ΔleuA::Ptuf-leuA_B018 IR(cg1121/1122)::Ptuf-leuA_B018 ΔiolR ilvN_fbr) Vogt and colleagues (2014) MV-KICF1 MV-LeuF1 derivative with chromosomal replacement of ATG start codon of ilvE by GTG start codon This study ΔilvE Δcg0018 ΔilvE derivative with cg0018 in-frame deletion This study ΔilvE Δcg1121 ΔilvE derivative with cg1121 in-frame deletion This study ΔilvE Δcg1219 ΔilvE derivative with cg1219 in-frame deletion This study ΔilvE Δcg1419 ΔilvE derivative with cg1419 in-frame deletion This study ΔilvE Δcg1658 ΔilvE derivative with cg1658 in-frame deletion This study ΔilvE Δcg2557 ΔilvE derivative with cg2557 in-frame deletion This study ΔilvE Δcg2676 ΔilvE derivative with cg2676 in-frame deletion This study ΔilvE Δcg3334 ΔilvE derivative with cg3334 in-frame deletion This study ΔilvE Δcg1121::cg1121 ΔilvE Δcg1121 derivative with re-integrated gene cg1121 into its wild-type locus This study E. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology Exchange of the ilvE start codon in plasmid-free L-leucine production strains To avoid the L-isoleucine auxotrophy of strain MV-Leu20 ΔilvE, we intended to reduce IlvE activity to a value that was high enough to provide L-isoleucine for growth, but low enough to allow KIC overproduction. For this purpose, the ATG start codon of ilvE was exchanged against GTG, which should decrease the translation rate (Becker et al., 2010) of the ilvE transcript and thereby reduce the speciﬁc IlvE activity. The start codon exchange was performed in the L-leucine producers MV-Leu20 and MV-LeuF1 (Table 1). The latter strain contains (i) three copies of the leuA_B018 gene in the chromosome under control of the tuf promoter, two of them replacing ltbR and the native leuA gene, (ii) a deletion of iolR (Klaffl et al., 2013) for enhanced glucose uptake and (iii) a feedback-resistant acetohydroxyacid synthase encoded by ilvN_fbr (Vogt et al., 2014). The strains SH-KIC20 (from MV-Leu20) and MV-KICF1 (from MV-LeuF1) resulting from the ilvE start codon exchange (Table 1) were cultivated in CGXII medium with 4% (w/v) glucose to test for KIC accumula- tion in the supernatant (Fig. 2). Fig. 2. Growth and KIC formation of different C. glutamicum strains in shake ﬂasks with CGXII minimal medium containing 4% (w/v) glucose. Fig. 2. Growth and KIC formation of different C. glutamicum strains in shake ﬂasks with CGXII minimal medium containing 4% (w/v) glucose. Fig. 2. Growth and KIC formation of different C. glutamicum strains in shake ﬂasks with CGXII minimal medium containing 4% (w/v) glucose. A. MV-Leu20 without supplements (growth, ■; KIC, □), MV-Leu20 ΔilvE supplemented with 1 mM L-isoleucine (growth, ●; KIC, ○) and SH-KIC20 without supplements (growth, ▲; KIC, △). B. MV-LeuF1 without supplements (growth, ▼; KIC, ▽), MV-KICF1 without supplements (growth, ◀; KIC, ◁) and MV-KICF1 with 1 mM L-isoleucine (growth, ◆; KIC, ◇). The data represent mean values and standard deviations obtained from three independent cultivations. g A. MV-Leu20 without supplements (growth, ■; KIC, □), MV-Leu20 ΔilvE supplemented with 1 mM L-isoleucine (growth, ●; KIC, ○) and SH-KIC20 without supplements (growth, ▲; KIC, △). B. MV-LeuF1 without supplements (growth, ▼; KIC, ▽), MV-KICF1 without supplements (growth, ◀; KIC, ◁) and MV-KICF1 with 1 mM L-isoleucine (growth, ◆; KIC, ◇). The data represent mean values and standard deviations obtained from three independent cultivations. Without supplementation of L-isoleucine, SH-KIC20 showed a strongly reduced growth rate of 0.08 ± 0.01 h−1 compared with the ancestor strain MV-Leu20 (0.31 ± 0.01 h−1). Deletion of ilvE in plasmid-free L-leucine production strains Since KIV and KMV are substrates of the tran- saminase IlvE (Marienhagen et al., 2005), the ilvE deletion leads to an accumulation of these keto acids. The presum- ably low concentrations of L-isoleucine and L-valine in strain MV-Leu20 ΔilvE may also contribute to overproduc- tion of KIV and KMV by reducing the feedback-inhibition of threonine dehydratase (encoded by ilvA) by L-isoleucine (Möckel et al., 1992) and of acetohydroxyacid synthase (encoded by ilvBN) by L-valine and L-isoleucine (Eggeling et al., 1987). As mentioned above, L-leucine formation in the absence of the transaminase IlvE is presumably due to © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 2-Ketoisocaproate production with C. glutamicum 5 2-Ketoisocaproate production with C. glutamicum 5 5 Fig. 2. Growth and KIC formation of different C. glutamicum strains in shake ﬂasks with CGXII minimal medium containing 4% (w/v) glucose. A. MV-Leu20 without supplements (growth, ■; KIC, □), MV-Leu20 ΔilvE supplemented with 1 mM L-isoleucine (growth, ●; KIC, ○) and SH-KIC20 without supplements (growth, ▲; KIC, △). B. MV-LeuF1 without supplements (growth, ▼; KIC, ▽), MV-KICF1 without supplements (growth, ◀; KIC, ◁) and MV-KICF1 with 1 mM L-isoleucine (growth, ◆; KIC, ◇). The data represent mean values and standard deviations obtained from three independent cultivations. high cytoplasmic KIC concentrations, allowing its conver- sion by other transaminases such as AvtA that have weak affinities for KIC (Marienhagen et al., 2005). To test this assumption, we measured the cytoplasmic KIC concen- trations in the wild type and strain MV-Leu20 ΔilvE. Cells were grown in CGXII medium with 4% (w/v) glucose, harvested in the early exponential phase (optical density at 600 nm (OD600 = 5), and cytoplasmic concentrations were determined as described (Paczia et al., 2012). The internal KIC concentration of the wild type was below the detection limit of 5 μM, whereas MV-Leu20 ΔilvE accumu- lated approximately 2.8 mM KIC inside the cell, corre- sponding to a more than 500-fold increase. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology Exchange of the ilvE start codon in plasmid-free L-leucine production strains This phenotype suggests that the ilvE start codon exchange reduced the availability of BCAAs and consequently the growth rate. Strain SH-KIC20 accumu- lated about 19 mM KIC in 49 h, which correlates with the KIC concentration produced by MV-Leu20 ΔilvE. The slower growth of SH-KIC20 led to a lowered volumetric productivity of 0.38 mmol l−1 h−1 in comparison to MV-Leu20 ΔilvE (approximately 0.8 mmol l−1 h−1). Strain MV-KICF1 showed a growth rate of only 0.03 ± 0.01 h−1, which is 85% lower than the one of the parent strain (μ = 0.20 ± 0.01 h−1), and yielded a maximal KIC con- centration of 33 mM after 95 h (Table 2). Supplementa- tion of the medium with 1 mM L-isoleucine enabled MV-KICF1 to reach the same growth rate (μ = 0.21 ± 0.01 h−1) as its parent MV-LeuF1 and to form 47 ±q 4 mM KIC (6.1 g l−1) after 32 h with a yield of 0.20 ± 0.02 mol KIC per mol of glucose and a pro- ductivity of 1.41 ± 0.13 mmol KIC l−1 h−1 (Table 2). Both SH-KIC20 and MV-KICF1 formed KIV, KMV and L-leucine as by-products (Table 2). The results described above demonstrate that the start codon exchange for reduction of IlvE activity was suc- cessful and allowed growth and KIC accumulation without supplementation of BCAAs; however, the produc- tion parameters were lower compared with supplemen- tation with 1 mM L-isoleucine (Table 2). A successful industrial application depends on high product yields combined with sufficient cell growth, resulting in 6 M. Vogt et al. 6 Table 2. Growth and production parameters of strains MV-KICF1 and SH-KIC20 in shake ﬂask cultivationsa,b. Parameter MV-KICF1 + 1 mM L-isoleucine MV-KICF1 without L-isoleucine SH-KIC20d without L-isoleucine Growth rate (h−1) 0.21 ± 0.01 0.03 ± 0.01 0.08 ± 0.01 KIC (mM) 46.7 ± 4.1 31.8 ± 2.1 18.8 ± 0.67 By-productsc: KIV (mM) 13.3 ± 2.2 19.0 ± 4.1 2.6 ± 0.3 KMV (mM) 8.8 ± 1.3 4.9 ± 0.6 8.7 ± 0.1 L-leucine (mM) 3.0 ± 0.2 10.3 ± 3.1 4.8 ± 0.2 Molar product yield (mol KIC per mol glucose) 0.204 ± 0.018 0.143 ± 0.010 0.084 ± 0.001 Volumetric productivity (mmol KIC l−1 h−1) 1.41 ± 0.13 0.34 ± 0.02 0.38 ± 0.02 a. Cultivations were performed in 500 ml baffled shake ﬂasks containing 50 ml CGXII minimal medium with 4% (w/v) glucose. Supplementation of L-isoleucine is indicated. b. 2-Ketoisocaproate transport Recently, Bückle-Vallant and colleagues (2014) described a plasmid-based C. glutamicum strain for the production of KIC. This strain is characterized by deletions of the genes ilvE, ltbR, prpC1 and prpC2, an exchange of the two gltA promoters (van Ooyen et al., 2011) by the mutated dapA promoter L1 (Vašicová et al., 1999) to reduce citrate synthase activity (van Ooyen et al., 2012), and plasmid-based overexpression of ilvBNCD and a leuA allele of Escherichia coli encoding a feedback-resistant IPMS. This strain accumulated up to 71 mM KIC when cultivated with glucose plus acetate as carbon sources and supplemented with 2 mM each of L-isoleucine and L-valine. Under cultivation conditions comparable to ours, i.e. without acetate, this strain reached KIC titres When MV-KICF1 was batch cultivated in a bioreactor as described (Vogt et al., 2014) using a medium sup- plemented with 1 mM L-isoleucine, comparable KIC titres as in shake ﬂasks of about 50 mM were achieved (data not shown). Interestingly, additional feeding of glucose in fed-batch experiments did not further increase the KIC concentrations and led to an arrest of cell growth and glucose consumption (data not shown). A possible expla- nation is a block of the L-leucine biosynthesis pathway by elevated KIC concentrations, as Bückle-Vallant and colleagues (2014) found competitive inhibition of IPMS by KIC and non-competitive inhibition of 3-isopropylmalate dehydratase by KIC. Additionally, a competitive inhibition Table 3. Putative transporter genes showing increased expression in a KIC producer. Gene Annotation mRNA ratioa (MV-Leu20 ΔilvE/ Wild type TMHb cg0018 putative membrane protein, conserved 3.5 9 cg1121 putative permease of the major facilitator superfamily 2.2 7 cg1219 putative membrane protein 3.5 10 cg1419 putative Na+-dependent transporter, bile acid:Na+ symporter BASS family 7.2 8 cg1658 putative permease of the major facilitator superfamily 35.6 12 cg2557 putative secondary Na+/bile acid symporter, bile acid:Na+ symporter BASS family 2.5 8 cg2676 putative ABC-type dipeptide/oligopeptide/nickel transport system, permease component 2.1 6 cg3334 putative arabinose efflux permease, MFS type 2.0 12 a. Transcriptome analyses of KIC producer MV-Leu20 ΔilvE in comparison to the wild type were performed using DNA microarrays as described (Vogt et al., 2014). Candidate transporter genes were chosen based on an mRNA ratio (MV-Leu20 ΔilvE/wild type) of > 2, an annotation as (putative) membrane or transporter proteins and the prediction of multiple transmembrane helices in the encoded proteins. Data represent mean values of at least two (maximum four) evaluable microarray experiments (P-value < 0.05). Exchange of the ilvE start codon in plasmid-free L-leucine production strains Mean values and standard deviations from three independent cultivations are shown. c. Concentrations of L-valine and L-isoleucine were below 2 mM. d. Cultivation of SH-KIC20 supplemented with 1 mM L-isoleucine was not tested. Table 2. Growth and production parameters of strains MV-KICF1 and SH-KIC20 in shake ﬂask cultivationsa,b. (54 ± 4 mM) and yields (0.22 mol per mol of glucose) in a similar range as strain MV-KICF1 when supplemented with L-isoleucine. In comparison, the biotransformation with Rhodococcus opacus transcribed by Zhu and colleagues (2011) reached about 10 mM KIC using 39 mM L-leucine as substrate. competitive productivity. Therefore, the addition of L-isoleucine is still important to improve growth of the constructed strains to reach better productivity values. As an alternative approach to adjust the IlvE activity to an optimal value for prototrophic growth with simultaneous KIC production, ilvE gene expression could be ﬁne-tuned by testing promoters with varying strength (Vašicová et al., 1999; Hammer et al., 2006). a. Transcriptome analyses of KIC producer MV-Leu20 ΔilvE in comparison to the wild type were performed using DNA microarrays as described (Vogt et al., 2014). Candidate transporter genes were chosen based on an mRNA ratio (MV-Leu20 ΔilvE/wild type) of > 2, an annotation as (putative) membrane or transporter proteins and the prediction of multiple transmembrane helices in the encoded proteins. Data represent mean values of at least two (maximum four) evaluable microarray experiments (P-value < 0.05). b. TMH, number of transmembrane helices predicted with the SOSUI engine version 1.11. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 2-Ketoisocaproate transport b. TMH, number of transmembrane helices predicted with the SOSUI engine version 1.11. Table 3. Putative transporter genes showing increased expression in a KIC producer. a. Transcriptome analyses of KIC producer MV-Leu20 ΔilvE in comparison to the wild type were performed using DNA microarrays as described (Vogt et al., 2014). Candidate transporter genes were chosen based on an mRNA ratio (MV-Leu20 ΔilvE/wild type) of > 2, an annotation as (putative) membrane or transporter proteins and the prediction of multiple transmembrane helices in the encoded proteins. Data represent mean values of at least two (maximum four) evaluable microarray experiments (P-value < 0.05). b. TMH, number of transmembrane helices predicted with the SOSUI engine version 1.11. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 2-Ketoisocaproate production with C. glutamicum 7 Fig. 3. Complementation of the effects on growth and KIC accumu- lation caused by deletion of cg1121 in strain C. glutamicum ΔilvE carrying pAN6-leuA_B018. A. Growth of strains ΔilvE with pAN6-leuA_B018 (■), ΔilvE Δcg1121 with pAN6-leuA_B018 (●), ΔilvE Δcg1121::cg1121 with pAN6-leuA_B018 (△) and ΔilvE Δcg1121 with pAN6-leuA_B018- cg1121 (▽) are shown. B. Maximal KIC concentrations reached after 32 h cultivation in 500 ml baffled shake ﬂasks with 50 ml CGXII minimal medium con- taining 4% (w/v) glucose and 0.1 mM IPTG at 30°C and 120 rpm on a rotary shaker. The deletion of cg1121 was complemented either by genomic reintegration of cg1121 (ΔilvE Δcg1121::cg1121 p p g of acetohydroxyacid synthase by KIV has been reported by Krause and colleagues (2010). To avoid a reduced ﬂux into the leucine synthesis pathway by competitive inhibi- tion of IPMS by KIC, the concentration of KIV within the cell needs to be increased and/or the concentration of KIC within the cell should be decreased. These concentrations are determined on one hand by the rates of synthesis and further metabolic conversion and on the other hand by the rates of export to and import from the supernatant. p p p Knowledge on KIC transport is very limited. Obviously, as shown by our studies and that of Bückle-Vallant and colleagues (2014), KIC can leave the cell, and previous studies by Groeger and Sahm (1987) demonstrated that KIC can enter the cell. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 2-Ketoisocaproate transport As previously described for L-isoleucine (Zittrich and Krämer, 1994), passive diffusion, carrier-mediated uptake and carrier-mediated excretion must be considered as possibilities for an amphiphilic solute like KIC to cross the cytoplasmic membrane. Due to its similarity to L-isoleucine, it seems likely that also KIC is able to diffuse across the membrane. However, a necessity to possess carriers for KIC import or KIC export is not obvious, in contrast to the advantage of having importers for amino acids and exporters for non- catabolizable amino acids. In C. glutamicum, the export of BCAAs and methionine is catalysed by the exporter BrnFE (Kennerknecht et al., 2002; Trötschel et al., 2005; Xie et al., 2012), but evidence is available that BrnFE is not involved in KIC export (Radespiel, 2010). The identiﬁcation of transporters can be beneﬁcial for biotechnological processes to increase productivity since transport of desired products into the medium often rep- resents a bottleneck. For example, export was identiﬁed as a limiting factor for L-isoleucine production with C. glutamicum (Morbach et al., 1996), and the production of this BCAA was improved by overexpression of the respec- tive transporter encoded by brnFE (Kennerknecht et al., 2002; Xie et al., 2012). Fig. 3. Complementation of the effects on growth and KIC accumu- lation caused by deletion of cg1121 in strain C. glutamicum ΔilvE carrying pAN6-leuA_B018. A. Growth of strains ΔilvE with pAN6-leuA_B018 (■), ΔilvE Δcg1121 with pAN6-leuA_B018 (●), ΔilvE Δcg1121::cg1121 with pAN6-leuA_B018 (△) and ΔilvE Δcg1121 with pAN6-leuA_B018- cg1121 (▽) are shown In order to test if a KIC exporter is present in C. glutamicum that might be useful to improve KIC over- production, we searched for genes showing increased mRNA levels during KIC production by performing com- parative transcriptome analyses using DNA microarrays as described previously (Vogt et al., 2014). KIC producer strain MV-Leu20 ΔilvE was compared with the wild type to determine differentially expressed genes, resulting in a list of eight candidate transporter genes (Table 3). Each of these genes was deleted in the C. glutamicum ΔilvE background, and the resulting double deletion mutants were transformed with pAN6-leuA_B018. When cultivated in glucose minimal medium with 0.1 mM IPTG, one of the eight strains, which contained a deletion of cg1121 (annotated as putative permease of the major facilitator superfamily), showed a reduced growth rate (0.23 ± 0.01 h−1) and a reduced maximal KIC titre A. Acknowledgements We thank Dr Nicole Paczia (IBG-1: Biotechnology, Forschungszentrum Jülich, Germany) for her help with the measurements of cytoplasmic 2-ketoisocaproate concentrations. Groeger, U., and Sahm, H. (1987) Microbial production of L-leucine from α-ketoisocaproate by Corynebacterium glutamicum. Appl Microbiol Biotechnol 25: 352–356. Hammer, K., Mijakovic, I., and Jensen, P.R. (2006) Synthetic promoter libraries – tuning of gene expression. Trends Biotechnol 24: 53–55. Conﬂict of interest Heissig, H., Urban, K.A., Hastedt, K., Zunkler, B.J., and Panten, U. (2005) Mechanism of the insulin-releasing action of α-ketoisocaproate and related α-keto acid anions. Mol Pharmacol 68: 1097–1105. None declared. 2-Ketoisocaproate transport Growth of strains ΔilvE with pAN6-leuA_B018 (■), ΔilvE Δcg1121 with pAN6-leuA_B018 (●), ΔilvE Δcg1121::cg1121 with pAN6-leuA_B018 (△) and ΔilvE Δcg1121 with pAN6-leuA_B018- cg1121 (▽) are shown. B. Maximal KIC concentrations reached after 32 h cultivation in 500 ml baffled shake ﬂasks with 50 ml CGXII minimal medium con- taining 4% (w/v) glucose and 0.1 mM IPTG at 30°C and 120 rpm on a rotary shaker. The deletion of cg1121 was complemented either by genomic reintegration of cg1121 (ΔilvE Δcg1121::cg1121 pAN6-leuA_B018) or by plasmid-borne expression of cg1121 (ΔilvE Δcg1121 pAN6-leuA_B018-cg1121). The data represent mean values and standard deviations obtained from three independent cultivations. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology M. Vogt et al. 8 Cooper, A.J., Ginos, J.Z., and Meister, A. (1983) Synthesis and properties of the α-keto acids. Chem Rev 83: 321– 358. (22.4 mM) compared with the reference strain ΔilvE pAN6-leuA_B018 (0.30 ± 0.01 h−1, 37 mM KIC). The phenotype could be complemented by reintegration of cg1121 into the genome of strain ΔilvE Δcg1121 or by plasmid-borne expression of cg1121 (Fig. 3). However, the speciﬁc KIC export rates (determined as described by Kennerknecht et al., 2002) of strain ΔilvE Δcg1121 (6.5 ± 1.0 nmol min−1 gCDW−1) were not signiﬁcantly reduced compared with that of the reference strain ΔilvE (8.0 ± 1.5 nmol min−1 gCDW−1) and the cytoplasmic KIC concentrations of the two strains at an OD600 of 5 were comparable. Therefore, the role of Cg1121 for growth and KIC production remains unclear and needs further investigations. Eggeling, L., and Bott, M. (2005) Handbook of Corynebacterium glutamicum. Boca Raton, FL, USA: Taylor & Francis. Eggeling, I., Cordes, C., Eggeling, L., and Sahm, H. (1987) Regulation of acetohydroxy acid synthase in Corynebacterium glutamicum during fermentation of α-ketobutyrate to L-isoleucine. Appl Microbiol Biotechnol 25: 346–351. Escobar, J., Frank, J.W., Suryawan, A., Nguyen, H.V., Van Horn, C.G., Hutson, S.M., and Davis, T.A. (2010) Leucine and α-ketoisocaproic acid, but not norleucine, stimulate skeletal muscle protein synthesis in neonatal pigs. J Nutr 140: 1418–1424. Frunzke, J., Engels, V., Hasenbein, S., Gätgens, C., and Bott, M. (2008) Co-ordinated regulation of gluconate catabolism and glucose uptake in Corynebacterium glutamicum by two functionally equivalent transcriptional regulators, GntR1 and GntR2. Mol Microbiol 67: 305–322. References Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 2-Ketoisocaproate production with C. glutamicum 9 Klaffl, S., Brocker, M., Kalinowski, J., Eikmanns, B.J., and Bott, M. (2013) Complex regulation of the phosphoenolpyruvate carboxykinase gene pck and char- acterization of its GntR-type regulator IolR as a repressor of myo-inositol utilization genes in Corynebacterium glutamicum. J Bacteriol 195: 4283–4296. sis reveals extended overﬂow metabolism in various micro- organisms. Microb Cell Fact 11: 122. g Park, J.H., and Lee, S.Y. (2010) Fermentative production of branched chain amino acids: a focus on metabolic engi- neering. Appl Microbiol Biotechnol 85: 491–506. Pátek, M. (2007) Branched-chain amino acids. In Amino Acid Biosynthesis – Pathways, Regulation and Metabolic Engi- neering. Wendisch, V.F. (ed.). Berlin, Germany: Springer, pp. 129–162. Krause, F.S., Blombach, B., and Eikmanns, B.J. (2010) Meta- bolic engineering of Corynebacterium glutamicum for 2-ketoisovalerate production. Appl Environ Microbiol 76: 8053–8061. Pátek, M., Krumbach, K., Eggeling, L., and Sahm, H. (1994) Leucine synthesis in Corynebacterium glutamicum: enzyme activities, structure of leuA, and effect of leuA inactivation on lysine synthesis. Appl Environ Microbiol 60: 133–140. Leclercq-Meyer, V., Marchand, J., Leclercq, R., and Malaisse, W.J. (1979) Interactions of α-ketoisocaproate, glucose and arginine in the secretion of glucagon and insulin from the perfused rat pancreas. Diabetologia 17: 121–126. Radespiel, T. (2010) Dissertation abstract. URL http:// kups.ub.uni-koeln.de/id/eprint/3156. Litsanov, B., Brocker, M., and Bott, M. (2012a) Toward homosuccinate fermentation: metabolic engineering of Corynebacterium glutamicum for anaerobic production of succinate from glucose and formate. Appl Environ Microbiol 78: 3325–3337. Radmacher, E., Vaitsikova, A., Burger, U., Krumbach, K., Sahm, H., and Eggeling, L. (2002) Linking central metabo- lism with increased pathway ﬂux: L-valine accumulation by Corynebacterium glutamicum. Appl Environ Microbiol 68: 2246–2250. Litsanov, B., Kabus, A., Brocker, M., and Bott, M. (2012b) Efficient aerobic succinate production from glucose in minimal medium with Corynebacterium glutamicum. Microb Biotechnol 5: 116–128. Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach, G., and Pühler, A. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of deﬁned deletions in the chro- mosome of Corynebacterium glutamicum. Gene 145: 69–73. Marienhagen, J., Kennerknecht, N., Sahm, H., and Eggeling, L. (2005) Functional analysis of all aminotransferase proteins inferred from the genome sequence of Corynebacterium glutamicum. J Bacteriol 187: 7639– 7646. Scheele, S., Oertel, D., Bongaerts, J., Evers, S., Hellmuth, H., Maurer, K.H., et al. References (2013) Secretory production of an FAD cofactor-containing cytosolic enzyme (sorbitol-xylitol oxidase from Streptomyces coelicolor) using the twin- arginine translocation (Tat) pathway of Corynebacterium glutamicum. Microb Biotechnol 6: 202–206. Mimitsuka, T., Sawai, H., Hatsu, M., and Yamada, K. (2007) Metabolic engineering of Corynebacterium glutamicum for cadaverine fermentation. Biosci Biotechnol Biochem 71: 2130–2135. Morbach, S., Sahm, H., and Eggeling, L. (1996) L-Isoleucine production with Corynebacterium glutamicum: further ﬂux increase and limitation of export. Appl Environ Microbiol 62: 4345–4351. Schneider, J., and Wendisch, V.F. (2011) Biotechnological production of polyamines by bacteria: recent achievements and future perspectives. Appl Microbiol Biotechnol 91: 17–30. Smith, K.M., Cho, K.M., and Liao, J.C. (2010) Engineering Corynebacterium glutamicum for isobutanol production. Appl Microbiol Biotechnol 87: 1045–1055. Möckel, B., Eggeling, L., and Sahm, H. (1992) Functional and structural analyses of threonine dehydratase from Corynebacterium glutamicum. J Bacteriol 174: 8065– 8072. Tauch, A., Götker, S., Pühler, A., Kalinowski, J., and Thierbach, G. (2002) The alanine racemase gene alr is an alternative to antibiotic resistance genes in cloning systems for industrial Corynebacterium glutamicum strains. J Biotechnol 99: 79–91. Niebisch, A., and Bott, M. (2001) Molecular analysis of the cytochrome bc1-aa3 branch of the Corynebacterium glutamicum respiratory chain containing an unusual diheme cytochrome c1. Arch Microbiol 175: 282–294. Okino, S., Noburyu, R., Suda, M., Jojima, T., Inui, M., and Yukawa, H. (2008) An efficient succinic acid production process in a metabolically engineered Corynebacterium glutamicum strain. Appl Microbiol Biotechnol 81: 459– 464. Trötschel, C., Deutenberg, D., Bathe, B., Burkovski, A., and Krämer, R. (2005) Characterization of methionine export in Corynebacterium glutamicum. J Bacteriol 187: 3786– 3794. Vašicová, P., Pátek, M., Nešvera, J., Sahm, H., and Eikmanns, B. (1999) Analysis of the Corynebacterium glutamicum dapA promoter. J Bacteriol 181: 6188–6191. van Ooyen, J., Emer, D., Bussmann, M., Bott, M., Eikmanns, B.J., and Eggeling, L. (2011) Citrate synthase in Corynebacterium glutamicum is encoded by two gltA tran- scripts which are controlled by RamA, RamB, and GlxR. J Biotechnol 154: 140–148. Vogt, M., Haas, S., Klaffl, S., Polen, T., Eggeling, L., van Ooyen, J., and Bott, M. (2014) Pushing product formation to its limit: metabolic engineering of Corynebacterium glutamicum for L-leucine overproduction. Metab Eng 22: 40–52. van Ooyen, J., Noack, S., Bott, M., Reth, A., and Eggeling, L. (2012) Improved L-lysine production with Corynebacterium glutamicum and systemic insight into citrate synthase ﬂux and activity. Biotechnol Bioeng 109: 2070–2081. Wieschalka, S., Blombach, B., Bott, M., and Eikmanns, B.J. References Ikeda, M., and Nakagawa, S. (2003) The Corynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl Microbiol Biotechnol 62: 99–109. Abe, S., Takayama, K.I., and Kinoshita, S. (1967) Taxonomi- cal studies on glutamic acid-producing bacteria. J Gen Appl Microbiol 13: 279–301. Inui, M., Kawaguchi, H., Murakami, S., Vertes, A.A., and Yukawa, H. (2004) Metabolic engineering of Corynebacterium glutamicum for fuel ethanol production under oxygen-deprivation conditions. J Mol Microbiol Biotechnol 8: 243–254. Aparicio, M., Bellizzi, V., Chauveau, P., Cupisti, A., Ecder, T., Fouque, D., et al. (2012) Keto acid therapy in predialysis chronic kidney disease patients: ﬁnal consensus. J Ren Nutr 22: S22–S24. Becker, J., and Wittmann, C. (2012) Systems and synthetic metabolic engineering for amino acid production – the heartbeat of industrial strain development. Curr Opin Biotechnol 23: 718–726. Kalinowski, J., Bathe, B., Bartels, D., Bischoff, N., Bott, M., Burkovski, A., et al. (2003) The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J Biotechnol 104: 5–25. Becker, J., Buschke, N., Bücker, R., and Wittmann, C. (2010) Systems level engineering of Corynebacterium glutamicum – reprogramming translational efficiency for superior pro- duction. Eng Life Sci 10: 430–438. Keilhauer, C., Eggeling, L., and Sahm, H. (1993) Isoleucine synthesis in Corynebacterium glutamicum: molecular analysis of the ilvB-ilvN-ilvC operon. J Bacteriol 175: 5595– 5603. Blombach, B., Riester, T., Wieschalka, S., Ziert, C., Youn, J.W., Wendisch, V.F., and Eikmanns, B.J. (2011) Corynebacterium glutamicum tailored for efficient isobutanol production. Appl Environ Microbiol 77: 3300– 3310. Kennerknecht, N., Sahm, H., Yen, M.R., Patek, M., Saier Jr, M.H., Jr, and Eggeling, L. (2002) Export of L-isoleucine from Corynebacterium glutamicum: a two-gene-encoded member of a new translocator family. J Bacteriol 184: 3947–3956. Burkovski, A. (2008) Corynebacteria: Genomics and Molecu- lar Biology. Norfolk, UK: Caister Academic Press. Kind, S., and Wittmann, C. (2011) Bio-based production of the platform chemical 1,5-diaminopentane. Appl Microbiol Biotechnol 91: 1287–1296. Bückle-Vallant, V., Krause, F.S., Messerschmidt, S., and Eikmanns, B.J. (2014) Metabolic engineering of Corynebacterium glutamicum for 2-ketoisocaproate pro- duction. Appl Microbiol Biotechnol 98: 297–311. Kirchner, O., and Tauch, A. (2003) Tools for genetic engineering in the amino acid-producing bacterium Corynebacterium glutamicum. J Biotechnol 104: 287– 299. Cann, A.F., and Liao, J.C. (2010) Pentanol isomer synthesis in engineered microorganisms. Appl Microbiol Biotechnol 85: 893–899. © 2014 The Authors. References (2013) Bio-based production of organic acids with Corynebacterium glutamicum. Microb Biotechnol 6: 87–102. Paczia, N., Nilgen, A., Lehmann, T., Gätgens, J., Wiechert, W., and Noack, S. (2012) Extensive exometabolome analy- © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology 10 M. Vogt et al. 10 by Rhodococcus opacus DSM 43250 with L-leucine as the substrate. Enzyme Microb Technol 49: 321–325. by Rhodococcus opacus DSM 43250 with L-leucine as the substrate. Enzyme Microb Technol 49: 321–325. Xie, X., Xu, L., Shi, J., Xu, Q., and Chen, N. (2012) Effect of transport proteins on L-isoleucine production with the L-isoleucine-producing strain Corynebacterium glutamicum YILW. J Ind Microbiol Biotechnol 39: 1549–1556. Zittrich, S., and Krämer, R. (1994) Quantitative discrimination of carrier-mediated excretion of isoleucine from uptake and diffusion in Corynebacterium glutamicum. J Bacteriol 176: 6892–6899. Yukawa, H., and Inui, M. (2013) Corynebacterium glutamicum: Biology and Biotechnology. Heidelberg, Germany: Springer. Zanchi, N.E., Gerlinger-Romero, F., Guimaraes-Ferreira, L., de Siqueira Filho, M.A., Felitti, V., Lira, F.S., et al. (2011) HMB supplementation: clinical and athletic performance- related effects and mechanisms of action. Amino Acids 40: 1015–1025. Table S1. DNA oligonucleotides used in this study. Supporting information Additional Supporting Information may be found in the online version of this article at the publisher’s web-site: Zhu, Y., Li, J., Liu, L., Du, G., and Chen, J. (2011) Production of α-ketoisocaproate via free-whole-cell biotransformation Table S1. DNA oligonucleotides used in this study. © 2014 The Authors. Microbial Biotechnology published by John Wiley & Sons Ltd and Society for Applied Microbiology
https://openalex.org/W4388617972	https://ejournal.uniramalang.ac.id/index.php/eduabdimas/article/download/2996/2243	Indonesian	null	PUZZLE EDUKASI PHBS BAGI ANAK PENYINTAS BENCANA DI SDN 1 ROGO KECAMATAN DOLO SELATAN KABUPATEN SIGI	Jurnal Edukasi Pengabdian Masyarakat	2,023	cc-by	3,117	PUZZLE EDUKASI PHBS BAGI ANAK PENYINTAS BENCANA DI SDN 1 ROGO KECAMATAN DOLO SELATAN KABUPATEN SIGI Pitriani¹, Kiki Sanjaya², Firmansyah³, Annisa Nur Rahmadani ⁴ 1,2,3,4Universitas Tadulako Email: ¹pitriarifinkl07@gmail.com 2kksanjaya92@gmail.com 3kksanjaya92@gmail.com 4kksanjaya92@gmail.com Corresponding author: Pitriani Universitas Tadulako pitriarifinkl07@gmail.com Kata Kunci: PHBS, puzzle, edukasi sanitasi, bencana, banjir. ABSTRACT In 2022 there were 1,524 flood events with a total of 851 dead, 8,726 injured, 46 missing and 5,492,046 affected victims who were displaced. One area that is affected by floods almost every year is Rogo Village. Disasters will be followed by the emergence of refugees who are not supported by adequate sanitation facilities, this can trigger an outbreak of epidemics, so that knowledge of PHBS, especially vulnerable groups such as children, needs to be increased. Behavior change requires a long and consistent stage. To accelerate the change process, efforts to increase knowledge are required. For this reason, it is necessary to increase education for students at SDN 1 Rogo, Sigi District, Central Sulawesi Province regarding the minimum PHBS steps that must be taken. To measure changes in students' knowledge regarding PHBS, a pre and post test will be carried out. The activity will be packaged in the form of PHBS video screenings and games using PHBS EDUCATION PUZZLES so that it will be more interesting for participants. This activity was attended by 38 grade IV-VI students at SDN 1 Rogo and was accompanied by 4 teachers. Based on the analysis of the results of the pre post test which was carried out on July 20 2023, it was found that there was an increase in students' knowledge of PHBS after receiving education through videos and games. The highest increase was in materials related to the 6 steps of Hand Washing with Soap. Keyword: PHBS, puzzle, sanitation education, disaster, flood. EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) PENDAHULUAN Berdasarkan Geoportal Data Bencana Indonesia yang dirilis BNPB pada tahun 2022 terdapat 1.524 kejadian banjir, dengan total korban meninggal 851 jiwa, luka- luka 8.726 orang, hilang 46 orang dan korban terdampak yang mengungsi sebanyak 5.492.046 orang (Geoportal BNPB, 2022). Salah satu daerah yang hampir setiap tahun terdampak banjir adalah Desa Rogo, Kecamatan Dolo Selatan, Kabupaten Sigi, Provinsi Sulawesi Tengah. Dimana Desa Rogo terdapat terdapat 552 KK dengan rincian 1.782 jiwa. Desa ini memerlukan waktu sekitar 2 jam dari pusat pemerintahan. Kondisi jalan di desa ini cukup memprihatinkan mengingat desa ini sering terjadi banjir. Berdasarkan data dari Puskesmas Baluase, diare merupakan salah satu penyakit yang masuk dalam daftar 10 penyakit tertinggi pada tahun 2020-2022. Badan Penanggulangan Bencana Daerah (BPBD) Kabupaten Sigi mencatat telah terjadi banjir bandang pada tanggal 29 Agustus 2021 dan banjir susulan pada tanggal 5 September 2021. Banjir bandang melanda pemukiman warga disertai material lumpur, batu dan kayu (Wijaya, 2021). Banjir bandang terjadi akibat meluapnya sungai di Desa Rogo setelah hujan deras (Erafzon Saptiyulda AS & Muhammad Arshandi, 2021). Bencana selalu diikuti dengan mobilisasi penduduk ke titik-titik pengungsian (Adiyoso, 2021). Kondisi pengungsian pada umumnya padat dan terkadang tidak didukung fasilitas sanitasi yang memadai. Jika hal ini dibiarkan maka dapat memicu bencana susulan berupa merebaknya wabah seperti diare, penyakit kulit dan penyakit menular lainnya (Ibrahim, 2020). Untuk itu, pengetahuan terkait PHBS (Perilaku Hidup Bersih dan Sehat) masyarakat khususnya kelompok rentan seperti anak-anak perlu ditingkatkan. Salah satu cara yang paling efektif dalam mensosialisasikan PHBS yaitu melalui lembaga Pendidikan (Direktorat Jenderal Pendidikan Anak Usia Dini, Pendidikan Dasar dan Menengah, 2021). Sekolah dapat menjadi tempat pembelajaran efektif, dan perubahan perilaku pada anak sekolah sejak dini diharapkan akan menjadi kebiasaan baik hingga usia dewasa. Perubahan perilaku memerlukan tahap yang cukup panjang dan konsisten, sehingga untuk mempercepat proses perubahan ini diperlukan upaya dalam peningkatan pengetahuan dan sarana pendukungnya (Pitriani dan Sanjaya, 2020). Peningkatan pengetahuan siswa dapat dilakukan melalui berbagai metode, penggunaan media sosial bahkan menjadi salah satu alternatif yang masif digunakan saat ini (Fitriyanti et al., 2021). Namun melihat terbatasnya akses internet pada kelompok sasaran maka kita perlu memanfaatkan metode lain yang diharapkan dapat memberikan dampak signifikan terhadap peningkatan pengetahuan mereka terkait PHBS. Pada kegiatan ini, kami merancang permainan puzzle dalam meningkatkan pemahaman siswa Sekolah Dasar di Desa Rogo terkait PHBS dalam upaya pencegahan penyakit menular khususnya pada kondisi sanitasi yang kurang baik akibat bencana (Agustin et al., 2021). ABSTRAK Tahun 2022 terdapat 1.524 merupakan kejadian banjir dengan total korban meninggal 851 jiwa, luka- luka 8.726 orang, hilang 46 orang dan korban terdampak yang mengungsi sebanyak 5.492.046 orang. Salah satu daerah yang hampir setiap tahun terdampak banjir adalah Desa Rogo. Bencana akan diikuti munculnya pengungsian yang tidak didukung fasilitas sanitasi memadai, hal ini dapat memicu merebaknya wabah, sehingga pengetahuan PHBS khususnya kelompok rentan seperti anak-anak perlu ditingkatkan. Perubahan perilaku memerlukan tahap panjang dan konsisten, untuk mempercepat proses perubahan diperlukan upaya peningkatan pengetahuan. Untuk itu, perlu dilakukan peningkatan edukasi kepada siswa-siswi di SDN 1 Rogo, Kabupaten Sigi Provinsi Sulawesi Tengah terkait langkah-langkah PHBS minimal yang harus dilakukan. Untuk mengukur perubahan pengetahuan siswa terkait PHBS akan dilakukan pre dan post test. Kegiatan akan dikemas dalam bentuk pemutaran video PHBS dan games menggunakan PUZZLE EDUKASI PHBS sehingga lebih menarik minat peserta. Kegiatan ini diikuti oleh 38 Siswa-siswi kelas IV-VI di SDN 1 Rogo dan didampingi 4 orang guru. Berdasarkan analisis hasil pre post test yang telah dilaksanakan pada tanggal 20 Juli 2023 diketahui bahwa terjadi peningkatan pengetahuan siswa terhadap PHBS setelah memperoleh edukasi melalui video dan games. Peningkatan tertinggi pada materi terkait 6 langkah Cuci Tangan Pakai Sabun (CTPS). 225 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat (P-ISSN. 2964-1381) Vol. 2 No.4, Oktober 2023 (E-ISSN. 2964-139X) EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) METODE PELAKSANAAN Kegiatan Pengabdian Kepada Masyarakat (PKM) ini dilaksanakan dengan penyuluhan dengan memanfaatkan media puzzle edukasi PHBS yang telah disipakan sebelumnya dan pemutaran video yaitu lagu Cuci Tangan Pakai Sabun (CTPS), lagu sikat gigi dan lagu ayo iasakan sarapan. Khalayak sasaran utama kegiatan pengabdian ini adalah siswa/siswi SDN Inpres 1 Desa Rogo Kecamatan Dolo Selatan Kabupaten Sigi kelas 4 dan 5 berjumlah 58 orang. Kepala sekolah dan guru-guru sebagai sasaran tambahan untuk membantu dalam peningkatan pengetahuan dan perubahan perilaku siswa siswi terkait PHBS di Sekolah dan nantinya diharapkan dapat terus mengingatkan siswa/siswi mereka tentang pentingnya PHBS, dan menjadikan materi PHBS sebagai bagian dari materi pembelajaran. Untuk mengukur pengetahuan awal peserta terkait PHBS terlebih dahulu dilakukan pre tes yang dilanjutkan dengan penyuluhan, pada sesi akhir dilakukan pos tes untu melihat besarnya peningkatan pengetahuan siswa setelah diberikasn edukasi PHBS. Sebagai tambahan juga diberikan kuis dan pembagian doorprize bagi siswa/(i) yang dapat menjawab pertanyaan yang diberikan pemateri. PENDAHULUAN Penggunaan media puzzle yang didesain berwarna dengan gambar-gambar menarik sangat sesuai dengan model pembelajaran siswa sekolah dasar. SDN 1 Rogo merupakan sekolah dasar yang berada di Desa Rogo dengan jumlah siswa aktif 62 orang dan guru 9 orang. Pengabdian ini bertujuan untuk memberikan edukasi kepada siswa-siswi penyintas bencana di SDN 1 Rogo, Kecamatan Dolo Selatan, Kabupaten Sigi, Provinsi Sulawesi Tengah terkait Langkah-langkah PHBS minimal yang harus dilakukan pada kondisi pasca bencana atau saat mereka berada di Sekolah. Mengingat banjir di desa tersebut, merupakan bencana berulang. Kegiatan pengabdian ini, 226 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) juga bentuk dukungan terhadap pencapaian Visi Misi Program Studi Kesehatan Masyarakat FKM UNTAD. Gambar 1. Hasil Pre-post Tes Edukasi PHBS pada Siswa/(i) SDN 1 Rogo HASIL DAN PEMBAHASAN Kegiatan pengabdian kepada masyarakat telah dilaksanakan Senin, 20 Juli 2023 pada pukul 08.30 – selesai. Acara dibuka langsung oleh Ibu Kepala Sekolah SDN 1 Rogo Ibu Nurjanah., S.Pd. Kegiatan penyuluhan diikuti oleh 34 siswa kelas IV-VI dan 4 orang guru yang mendampingi. Kegiatan diawali dengan pre tes dan ditutup dengan post tes untuk mengukur sejauh mana materi yang diberikan dapat meningkatkan pengetahuan peserta. Hasil analisis nilai pre-post test dapat dilihat pada Gambar 1. Gambar 1. Hasil Pre-post Tes Edukasi PHBS pada Siswa/(i) SDN 1 Rogo 05 10 15 20 25 30 35 40 Arti PHBS Conto h PHBS Berap akali sikat gigi Wakt u penti ng sikat gigi Arti CTPS 6 langk ah CTPS Wakt u penti ng CTPS Cara poton g kuku Penti ngnya sarap an Akiba t tidak sarap an Penti ngnya olahra ga teratu r Jenis jajana n tidak sehat PRE TES 2 0 18 10 0 0 0 10 2 4 6 8 POSTTES 32 34 36 36 30 32 30 36 32 36 36 36 JUMLAH SISWA Pre Post Tes Pengetahuan PHBS Siswa(i) SDN 1 Rogo Pre Post Tes Pengetahuan PHBS Siswa(i) SDN 1 Rogo Gambar 1. Hasil Pre-post Tes Edukasi PHBS pada Siswa/(i) SDN 1 Rogo 227 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) Berdasarkan hasil analisis pre-pos test pada peserta penyuluhan diketahui bahwa terdapat beberapa pertanyaan yang sebagian besar dijawab dengan benar oleh peserta pada post test. Pada pertanyaan arti PHBS pada pre tes hanya 2 siswa (5.5%) menjawab benar dan pada post tes meningkat menjadi 32 siswa (88.8%), artinya terjadi peningkatan 83.3% setelah dilakukan edukasi PHBS menggunakan PUZZLE EDUKASI dan pemutaran video PHBS. Sedangkan untuk pertanyaan terkait contoh-contoh penerapan PHBS di lingkungan sekolah, semua peserta tidak mengetahui sebelum dilakukan penyuluhan, namun setelah penyuluhan 34 siswa (94.4%) telah mampu menjawab dengan benar dan memberikan contoh PHBS di Sekolah. Hal ini sejalan dengan temuan (Pitriani dkk, 2022) yang merilis hasil pre test siswa kelas 4, 5 dan 6 di SD Inpres Rogo terkait PHBS, pada dasarnya siswa telah mengetahui defenisi PHBS namun mereka sangat minim informasi terkait indikator PHBS, langkah- langkah CTPS (Cuci Tangan Pakai Sabun) sesuai standar dan syarat jajanan sehat. HASIL DAN PEMBAHASAN Terkait berapa kali kita harus sikat gigi dalam satu hari pada pre test 18 siswa (50%) telah menjawab dengan benar dan meningkat menjadi 36 siswa (100%) pada post tes. Sedangkan untuk waktu penting sikat gigi pada pre tes hanya 10 siswa (27.8%) yang menjawab dengan benar dan meningkat menjadi 36 siswa (100%) pada post tes, ini menunjukkan bahwa edukasi yang diberikan telah mampu meningkatkan pengetahuan siswa terkait pentingnya sikat gigi. Pada pertanyaan terkait arti CTPS, 6 langkah CTPS dan waktu-waktu penting untuk CTPS pada pre tes semua siswa tidak mengetahuinya, namun setelah penyuluhan pengetahuan mereka terkait arti singakatan CTPS dan waktu penting untuk kita melakukan CTPS telah meningkat dimana 30 siswa (83.3%) telah mampu menjawab dengan benar, sedangkan untuk 6 langkah CTPS terdapat 32 siswa (88.8%) yang mampu menjawab dengan benar. Berdasarkan data pada grafik 1 ini, membuktikan bahwa edukasi PHBS menggunakan PUZZLE EDUKASI diserta pemutaran video tentang lagu Cuci Tangan Pakai Sabun telah mampu meningkatkan pengetahuan siswa akan pentingnya CTPS dan bagaimana cara melakukannya. Perilaku CTPS yang baik dapat mencegah kejadian diare dan ISPA yang rentan dialami oleh anak-anak khususnya anak usia sekolah. Sehingga penting untuk mengajarkan anak sejak dini mengenai CTPS yang baik dan benar (Prasetya, 2022). Untuk pertanyaan terkait cara memotong kuku yang baik dan benar hanya 10 siswa (27.8%), yang menjawab dengan benar, namun setelah penyuluhan semua siswa telah dapat menjawab dengan benar. Kuman ada di mana-mana dan kita tidak bisa melihatnya dengan mata telanjang. Kuman yang menempel di tangan dapat berpindah ke benda yang kita pegang atau sentuh, ke makanan yang dipegang atau langsung ke mulut ketika menyuap, ke mulut, ketika kita ke meletakkan tangan ke mulut, ke mata ketika kita menggosok-gosok mata, ke saluran pernapasan karena terhisap waktu kita memegang hidung atau mengupil, ke orang lain ketika kita bersentuhan atau bersalaman. Untuk itu anak perlu potong kuku secara teratur, sehingga kuman tidak bersarang di tangan (UNICEF, 2020). Pada pertanyaan terkait pentingnya sarapan pada pre tes hanya 2 siswa (5.5%) yang mampu menjawab dengan benar, namun setelah penyuluhan meningkat menjadi 32 siswa (88.8%). Sedangkan untuk pertanyaan terkait akibat atau dampak negative tidak sarapan pada pre tes hanya 4 siswa (11.1%) yang menjawab dengan benar, namun setelah penyuluhan semua siswa telah mampu menjawab dengan benar. Pada pertanyaan jenis 228 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. HASIL DAN PEMBAHASAN 2964-139X) jajanan tidak sehat terdapat 8 siswa (22.2%) siswa yang telah mampu menjawab dengan benar dan setelah penyuluhan semua siswa telah mampu menjawab dengan benar. Sarapan merupakan hal penting dalam peningkatan kemampuan literasi anak usia sekolah, Mawarni (2018), dalam publikasinya menjelaskan peningkatan konsentrasi belajar pada akhirnya dapat meningkatkan prestasi belajar anak dapat dicapai dengan berbagai cara, salah satunya adalah membiasakan makan pagi atau sarapan. Sarapan mempunyai peranan penting bagi anak sekolah usia 6-14 tahun, yaitu untuk pemenuhan gizi di pagi hari, dimana anak-anak berangkat ke sekolah dan mempunyai aktivitas yang sangat padat di sekolah. Apabila anak- anak terbiasa sarapan pagi, maka akan berpengaruh terhadap kecerdasan otak, terutama daya ingat anak sehingga dapat mendukung prestasi belajar anak ke arah yang lebih baik. Pada pertanyaan bagaimana pendapat siswa tentang pentingnya berolahraga secara teratur terdapat 6 siswa (16.7%) yang menjawab dengan benar dan setelah penyuluhan semua siswa telah mampu menjawab dengan benar. Artinya edukasi dan motivasi yang diberikan kepada siswa telah mampu meningkatkan pengetahuan mereka akan pentingnya berolahraga secara teratur. Olahraga merupakan sebuah proses kegiatan yang sistematis untuk mendorong membina serta mengembangkan potensi jasmani, rohani dan sosial (Rubiyatno, 2021). Olahraga merupakan sebuah wadah bagi manusia untuk mengeksplorasi pengalaman geraknya dengan olahraga individu akan menjadi bugar serta kualitas hidup menjadi lebih baik tak terkecuali pada anak usia dini sekalipun mereka juga sedini mungkin harus diperkenalkan oleh aktivitas olahraga atau aktivitas jasmani walaupun itu hanya olahraga yang sifat nya tidak terstruktur seperi jalan, bersepeda, bermain lompat tali dan berlari-larian dengan melakukan aktivitas gerak seperti itu motorik anak akan lebih baik serta tumbuh kembang mereka menjadi optimal. Berdasarkan grafik 1 secara keseluruhan dapat diketahui bahwa bahwa penyuluhan dengan menggunakan media audio visual dapat membantu meningkatkan pengetahuan siswa/(i) terkait materi yang disampaikan. Hal ini sejalan dengan temuan Purnamasari (2020) bahwa sosialisasi menggunakan media audio visual dapat meningkatkan pengetahuan masyarakat terkait penyelahgunaan narkoba di Kediri, penggunaan berbagai media elektronik dapat menjadi alternatif dalam menjangkau masyarakat lebih luas. Disamping itu penggunaan game berupa PUZZLE EDUKASI yang memuat materi-materi terkait PHBS telah mampu meningkatkan minat siswa dalam memperhatikan materi yang diberikan. Penggunaan games sebagai media edukasi dapat menjadi alternatif untuk menarik minat siswa dan metode ini terbukti efektif dalam meningkatkan pengetahuan siswa sekolah dasar, salah satu yang telah diuji yaitu penggunaan permainan monopoli dalam meningkatkan pengetahuan gizi seimbang pada siswa Sekolah Dasar di Kabupaten Bogor (Fitriyani dkk, 2021). HASIL DAN PEMBAHASAN Pada kegiatan ini, kami merancang permainan puzzle dalam meningkatkan pemahaman siswa Sekolah Dasar di Desa Rogo terkait PHBS dalam upaya pencegahan penyakit menular khususnya pada kondisi sanitasi yang kurang baik akibat bencana. Puzzle merupakan permainan menyusun potongan-potongan gambar dan huruf yang dapat membantu siswa dalam mengembangakan keterampilan motorik dan kognitif untuk mengasah daya fikir siswa terkait obyek atau materi yang diberikan (Agustina dkk, 2021). Penggunaan permainan puzzle tentunya akan menarik minat siswa dalam belajar, 229 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) menggunakan metode yang tepat dalam memberikan edukasi menjadi penting demi efektivitas proses pembelajaran. Penggunaan media puzzle yang didesain berwarna dengan gambar-gambar menarik sangat sesuai dengan model pembelajaran siswa sekolah dasar. Gambar 2. Dokumentasi Pelaksanaan Kegiatan Penyuluhan PHBS Gambar 2. Dokumentasi Pelaksanaan Kegiatan Penyuluhan PHBS UCAPAN TERIMA KASIH Terima kasih kepada FKM UNTAD sebagai pihak yang mendanai pelaksanaan kegiatan PKM ini dan kepada Kepala Sekolah dan guru-guru SDN Inpres Rogo yang telah berpartisi aktif dalam pelaksanan kegiatan PKM ini. SIMPULAN Pelaksanaan pengabdian kepada masyarakat ini telah dilaksanakan pada tanggal 20 Juli 2023. Kegiatan penyuluhan diikuti 36 siswa/siswi kelas V dan VI dan 4 orang guru SDN Rogo secara off-line di kelas. Berdasarkan hasil pre-post test diketahui secara keseluruhan materi penyuluhan menggunakan PUZZLE EDUKASI PHBS & pemutaran video edukasi PHBS yang diberikan dapat meningkatkan pengetahuan siswa terkait PHBS di sekolah, hal ini dilihat dari peningkatan nilai post test setelah dilakukan penyuluhan. Peningkatan pengetahun paling tinggi pada materi terkait contoh penerapan PHBS di sekolah nilai meningkat dari 0% menjadi 94.4%. Sedangkan pemahaman siswa-siswi waktu penting CTPS masih perlu untuk ditingkatkan, meskipun peningkatan pengetahuan peserta setelah penyuluhan sudah cukup baik. DAFTAR RUJUKAN Adiyoso, Wignyo. Urgensi Pendekatan Multi dan inter disiplin Ilmu dalam Penaggulangan Bencana. BAPPENAS Working Paper. 2021: Volume 4 Nomor 2. DOI https://doi.org/10.47266/bwp.v4i2.104. Adiyoso, Wignyo. Urgensi Pendekatan Multi dan inter disiplin Ilmu dalam Penaggulangan Bencana. BAPPENAS Working Paper. 2021: Volume 4 Nomor 2. DOI https://doi.org/10.47266/bwp.v4i2.104. Agustin, M.D., Angga, S., Maruf, M.F., 2021. Pengembangan Media Puzzle Berganda Sebagai Supporting Daya Ingat Menghafal Huruf Alfabet Siswa Kelas I SDNegeri 2 Sengon. Jurnal Riset dan Inovasi Pendidikan Dasar 2. Agustin, M.D., Angga, S., Maruf, M.F., 2021. Pengembangan Media Puzzle Berganda Sebagai Supporting Daya Ingat Menghafal Huruf Alfabet Siswa Kelas I SDNegeri 2 Sengon. Jurnal Riset dan Inovasi Pendidikan Dasar 2. Badan Meterology, Klimatology dan Geofisika, 2022. Informasi Perubahan Normal Curah Hujan. BMKG, Jakarta. Badan Meterology, Klimatology dan Geofisika, 2022. Informasi Perubahan Normal Curah Hujan. BMKG, Jakarta. 230 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 EDUABDIMAS: Jurnal Edukasi Pengabdian Masyarakat Vol. 2 No.4, Oktober 2023 (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) (P-ISSN. 2964-1381) (E-ISSN. 2964-139X) Direktorat Jenderal Pendidikan Anak Usia Dini, Pendidikan Dasar Dan Menengah, 2021. Perilaku Hidup Bersih dan Sehat di Sekolah. Erafzon Saptiyulda AS dan Muhammad Arshandi. Bantuan Kemanusian Civitas Akademika POLTEKKES KEMENKES Palu Untuk Korban Banjir Bandang Rogo, Kabupaten Sigi Sulawesi Tengah. Jurnal PKM Poltekia. 2022: Volume 3 Nomor 1. Fitriyanti, R., Sriprahastuti, B., Cicih, L.H.M., 2021. Intervensi Permainan Monopoli dan Diskusi Gizi Seimbang untuk Meningkatkan Pengetahuan dan Sikap Siswa Sekolah Dasar di Kabupaten Bogor. Jurnal Nutrition College 10. https://doi.org/10.14710/jnc.v10i3.30772 Geoportal BNPB, G.B., 2022. Time Line Bencana Tahun 2022. Jakarta. Ibrahim, Sri A. 2020. Laporan Pengabdian Kepada Masyarakat “Perilaku Hidup Bersih dan Sehat (PHBS) di Era New Normal”. Program Studi Ilmu Keperawatan Fakultas Olahraga dan Kesehatan Universitas Gorontalo. Mawarni, Elita E. Edukasi Gizi "Pentingnya Sarapan Bagi Anak Sekolah". Warta Pengabdi. 2018 Volume 11 Issue 4 (pp 97-107). Pitriani, P., Sanjaya, K., 2020. Buku Ajar Dasar Kesehatan Lingkungan. Nasmedia, Makassar. Prasetyo, Ekawaty. Health Education On The Importance Of Washing Hands With Soap at SDN 10 Dungaliyo. JPKM. 2022: Volume 3 Nomor 1.E-ISSN 2774-3519. Rubiyatno. 2020. Peranan Aktivitas Olahraga Bagi Tumbuh Kembang Anak. Program Studi Pendidikan Jasmani, Kesehatan, dan Rekreasi IKIP-PGRI Pontianak. UNICEF. 2020. Buku Saku Program PHBS di Layanan PAUD. Direktorat Pembinaan Pendidikan Anak Usia Dini Direktorat Jenderal Pendidikan Anak Usia Dini dan Pendidikan Masyarakat Kementerian Pendidikan dan Kebudayaan. Wijaya, Chandra. Potensi Banjir Bandang pada Wilayah Sigi Sulawesi Tengah Indonesia. Jurnal Borneo Enginering. 2021: Volume 5 Nomor 2. DOI: https://doi.org/10.35334/be.v5i2.1838 2. 231 231
https://openalex.org/W3121209667	https://link.springer.com/content/pdf/10.1007/s00182-015-0462-1.pdf	English	null	Pillage games with multiple stable sets	International journal of game theory	2,015	cc-by	11,584	JEL Classiﬁcation C63 · C71 · P14 JEL Classiﬁcation C63 · C71 · P14 Int J Game Theory (2015) 44:993–1013 DOI 10.1007/s00182-015-0462-1 Int J Game Theory (2015) 44:993–1013 DOI 10.1007/s00182-015-0462-1 S. MacKenzie NICTA, University of New South Wales, Sydney, Australia M. Kerber School of Computer Science, University of Birmingham, Birmingham, UK Pillage games with multiple stable sets Simon MacKenzie · Manfred Kerber · Colin Rowat Accepted: 12 January 2015 / Published online: 15 March 2015 Accepted: 12 January 2015 / Published online: 15 March 2015 © The Author(s) 2015. This article is published with open access at Springerlink.com p y © The Author(s) 2015. This article is published with open access at Springerlink.com Abstract We prove that pillage games (Jordan in J Econ Theory 131.1:26–44, 2006, “Pillage and property”, JET) can have multiple stable sets, constructing pillage games with up to 2 n−1 3 stable sets, when the number of agents, n, exceeds four. We do so by violating the anonymity axiom common to the existing literature to establish a power dichotomy: for all but a small exceptional set of endowments, powerful agents can overcome all the others; within the exceptional set, the lesser agents can defend their resources. Once the allocations giving powerful agents all resources are included in a candidate stable set, deriving the rest proceeds by considering dominance relations over the ﬁnite exceptional sets—reminiscent of stable sets’ derivation in classical cooperative game theory. We also construct a multi-good pillage game with only three agents that also has two stable sets. Keywords Pillage games · Cooperative game theory · Core · Stable sets JEL Classiﬁcation C63 · C71 · P14 JEL Classiﬁcation C63 · C71 · P14 C. Rowat (B) Department of Economics, University of Birmingham, Birmingham, UK e-mail: c.rowat@bham.ac.uk S. MacKenzie NICTA, University of New South Wales, Sydney, Australia M. Kerber School of Computer Science, University of Birmingham, Birmingham, UK 1 Introduction Pillage games are cooperative games which allow a coalition’s power to depend on its resources as well as its membership. As cooperative games, they do not require 12 3 3 994 S. MacKenzie et al. agents to follow any predetermined game form, making them well-suited to the study of power struggles without third party enforcement, in which any proposed game form may itself be contested. Allowing transferrable resources to contributed to both power and value is a similarly appealing extension to standard cooperative games. agents to follow any predetermined game form, making them well-suited to the study of power struggles without third party enforcement, in which any proposed game form may itself be contested. Allowing transferrable resources to contributed to both power and value is a similarly appealing extension to standard cooperative games. Concretely, consider coalition formation problems. Non-cooperative analyses typ- ically specify rules for determining a proposer and a sequence of respondents, which often seem contrived. Standard cooperative analyses, on the other hand, do not recog- nize that prospective coalition partners may be attractive not just for intrinsic reasons, but for their transferrable holdings. Google, for example, seeking to defend hard- ware manufacturers using its Android operating system from Apple’s legal attacks, has sold patents to HTC, allowing HTC to—days later—launch its own patent case against Apple (Kwong 2011; Milford and Decker 2011). In international politics, great powers often transfer power—in the form of military aid—to their clients. Such behav- ior is pervasive and intuitively obvious, but poses problems for the standard analyses mentioned. Analysis of pillage games has focused on their stable sets: these are both intrin- sically interesting, and have been shown to be equivalent to a (farsighted) core in expectation (Jordan 2006). Thus, if the property of being undominated is an appealing one, and if forward looking agents are seen as natural, stable sets are a compelling solution concept. 1 Lucas (1992, pp. 562, 566) observed that this result ended hopes of ﬁnding an algorithm for constructing stable sets. Aumann (1985) had already lamented that “[o]ther than a small number of elementary truisms …there is no theory, no tools, certainly no algorithms” for computing stable sets. 2 Beardon and Rowat (2013) extended the original result to multi-good pillage games, showing that stable sets had measure zero, and bounding their Hausdorff dimension. 2 Beardon and Rowat (2013) extended the original result to multi-good pillage games, showing that stab sets had measure zero, and bounding their Hausdorff dimension. 1 Introduction There is evidence that pillage games’ stable sets behave better than those of other cooperative games: they are small, and often unique—in sharp contrast to the famous ‘signature’ example of Shapley (1959) which constructed a game in characteristic function form with an inﬁnite number of stable sets, each potentially inﬁnite.1 Both stable sets’ small size and their apparent uniqueness owe to pillage games’ structure: coalitions’ power must satisfy a coalitional monotonicity axiom (larger coalitions, in the sense of set inclusion, are more powerful) as well as weak and strong versions of a resource monotonicity axiom (coalitions whose members have more resources are more powerful). On size, Jordan (2006) used the strong resource monotonicity axiom to prove that pillage games’ stable sets were ﬁnite.2 Kerber and Rowat (2011) further exploited the same axiom to tighten Jordan’s bound to a Ramsey number. Saxton (2011) also made use of the coalitional monotonicity axiom to tighten the bound yet further to one that nonetheless remained doubly-exponential. Kerber and Rowat (2014) identiﬁed three axioms satisﬁed by all examples of pillage games previously studied—anonymity (or symmetry), continuity, and responsiveness—and derived a tight bound of 15 alloca- tions for three agent pillage games satisfying these additional axioms. On uniqueness, Jordan (2006) proved that the wealth is power pillage game—his leading example—had a unique stable set; he did not ﬁnd examples of multiple sta- ble sets. Kerber and Rowat (2014) proved uniqueness in three agent pillage games 123 Pillage games with multiple stable sets 995 satisfying the three additional axioms; a draft of their paper that preceded the present paper’s results (Kerber and Rowat 2009) also conjectured that pillage games’ monotonicity axioms might lead to ﬁne balances of power, such that stable sets either existed uniquely, or did not exist at all. This paper presents a family of counter-examples to that uniqueness conjecture, each more general than its predecessor. All extend the majority pillage game (or strength in numbers game) analyzed by Jordan and Obadia (2014) and Jordan (2006). In the classic majority game (Lucas 1992), domination only occurs if one coalition has strictly more members than the other. For three agent majority games, there are a continuum of stable sets, each containing a continuum of allocations such that one agent’s resources are held constant; the choice between two allocations within such sets pits one agent against one other, satisfying internal stability. 1 Introduction By creating these to exclude the possibility of dominance across them, stable sets are constructed by independently picking either 12 3 996 S. MacKenzie et al. the even or odd elements from each such exceptional set. Thus, an n-agent game yields 2 n−1 3 stable sets. Section 5 shows how to introduce dominance relations across exceptional sets, making choices of even or odd elements from one dependent on choices from another. This can arbitrarily reduce the number of stable sets below 2 n−1 3 Finally, Sect. 6 constructs a multi-good pillage game with two stable sets, but only three agents. Multi-good pillage games (Rowat 2009) add qualitatively to the original single-good pillage game by treating consensual trade in the same framework as non-consensual ‘pillage’. In this richer environment, dominance can be symmetric, allowing construction of two-element exceptional sets for two agents, each element of which dominates the other. In single good pillage games, domination cannot be symmetric (Jordan 2006), making four-element exceptional sets the smallest that can support two stable sets. The Appendix collects proofs. An earlier version of the paper (MacKenzie et al. 2013) contained more detail at various points. In conclusion, this paper replaces the conjecture that pillage games have no more than one stable set with new questions. First, do the additional axioms of anonymity, continuity and responsiveness—sufﬁcient for uniqueness with three agents (Kerber and Rowat 2014)—guarantee that larger pillage games also have no more than one stable set? As the constructions presented here rely on violating anonymity, is that condition alone sufﬁcient for uniqueness? Finally, as the present constructions only produce ﬁnite numbers of stable sets, are there pillage games capable of producing inﬁnite sets of stable sets? 1 Introduction Three agent majority games also have a degenerate stable set containing three allocations, each of which splits the resource equally between two of the agents. By contrast, the majority pillage game breaks ties in coalitional size in favour of the coalition with greater aggregate resource holdings; this additional mode of dominance eliminates the majority game’s continuum stable sets. The examples presented here extend the majority pillage game by violating the anonymity axiom that grants each agent identical intrinsic strength. Insofar as most empirical applications will not involve intrinsically identical agents, it is clearly impor- tant that the theory be developed to handle violations of anonymity. The present constructions all include one or more powerful agents with a large, ﬁxed intrinsic strength, and many lesser agents with variable—but lower—strengths. Speciﬁcally, they all exhibit the following power dichotomy: for all but a handful of exceptional allocations, the powerful agents are more powerful than the lesser agents; however, at those exceptional allocations, the lesser agents just match the powerful agents. The logic behind these constructions is that stable sets must satisfy both internal stability—harder for large sets—and external stability—harder for small sets. As pil- lage games’ stable sets are small, the existence of powerful agents with the ability to dominate all but a small number of exceptional allocations overcomes the usual problem with external stability. Once the allocations sharing the endowment between the powerful agents are included in candidate stable sets, the problem of constructing stable sets reduces to one of assessing stability over the ﬁnite exceptional sets. This problem is both straight- forward, and reminiscent of classical cooperative analysis (see, for example, Lucas (1971)). Section 3 presents the ﬁrst counter-example, a four-agent game whose excep- tional set is deﬁned to create a cyclical dominance relation among its four elements. Thus, either the even or odd elements of the exceptional set may belong to a stable set, yielding two stable sets. While most of the examples use power functions violating the continuity axiom, we show that this is an expositional aid rather than an intrinsic feature of the examples. Section 3.1 applies the same construction to a pillage game with an empty core. Section 4 generalizes the initial example by deﬁning an exceptional set for each non-overlapping set of three lesser agents. and external stability, S ∪D (S) = X. (ES) (ES) The conditions combine to yield S = X\D (S). While stable sets may not exist, or may be non-unique in general cooperative games, the core necessarily belongs to any stable set; when the core also satisﬁes external stability, it is the unique stable set. 2 Pillage games (CX) π (C x) is continuous in x (AN) if σ : I →I is a 1:1 onto function permuting the agent set, then i ∈C ⇔ σ (i) ∈C′, and xi = x′ σ(i) then π (C, x) = π C′, x′ . (CX) π (C, x) is continuous in x. (RE) if i /∈C and π ({i} , x) > 0 then π (C ∪{i} , x) > π (C, x). ) ( ) RE) if i /∈C and π ({i} , x) > 0 then π (C ∪{i} , x) > π (C, x). Axiom AN is an anonymity axiom which means that power does not depend on the identity of agents, merely their cardinality and resources.3 Axiom RE is the respon- siveness axiom of Jordan (2009): the addition of an agent which has power even as a singleton strictly increases the power of its new coalition. With continuity axiom CX, these ensure that three agent pillage games have no more than one stable set, containing no more than 15 allocations (Kerber and Rowat 2014). An allocation y dominates an allocation x, written y x, iff π (W, x) > π (L, x) ; where W ≡{i \|yi > xi } is the win set and L ≡{i \|xi > yi } is the lose set. For Y ⊆X, let where W ≡{i \|yi > xi } is the win set and L ≡{i \|xi > yi } is the lose set. For Y ⊆X let where W ≡{i \|yi > xi } is the win set and L ≡{i \|xi > yi } is the lose set. For Y ⊆X, let be the dominion of Y, the set of allocations dominated by an allocation in Y. Similarly, U (Y) = X\D (Y), is the set of allocations undominated by any allocation in Y. The core, K, is the set of undominated allocations, U (X) = X\D (X). A set of allocations, S ⊆X, is a stable set iff it satisﬁes internal stability, be the dominion of Y, the set of allocations dominated by an allocation in Y. Similarly, U (Y) = X\D (Y), is the set of allocations undominated by any allocation in Y. The core, K, is the set of undominated allocations, U (X) = X\D (X). 2 Pillage games A set of allocations, S ⊆X, is a stable set iff it satisﬁes internal stability, S ∩D (S) = ∅; (IS) (IS) S ∩D (S) = ∅; 3 Jordan and Obadia (2014) called this axiom ‘symmetry’, following the terminology of CF games (Lucas 1971). We regard ‘anonymity’ as more precise, as it restricts the symmetry to agents’ identity: intrinsically identical agents may differ in power when their resources differ. 2 Pillage games Let I = {0, . . . , n −1} be a ﬁnite set of agents; when indexed by i, j, and k, these refer to distinct agents. An allocation divides a unit resource among them, so that the feasible set of allocations is a compact, continuous n −1 dimensional simplex: X ≡ {xi}i∈I xi ≥0, i∈I xi = 1 . Let ⊂denote a proper subset, and use ⊆to allow the possibility of equality. A power function is deﬁned over subsets of agents and allocations, so that π : 2I × X →R satisﬁes: (WC) if C ⊂C′ ⊆I then π C′, x ≥π (C, x) ∀x ∈X; (WC) if C ⊂C′ ⊆I then π C′, x ≥π (C, x) ∀x ∈X; (WR) if yi ≥xi∀i ∈C ⊆I then π (C, y) ≥π (C, x); and (SR) if ∅̸= C ⊆I and yi > xi∀i ∈C then π (C, y) > π (C, x). (WR) if yi ≥xi∀i ∈C ⊆I then π (C, y) ≥π (C, x); and y y (SR) if ∅̸= C ⊆I and yi > xi∀i ∈C then π (C, y) > π (C, x). (SR) if ∅̸= C ⊆I and yi > xi∀i ∈C then π (C, y) > π (C, x). Axiom WC requires weak monotonicity in coalitional inclusion; WR requires weak monotonicity in resources; SR requires strong monotonicity in resources. Axiom WC requires weak monotonicity in coalitional inclusion; WR requires weak monotonicity in resources; SR requires strong monotonicity in resources. y ; q g A pillage game is then a triple, (n, X, π). y q g A pillage game is then a triple, (n, X, π). Kerber and Rowat (2014) noted that all examples of power functions in Jordan (2006) satisﬁed the following additional axioms: Kerber and Rowat (2014) noted that all examples of power functions in Jordan (2006) satisﬁed the following additional axioms: 123 Pillage games with multiple stable sets 997 (AN) if σ : I →I is a 1:1 onto function permuting the agent set, then i ∈C ⇔ σ (i) ∈C′, and xi = x′ σ(i) then π (C, x) = π C′, x′ . 4 Fully expressed, the lower bound is 2q + 2δ + 1 3 (1 −3δ) > q + δ + 2 3 (1 −3δ). 5 While the allocations in E can be written more simply—e.g. a = 0, δ, 2 3 −δ, 1 3 —the current presen- tation will be clearer in the more general constructions. 3 Four agents and two stable sets This section presents a power function based on the strength in numbers (SIN) power function (Jordan 2006), π (C, x) = i∈C [xi + v] , v > 1. 3 Jordan and Obadia (2014) called this axiom ‘symmetry’, following the terminology of CF games (Lucas 1971). We regard ‘anonymity’ as more precise, as it restricts the symmetry to agents’ identity: intrinsically identical agents may differ in power when their resources differ. 998 S. MacKenzie et al. This deﬁnes a majority pillage game (Jordan and Obadia 2014) in which the larger coalition is always the more powerful, with ties in coalitional size resolved by com- paring their aggregate resource holdings. Our version extends the original in two ways. First, it violates anonymity axiom AN by granting the zeroth agent an intrin- sic strength much greater than each of the others’. Second, it allows these intrinsic strengths to depend on coalitions and their resource holdings; in particular, it creates a small exceptional set of allocations, outside of which the zeroth agent alone is stronger than the other three together; inside it, however, they can resist the zeroth. This power dichotomy is the key to all of the following examples. Our power function is therefore: π (C, x) = i∈C [xi + vi (C, x)] ; (1) (1) where v0 (C, x) = 10, and where v0 (C, x) = 10, and vi (C, x) = ¯q if {1, 2, 3} ⊆C and x ∈E q otherwise ∀i ∈{1, 2, 3} (2) (2) where ¯q ≡3, q is bounded by where ¯q ≡3, q is bounded by where ¯q ≡3, q is bounded by where ¯q ≡3, q is bounded by 1 3 −2δ < q < ¯q (3) 1 3 −2δ < q < ¯q (3) with 0 < δ < 1 6, and the exceptional set is with 0 < δ < 1 6, and the exceptional set is E ≡{a, b, c, d} , where a ≡ 0, δ, δ + 2 3(1 −3δ), δ + 1 3(1 −3δ) b ≡ 0, δ, δ + 1 3(1 −3δ), δ + 2 3(1 −3δ) c ≡ 0, δ + 1 2(1 −3δ), δ, δ + 1 2(1 −3δ) d ≡ 0, δ + 1 2(1 −3δ), δ + 1 2(1 −3δ), δ . 123 Lemma 1 Equations 1 and 2 deﬁne a power function. Proof Equation 1 deﬁnes a function: the auxiliary conditions partition the domain, and each condition maps onto a single-valued term. Proof Equation 1 deﬁnes a function: the auxiliary conditions partition the domain and each condition maps onto a single-valued term. Now demonstrate that function 1 satisﬁes the three pillage axioms. As, on X\E, the function is the SIN power function, the axioms need only be tested on the exceptional set, E, and across that set and its complement. WC As x is held constant, only increases in C to C′ for x ∈E need to be considered. There are three cases to consider: 1. not all i ∈{1, 2, 3} are in C or in C′. In this case, each agent i in C′\C contributes xi + q > 0 to coalitional power if i > 0 and xi + 10 > 0 if i = 0. As π C′, x is additive across agents, and as the contributions of the agents in C do not change, the result holds in this case. 2. not all i ∈{1, 2, 3} are in C but {1, 2, 3} = C′. Now each agent i in C′\C contributes xi + ¯q > 0; further, the contributions of those i in C are increased from xi +q to xi + ¯q. Again, π C′, x > π (C, x), consistent with the result. 3. {1, 2, 3} ∈C so that C′ = I. As the newcomer, i = 0 adds x0 + 10, and as the others’ contributions do not change, the axiom is satisﬁed. WR Now there are two cases to consider: 1. 0 ∈C, which subdivides into: (a) {1, 2, 3} ⊆C, which implies that yi ≥xi∀i ∈I. By the aggregate endowment constraint, this implies that y = x, for which axiom WR is trivially satisﬁed. y (b) {1, 2, 3} ⊈C, so that vi (C, y) = vi (C, x) ∀i ∈I, allowing these terms to cancel when calculating π (C, y) −π (C, x) = i∈C (yi −xi). That yi ≥xi for all i ∈C then upholds the result. (b) {1, 2, 3} ⊈C, so that vi (C, y) = vi (C, x) ∀i ∈I, allowing these terms to cancel when calculating π (C, y) −π (C, x) = i∈C (yi −xi). That yi ≥xi for all i ∈C then upholds the result. 3 Four agents and two stable sets (4) (4) Thus, the upper bound on q in inequality 3 prevents the three lesser agents overpow- ering the zeroth, while lower bound ensures that any two weak agents can overpower the third at any x ∈E.4,5 The δ parameter ensures that the three weaker agents have positive endowments in the exceptional allocations, and will therefore resist pillage to t0; its upper bound ensures that q > 0, whose importance shall be seen below. The power dichotomy here owes to this inequality: The power dichotomy here owes to this inequality: here owes to this inequality: 3q + 1 < v0 (C, x) = 3¯q + 1. (5) (5) 4 Fully expressed, the lower bound is 2q + 2δ + 1 3 (1 −3δ) > q + δ + 2 3 (1 −3δ). 5 While the allocations in E can be written more simply—e.g. a = 0, δ, 2 3 −δ, 1 3 —the current presen- tation will be clearer in the more general constructions. 123 Pillage games with multiple stable sets 999 This will be seen to allow, on the one hand, the powerful zeroth agent to seize the resources of the other three for any non-exceptional allocation while, on the other, never allowing the lesser agents to seize the zeroth agent’s resources. It may seem that the discontinuity in the three agents’ intrinsic power at the excep- tional set risks violating resource monotonicity. This is not the case: Lemma 1 Equations 1 and 2 deﬁne a power function. Lemma 1 Equations 1 and 2 deﬁne a power function. 2. 0 /∈C further subdivides into two non-trivial cases: (a) {1, 2, 3} = C, which implies that, if x ∈E, then y = x: i∈C xi = 1 for x ∈E, and yi ≥xi for all i ∈C with i∈C yi = 1. Then, for all i ∈C, vi (C, y) = vi (C, x) so that π (C, y) −π (C, x) depends only on relative resource holdings, as above. (b) {1, 2, 3} ̸= C, implying that {1, 2, 3} ⊈C so that for all i ∈C, it is again the case that vi (C, y) = vi (C, x). (b) {1, 2, 3} ̸= C, implying that {1, 2, 3} ⊈C so that for all i ∈C, it is again the case that vi (C, y) = vi (C, x). ⊓⊔ ⊓⊔ SR The proof is analogous to that for WR. ⊓⊔ SR The proof is analogous to that for WR. Following Jordan (2006), let ti ∈X be a tyrannical allocation such that ti i = 1 and ti j = 0. Then: Following Jordan (2006), let ti ∈X be a tyrannical allocation such that ti i = 1 and ti j = 0. Then: ti j = 0. Then: Lemma 2 For the pillage game deﬁned by Eqs. 1 and 2, the core is t0 . 123 3 1000 S. MacKenzie et al. Proof First establish that t0 belongs to the core: Proof First establish that t0 belongs to the core: Proof First establish that t0 belongs to the core: π {0} , t0 = 1 + 10 = 11 > 3q = π {1, 2, 3} , t0 . Second, show that t0 dominates all allocations, y, in X\ E ∪ t0 . Noting that i t0 i > yi = {0} and i yi > t0 i ⊆{1, 2, 3} we calculate Second, show that t0 dominates all allocations, y, in X\ E ∪ t0 . Noting that i t0 i > yi = {0} and i yi > t0 i ⊆{1, 2, 3} we calculate π ({0} , y) = y0 + 10 ≥10 > (1 −y0) + 3q = π ({1, 2, 3} , y) ≥π i yi > t0 i , y . Lemma 1 Equations 1 and 2 deﬁne a power function. Third, show that t0 does not dominate any z ∈E: π i t0 i > zi , z = π ({0} , z) = 10 = 1 + 3 · 3 = π ({1, 2, 3} , z) = π i zi > t0 i , z . z = π ({0} , z) = 10 = 1 + 3 · 3 = π ({1, 2, 3} , z) = π i zi > t0 i , z . Finally, for each allocation z in E, it can be conﬁrmed that there is another, w ∈E, such that w z. Concretely: Finally, for each allocation z in E, it can be conﬁrmed that there is another, w ∈E, such that w z. Concretely: (6) (6) (Consider, for example, a d: W = {2, 3} and L = {1} as direct implications o δ ∈ 0, 1 6 ; thus, π (W, d) = 2q + 1 2 + 1 2δ > q + 1 2 −1 2δ = π (L, d).) (Consider, for example, a d: W = {2, 3} and L = {1} as direct implications of δ ∈ 0, 1 6 ; thus, π (W, d) = 2q + 1 2 + 1 2δ > q + 1 2 −1 2δ = π (L, d).) (Consider, for example, a d: W = {2, 3} and L = {1} as direct implications of δ ∈ 0, 1 6 ; thus, π (W, d) = 2q + 1 2 + 1 2δ > q + 1 2 −1 2δ = π (L, d).) 6 2 2 2 2 Thus, the only undominated allocation is t0, establishing the result. ⊓⊔ ⊓⊔ As shall be seen, the cores in our examples, when they are non-empty, are small, but have large dominions—precisely the conditions that led Lucas to discover games in characteristic function form that had no stable sets (Lucas 1971). Here, the condition contributes to multiplicity of stable sets. Figure 1 illustrates some of the dominance relations generated by power function 1: arrows ﬂow from dominating to dominated allocations. Their cyclical behaviour is key to the counter-example. Theorem 1 The pillage game deﬁned by functions 1 and 2 yields exactly two stable sets, S1 ≡ t0, a, b , and S2 ≡ t0, c, d . Lemma 1 Equations 1 and 2 deﬁne a power function. Proof First conﬁrm that S1 and S2 are stable. As established in Lemma 2, t0 dominates all allocations in X\ E ∪ t0 , and no allocations in E. As a, b are in E, and neither a b nor b a is true, S1 is internally stable. The cyclical dominance within E ensures that {c, d} ∈D ({a, b}), establishing external stability. Thus, S1 is stable. A parallel argument establishes that S2 is as well. To prove that there are no other stable sets, denote any stable set of the game by Si. In any cooperative game, the core belongs to any such stable set. In this case, t0 ∈Si. As allocations dominated by the core, D (K), cannot belong to a stable set, the problem of ﬁnding another stable set reduces to that of considering inclusions and exclusions 123 123 1001 Pillage games with multiple stable sets Fig. 1 D t0 = X\ t0, a, b, c, d and a d b c a Fig. 1 D t0 = X\ t0, a, b, c, d and a d b c a of the allocations in E. As t0 does not dominate them, the core is not a stable set on its own. By the cyclical dominance, suppose that one other element, a, also belonged to Si; this is insufﬁcient for external stability as it leaves b and c undominated. Adding b to a yields S1, which was already identiﬁed; as that satisﬁes external stability, adding either cor d wouldthenviolateinternalstability.Allocation c(andsupersetscontaining it) cannot be added to a without violating internal stability. Finally, attempting to add allocation d (or supersets) to a also violates internal stability. Identical reasoning beginning, in turn, with each of b, c and d establishes that no further stable sets exist. ⊓⊔ ⊓⊔ All the examples of pillage games studied by Jordan (2006), Jordan and Obadia (2014) and Kerber and Rowat (2014)—including a discontinuous example in Kerber and Rowat (2014)—were such that each allocation, x, generated a dominion, D (x), whose components had no holes. In the present example, the core’s dominion has holes at the edge of the simplex—the exceptional set. Once the core is included in a candidate stable set, analysis resembles classical analysis of games in characteristic function form. 6 Lucas (1992, p. 586) argued that the stable set became a particularly important solution concept when the core was empty, as the core was generally a good solution concept, and its emptiness often implied the emptiness of related solution concepts. Lucas (1992, p. 552) also noted that the initial focus of von Neumann and Morgenstern (1953) was on constant sum essential games in characteristic function form, all of which have empty cores. Lemma 1 Equations 1 and 2 deﬁne a power function. The preceding example can be replicated with a continuous power function, and an empty core, isolating violations of the anonymity axiom as the common feature in all of our constructions. Constructing an example in which the power of the lesser agents increases continuously to that of the zeroth agent as the allocations approach the exceptional set merely requires replacing Eq. 2 with vi (C, x) = max { f (r (a, x)) , f (r (b, x)) , f (r (c, x)) , f (r (d, x)) , q} if {1, 2, 3} = C q otherwise (7) where r (z, x) ≡ i∈I \|zi −xi\| is the rectilinear distance between x and z, and f (r (y, x)) ≡¯q −10 (¯q −q)r (y, x) ; (8) (8) 12 3 3 1002 S. MacKenzie et al. where q and ¯q are as deﬁned on page 7. Worked details can be found in MacKenzie et al. (2013). where q and ¯q are as deﬁned on page 7. Worked details can be found in MacKenzie et al. (2013). 7 The Bondareva–Shapley theorem provides a parallel result for games in characteristic function form. 3.1 Two stable sets with an empty core The emptiness or non-emptiness of the core is often linked to other phenomena in cooperative games; however, we now extend the construction above to produce multi- ple stable sets in pillage games with empty cores.6 To do so, triplicate agent 0 to create agents 00 and 000. The exceptional set again has four elements, E ≡{a, b, c, d}: for the lesser agents, i > 0, the allocations remain identical to the original ones deﬁned by Eq. 4; the triplicated zeroth agents each receive nothing under the exceptional allo- cations, x0 = x00 = x000 = 0∀x ∈E. Finally, modify the membership premium deﬁned in Eq. 2 so that (for this subsection alone) ¯q = 19 3 rather than 3; maintain the upper bound q < 3. Let split allocations, si, j, divide the resource equally between i and j, so that si, j i = si, j j = 1 2; s0,00, s0,000 and s00,000 shall play the same role as t0 did in the original example, although they are not core allocations. Theorem 2 For the six agent pillage game deﬁned by modiﬁed Eqs. 4, 2 and original Eq. 1, the core is empty, and the only stable sets are S1 = s0,00, s0,000, s00,000, a, b and S2 = s0,00, s0,000, s00,000, c, d . The theorem’s proof relies on the following deﬁnition and lemma. Denote the set of allocations partitioning the endowment between the zero agents by The theorem’s proof relies on the following deﬁnition and lemma. Denote the set of allocations partitioning the endowment between the zero agents by Z ≡{z ∈X \|z0 + z00 + z000 = 1} . The lemma is then: Lemma 3 If f is an allocation that belongs to neither the exceptional set, E, nor to the zero simplex, Z, then f cannot belong to any stable set in the pillage game of Theorem 2. Proof of Theorem 2 Proposition 2.6 of Jordan (2006) states that a pillage game has an empty core if no agent can defend itself from pillage even when it possesses the entire allocation.7 Here, conﬁrming that the strongest agents cannot defend themselves also conﬁrms that the core is empty: π {0} , t0 = 10 + 1 < 10 + 10 + 3q = π I\ {0} , t0 . 6 Lucas (1992, p. 3.1 Two stable sets with an empty core 586) argued that the stable set became a particularly important solution concept when the core was empty, as the core was generally a good solution concept, and its emptiness often implied the emptiness of related solution concepts. Lucas (1992, p. 552) also noted that the initial focus of von Neumann and Morgenstern (1953) was on constant sum essential games in characteristic function form, all of which have empty cores. 123 1003 Pillage games with multiple stable sets Second, establish internal stability. Dominance within the exceptional set remains as described in expression 6. Thus, an internally stable set may contain only one of {a, b} and {c, d}. By symmetry, it only remains to show that neither a s0,00 nor s0,00 a: π {1, 2, 3} , s0,00 = 3q < 10 + 10 + 1 = π {0, 00} , s0,00 ; π ({0, 00} , a) = 10 + 10 = 3 · 19 3 + 1 = π ({1, 2, 3} , a) . Third, establish external stability. By expression 6, including either {a, b} or {c, d} ensures that all excluded allocations in E are dominated by an included one. For allocations outside E, the power of a coalition formed of any two of the zero agents is at least 20. The strongest opposition that they can face has power 10 + 3q + 1 < 20. Any allocation not in E which does not split the unit endowment between two of the zero agents therefore allows formation of a coalition of two zero agents who stand to beneﬁt by splitting the endowment equally between themselves; such equal splits belong to the proposed stable sets. As, by the previous logic, these two agents can overpower their opposition, external stability holds. Finally, show that no other stable sets are possible. By Lemma 3, only allocations in E ∪Z may belong to any stable set. As its proof establishes that no allocation in E dominates one in Z, ES on Z must be satisﬁed using only allocations from Z itself. As the allocations dominating the rest of Z must also be internally stable amongst themselves, and as the game played over Z is a three agent majority pillage game, the result follows directly from Theorem 3.4 of Jordan and Obadia (2014), which proved a unique stable set equal to s12, s23, s13 in the three agent majority pillage game. 3.1 Two stable sets with an empty core ⊓⊔ The pillage game above is based on a majority pillage game played among the three zero agents, which has s0,00, s0,000, s00,000 as its unique stable set. Adding the three lesser agents thus creates new stable sets containing the original one. 4 Independent exceptional sets: 2 n−1 3 stable sets 4 Independent exceptional sets: 2 n−1 3 stable sets This section extends the original construction, deﬁning exceptional sets for each dis- joint set of three consecutive lesser agents. As these are designed to ensure that there are no dominance relations across sets, stable sets may be constructed by indepen- dently choosing the even or odd allocations in each of these, yielding 2 n−1 3 stable sets when n −1 is a multiple of three. Deﬁne the index J = 1, . . . , n−1 3 so that the family of sets of three consecutive, disjoint agents is Pj j∈J, with Pj = {k,l, m}, where k = 3( j −1) + 1, l = 3( j −1) + 2 and m = 3( j −1) + 3. The exceptional set belonging to each j ∈J is E j ≡ a j, b j, cj, d j ⊂X, where a jk, a jl, a jm = δ, δ + 2 3(1 −3δ), δ + 1 3(1 −3δ) , b jk, b jl, b jm = δ, δ + 1 3(1 −3δ), δ + 2 3(1 −3δ) , 3 3 1004 S. MacKenzie et al. pendent exceptional c jk, c jl, c jm = δ + 1 2(1 −3δ), δ, δ + 1 2(1 −3δ) , d jk, d jl, d jm = δ + 1 2(1 −3δ), δ + 1 2(1 −3δ), δ , Fig. 2 Independent exceptional sets Fig. 2 Independent exceptional sets c jk, c jl, c jm = δ + 1 2(1 −3δ), δ, δ + 1 2(1 −3δ) , d jk, d jl, d jm = δ + 1 2(1 −3δ), δ + 1 2(1 −3δ), δ , and a ji, b ji, c ji, d ji = 0 for all i /∈Pj. As before, Eq. 1 deﬁnes the base of the power function. Now generalize the power dichotomy so that—for q bounded below by the inequality in 3—its upper bound, ¯q and v0 (C, x) are determined by (n −1) q + 1 < v0 (C, x) = 3¯q + 1. (9) (9) Again, this allows the zeroth agent to seize all resources other than those in exceptional allocations, while always protecting its own resources from seizure. 4 Independent exceptional sets: 2 n−1 3 stable sets The other vi (C, x) are again deﬁned to give the lesser agents less intrinsic power except when allied at their exceptional allocations: vi (C, x) = ¯q if ∃j : i ∈Pj and Pj ⊆C and x ∈E j q otherwise . (10) (10) Lemma 1 generalizes as expected: Lemma 4 Equations 1 and 10 deﬁne a power function. This section’s main result is: This section’s main result is: Theorem 3 The n-agent pillage game deﬁned by Eqs. 1 and 10 yields exactly 2 n−1 3 stable sets, each containing t0 and exactly one of Sj,1 ≡ a j, b j or Sj,2 ≡ cj, d j for each j ∈J. The construction is illustrated for six agents in Fig. 2. Before proving the theorem, we state and prove an auxiliary lemma: 123 Pillage games with multiple stable sets 1005 Lemma 5 For all distinct Ei and E j, there are no x ∈Ei and y ∈E j such that x y. Lemma 5 For all distinct Ei and E j, there are no x ∈Ei and y ∈E j such that x y. Proof of Theorem 3 Internal stability: each individual Sj,1 and Sj,2 is internally stable by Theorem 1. Lemma 5 proves that the series of such sets deﬁned in the theorem are also internally stable. Finally, to show that t0 does not dominate any of the exceptional allocations, it sufﬁces to show that π ({0} , ai) = v0 (C, ai) = 3¯q + 1 = π (Pi, ai) . External stability: the core is again t0 ; it dominates all other allocations outside the E j. As in Theorem 1, the elements in any Sj,1 and Sj,2 dominate the remaining allocations in the corresponding E j; as the theorem constructs a set from t0 and one of these for each E j, external stability follows. j Finally, show that these are the only possibilities. As already noted, the core belongs to any stable set. This leaves the allocations in the E j to be included or excluded. As, by Lemma 5, no allocation in one exceptional set can dominate an allocation in another, each a j must either be included, or dominated by an included allocation from E j, reducing the problem to that in Theorem 1. ⊓⊔ 8 See MacKenzie et al. (2013) for the general construction. 5 Dependency: arbitrary numbers of stable sets Theorem 3 restricted the number of stable sets to a power of two. By violating Lemma 5 to allow dominance across the exceptional sets, choices about inclusion of allocations in one exceptional set become dependent on these choices in other exceptional sets, reducing the possible number of stable sets. This section builds a six agent game with three stable sets to illustrate.8 Deﬁne the sets ˜E1 = {a1, b1, c1, d1} and ˜E2 = {a2, b2, c2, d2}, where 0 < δ < 1 6, 0 < α < 1 6, and a1 = 0, δ, δ + 2 3 (1 −3δ) , δ + 1 3 (1 −3δ) , 0, 0 b1 = 0, δ, δ + 1 3 (1 −3δ) , δ + 2 3 (1 −3δ) , 0, 0 c1 = 0, δ + 1 2 + α (1 −3δ) , δ, δ + 1 2 −α (1 −3δ) , 0, 0 d1 = 0, δ + 1 2 + α (1 −3δ) , δ + 1 2 −α (1 −3δ) , δ, 0, 0 a2 = 0, δ, 0, 0, δ + 2 3 (1 −3δ) , δ + 1 3 (1 −3δ) b2 = 0, δ, 0, 0, δ + 1 3 (1 −3δ) , δ + 2 3 (1 −3δ) c2 = 0, δ + 1 2 (1 −3δ) , 0, 0, δ, δ + 1 2 (1 −3δ) d2 = 0, δ + 1 2 (1 −3δ) , 0, 0, δ + 1 2 (1 −3δ) , δ . Three parameters therefore deﬁne the exceptional sets. As before, δ ensures that excep- tional allocations are split between three lesser agents. Now, as agent i = 1 receives Three parameters therefore deﬁne the exceptional sets. As before, δ ensures that excep- tional allocations are split between three lesser agents. Now, as agent i = 1 receives 3 1006 S. MacKenzie et al. resources in both ˜E1 and ˜E2, it has strict preferences across exceptional sets. The new parameter, α, overturns Lemma 5 to establish dominance relations across the excep- tional sets. Its upper bound of 1 6 reﬂects the requirement that agent 2 (resp. 3) gain from the transition from b1 to d1 (resp. where ¯q is as deﬁned in Eq. 9. where ¯q is as deﬁned in Eq. 9. where ¯q is as deﬁned in Eq. 9. Theorem 4 The six agent pillage game deﬁned by Eqs. 1 and 13 yields exactly thre stable sets, t0, a1, b1, a2, b2 , t0, a1, b1, c2, d2 and t0, c1, d1, a2, b2 . Theorem 4 The six agent pillage game deﬁned by Eqs. 1 and 13 yields exactly three stable sets, t0, a1, b1, a2, b2 , t0, a1, b1, c2, d2 and t0, c1, d1, a2, b2 . Theorem 4 The six agent pillage game deﬁned by Eqs. 1 and 13 yields exactly three stable sets, t0, a1, b1, a2, b2 , t0, a1, b1, c2, d2 and t0, c1, d1, a2, b2 . The proof uses the following lemmas: The proof uses the following lemmas: Lemma 6 For each i ∈{1, 2}, ai di bi ci ai. Lemma 6 For each i ∈{1, 2}, ai di bi ci ai. Lemma 7 No allocation in ˜E1 is dominated by an allocation in ˜E2. Lemma 8 Neither a2 nor b2 is dominated by an allocation in ˜E1. Lemma 9 Neither c2 nor d2 is dominated by either a1 or b1. Lemma 10 Across exceptional sets, the following hold: c1 c2, d1 c2, c1 d2 and d1 d2. Figure 3 depicts the dominance relations established by Lemmas 6 to 10. Proof of Theorem 4 As before, the core is t0, which must belong to any stable set. Similarly, by deﬁnition of v0 (C, x) and ¯q, the remaining set of allocations undomi- nated by it are ˜E1 and ˜E2. Thus, any stable set must be the union of t0 and sets stable on the restriction of the allocation space to ˜E1 and ˜E2. Lemmas 6–10, which establish all dominance relations among and between the members of ˜E1 and ˜E2, make the task of ﬁnding stable sets on these restricted spaces trivial. As previously, for each i = 1, 2, ai pairs with bi and ci with di: a stable set including one of a pair but not the other will—by Lemma 6—violate either internal stability or external stability. First consider including a1 and b1, which dominate c1 and d1, but do not dominate allocations in ˜E2. 5 Dependency: arbitrary numbers of stable sets from a1 to c1), as it did previously. Finally, two further bounds are added to q (in addition to inequality 3), also assisting in establishing dominance relations across exceptional sets: 1 2 (1 −3δ) + δ + 3q > 1 2 (1 −3δ) + 2δ + 2q ⇒q > δ; (11) (1 −3δ) + 2δ + 2q ≥δ + 3q ⇒1 −2δ ≥q; (12) (11) (12) Otherwise, the power function’s base remains deﬁned by Eq. 1 and v0 (C, x) by Eq. 9. The remaining vi (C, x), for all i > 0, are then expressed symmetrically as Otherwise, the power function’s base remains deﬁned by Eq. 1 and v0 (C, x) by Eq. 9. The remaining vi (C, x), for all i > 0, are then expressed symmetrically as vi (C, x) = ¯q if {1, 2, 3} ⊆C ∧x ∈˜E1 ∨ {1, 4, 5} ⊆C ∧x ∈˜E2 q otherwise ;(13) where ¯q is as deﬁned in Eq. 9. Thus, as before, either {a2, b2} or {c2, d2} may be added to the candidate stable set, completing it. This produces the theorem’s ﬁrst two stable sets. 123 1007 Pillage games with multiple stable sets Fig. 3 Dominance within and between ˜E1 and ˜E2 Fig. 3 Dominance within and between ˜E1 and ˜E2 Now include c1 and d1 instead. This excludes c2 and d2, forcing the inclusion of a2 and b2, the theorem’s third stable set. ˜ Finally, no other stable sets are possible: t0 dominates all allocations outside ˜E1 and ˜E2. Within these exceptional sets, a1 must either be included or excluded, both of which have been considered above. ⊓⊔ 6 A three agent counter-example with multiple goods While Jordan noted that domination is asymmetric in the original pillage environment, Rowat (2009) showed that it may be symmetric in a multi-good pillage game. This allows construction of exceptional sets of two allocations, each dominating the other. Economically, the multi-good pillage game allows one to model trade as a form of pillage with an empty lose set, allowing analysis of trade in the absence of externally enforced property rights. The objects of the single good pillage game generalise naturally to the case of m > 1 goods, each in unitary endowment. The set of allocations becomes X = (x1, . . . , xn) : i x1i = · · · = i xmi = 1, 0 ≤x ji ≤1 ⊂ Rmn , a subset of Rmn. The allocation of each agent i is xi = (x1i, . . . , xmi), where 0 ≤ x ji ≤1 for each i and j. Allocations of the goods are vectors, x = (x1, . . . , xn); as inequality no longer provides a complete order, they are compared according to a preference relation, ≿i, deﬁned on Rm +; denote strict preference by ≻i and indifference by ∼i. The axioms deﬁning coalitional power extend naturally to the multi-goods environ- ment, by replacing the original x ∈Rn with x ∈Rmn. The deﬁnition of dominance remains unchanged, but the win and lose sets are now formed on the basis of strict 12 3 3 1008 S. MacKenzie et al. Fig. 4 An example of cyclicity: a b a preference rather than strict inequality. Determining whether x y thus requires con- struction of W ≡ i xi ≻i yi and L ≡ i yi ≻i xi . The deﬁnition of stability is unchanged. Fig. 4 An example of cyclicity: a b a Fig. 4 An example of cyclicity: a b a Fig. 4 An example of cyclicity: a b a preference rather than strict inequality. Determining whether x y thus requires con- struction of W ≡ i xi ≻i yi and L ≡ i yi ≻i xi . The deﬁnition of stability is unchanged. Figure 4 illustrates a two agent counter-example to acyclicity. To form multiple stable sets, add a zeroth agent deﬁned by To form multiple stable sets, add a zeroth agent deﬁned by To form multiple stable sets, add a zeroth agent deﬁned by u0 (x10, x20) = (x10) 1 2 (x20) 1 2 ; v0 (C, x) = 7. u0 (x10, x20) = (x10) 1 2 (x20) 1 2 ; v0 (C, x) = 7. u0 (x10, x20) = (x10) 1 2 (x20) 1 2 ; v0 (C, x) = 7. These values again maintain the power dichotomy: 2q +1 < v0 (C, x) = 2 ¯q +1 = 7, for all C ⊆I and x ∈X. Theorem 5 In the three agent, multigood pillage game deﬁned in this section, the core is t0 , and there are exactly two stable sets, t0, a and t0, b . Proof First show that t0 is undominated: Proof First show that t0 is undominated: π {0} , t0 = 7 + 1 = 8 > 2q = π {1, 2} , t0 . To show that no other allocation is undominated, we have already eliminated a and b, which dominate each other. This leaves X\ t0, a, b , all of which are dominated by t0. Consider x ∈X\ t0, a, b with x1, x2 ̸= 0. Then: π ({0} , x) = 2 7 (1 −x11 −x12) + 5 7 (1 −x21 −x22) + 7 > 7 > 2 7 (x11 + x12) + 5 7 (x21 + x22) + 2q = π ({1, 2} , x) If either x1 or x2 sets either component to zero, xmi = 0, then the coalition opposing the zeroth agent is even weaker. Finally, we show that the game has the two stable sets stated in the theorem. As before, the core must belong to any stable set. Once t0 is included, only the two named allocations must be considered. As they dominate each other, both cannot be without violating IS; for the same reason, including either satisﬁes ES. ⊓⊔ 6 A three agent counter-example with multiple goods In it, a = 19 20, 7 25 , 1 20, 18 25 and b = 1 20, 18 25 , 19 20, 7 25 , with ˆE ≡{a, b} . Agents’ prefer- ences are represented by the utility functions u1 (x11, x21) = (x11) 1 3 (x21) 2 3 u2 (x12, x22) = (x12) 2 3 (x22) 1 3 . In the ﬁgure, ∼i denotes their indifference curves. Their power is described by an extension of the majority pillage game’s power function (Rowat 2009): In the ﬁgure, ∼i denotes their indifference curves. Their power is described by an extension of the majority pillage game’s power function (Rowat 2009): π (C, x) = i∈C 2 7x1i + 5 7x2i + vi (C, x) ; (14) (14) and vi (C, x) = ¯q {1, 2} ⊆C and x ∈ˆE q otherwise ; for 1 < q < ¯q = 3, maintaining the power dichotomy. The ﬁgure indicates the locus of allocations such that π ({1} , x) = π ({2} , x) by a dashed line. Thus, a1 ≻1 b1, b2 ≻2 a2, π ({1} , b) > π ({2} , b) and π ({2} , a) > π ({1} , a) so that a b a. This demonstrates symmetry, the minimal example of cyclicity. Intu- itively, goods contribute linearly to agents’ power, but non-linearly to utility, creating trade-offs between power and utility. 123 1009 Pillage games with multiple stable sets 7 Discussion The examples above all violate anonymity axiom AN, allowing powerful agents to dominate all but a small, ﬁnite set of allocations even if they initially hold none of the resource. The resulting multiplicity, while unexpected, is practically disappoint- ing: pillage games’ independence of game forms make them appealing for analyses of environments with weak institutions; however, as intrinsic asymmetries between agents are likely to be a feature of empirical work, this paper’s link between violat- ing axiom AN and multiple stable sets suggests that pillage games may not offer as precise predictions as previously hoped. Selten (1987) discussed the interpretation of set-valued solutions in empirical cooperative game theory more generally; Leonard (2010, pp. 244, 260–262) presented early concerns about testable implications of the stable sets. Nevertheless, it remains the case that stable sets of pillage games—when they exist—are ﬁnite, and the worst example that we have been able to construct only produces exponentially many stable sets as a function of the number of agents. 12 12 3 3 1010 S. MacKenzie et al. The introduction compared the classic majority game in characteristic function form to the majority pillage game, showing that the ability to use resource holdings to break ties arising purely from coalitional membership reﬁned the stable sets of the former. Other games in characteristic function form may be similarly extended when the characteristic function, v (C), is superadditive, so that v C ∪C′ ≥v (C)+v C′ for disjoint coalitions C and C′. Consider the n = 3 veto-power or market game in Lucas (1992), in which v ({1, 2, 3}) = v ({1, 2}) = v ({2, 3}) = 1; v ({1, 3}) = v ({1}) = v ({2}) = v ({3}) = 0. To create a pillage game extension, replace the power function in Eq. 1 with π (C, x) = v (C) + i∈C xi; where v (C) is deﬁned above, ∈ 0, minC,C′ v (C) −v C′ s.t.v (C) ̸= v C′ . Thus, ensures that resource holdings only inﬂuence dominance when opposing coalitions alone are evenly balanced. 9 For the classic game, see Lucas (1992). For the pillage game, π {2} , t2 = > 0 = π {1, 3} , t2 ; for any x ∈X such that x2 < 1, i = 2 can successfully pillage any agent holding resources by allying itself with the other. 10 See Kerber and Rowat (2014) for a less condensed version of this sort of reasoning. Appendix Proof of Lemma 3 Proceed by contradiction, assuming that at least one such alloca- tion, f , exists. Each such f therefore induces an open triangle, T ( f ), on Z such that each zero agent has strictly more wealth than in f . As allocations in T ( f ) cannot be dominated by f , ES requires either that they belong to a stable set, or that they are dominated by another member of a stable set. As, by Jordan (2006, Theorem 2.9), stable sets are ﬁnite, and T ( f ) has positive area, there must exist at least one member of a stable set that individually dominates a positive measure of T ( f ). We now attempt to ﬁnd such an allocation. Exceptional allocations, in E, cannot dominate any z ∈T ( f ) ⊆Z: the attempt would set L = {0, 00, 000}, and therefore π (L, z) = 31 > 19 ≥π (W, z). For x to dominate a z in T ( f ), x must belong to X\E. There are only three ways in which such domination may occur: 1. Coalitions in which fewer zero agents belong to W than do to L. No x can dominate a z in this way: as all three lesser agents are collectively less powerful than a single zero agent, x z in this way requires that W include more zero agents than L does. 2. Coalitions in which the same number of zero agents belong to W and L. As z is in Z it shares the resource between the zero agents; for any distinct x, at least one such agent must belong to L; as there are only three zero agents, the coalitions cannot each contain two or more of them. Thus, the two coalitions each contain exactly one zero agent, with the third remaining indifferent. The third agent’s indifference restricts the possible x that may dominate z to, at most, lines along Z; a ﬁnite number of these cannot cover T ( f ). 3. Coalitions in which more zero agents belong to W than do to L. As in the previous case, at least one zero agent must belong to L; as there are only three zero agents, it follows that two must belong to W. 7 Discussion In both the classic market game and its pillage extension, the core is the singleton t2.9 In the classic market game, any continuous curve in the simplex from t2 to any (x1, 0, 1 −x1) satisfying the Lipschitz condition y2 < z2 ⇒(y1 ≥z1) ∧(y3 ≥z3) (15) (15) for all y and z on the curve is a stable set. In the pillage extension, the unique stable set is t2, s12, s13, s23, q2 , where q2 ≡ 1 4, 1 2, 1 4 : t2 must be included; all x such that x2 > 1 2 are dominated by t2, so excluded; none of the remaining allocations dominate s12, s23 and q2, so they must belong to any stable set; the only allocation that remains undominated is s13, which must therefore be included in any stable set.10 Thus, while there is a unique stable set in the pillage market game, it is not—unlike that of the introduction’s majority pillage game—a subset of one of the inﬁnitely many stable sets in the market game: in breaking coalitional ties in dominance, the pillage extension introduces dominance into previously internally stable sets, forcing the removal of some such allocations which may have contributed to external stability, forcing—in this example—the inclusion of s12 and s23. Finally, note that the construction based on the power function in Eq. 1 breaks down as n →∞, as inequality 9 would require ¯q →∞as well. For this to hold for any n requires something like a power function that sets q to be a decreasing function of n, with limn→∞qn = 0; this, in turn, disables the requisite dominance among the 12 123 1011 Pillage games with multiple stable sets exceptional allocations in E. Thus, the constructions presented here do not extend to inﬁnite agents, and cannot be used to generate inﬁnite numbers of stable sets. exceptional allocations in E. Thus, the constructions presented here do not extend to inﬁnite agents, and cannot be used to generate inﬁnite numbers of stable sets. Acknowledgments We are grateful to Jim Jordan, the audience at SAET 2013, and two referees for com- ments, and to the EPSRC for ﬁnancial support (EP/J007498/1); Rowat thanks Birkbeck for its hospitality. 7 Discussion 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. Appendix When this is so, any x such that x z implies that x f : the set of zero agents beneﬁting from a transition from f to x contains at least the two beneﬁting from the x z operation; by deﬁnition, as f /∈E, the two zero agents are able to overcome all other agents. The presence of both f and x in a candidate stable set therefore violates IS. These cases are exhaustive: as z ∈Z, any distinct x ensures that at least one zero agent belongs to L. To summarize: any f /∈E ∪Z belonging to a stable set is dominated by a z ∈ T ( f ); ES therefore requires the existence of at least one x in any stable set containing f that dominates x; the three cases above show that dominating T ( f ), an object of 12 3 1012 S. MacKenzie et al. positive measure, therefore requires the support of two of the zero agents; this, in turn, causes x f , violating IS. ⊓⊔ positive measure, therefore requires the support of two of the zero agents; this, in turn, causes x f , violating IS. ⊓⊔ Proof of Lemma 5 A possible move from y ∈E j to x ∈Ei generates W = Pi and L = Pj. Thus, x y is equivalent to π (Pi, y) = i∈Pi q = 3q < 10 = j∈Pj y j + ¯q = π Pj, y a contradiction. ⊓⊔ Proof of Lemma 6 Comparison of any two allocations in ˜E1 does not allow {1, 2, 3} to be a subset of either W or L. Thus, the second clause of Eq. 13 ensures that vi (C, x) = q for any x ∈˜E1. (Similarly, for ˜E2, comparison splits agents {1, 4, 5}, so that vi (C, x) = q for any x ∈˜E2.) Thus, the result is ensured by constraint 3, which ensures that—for any such allocation and coalitions—any two-member coalition is more powerful than any one-member agent. ⊓⊔ Proof of Lemma 7 Let x ∈˜E2 and W ≡{i ∈I\|xi > c1,i} and L ≡{i ∈I\|xi < c1,i}. Then W and L do not depend on the choice of x as all four possibilities yield the same results: W = {4, 5} and L = {1, 2, 3}. Appendix Formally, this requires that Proof of Lemma 9 For c2 or d2 to be dominated by either a1 or b1, coalition {1, 4, 5} must be weaker in c2 or d2 than is coalition {2, 3}. Formally, this requires that π ({1, 4, 5} , c2) = 3¯q + 1 < 2q = π ({2, 3} , c2) ; contradiction as ¯q > q. ⊓⊔ ⊓⊔ a contradiction as ¯q > q. a contradiction as ¯q > q. Proof of Lemma 10 We prove only the ﬁrst dominance relationship; the rest follow by symmetry. Deﬁning W ≡ i ∈I c1,i > c2,i and L ≡ i ∈I c1,i < c2,i sets W = {1, 2, 3} and L = {4, 5}. The coalitions’ powers are therefore π (W, c2) = δ + 1 2 (1 −3δ) + 3q > 2δ + 1 2 (1 −3δ) + 2q = π (L, c2) ; where the inequality follows from constraint 11. This is equivalent to c1 c2, the result. ⊓⊔ Appendix Therefore to show that c1 is not dominated by any allocation from ˜E2, one computes the relevant powers: π (W, c1) = 2q < 1 + 3¯q = π (L, c1) which holds as q < ¯q. By symmetry, the same argument holds for d1. which holds as q < ¯q. By symmetry, the same argument holds for d1. For a1 and b1, two cases are identiﬁed. First consider y ∈{c2, d2} so that W ≡ i ∈I yi > a1,i and L ≡ i ∈I yi < a1,i . This yields W = {1, 4, 5} and L = {2, 3}, so that which holds as q < ¯q. By symmetry, the same argument holds for d1. For a1 and b1, two cases are identiﬁed. First consider y ∈{c2, d2} so that W ≡ i ∈I yi > a1,i and L ≡ i ∈I yi < a1,i . This yields W = {1, 4, 5} and L = {2, 3}, so that π (W, a1) = 3q + δ ≤(1 −δ) + 2q = π (L, a1) whose inequality is equivalent to constraint 12. Again, symmetry extends the result to b1. The second case considers elements z ∈{a2, b2}. Deﬁning W and L as above sets W = {4, 5} and L = {2, 3}. As this case differs from the previous only by removing i = 1 from W—reducing π (W, a1) to 2q—coalitional monotonicity axiom WC guarantees that a1 is not dominated by any element from ˜E2. Symmetry again applies the result to b1. ⊓⊔ ⊓⊔ ⊓⊔ Proof of Lemma 8 The sections of the proof of Lemma 7 corresponding to alloca- tions a1 and b1 establish this result as well if allocations in ˜E1 are replaced by the corresponding allocations in ˜E2 and vice versa, and if coalition {2, 3} is replaced by coalition {4, 5}, and coalition {1, 4, 5} by coalition {1, 2, 3}. ⊓⊔ ⊓⊔ 123 1013 Pillage games with multiple stable sets Proof of Lemma 9 For c2 or d2 to be dominated by either a1 or b1, coalition {1, 4, 5} must be weaker in c2 or d2 than is coalition {2, 3}. References Aumann RJ (1985) What is game theory trying to accomplish? In: Arrow K, Honkapohja S (eds) Frontiers of economics. Basil Blackwell, Oxford Aumann RJ (1985) What is game theory trying to accomplish? In: Arrow K, Honkapohja S (eds) Frontiers of economics. Basil Blackwell, Oxford Beardon AF, Rowat C (2013) Efﬁcient sets are small. J Math Econ 49(5):367–374 Jordan JS (2006) Pillage and property. J Econ Theory 131(1):26–44 Jordan JS (2009) Power and efﬁciency in production pillage games. Rev Econ Des 13(3):171–193 Jordan JS, Obadia D (2014) Stable sets in majority pillage games. Int J Game Theory 1–14. doi:10.100 s00182-014-0440-z s00182-014-0440-z Kerber M, Rowat C (2009) Stable sets in three agent pillage games. Working Paper 09–07. University of Birmingham, Department of Economics Kerber M, Rowat C (2011) A Ramsey bound on stable sets in Jordan pillage games. Int J Game Theory 40(3):461–466 Kerber M, Rowat C (2014) Sufﬁcient conditions for unique stable sets in three agent pillage games. Ma Soc Sci 69:69–80 Kwong R (2011) Google helps HTC in patent battle. In: Financial times Leonard R (2010) von Neumann, Morgenstern and the creation of game theory: from chess to social science, Leonard R (2010) von Neumann, Morgenstern and the creatio 1900–1960. Cambridge University Press, Cambridge ( ) , g 1900–1960. Cambridge University Press, Cambridge Lucas WF (1971) Some recent developments in n-person game theory. SIAM Rev 13(4):491–523 Lucas WF (1992) von Neumann–Morgenstern stable sets. In: Aumann RJ, Hart S (eds) Handbook of theory with economic applications, vol 1. Elsevier, Amsterdam, pp 543–590 Chap. 17 ucas WF (1992) von Neumann–Morgenstern stable sets. In: Aumann RJ, Hart S (eds) Handbook of game ucas W ( 99 ) vo Neu a o ge ste stab e sets. : u a J, a t S (eds) a dboo o g theory with economic applications, vol 1. Elsevier, Amsterdam, pp 543–590 Chap. 17 MacKenzie S, Kerber M, Rowat C (2013) Pillage games with multiple stable sets. Discussion Paper 13–0 University of Birmingham, Department of Economics Milford P, Decker S (2011) HTC sues apple using google patents bought last week as battle escalates. In: Bloomberg Rowat C (2009) An exchange economy with pillage. mimeo xton D (2011) Strictly monotonicmultidimensional sequences and stable sets in pillage games. J Comb Theory Ser A 118(2):510–524 Selten R (1987) Equity and coalition bargaining in experimental threeperson games. In: Roth AE (ed) Laboratory experiments in economics: six points of view. Cambridge University Press, Cambridge, pp 42–98 Chap. 3 pp p Shapley LS (1959) A solution containing an arbitrary closed component. In: Tucker AW, Luce RD (eds) Contribution to the theory of games. Vol. IV. Annals of Mathematical Studies 40. Princeton University Press, Princeton, pp. 87–93 von Neumann J, Oskar M (1953) Theory of games and economic behavior, 2nd edn. Princeton University Press, Princeton 12 3
https://openalex.org/W4239054646	https://osf.io/b7h4a/download	English	null	Crossing Rivers Problem Solution with Breadth-First Search Approach	null	2,017	cc-by	2,392	Crossing Rivers Problem Solution with Breadth-First Search Approach R Ratnadewi1, E M Sartika1, R Rahim2, B Anwar3, M Syahril3, and H Winata3 1Electrical Engineering, Maranatha Christian University, Jalan Prof. Drg. Surya Sumantri No.65, Bandung 40164, Indonesia 2Department of Health Information, Akademi Perekam Medik dan Infokes Imelda, Jl. Bilal Ujung Medan 20116, Indonesia 3Department System Computer and Information System, STMIK Triguna Dharma, Jl. Jenderal Abdul Haris Nasution No.73, Medan 20219, Indonesia usurobbi85@zoho.com Abstract. Crossing river problem is one of the problems state and space that represents a state by using a rule to define the problem, to find solutions to state and space problem can use breadth first search method by conducting a search based on rule and condition given, it is expected with this research crossing river problem could be solved properly and optimally and also this research can be the basis of solving other similar problems by using breadth first search method, to model the situation and the problem space using Breadth First Search made an application with a programming language. 3. Breadth-First Search In the Breadth-First Search method, all nodes at level n will be visited first before visiting nodes at level n + 1 [3] [6] [7]. The search starts from the root node and continues to the 1st level from left to right, then moves to the next step Moreover so on until the discovery of a solution [5] [6] [7] The advantages of this breadth-first search method are [6]: The advantages of this breadth-first search method are [6]: a. Will not get stuck. ere is one solution, then a breadth-first search will find it. Also, if there is more than one solution, i i l i ill b f d b. If there is one solution, then a breadth-first search will find it. Also, if there is more than one solution, then a minimum solution will be found. The weakness of this method [6]: a. Requires enough memory, because it stores all the nodes in one tree. b. It takes a long time because it will test n level to get the solution at level (n + 1). 2. State and Space Problem 2. State and Space Problem Systems that use artificial intelligence will try to provide output in the form of a solution of a problem ased on a collection of existing knowledge [1] [4] [5] AI SYSTEM Problem Knowledge Inference Engine Solution Figure 1. Systems that use AI p Systems that use artificial intelligence will try to provide output in the form of a solution of a problem based on a collection of existing knowledge [1] [4] [5] AI SYSTEM Problem Knowledge Inference Engine Solution Figure 1. Systems that use AI Inference Engine Figure 1. Systems that use AI Input given to systems using artificial intelligence is a problem, on the system must be equipped with a set of existing knowledge in the knowledge base. The system must have an inference engine to be able to draw inferences based on fact or knowledge. The given output is a problem solution as a result of inference [1] [4] [6]. In general, to build a system capable of solving problems, four things need to be considered: a. Define the problem appropriately. This definition includes precise specifications of the initial state and the expected solution. a. Define the problem appropriately. This definition includes precise specifications of the initial state and the expected solution. p b. Analyze the problem and look for some appropriate problem-solving techniques. c. Represents the necessary knowledge to solve the problem. c. Represents the necessary knowledge to solve the problem. epresents the necessary knowledge to solve the p c. Represents the necessary knowledge to solve the problem. d. Choosing the best problem-solving technique 1. Introduction A state is a representation of a state at a time or decryption of system configuration [1] [2]. State and space are all possible states and usually described as networks with vertices being the state and edge representing possible changes [1]. The representation of state and space allows the formal definition of a problem as a matter of changing status by using a set of operators (rules) and also defining a problem as a search i.e. searching the path in state and space from the initial state to the goal state [1] [2]. Problem state and space [1] is a common problem in the field of Artificial Intelligence, Searching Process and also Machine Learning, some examples of state and space problems such as 8-Puzzle, Cannibals and Missionaries Problem and also Water Jug Problem, and on the results of this study conducted experiments with create a problem called Crossing River Problem. The illustration of this Crossing River Problem is as follows, a family will cross to the other side of the river along with police and criminals, as for the variables of crossing rivers problem consisting of father, mother, son 1, son 2, daughter 1, daughter 2, criminals and police. This problem will be modified so that it can be used to solve all problems that are almost similar to this issue. This issue could be resolved by using several stages, the first stage starts from the identification of the state space, by declaring this problem by determining the variables contained in this problem. Then, specify the rules included in the problem of initial conditions and conditions of purpose, crossing rivers problem solved using the help of tracking tree. Possible conditions are described in the tree structure starting with the first state as the root of the tree. The process is continued by describing the next state that could be created from that state. The process continues until a goal state is obtained or no new conditions can be developed. Breadth-first search (BFS) search method is used to perform checking procedures whether every action carried out on any circumstances violates pre-made rules [3]. This procedure ensures that new conditions to be established do not violate existing rules so that the desired goal state can be obtained, to facilitate the experiment of using Breadth-First Search algorithm, an application was created to model the BFS simulation process to solve the problem created. 4. Result and Discussion Breadth First Search algorithm completion on crossing rivers problem. The illustration of this problem is as follows, a family will cross to the other side of the river along with police and criminals, as for the variables of crossing rivers problem consisting of father, mother, son 1, son 2, daughter 1, daughter 2, criminals and police, Table of variables and rules can be seen in table 1 and table 2. Table 1. Variable Crossing Rivers Problem Variable ID Variable Name Could Raft Driver AY Father True IB Mother True AP1 Daughter 1 False AL1 Son 1 False AL2 Son 2 False AP2 Daughter 2 False PL Police True PJH Criminal False Table 1. Variable Crossing Rivers Problem Table 2. Crossing Rivers Problem Rules Table 2. Crossing Rivers Problem Rules No Rule 1 Father will beat Girl 1 if Mother is not in place 2 Father will beat Girl 2 if Mother is not there. 3 Mother will hit Son 1 if Dad is not in place 4 Mother will beat Son 2 if Dad is not in place 5 The criminals will hit Sons 1 if the Police are not in place 6 Criminals will hit Sons 2 if Police are not in place. 7 Criminals will beat Girl 1 if Police are not in place 8 Criminals will beat Girl 2 if Police are not in place. 9 The criminals will hit Dad if the Police are not in place 10 Criminals will hit mother if the Police are not in place Any state of the Crossing Rivers Problem could describe by (criminals, police, children, father and mother) and raft positions. The initial state and state of destination could describe as: Any state of the Crossing Rivers Problem could describe by (criminals, police, children, father and mother) and raft positions. The initial state and state of destination could describe as: A. Initial state:  Left area = (1, 1, 1, 1, 1, 1, 1, 1).  Right area = (0,0,0,0,0,0,0,0)  The raft position is on the left. That is, there are fathers, mothers, sons, daughters, police, and criminals in the left.  The raft position is on the left. That is, there are fathers, mothers, sons, daughters, police, and criminals in the left. B. Goal state:  Left area = (0, 0, 0, 0, 0, 0, 0, 0). 4. Result and Discussion  Right area = (1,1,1,1, 1,1,1,1)  The position of the raft is on the right. That is, there are fathers, mothers, sons, daughters, police, and criminals in the right area B. Goal state:  Left area = (0, 0, 0, 0, 0, 0, 0, 0).  Right area = (1,1,1,1, 1,1,1,1) The position of the raft is on the right. That is, there are fathers, mothers, sons, daughters, polic and criminals in the right area The process of development and search is extended by forming a new state from existing branches until the solution is achieved. The process of establishing a new state also performs checking procedures whether the situation has been previously set up in the tree. If it is already there, a new branch is not formed, and if it is not, then a new branch is formed. If there is no branch that can be developed again and it has not found a solution, then the problem being solved has no solution, the application of Breadth First Search algorithm on the problem of Crossing Rivers Problem is to find a solution how to move all the variables on the left to the right with Follow the predefined rule. From the variables and rules contained in the Crossing Rivers Problem, it can be concluded that some action can be done by keeping the existing regulations. The actions that can be done can be seen in Table 3. Table 3. Solution Crossing Rivers Problem Left Side Action Solution Right Side (1,1,1,1,1,1,1,1) - (0,0,0,0,0,0,0,0) (1,1,1,1,1,1,0,0) Police and Criminals cross to the right. (0,0,0,0,0,0,1,1) (1,1,1,1,1,1,0,1) The police crossed to the left. (0,0,0,0,0,0,1,0) (0,1,1,1,1,1,0,0) Police and Son 1 crossed to the right. (1,0,0,0,0,0,1,1) (0,1,1,1,1,1,1,1) Police and Criminals cross to the left. (1,0,0,0,0,0,0,0) (0,0,1,1,0,1,1,1) Father and Son 2 cross to the right. (1,1,0,0,1,0,0,0) (0,0,1,1,1,1,1,1) Dad crossed to the left. (1,1,0,0,0,0,0,0) (0,0,1,1,0,0,1,1) Mom and Dad crossed to the right. (1,1,0,0,1,1,0,0) (0,0,1,1,0,1,1,1) Mother crossed to the left. (1,1,0,0,1,0,0,0) (0,0,1,1,0,1,0,0) Police and Criminals cross to the right. (1,1,0,0,1,0,1,1) (0,0,1,1,1,1,0,0) Dad crossed to the left. (1,1,0,0,0,0,1,1) (0,0,1,1,0,0,0,0) Mom and Dad crossed to the right. (1,1,0,0,1,1,1,1) (0,0,1,1,0,1,0,0) Mother crossed to the left. (1,1,0,0,1,0,1,1) (0,0,0,1,0,0,0,0) Mother and Daughter 1 cross to right. (1,1,1,0,1,1,1,1) (0,0,0,1,0,0,1,1) Police and Criminals cross to the left. (1,1,1,0,1,1,0,0) (0,0,0,0,0,0,1,0) Police and Girl 2 cross to the right. (1,1,1,1,1,1,0,1) (0,0,0,0,0,0,1,1) The police crossed to the left. (1,1,1,1,1,1,0,0) (0,0,0,0,0,0,0,0) Police and Criminals cross to the right. 5. Conclusion After solving the problem of crossing rivers problem by using breadth-first search method can be concluded that a problem consists of variables and rules that can be utilized as an initial state to get the desired solution to achieve goal state, the problem used as an example in this paper could be used as a Preliminary solutions for various similar cases and application to computing systems will not be too difficult if the variables and rules already know. 4. Result and Discussion (1,1,1,1,1,1,1,1) The search for a solution in the tree tracing structure uses the search method of breadth-first search, so the solution (Table 3) obtained is the shortest and best option, some problem solving simulation Crossing River Problem made using programming language can be seen figure 2 and figure 3. Figure 2. Initial State from Cross River Problem Figure 2. Initial State from Cross River Problem Figure 2. Initial State from Cross River Problem Figure 2 shows the initial condition of the simulation used to solve the problem of state and space crossing river problem with information of variable, rule and settlement process obtained from Breadth First Search algorithm. Figure 3. Simulation State and Space Cross River Problem Figure 3. Simulation State and Space Cross River Problem Figure 3. Simulation State and Space Cross River Problem Figure 3 is a simulation process to solve the crossing river problem by performing the procedure accordance Breadth First Search algorithm. Figure 3 is a simulation process to solve the crossing river problem by performing the procedure accordance Breadth First Search algorithm. Figure 3 is a simulation process to solve the crossing river problem by performing the procedure accordance Breadth First Search algorithm. [1] J. C. Trigeassou and N. Maamri, "State space modeling of fractional differential equations and the initial condition problem," in 6th International Multi-Conference on Systems, Signals and Devices, Djerba, Tunisia , 2009. [1] J. C. Trigeassou and N. Maamri, "State space modeling of fractional differential equations and the initial condition problem," in 6th International Multi-Conference on Systems, Signals and Devices, Djerba, Tunisia , 2009. [2] B. Bonet and H. Geffner, "Planning as heuristic search," Journal Artificial Intelligence, vol. 129, no. 1-2, pp. 5-33, 201. [2] B. Bonet and H. Geffner, "Planning as heuristic search," Journal Artificial Intelligence, vol. 129, no. 1-2, pp. 5-33, 201. References [1] J. C. Trigeassou and N. Maamri, "State space modeling of fractional differential equations and the initial condition problem," in 6th International Multi-Conference on Systems, Signals and Devices, Djerba, Tunisia , 2009. [2] B. Bonet and H. Geffner, "Planning as heuristic search," Journal Artificial Intelligence, vol. 129, no. 1-2, pp. 5-33, 201. [3] S. Beamer, K. Asanović and D. Patterson, "Direction-Optimizing Breadth-First Search," Scientific Programming, vol. 21, no. 3-4, pp. 137-148, 2013. [4] T. Kim and K. Chwa , "Parallel algorithms for a depth first search and a breadth first search," International Journal of Computer Mathematics, vol. 19, no. 1, pp. 39-54, 2007. [5] A. Kaur, P. Sharma and A. Verma, "A appraisal paper on Breadth-first search, Depth-first search and Red black tree," International Journal of Scientific and Research Publications, vol. 4, no. 3, pp. 1-3, 2014. [6] R. K. Ghosh and G. P. Bhattacharjee , "Parallel breadth-first search algorithms for trees and graphs," International Journal of Computer Mathematics, vol. 15, no. 1-4, pp. 255-268, 2007. [7] R. Zhou and E. A. Hansen, "Breadth-ﬁrst heuristic search," Journal Artificial Intelligence, vol. 170, no. 4-5, pp. 385-408, 2006.
https://openalex.org/W3017045547	https://europepmc.org/articles/pmc7291140?pdf=render	English	null	Consequences of Lmna Exon 4 Mutations in Myoblast Function	Cells	2,020	cc-by	12,968	cells cells Consequences of Lmna Exon 4 Mutations in Myoblast Function Déborah Gómez-Domínguez 1, Carolina Epifano 2, Fernando de Miguel 1,3, g p g Albert García Castaño 4, Borja Vilaplana-Martí 1, Alberto Martín 1, Sandra Amarilla-Quintana 5, Anne T Bertrand 6 , Gisèle Bonne 6 , Javier Ramón-Azcón 4,7, Miguel A Rodríguez-Milla 1 and Ignacio Pérez de Castro 1,* Albert García Castaño 4, Borja Vilaplana-Martí 1, Alberto Martín 1, Sandra Amarilla-Quintana 5, Anne T Bertrand 6 , Gisèle Bonne 6 , Javier Ramón-Azcón 4,7, Miguel A Rodríguez-Milla 1 and Ignacio Pérez de Castro 1,* 1 Instituto de Investigación de Enfermedades Raras, Instituto de Salud Carlos III, Ctra. Majadahonda-Pozuelo km2.2, E-28029 Madrid, Spain; d.gomez@isciii.es (D.G.-D.); ferdemiguelpe@gmail.com (F.d.M.); bvilaplana@isciii.es (B.V.-M.); almmartin@isciii.es (A.M.); rmilla@isciii.es (M.A.R.-M.) 2 Fundación Andrés Marcio, niños contra la laminopatía, C/Núñez de Balboa, 11, E-28001 Madrid, Sp cepifano@hotmail.com 3 3 Universidad Europea de Madrid, C/ Tajo, s/n, E-28670 Villaviciosa de Odón, Spain 4 Institute for Bioengineering of Catalonia (IBEC), C/Baldiri Reixac, 10-12, E-08028 Barcelona, Spain; agarciac@ibecbarcelona.eu (A.G.C.); jramon@ibecbarcelona.eu (J.R.-A.) 5 Fundación de Investigación HM Hospitales, Plaza del Conde Valle Suchil, 2, E-28015 Madrid, Spain; amarillaquintana@gmail.com q g 6 UMRS 974, Center of Research in Myology, Institut de Myologie, Sorbonne Université, INSERM, 75013 Paris, France; a.bertrand@institut-myologie.org (A.T.B.); g.bonne@institut-myologie.org (G.B 7 ICREA-Institució Catalana de Recerca i Estudis Avançats, 08010 Barcelona, Spain * Correspondence: iperez@isciii.es; Tel.: +34-918223188 1. Introduction Lamins are nuclear intermediate ﬁlaments proteins that form a meshwork underlying the inner side of the nuclear membrane [1]. Nuclear lamins have been associated with a wide spectrum of cellular functions including structural support, mechanosensing, chromatin organization, transcription regulation, replication, nuclear assembly, and nuclear pore complex activity [2–5]. Laminopathies are a group of human rare diseases mainly associated with diﬀerent mutations on the Lamin A/C gene (LMNA). Lamin A/C proteins are main components of the lamin. Laminopathies include at least 15 diﬀerent diseases that are divided in 4 diﬀerent categories depending on the aﬀected tissues: striated muscle, adipose tissue, peripheral nerves, or multiple tissues [6]. This last group includes Hutchinson–Gilford progeria syndrome (HGPS), which is characterized by aging-like symptoms emerging in childhood. g g The LMNA gene contains 12 exons and codes for 2 diﬀerent proteins, Lamin A and C. Both are members of the intermediate ﬁlament group of proteins, which are characterized for the presence of a short N-terminal domain, a central domain composed of four α-helical domains that are separated by three linker regions and a globular C-terminal region. To date, more than 400 mutations in LMNA have been reported in patients (http://www.umd.be/LMNA/). These mutations are widely distributed throughout the entire gene. Moreover, there is no association between the gene location of these mutations and the wide range of phenotypes included in this group of rare diseases. Two main hypotheses have been postulated to explain the development and progression of laminopathies [7]. One of them assumes that LMNA mutations provoke structural and mechanical changes that are responsible for the development of the disease, whereas the other relies on functional abnormalities induced by changes in gene expression patterns and diﬀerentiation programs. Unfortunately, little is known about the exact mechanisms underlying the eﬀects induced by the majority of the laminopathy-causing LMNA mutations. No curative treatment is currently available for any type of laminopathy. A number of pre-clinical studies have explored the therapeutic potential of diﬀerent compounds that modulate the Mechanistic Target Of Rapamycin Kinase (mTOR) pathway [8,9], the mitogen-activated protein kinase (MAPK) cascade [10,11], and the epigenetic regulator N-Acetyltransferase 10 (NAT10) [12], among others. Gene therapy approaches are being studied as well. In this regard, the most promising results have been obtained for HGPS by introducing frameshift mutations in the LMNA gene using CRISPR/Cas9 (Clustered Regularly Interspaced Short Palindromic Repeats/CRISPR associated protein 9) [13,14]. Received: 13 April 2020; Accepted: 16 May 2020; Published: 21 May 2020 Abstract: Laminopathies are causally associated with mutations on the Lamin A/C gene (LMNA). To date, more than 400 mutations in LMNA have been reported in patients. These mutations are widely distributed throughout the entire gene and are associated with a wide range of phenotypes. Unfortunately, little is known about the mechanisms underlying the eﬀect of the majority of these mutations. This is the case of more than 40 mutations that are located at exon 4. Using CRISPR/Cas9 technology, we generated a collection of Lmna exon 4 mutants in mouse C2C12 myoblasts. These cell models included diﬀerent types of exon 4 deletions and the presence of R249W mutation, one of the human variants associated with a severe type of laminopathy, LMNA-associated congenital muscular dystrophy (L-CMD). We characterized these clones by measuring their nuclear circularity, myogenic diﬀerentiation capacity in 2D and 3D conditions, DNA damage, and levels of p-ERK and p-AKT (phosphorylated Mitogen-Activated Protein Kinase 1/3 and AKT serine/threonine kinase 1). Our results indicated that Lmna exon 4 mutants showed abnormal nuclear morphology. In addition, levels and/or subcellular localization of diﬀerent members of the lamin and LINC (LInker of Nucleoskeleton and Cytoskeleton) complex were altered in all these mutants. Whereas no signiﬁcant diﬀerences were observed for ERK and AKT activities, the accumulation of DNA damage was associated to the Lmna p.R249W mutant myoblasts. Finally, signiﬁcant myogenic diﬀerentiation defects were detected in the Lmna exon 4 mutants. These results have key implications in the development of future therapeutic strategies for the treatment of laminopathies. Keywords: LMNA; laminopathy; CRISPR; nuclear envelope Keywords: LMNA; laminopathy; CRISPR; nuclear envelope Cells 2020, 9, 1286; doi:10.3390/cells9051286 www.mdpi.com/journal/cells www.mdpi.com/journal/cells Cells 2020, 9, 1286 2 of 21 1. Introduction The study of these mutants allowed us to ﬁnd critical information on the importance of Lmna exon 4 in myoblast biology, speciﬁcally in the integrity of the nucleus, the protein levels, and sub-cellular location of nuclear envelope proteins and the capacity to diﬀerentiate to myogenic ﬁbers. Given the fact that LMNA mutations are causally associated with several rare diseases, this work provides new and valuable information for a better understanding and future treatment of these diseases with no cure. 2. Materials and Methods Details of resources used in the research, including antibodies, cell culture media, reagents, and software, are listed in Table 1. 2.1. Cell Lines and 3D Model For the majority of our experiments, we used C2C12 myoblasts, which are myoblasts originally established from satellite cells derived from the thigh muscle of a female C3H murine donor following a crush injury [21]. C2C12 cells and selected clones with mutations in exon 4 of Lmna have been grown at 37 ◦C of temperature and 5% CO2 conditions. The culture medium of myoblasts was based on Dulbecco’s modiﬁed Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Three human myoblast cell lines, previously described [18], were used in this work. Two were controls isolated from 25 and 38 year old healthy individuals (C25CL48 and AB1079, respectively). A third myoblast cell line was isolated from a 3 year old L-CMD patient who carried a LMNA p.R249W mutation. Human myoblasts were grown at a 37 ◦C and 5% CO2 conditions, in 1 volume of medium 199 with four volumes of DMEM with 20% FBS and 1% penicillin/streptomycin, and supplemented with 25 µg/mL fetuin, 5 ng/mL human epidermal growth factor (hEGF), 0.5 ng/mL basic ﬁbroblast growth factor (bFGF), 5 µg/mL insulin, and 0.2 µg/mL dexamethasone. 3D muscle models were fabricated by photo-molding technique, as we have described previously [22] (Figure S2). Gelatin-based polymer solution (see Appendix A) was mixed with a C2C12 cell suspension to reach the ﬁnal cell density of 25 × 106 cell/mL. Brieﬂy, a 6 µL drop of cell-laden polymer was placed on a functionalized coverslip (Appendix A), and a microstructured silicone (PDMS) stamp was pressed lightly on top, ﬁlling the microchannels with the solution. The sizes of PDMS stamp consisted of grooves of 100 µm, which were 100 µm in height with and ridges of 100 µm. The hydrogel was photo-crosslinked using a Ultraviolet Product (UVP) Crosslinker (Model CL-1000L, 365 nm, 40 W, from Analytik Jena U.S., Upland, CA, USA) by the exposure of 30 s under UV light. After carefully removing the stamp, the micro-structured cell-laden hydrogels were incubated for 1 day with culture medium. Due to high cell density in the hydrogel, growth medium was changed to diﬀerentiation medium, based in DMEM high glucose, supplemented with 2% horse serum and 1% penicillin/streptomycin, after 1 day to induce diﬀerentiation into myotubes. 1. Introduction Besides these advances, we are still far from a cure for the majority of the laminopathies. More information is therefore needed about the molecular and cellular defects induced for each disease-associated LMNA mutation in order to obtain the therapeutic strategies that better ﬁt each case. p g Here, we explored whether mutations in Lmna exon 4 aﬀect the biology and function of mouse myoblasts. This exon codes for a linker domain located between the two central α-helical domains of Lamins A and C (Figure S1). Although up to 47 mutations in LMNA exon 4 have been reported with diﬀerent disorders linked to this gene (http://www.umd.be/LMNA/), little is known about the cellular and molecular changes induced by them that are causally related with laminopathy development and progression. Thus, although LMNA p.R249W is the most frequent mutation for LMNA-associated congenital muscular dystrophy, a severe type of laminopathy [15], and has been found to be related to abnormal nuclear morphology and myogenic diﬀerentiation, mislocalization of Lamin B, altered sense of microenvironment stiﬀness, and DNA damage [16–19], the causal connections between this mutation and the disease remain mainly unknown. On the other hand, because LMNA exons 3 and 5 are in frame, exon 4 skipping might be considered as a therapeutic strategy for all those laminopathies associated with LMNA exon 4 mutations. Although it has been reported that exon skipping could be successfully used for the treatment of diseases associated with LMNA mutations in exon 5 [20], the potential of this approach has not been tested yet for LMNA exon 4. Using mouse C2C12 myoblasts and CRISPR/Cas9, we generated a collection of Lmna exon 4 mutants that were classiﬁed as Lmna-null, Lmna-R249W, or carriers of Lmna exon4 in-frame deletions. 3 of 21 Cells 2020, 9, 1286 The study of these mutants allowed us to ﬁnd critical information on the importance of Lmna exon 4 in myoblast biology, speciﬁcally in the integrity of the nucleus, the protein levels, and sub-cellular location of nuclear envelope proteins and the capacity to diﬀerentiate to myogenic ﬁbers. Given the fact that LMNA mutations are causally associated with several rare diseases, this work provides new and valuable information for a better understanding and future treatment of these diseases with no cure. 2.2. Generation of Clones with Mutations in Lmna Exon 4 Using CRISPR/Cas9 Eric Schirmer 1:500 1:200 Anti-rabbit p44/42 Mitogen-Activated Protein Kinase 1/3 (Erk1/2)-137F5 Cell Signaling (Danvers, MA, USA) (#4695) 1:1000 Anti-rabbit phospho-p44/42 MAPK Erk1/2 Thr202/Tyr204 D13.14.4E Cell Signaling (Danvers, MA, USA) (#4370) 1:1000 Anti-mouse Akt (pan) 40D4 Cell Signaling (Danvers, MA, USA) (#2920) 1:1000 Anti-rabbit phospho-Akt (Ser473) D9E Cell Signaling (Danvers, MA, USA) (#4060) 1:1000 Anti-mouse phospho-histone H2A.X (Ser139) Millipore (Burlington, MA, USA) (05-636-I) 1:500 Anti-MYH7 (Myosin Heavy Chain 7) Thermoﬁsher (Waltham, MA, USA) (PA5-69132) 1:200 MF20 (Mouse Monoclonal Anti-Myosin Heavy Chain Antibody) DSHB Hybridoma (Iowa City, Iowa, USA) 1:50 Rhodamine phalloidin Thermoﬁsher (Waltham, MA, USA) (R415) 1:40 HRP-labelled anti-mouse secondary antibody GE Healthcare (Chicago, Illinois, USA) (NA931-1ML) 1:5000 HRP-labelled anti-rabbit secondary antibody GE Healthcare (Chicago, Illinois, USA) (NA934-1ML) 1:5000 Goat anti-mouse Alexa Fluor 488 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32723) 1:500 Goat anti-rabbit Alexa Fluor 594 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32740) 1:500 Goat anti-rabbit Alexa Fluor 488 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32731) 1:500 Goat anti-rabbit Alexa Fluor 488 (3D models) Thermoﬁsher (Waltham, MA, USA) (A11034) 1:200 CELL CULTURE MEDIA REAGENTS d PLASMIDS ANTIBODIES Name Source (identiﬁer) Western Blot Immuno-Fluorescence Anti-mouse Lamin A/C (E-1) Santa Cruz Biotechnology (Dallas, Texas, USA) (sc-376248) 1:3000 1:500 Anti-mouse α-tubulin Sigma-Aldrich (St. Louis, MI, USA) (T9026) 1:5000 Anti-rabbit lamin B1 Abcam (Cambridge, UK) (ab16048) 1:100 Anti-rabbit emerin (D3B9G) Cell Signaling (Danvers, MA, USA) (#30853) 1:100 Anti-mouse Sad1 And UNC84 Domain Containing 1 (SUN1) BBmSun1 IgC2b X12.11 Donated by Dr. Colin Stewart 1:50 1:50 Anti-rabbit SUN2 (Sad1 And UNC84 Domain Containing 2) 11905 Donated by Dr. 2.2. Generation of Clones with Mutations in Lmna Exon 4 Using CRISPR/Cas9 Eric Schirmer 1:500 1:200 Anti-rabbit p44/42 Mitogen-Activated Protein Kinase 1/3 (Erk1/2)-137F5 Cell Signaling (Danvers, MA, USA) (#4695) 1:1000 Anti-rabbit phospho-p44/42 MAPK Erk1/2 Thr202/Tyr204 D13.14.4E Cell Signaling (Danvers, MA, USA) (#4370) 1:1000 Anti-mouse Akt (pan) 40D4 Cell Signaling (Danvers, MA, USA) (#2920) 1:1000 Anti-rabbit phospho-Akt (Ser473) D9E Cell Signaling (Danvers, MA, USA) (#4060) 1:1000 Anti-mouse phospho-histone H2A.X (Ser139) Millipore (Burlington, MA, USA) (05-636-I) 1:500 Anti-MYH7 (Myosin Heavy Chain 7) Thermoﬁsher (Waltham, MA, USA) (PA5-69132) 1:200 MF20 (Mouse Monoclonal Anti-Myosin Heavy Chain Antibody) DSHB Hybridoma (Iowa City, Iowa, USA) 1:50 Rhodamine phalloidin Thermoﬁsher (Waltham, MA, USA) (R415) 1:40 HRP-labelled anti-mouse secondary antibody GE Healthcare (Chicago, Illinois, USA) (NA931-1ML) 1:5000 HRP-labelled anti-rabbit secondary antibody GE Healthcare (Chicago, Illinois, USA) (NA934-1ML) 1:5000 Goat anti-mouse Alexa Fluor 488 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32723) 1:500 Goat anti-rabbit Alexa Fluor 594 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32740) 1:500 Goat anti-rabbit Alexa Fluor 488 Thermoﬁsher Scientiﬁc (Waltham, MA, USA) (A32731) 1:500 Goat anti-rabbit Alexa Fluor 488 (3D models) Thermoﬁsher (Waltham, MA, USA) (A11034) 1:200 CELL CULTURE MEDIA REAGENTS and PLASMIDS C2C12 myoblasts/SIGMA (St. Louis, MI, USA) (91031101) pX459 vector (pSpCas9(BB)-2A-Puro)/Addgene (Watertown, MA, USA) (#62988) DMEM (Dulbecco’s modiﬁed Eagle’s medium) high glucose/Invitrogen (Waltham, MA, USA) (61965-026) Lipofectamine 3000/Invitrogen (Waltham, MA, USA) (L3000015) FBS (fetal bovine serum)/Sigma-Aldrich (St. Louis, MI, USA) (#F7524-500ML) Puromycin/InvivoGen (San Diego, CA, USA) (ant-pr-1) Penicillin/streptomycin/Lonza (Basel, Switzerland) (#DE17-602E) DNA polymerase/NZYTech (Lisbon, Portugal) (MB354) Medium 199/Invitrogen (Waltham, MA, USA) (41150020) MiSeq DNA/Illumina (San Diego, CA, USA) (MS-102-2003) Fetuin/Life Technologies (Waltham, MA, USA) (10344026) BCA system/Pierce (Waltham, MA, USA) (23227) hEGF/Life Technologies (Waltham, MA, USA) (PHG0311) ECL western blotting system/Thermo Fisher Scientiﬁc (Waltham, MA, USA) (Pierce 32106) bFGF/Life Technologies (Waltham, MA, USA) (PHG0026) Methanol/Panreac AppliChem (Barcelona, Spain) (#131091.1612) Insuline/Sigma (St. Louis, MI, USA) (91077C-1G) BSA (bovine serum albumin)/Sigma-Aldrich (St. Louis, MI, USA) (#A7906) Dexamethasone/Sigma (St. Louis, MI, USA) (D4902-100mg) PBS (phosphate-buﬀered saline)/Lonza (Basel, Switzerland) (#BE17-515Q) Horse serum/Thermoﬁsher (Waltham, MA, USA) (#26050-088) Goat serum/Sigma-Aldrich (St. Louis, MI, USA) (#G9023-10ML) SOFTWARE AND PLATFORMS Donkey serum/Sigma-Aldrich (St. Louis, MI, USA) (#D9663-10ML) CRISPResso (http://crispresso.pinellolab.partners.org/) Triton X-100/Sigma-Aldrich (St. Louis, MI, USA) (#X100-1L) TIDE (https://tide.deskgen.com/) Prolong Gold with 4′,6-Diamidino-2-Phenylindole (DAPI)/Cell Signalling Technology (Danvers, MA, USA) (P36935) ImageJ (U.S. National Institutes of Health, Bethesda, Maryland, USA) TBS (Tris-buﬀered saline)/Canvax Biotech (Cordoba, Spain) (BR0042) Prism 8 (GraphPad Software, Inc) Hoechst 33324/Thermo Fisher Scientiﬁc (Waltham, MA, USA) (H3570) 2.3. DNA Sequencing and Bioinformatic Analysis 2.2. Generation of Clones with Mutations in Lmna Exon 4 Using CRISPR/Cas9 All single guide RNAs (sgRNAs) were designed to target exon 4 of the LMNA gene using the Breaking Cas Design tool (https://bioinfogp.cnb.csic.es/tools/breakingcas/) [23]. Three sgRNAs for the Lmna exon 4 (sg10, sg11, and sg12; Table 2) were cloned in the pX459 vector. The C2C12 cells were transfected, using Lipofectamine 3000, with the pX459-sgRNA vector with the corresponding template (ssODN) in low (30 pmol) or high (300 pmol) doses. Templates included forms mutated only in the Protospacer Adjacent Motif (PAM) (mut) or mutated in the PAM and in the Protospacer (mut2) (Table 2). After 48 hours of transfection, puromycin (2 µg/mL) was added for 5 days. The selected cells were grown until a pool was obtained per condition. The clones were picked up after seeding at low density of the pools. The pools and clones were expanded to allow the complete analysis. For the control samples, we followed the same process but using an empty px459 vector and no templates. Cells 2020, 9, 1286 4 of 21 Table 1. Key resources used in this work. Table 1. Key resources used in this work. ANTIBODIES Name Source (identiﬁer) Western Blot Immuno-Fluorescence Anti-mouse Lamin A/C (E-1) Santa Cruz Biotechnology (Dallas, Texas, USA) (sc-376248) 1:3000 1:500 Anti-mouse α-tubulin Sigma-Aldrich (St. Louis, MI, USA) (T9026) 1:5000 Anti-rabbit lamin B1 Abcam (Cambridge, UK) (ab16048) 1:100 Anti-rabbit emerin (D3B9G) Cell Signaling (Danvers, MA, USA) (#30853) 1:100 Anti-mouse Sad1 And UNC84 Domain Containing 1 (SUN1) BBmSun1 IgC2b X12.11 Donated by Dr. Colin Stewart 1:50 1:50 Anti-rabbit SUN2 (Sad1 And UNC84 Domain Containing 2) 11905 Donated by Dr. 2.3. DNA Sequencing and Bioinformatic Analysis To identify mutations at Lmna exon 4, MiSeq DNA sequencing was carried out in the Genomic Unit of Instituto de Salud Carlos III. DNA was isolated from the diﬀerent pools and clones. Next, Illumina adapters were added to the target region of the Cas9 nuclease by PCR using the DeepSeq-Fw and DeepSeq-Rv primers (Table 2). The resulting product was used for a second PCR, where the speciﬁc indexes were added to every sample. Genomic reads in FASTQ format were analyzed using the CRISPResso platform. This software isolates the part of each read spanning the chosen region, highlights small insertions/deletions, and outputs a count of each regional sequence. We then analyzed the percentage of the sequences showing regional diﬀerences in control and with the diﬀerent sgRNA-exon 4 of LMNA-transduced 5 of 21 Cells 2020, 9, 1286 samples. All the pools and clones were also analyzed by Sanger sequencing of PCR products ampliﬁed using Lmna exon 4-speciﬁc primers Sanger-mLmna_Ex4_Fw and Sanger-mLmna_Ex4_Rv (Table 2). Sanger sequences were analyzed using the TIDE platform. Table 2. DNA sequences used in this work. Name and 5′ to 3′ Sequence sg10: CCGGCTGGCAGATGCCCTGCAGG sg11: GCGGGCTCAGCATGAGGACCAGG sg12: GGTCCACCTTGTCATATTCTTCC ssODNmex4g10: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCT GGGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g11: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAGGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g12: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAGGTGGAACAGTATAA GAAGGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT ssODNmex4g10mut: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAAGAGCTCT GGGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g10mut2: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGTCTTGCCGACGCACTTCAAGAGCTCTG GGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g11mut: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAAGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g11mut2: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCACAACACGAAGATCAAGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g12mut: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGATCAGGTGGAACAGTATAA GAAGGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT ssODNmex4g12mut2: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGATCATGTTGAGCAATACAA AAAAGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT DeepSeq-Fw: TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGAGGCGAGTGGATGCTGAG DeepSeq-Rv: GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGTCAATGCGGATTCGAGACT Sanger-mLmna_Ex4_Fw: CCAGGCTAAGCGAGGGCTGC Sanger-mLmna_Ex4_Rv: CCTGAGGAAGGCATCCCTGG Table 2. DNA sequences used in this work. Name and 5′ to 3′ Sequence sg10: CCGGCTGGCAGATGCCCTGCAGG sg11: GCGGGCTCAGCATGAGGACCAGG sg12: GGTCCACCTTGTCATATTCTTCC ssODNmex4g10: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCT GGGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g11: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAGGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g12: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAGGTGGAACAGTATAA GAAGGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT ssODNmex4g10mut: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAAGAGCTCT GGGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g10mut2: GTGGAGATCGATAACGGGAAGCAGCGAGAGTTTGAGAGCCGTCTTGCCGACGCACTTCAAGAGCTCTG GGCTCAGCATGAGGACCAGGTGGAACAGTATAAGAAGGAGCT ssODNmex4g11mut: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGACCAAGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g11mut2: GAGTTTGAGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCACAACACGAAGATCAAGTGGAAC AGTATAAGAAGGAGCTAGAAAAGACATACTCCGCCAAGGTGCT ssODNmex4g12mut: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGATCAGGTGGAACAGTATAA GAAGGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT ssODNmex4g12mut2: AGAGCCGGCTGGCAGATGCCCTGCAGGAGCTCTGGGCTCAGCATGAGGATCATGTTGAGCAATACAA AAAAGAGCTAGAAAAGACATACTCCGCCAAGTGCTGGCCTCAT DeepSeq-Fw: TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGAGGCGAGTGGATGCTGAG DeepSeq-Rv: GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGTCAATGCGGATTCGAGACT Sanger-mLmna_Ex4_Fw: CCAGGCTAAGCGAGGGCTGC Sanger-mLmna_Ex4_Rv: CCTGAGGAAGGCATCCCTGG Table 2. DNA sequences used in this work. 2.4. Western Blot Analysis Cells were incubated in blocking solution (0.1% bovine serum albumin (BSA), 1% FBS, and 1% horse serum in phosphate-buﬀered saline (PBS), in the case of SUN2; 1% BSA, 2.5% goat serum, 2.5% donkey serum, and 0.3% Triton X-100 in PBS, in the cases of lamin A/C, lamin B1, and emerin) for 30 min. Primary and secondary antibodies used for immunoﬂuorescence detection are detailed in Table 1. Coverslips were mounted using Prolong Gold (ThermoFisher Scientiﬁc; Waltham, MA, USA) with DAPI, and images were captured using a Leica TCS SP5 confocal microscope (Leica, Wetzlar, Germany) and a 63× (HCX PL APO 63× 1.4 OIL). For the 3D models, cells were ﬁxed in a 10% formalin solution (Sigma-Aldrich, St. Louis, MI, USA). Then, hydrogels were washed with Tris-buﬀered saline (TBS), permeabilized with 0.1% Triton X-100 in TBS for 15 min, and blocked with a blocking buﬀer consisting of 0.3% Triton X-100 and 3% donkey serum in TBS for 2 h. Afterwards, samples were washed with TBS and incubated with anti-MYH7 in blocking buﬀer (O/N, 4 ◦C). Then, samples were washed with TBS and incubated with rhodamine phalloidin and goat anti-rabbit A488 in blocking buﬀer (O/N, 4 ◦C). After washing with TBS, nuclei were counterstained with 1 µM DAPI for 15 min. Cells (5000 cells per well) were plated into a 96-well cell culture black microplate with a glass bottom (Greiner Bio-one, Kremsmünster, Austria). After 24 h, the cells were ﬁxed with methanol for 5 min at −20 ◦C. The ﬁxed cells were incubated with 2 µL/mL Hoechst 33324 in PBS for 10 min at 37 ◦C. Finally, nuclear morphology was evaluated by determining the circularity index using the Cytell Cell Imaging System (GE Healthcare Life Sciences, Marlborough, MA, USA). The circularity values were between 0 and 1, with 1 being the value of a perfect circle. For the nuclei morphology in the 3D muscle model, confocal microscopy images were analyzed using ImageJ software (National Institutes of Health and the Laboratory for Optical and Computational Instrumentation, Madison, WI, USA). Myoblast diﬀerentiation was induced by growth medium exchange for the diﬀerentiation medium (DMEM with 2% horse serum and 1% of penicillin/streptomycin). The cells were diﬀerentiated for 5 days. 2.4. Western Blot Analysis SDS-PAGE was performed using standard procedures. Brieﬂy, cells were disrupted with RadioImmunoPrecipitation Assay (RIPA) buﬀer (150 mM NaCl, 1% Nonyl Phenoxypolyethoxylethanol (NP-40), 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl (pH 7.5), and protease and phosphatase inhibitors). The lysates were sonicated with the ultrasonic processor UPH100H (Hielscher, Teltow, Germany) (three pulses of 5 s at 100% amplitude) to allow dissociation of proteins. Samples were centrifuged at 14,000 rpm at 4 ◦C for 15 min. Protein concentration was quantiﬁed with the BCA system. Table 1 lists the references and dilutions of the primary and Horseradish Peroxidase (HRP)-labelled SDS-PAGE was performed using standard procedures. Brieﬂy, cells were disrupted with RadioImmunoPrecipitation Assay (RIPA) buﬀer (150 mM NaCl, 1% Nonyl Phenoxypolyethoxylethanol (NP-40), 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl (pH 7.5), and protease and phosphatase inhibitors). The lysates were sonicated with the ultrasonic processor UPH100H (Hielscher, Teltow, Germany) (three pulses of 5 s at 100% amplitude) to allow dissociation of proteins. Samples were centrifuged at 14,000 rpm at 4 ◦C for 15 min. Protein concentration was quantiﬁed with the BCA system. Table 1 lists the references and dilutions of the primary and Horseradish Peroxidase (HRP)-labelled SDS-PAGE was performed using standard procedures. Brieﬂy, cells were disrupted with RadioImmunoPrecipitation Assay (RIPA) buﬀer (150 mM NaCl, 1% Nonyl Phenoxypolyethoxylethanol (NP-40), 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris-HCl (pH 7.5), and protease and phosphatase inhibitors). The lysates were sonicated with the ultrasonic processor UPH100H (Hielscher, Teltow, Germany) (three pulses of 5 s at 100% amplitude) to allow dissociation of proteins. Samples were centrifuged at 14,000 rpm at 4 ◦C for 15 min. Protein concentration was quantiﬁed with the BCA system. Table 1 lists the references and dilutions of the primary and Horseradish Peroxidase (HRP)-labelled 6 of 21 Cells 2020, 9, 1286 secondary antibodies used for Western blot analyses. Detection was carried out using an ECL western blotting system. secondary antibodies used for Western blot analyses. Detection was carried out using an ECL western blotting system. 2.5. Immunoﬂuorescence Microscopy, Nuclear Morphology Analyses, and Myogenic Diﬀerentiation Studies For 2D cultures, cells plated onto coverslips were ﬁxed with methanol for 5 min at −20 ◦C. 2.4. Western Blot Analysis The cells (1 × 105 cells per well) were ﬁxed with methanol for 5 min at −20 ◦C, permeabilized with 0.05% Triton X-100 in PBS for 5 min at room temperature, and washed with PBS for 5 min. Then, coverslips were blocked in blocking solution (15% FBS in PBS) for 30 min at room temperature. The cells were incubated with primary antibody against myosin heavy chain (MF20) for 1 h at room temperature. The cells were washed with PBS three times for 5 min, incubated with secondary antibody anti-mouse Ig for 45 min at room temperature and in the dark, and washed with PBS three times for 5 min. Coverslips were mounted using Prolong Gold (Thermo Fisher Scientiﬁc) with DAPI. Stained myotubes were visualized in a Leica DM4B microscope (Leica-Microsystems, Wetzlar, Germany) using the 10× (N PLAN 10×/0.25 DRY) and 20× objectives (HC PL FLUOTAR 20×/0.55 DRY). 2.6. Quantiﬁcation and Statistical Analysis 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells In order to produce Lmna exon 4 mutants by CRISPR/Cas9 technology, we designed three sgRNAs to target the Cas9 endonuclease to this Lmna region (Figure 1A). Following the scheme depicted in Figure 1B and detailed in Section 2, we produced a total of 18 pools and 210 clones in which CRISPR/Cas9 activity (insertions and deletions, Indels) was measured by TIDE and deep sequence analyses (Table S1). It is interesting to note that each sgRNA induced a diﬀerent Indel pattern (Figure S3). One single nucleotide deletion was the most frequent event when using sg12 (present in 77% of the clones), whereas a wider spectrum of deletions and insertions was observed for sg10 and sg11 guides. Interestingly, deletions of 3 and 6 nucleotides, which produce in-frame deletions mutants, were mainly observed in sg11-transfected cells (they were detected in 20% and 35% of the clones, respectively). Sequences with no Indels were detected in 3%, 5%, and 23% of the clones transfected with s10, sg11, and sg12 guides, respectively. We also aimed at introducing the R249W mutation, one of the human variants associated with a severe type of laminopathy, LMNA-associated congenital muscular dystrophy. In all the transfections, we incorporated a donor template including the mutant nucleotide. Although ﬁve clones were positive for the R249W mutation, only one (g11H-D6; Table 3) was free of additional nucleotide changes and, therefore, allowed the expression of this mutant form. Figure 1. Generation of Lamin A/C gene (Lmna) exon 4 mutations using CRISPR/Cas technology. (A) Scheme for exon 4 Lmna sgRNA localization. (B) Work-ﬂow followed for the generation of C2C12 cells carrying mutations in Lmna exon 4. Figure 1. Generation of Lamin A/C gene (Lmna) exon 4 mutations using CRISPR/Cas technology. (A) Scheme for exon 4 Lmna sgRNA localization. (B) Work-ﬂow followed for the generation of C2C12 cells carrying mutations in Lmna exon 4. Figure 1. Generation of Lamin A/C gene (Lmna) exon 4 mutations using CRISPR/Cas technology. (A) Scheme for exon 4 Lmna sgRNA localization. (B) Work-ﬂow followed for the generation of C2C12 cells carrying mutations in Lmna exon 4. A total of 113 clones, named as “null”, only carried frameshift mutations that produced truncated proteins due to premature stop codons. 2.6. Quantiﬁcation and Statistical Analysis The levels of Lamin A/C, Lamin B1, Emerin, SUN1, and SUN2 at the nucleus were analyzed using ImageJ software (National Institutes of Health). In brief, individual lines of constant length were plotted perpendicular to the nucleus and used to evaluate peak ﬂuorescence intensity using the PlotProﬁle feature from ImageJ. Random individual plot proﬁles of conditions were generated and quantiﬁed at the nuclei using the PlotProﬁle feature from ImageJ. In every set of experiments, all immunoﬂuorescence images were taken from single confocal acquisitions under the same exposure conditions. All the graphs and statistical analyses were carried out with GraphPad 8.0 software (GraphPad, San Diego, CA, USA). Error bars represent standard error of the mean (SEM). For statistical analyses, the level of signiﬁcance was determined by two-tailed, unpaired Mann–Whitney U test or Student’s t-test. 7 of 21 7 of 21 Cells 2020, 9, 1286 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells Sixteen clones, classiﬁed as “delta”, were characterized by the only presence of triplet deletions compatible with the formation of Lamin A/C proteins with deletions of 1, 7, 14, 15, or 57 amino acids (Figure S1). Ten clones that were transfected with Cas9 expressing vectors with no sgRNAs showed no mutations in Lmna exon 4 and were used and named 8 of 21 Cells 2020, 9, 1286 as “controls”. Table 3 lists the control, null, delta, and R249W clones that were selected for a detailed study on the eﬀects caused by the mutations in Lmna exon 4. as “controls”. Table 3 lists the control, null, delta, and R249W clones that were selected for a detailed study on the eﬀects caused by the mutations in Lmna exon 4. 3.2. Components of the Nuclear Lamina Were Abnormally Distributed in Lmna Exon 4 Mutant Myoblasts, Lamin A/C Abnormal Expression, and Subcellular Localization in Lmna Exon 4 Mutants We ﬁrst analyzed the total protein levels and nuclear sub-cellular localization of Lamin A/C. As shown in Figure 2A, all the Lmna exon 4 mutants were characterized by a signiﬁcant reduction in the amount of both Lamin A and C proteins. This feature was accompanied by an abnormal subcellular localization of Lamin A/C in all delta and R249W mutants (Figure 2B). Whereas control and R249W myoblasts showed a predominant Lamin A/C signal in the nuclear periphery, delta clones showed no diﬀerences in Lamin A/C intensity between perinuclear and nucleoplasmic regions. Moreover, in the diﬀerent delta clones, we were able to detect a wide spectrum of abnormalities regarding Lamin A/C staining including honeycomb-like patterns, asymmetric distribution, capping, and bright foci (Figure S4). Although Lamin A/C subcellular localization in R249W cells followed the same pattern found in control myoblasts, a detailed analysis showed lower ﬂuorescence intensity of Lamin A/C at the nuclear periphery when compared with the control clones (Figure 2B). To investigate the consequences of the Lamin A/C mislocalization and low expression in Lmna exon 4 mutants, we studied, using single confocal imaging, the sub-cellular localization of Lamin B1 and Emerin, two main inner nuclear membrane proteins. In control clones, Lamin B1 accumulated at the nuclear periphery and was also detected, to a lesser extent, at the nucleoplasm. 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells Lamin B1 distribution was slightly altered in Lmna exon 4 mutants, which did not show a sharp contrast between the Lamin B1 signal detected in Lamin area and nucleoplasm (Figure 2B). A signiﬁcant eﬀect was also expected for Emerin because the localization of this protein is dependent on Lamin A/C [24]. Indeed, Emerin sub-cellular distribution was aﬀected in null, delta, and R249W clones (Figure 2B). In all the cases, Emerin extranuclear signal was compatible with the endoplasmic reticulum pattern previously reported for Lmna null and mutant cells [25–27]. Our results indicated that a reduction and abnormal sub-cellular localization of the main lamina components—Lamin A/C, Lamin B1, and Emerin—was common to all the Lmna exon 4 mutants analyzed. Table 3. Molecular properties of the selected clones. ID Type CDS (1) Protein (Expected) 459-A6 Control wt/wt p.(666)/p.(666) (665 aa) 459-B4 wt/wt p.(666)/p.(666) (665 aa) g10L-A2 Null c. 734_735del p.(252) (251 aa) g11Hm-8 c. [757_758ins] [758_810del] p.(263) (262 aa) g12Hm2-3 c. [767del] [811subsC > G] p.(263) (262 aa) g12Lm2-3 c. [739_740ins] [740_761del] [766_767GT > AG] p.(273) (272 aa) g12Lm2-5 c. [del767]/c. [del769] p.(263) (262 aa)/p.(263) (262 aa) g10L-A1 Delta1 c.734_736del (loss of 3 nt) p.Leu245del (loss of 1 aa) g11Hm2-1 Delta7-1 c.754_774del (loss of 21 nt) p.His252_Gln258del (loss of 7 aa) g11Hm2-10 Delta7-10 c.754_774del (loss of 21 nt) p.His252_Gln258del (loss of 7 aa) g10Hm-5 Delta14 c.707_748del (loss of 42 nt) p.Glu236_Arg249del (loss of 14 aa) g12Hm-6 Delta15 c.766_810del (loss of 45 nt) p.Val256_Lys270del (loss of 15 aa) g10H-A4 Delta57 c. [726_727ins] [727_1967del]/c. 640_810del p.(252) (251 aa)/p.Glu214_Lys270del (loss of 57 aa) g11H-D6 R249W c. [744_745subsGC > CT] [750subsT > G] p.Arg249Tryp (665 aa) 1 Obtained by deep sequencing. Table 3. Molecular properties of the selected clones. Table 3. Molecular properties of the selected clones. 9 of 21 Cells 2020, 9, 1286 Figure 2. Components of the nuclear lamin were abnormally distributed in Lmna exon 4 mutant myoblasts. (A) Lamin A and C levels were reduced in Lmna exon 4 mutants. Quantiﬁcation of Lamin A and C protein levels. Left pictures show one of the three independent biological replicates analyzed. The other replicates are included in Figure S5. Quantiﬁcation of Lamin A/C signal in the three independent analyses, by mutant type, is shown on the right. Bars are mean ± SEM. Two-tailed, unpaired Student’s t-test: * p < 0.0001, * p = 0.003, n = 4. 3.1. Generation of Lmna Exon 4 Mutations in C2C12 Cells (B) Mouse myoblasts were immunostained for Lamin A/C, Lamin B1, and Emerin. Representative single confocal images from nuclei are shown for each Lmna group. Extra-nuclear signal was observed for Emerin in null, delta, and R249W mutants (arrows). Scale bar = 8 µm. Right plots show average ﬂuorescence intensity across nuclei by mutant type (control (blue): three clones; null (purple): ﬁve clones; delta (orange): six clones; R249W (green): one clone; n = 5 nuclei per clone). Peaks at both ends of the plot represent peripheral staining. Representative images from other clones are shown in Figure S4 (Lamin A/C), Figure S6 (Lamin B1), and Figure S7 (Emerin). Figure 2. Components of the nuclear lamin were abnormally distributed in Lmna exon 4 mutant myoblasts. (A) Lamin A and C levels were reduced in Lmna exon 4 mutants. Quantiﬁcation of Lamin A and C protein levels. Left pictures show one of the three independent biological replicates analyzed. The other replicates are included in Figure S5. Quantiﬁcation of Lamin A/C signal in the three independent analyses, by mutant type, is shown on the right. Bars are mean ± SEM. Two-tailed, unpaired Student’s t-test: ** p < 0.0001, * p = 0.003, n = 4. (B) Mouse myoblasts were immunostained for Lamin A/C, Lamin B1, and Emerin. Representative single confocal images from nuclei are shown for each Lmna group. Extra-nuclear signal was observed for Emerin in null, delta, and R249W mutants (arrows). Scale bar = 8 µm. Right plots show average ﬂuorescence intensity across nuclei by mutant type (control (blue): three clones; null (purple): ﬁve clones; delta (orange): six clones; R249W (green): one clone; n = 5 nuclei per clone). Peaks at both ends of the plot represent peripheral staining. Representative images from other clones are shown in Figure S4 (Lamin A/C), Figure S6 (Lamin B1), and Figure S7 (Emerin). 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 The abnormalities described for all the Lmna exon 4 mutants were compatible with a defective lamin that might aﬀect the nuclear membrane integrity. To test this hypothesis, we investigated the nuclear morphology of all the clones analyzed in this work. Whereas nuclei of controls are circular or ovoid, mutant clones are characterized by irregular nuclei with multiple herniations and, more frequently, by elongated nuclei. To quantify these diﬀerences in terms of nuclear morphology, we calculated the circularity index of the nuclei of myoblasts asynchronously growing in 2D cultures. As expected, all the mutants showed a signiﬁcant reduction in the circularity index of their nuclei (Figure 3A). To study nuclear morphology in a more physiologic condition, we encapsulated the myoblasts in a mimetic 3D architecture that recreated a native skeletal muscle (Figure S2A). Under this 3D cellular Cells 2020, 9, 1286 10 of 21 organization, all type of clones formed highly elongated and aligned ﬁbers (Figure S2B). For both control and mutant clones, the circularity was reduced when compared to the values obtained in 2D experiments (Figure 3B). This eﬀect was probably caused by the geometric constriction of the pattern and the compliant nature of the hydrogel used to create the 3D structure. Importantly, circularity was signiﬁcantly reduced in mutant clones when they were compared to controls (Figure 3B). g y y p ( g ) These results demonstrated that mutations in Lmna exon 4 negatively aﬀect nuclear morphology in myoblasts growing in both 3D and 2D conditions. Figure 3. Nuclear morphology was altered in mouse myoblasts carrying Lmna exon 4 mutations. (A) Circularity index was calculated in nuclei from clones asynchronously growing in 2D cultures (** p < 0.0001; n > 5000 nuclei per clone). (B) Plot shows the circularity index from clones encapsulated in a 3D pattern. Data were compared using an unpaired Mann–Whitney U test ( p < 0.0001). Blue, purple, green, and orange colors indicate control, null, R249W, and delta clones, respectively. 4 Ab l M i Diﬀ ti ti W C t All M bl t C i L E 4 M t ti Figure 3. Nuclear morphology was altered in mouse myoblasts carrying Lmna exon 4 mutations. 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 (A) Circularity index was calculated in nuclei from clones asynchronously growing in 2D cultures ( p < 0.0001; n > 5000 nuclei per clone). (B) Plot shows the circularity index from clones encapsulated in a 3D pattern. Data were compared using an unpaired Mann–Whitney U test ( p < 0.0001). Blue, purple, green, and orange colors indicate control, null, R249W, and delta clones, respectively. Figure 3. Nuclear morphology was altered in mouse myoblasts carrying Lmna exon 4 mutations. (A) Circularity index was calculated in nuclei from clones asynchronously growing in 2D cultures ( p < 0.0001; n > 5000 nuclei per clone). (B) Plot shows the circularity index from clones encapsulated in a 3D pattern. Data were compared using an unpaired Mann–Whitney U test (** p < 0.0001). Blue, purple, green, and orange colors indicate control, null, R249W, and delta clones, respectively. Abnormal Myogenic Diﬀerentiation Was Common to All Myoblasts Carrying Lmna Exon 4 Mutations 3.4. Abnormal Myogenic Diﬀerentiation Was Common to All Myoblasts Carrying Lmna Exon 4 Mutations LMNA mutations have been associated not only with structural abnormalities, such as the nuclear integrity defects described above, but also with functional deﬁciencies, including diﬀerentiation capacity to myogenic ﬁbers [28,29]. To test if this was also the case for the mutants generated in this work, we induced myogenic diﬀerentiation in all the myoblasts under study. In a ﬁrst set of experiments, we studied the diﬀerentiation capacity of control and mutant clones growing in 2D conditions. As shown in Figure 4A, 5 days post-induction of diﬀerentiation, control myoblasts formed myogenic ﬁbers that were positive for myosin heavy chain (MHC). However, in null, R249W and the majority of the delta clones, we were not able to detect either myogenic ﬁbers or MHC expression. Interestingly, delta57 clone, which is missing the whole Lmna exon 4, showed a moderate formation of ﬁbers and MHC expression. Similar results were obtained from myoblasts growing in 3D structures (Figure 4B). In this case, MHC was only detected in control clones as well as in delta57 and R249W mutants. Importantly, although R249W cells were positive for MHC, they showed an abnormal distribution of this protein that could indicate the formation of abnormal myogenic ﬁbers. Our results indicate that Lmna exon 4 mutations signiﬁcantly aﬀect the capacity of C2C12 cells to diﬀerentiate to myogenic ﬁbers. 11 of 21 Cells 2020, 9, 1286 Figure 4. 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 These proteins, together with Lamin A/C, Emerin, and Nesprins, among others, constitute the LINC complex, a molecular link that connects the nucleus to the cyto-skeleton, and have been associated with muscular dystrophies [30]. The total protein levels of SUN1 and SUN2 were not altered in any of the mutant types in null, delta, or R249W (Figure 5A). However, their subcellular distribution was signiﬁcantly aﬀected in all the mutants when compared with controls (Figure 5B). Whereas SUN1 and SUN2 accumulated in the nuclear envelope of control myoblasts, they were barely detected in the same region of Lmna exon 4 mutants. Figure 5. SUN1 and SUN2 protein levels and sub-cellular localization in Lmna exon 4 mutant myoblasts. (A) Total amount of SUN1 and SUN2 proteins was determined by Western blot. One of the three biological replicates is shown. The rest of the Western blots are shown in Figure S5. Quantiﬁcation, by mutant type, of SUN1 and SUN2 signals for the three independent analyses is shown in the graphs on the right. Bars are mean ± SEM, n = 3. No signiﬁcant diﬀerences were detected when mutants were compared with controls using a two-tailed, unpaired Student’s t-test. Representative, single, confocal, images of SUN1 (B) and SUN2 (C) detected by immunostaining are shown. A decreased nuclear envelope signal was observed in null, delta, and R249W mutants. Scale bar = 8 µm. Right plots show average ﬂuorescence intensity across nuclei by mutant type (control (blue): three clones; null (purple): ﬁve clones; delta (orange): six clones; R249W (green): one clone; n = 5 nuclei per clone). Peaks at both ends of the plot represent peripheral staining. Representative images for SUN1 and SUN2 in all the clones are shown in Figures S9 and S10, respectively. Figure 5. SUN1 and SUN2 protein levels and sub-cellular localization in Lmna exon 4 mutant myoblasts. (A) Total amount of SUN1 and SUN2 proteins was determined by Western blot. One of the three biological replicates is shown. The rest of the Western blots are shown in Figure S5. Quantiﬁcation, by mutant type, of SUN1 and SUN2 signals for the three independent analyses is shown in the graphs on the right. Bars are mean ± SEM, n = 3. No signiﬁcant diﬀerences were detected when mutants were compared with controls using a two-tailed, unpaired Student’s t-test. Representative, single, confocal, images of SUN1 (B) and SUN2 (C) detected by immunostaining are shown. 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 Myogenic diﬀerentiation was impaired in mouse myoblasts with mutations in Lmna exon 4. (A) Myoblasts growing in 2D conditions were induced to diﬀerentiate to myogenic ﬁber as described in Section 2. Representative pictures show myosin heavy chain (MHC, green) and nuclei (DAPI, blue). Scale bar = 60 µm. (B) Myogenic ﬁber formation was also aﬀected in 3D models. Representative pictures show, at 7 days post-diﬀerentiation, myosin heavy chain 7 (green), nuclei (blue), and phalloidin staining (red). Control and null samples correspond to 459-A6 and g12Lm2-3 clones. Scale bar = 50 µm. Single color channels for this panel are shown in Figure S8. Figure 4. Myogenic diﬀerentiation was impaired in mouse myoblasts with mutations in Lmna exon 4. (A) Myoblasts growing in 2D conditions were induced to diﬀerentiate to myogenic ﬁber as described in Section 2. Representative pictures show myosin heavy chain (MHC, green) and nuclei (DAPI, blue). Scale bar = 60 µm. (B) Myogenic ﬁber formation was also aﬀected in 3D models. Representative pictures show, at 7 days post-diﬀerentiation, myosin heavy chain 7 (green), nuclei (blue), and phalloidin staining (red). Control and null samples correspond to 459-A6 and g12Lm2-3 clones. Scale bar = 50 µm. Figure 4. Myogenic diﬀerentiation was impaired in mouse myoblasts with mutations in Lmna exon 4. (A) Myoblasts growing in 2D conditions were induced to diﬀerentiate to myogenic ﬁber as described in Section 2. Representative pictures show myosin heavy chain (MHC, green) and nuclei (DAPI, blue). Scale bar = 60 µm. (B) Myogenic ﬁber formation was also aﬀected in 3D models. Representative pictures show, at 7 days post-diﬀerentiation, myosin heavy chain 7 (green), nuclei (blue), and phalloidin staining (red). Control and null samples correspond to 459-A6 and g12Lm2-3 clones. Scale bar = 50 µm. Single color channels for this panel are shown in Figure S8. Cells 2020, 9, 1286 12 of 21 3.5. Expression of Lmna Exon 4 Mutants Was Associated with Abnormal Subcellular Localization of SUN1 and SUN2 Expression of Lmna Exon 4 Mutants Was Associated with Abnormal Subcellular Localization of SUN1 SUN2 In an eﬀort to understand the mechanisms underlying the nuclear morphology and myogenic diﬀerentiation defects that characterize the Lmna exon 4 mutants, we studied the status of SUN1 and SUN2. 3.6. MAPK Signaling Was Not Altered in Lmna Exon 4 Mutant Myoblasts Thanks to the analysis of the LmnaH222P mouse model, a connection between LMNA mutations and signaling pathways has been discovered [31]. Speciﬁcally, enhanced phosphorylation of ERK1/2 and AKT has been reported in the heart of LmnaH222P mice [10,32]. Therefore, we decided to study the status of these key signaling molecules in Lmna exon 4 mutants. Although the activation of ERK1/2, estimated by the quantiﬁcation of phosphorylated-ERK1/2, was very heterogeneous for inter and intra clones (Figure S5), no signiﬁcant diﬀerences were observed when control, null, delta, and R249W groups were compared (Figure 6A). Likewise, there were no signiﬁcant diﬀerences in AKT activation (Figure 6B). Figure 6. No diﬀerences in ERK1/2 and AKT activation were detected in Lmna exon 4 mutant myoblasts. (A) Representative immunoblot for the detection of phosphorylated (pERK) and total (ERK) ERK1/2 in Lmna exon 4 mutants. (B) Representative immunoblot for the detection of phosphorylated (pAKT) and total (AKT) AKT. Detection of α-tubulin was used in all the cases as loading control. Data in bar graphs are mean ± SEM of all the clones of each type, n = 3. No signiﬁcant diﬀerences were detected when mutants were compared with controls using a two-tailed, unpaired Student’s t-test. The remainder of the Western blots analyzed in (A) and (B) are shown in Figure S5. Therefore, the Lmna exon 4 mutations studied here were not characterized by homogeno Figure 6. No diﬀerences in ERK1/2 and AKT activation were detected in Lmna exon 4 mutant myoblasts. (A) Representative immunoblot for the detection of phosphorylated (pERK) and total (ERK) ERK1/2 in Lmna exon 4 mutants. (B) Representative immunoblot for the detection of phosphorylated (pAKT) and total (AKT) AKT. Detection of α-tubulin was used in all the cases as loading control. Data in bar graphs are mean ± SEM of all the clones of each type, n = 3. No signiﬁcant diﬀerences were detected when mutants were compared with controls using a two-tailed, unpaired Student’s t-test. The remainder of the Western blots analyzed in (A) and (B) are shown in Figure S5. Figure 6. No diﬀerences in ERK1/2 and AKT activation were detected in Lmna exon 4 mutant myoblasts. (A) Representative immunoblot for the detection of phosphorylated (pERK) and total (ERK) ERK1/2 in Lmna exon 4 mutants. (B) Representative immunoblot for the detection of phosphorylated (pAKT) and total (AKT) AKT. Detection of α-tubulin was used in all the cases as loading control. 3.3. Nuclear Morphology Abnormalities Were Common for All the Clones with Mutations in Lmna Exon 4 A decreased nuclear envelope signal was observed in null, delta, and R249W mutants. Scale bar = 8 µm. Right plots show average ﬂuorescence intensity across nuclei by mutant type (control (blue): three clones; null (purple): ﬁve clones; delta (orange): six clones; R249W (green): one clone; n = 5 nuclei per clone). Peaks at both ends of the plot represent peripheral staining. Representative images for SUN1 and SUN2 in all the clones are shown in Figures S9 and S10, respectively. Cells 2020, 9, 1286 Cells 2020, 9, 1286 13 of 21 These results, together with the ones obtained for the localization of Emerin and Lamin proteins, indicated that the defects induced at the nuclear envelope by the Lmna exon 4 mutations signiﬁcantly impact the subcellular localization of LINC complex components. 3.6. MAPK Signaling Was Not Altered in Lmna Exon 4 Mutant Myoblasts Data in bar graphs are mean ± SEM of all the clones of each type, n = 3. No signiﬁcant diﬀerences were detected when mutants were compared with controls using a two-tailed, unpaired Student’s t-test. The remainder of the Western blots analyzed in (A) and (B) are shown in Figure S5. Figure 6. No diﬀerences in ERK1/2 and AKT activation were detected in Lmna exon 4 mutant myoblasts. Therefore, the Lmna exon 4 mutations studied here were not characterized by homogenous and signiﬁcant changes in the activation of the signaling proteins ERK1/2 and AKT. 3.7. High Levels of DNA Damage Were Associated with R249W Mutation We also wanted to study the eﬀect of Lmna exon 4 mutations at the DNA level, as progeroid and striated muscle laminopathies have been associated with high levels of DNA damage [19,33–35]. 14 of 21 14 of 21 Cells 2020, 9, 1286 Quantiﬁcation of DNA damage by detection of γH2AX (phosphor-H2A.X Variant Histone) in asynchronously growing myoblasts revealed elevated levels of DNA damage in the R249W clone, whereas no signal was detected in the rest of mutants (null and delta) or controls (Figure 7A). To further analyze the connection between the R249W mutation and DNA damage, we also explored it in human myoblasts carrying a LMNA p.R249W mutation causing L-CMD. Interestingly, the human R249W mutant showed a clear increase in γH2AX levels when compared with the controls (Figure 7B). These results indicate that high levels of DNA damage are a speciﬁc feature of LMNA R249W mutant showed a clear increase in γH2AX levels when compared with the controls (Figure 7B). These results indicate that high levels of DNA damage are a speciﬁc feature of LMNA p.R249W myoblasts. These results indicate that high levels of DNA damage are a speciﬁc feature of LMNA p.R249W myoblasts. Figure 7. High levels of DNA damage were speciﬁcally associated with LMNA R249W mutation. (A) Representative immunoblot showing γH2AX levels in Lmna exon 4 mutant myoblasts. (B) Immunoblot showing γH2AX levels in human myoblasts from two controls and one patient carrying the p.R249W mutation. Detection of α-tubulin was used in all the cases as loading control. Data in bar graphs are mean ± SEM, n = 3. In (A) and (B), data were compared using a two-tailed, unpaired Student’s t-test. * p < 0.05. The remainder of the Western blots analyzed in (A) and (B) are shown in Figure S5. Figure 7. High levels of DNA damage were speciﬁcally associated with LMNA R249W mutation. (A) Representative immunoblot showing γH2AX levels in Lmna exon 4 mutant myoblasts. (B) Immunoblot showing γH2AX levels in human myoblasts from two controls and one patient carrying the p.R249W mutation. Detection of α-tubulin was used in all the cases as loading control. Data in bar graphs are mean ± SEM, n = 3. In (A) and (B), data were compared using a two-tailed, unpaired Student’s t-test. * p < 0.05. The remainder of the Western blots analyzed in (A) and (B) are shown in Figure S5. 4. Discussion 4.1. Comparison of Lmna Exon 4 Mutants Generated in This Study with Previously Reported LMNA Mutants In this study, using CRISPR/Cas9 technology, we analyzed a collection of Lmna exon 4 mutants in C2C12 mouse myoblasts. Three diﬀerent types of mutants were generated: null mutants, due to frameshift mutations; delta mutants, caused by in-frame deletions; and an already known R249W mutant. Interestingly, none of the delta and null mutants generated in this work have been previously reported in patients. Upon studying a number of molecular, cellular, and functional features, we can conclude that the subcellular localization of key members of the LINC complex and nuclear envelope, including Lamin A/C, Lamin B1, Emerin, SUN1, and SUN2, as well as the nuclear membrane integrity and myogenic diﬀerentiation capacity, were negatively aﬀected. 4.1. Comparison of Lmna Exon 4 Mutants Generated in This Study with Previously Reported LMNA Mutants 4.1. Comparison of Lmna Exon 4 Mutants Generated in This Study with Previously Reported LMNA Mutants In this study, using CRISPR/Cas9 technology, we analyzed a collection of Lmna exon 4 mutants in C2C12 mouse myoblasts. Three diﬀerent types of mutants were generated: null mutants, due to frameshift mutations; delta mutants, caused by in-frame deletions; and an already known R249W mutant. Interestingly, none of the delta and null mutants generated in this work have been previously reported in patients. Upon studying a number of molecular, cellular, and functional features, we can conclude that the subcellular localization of key members of the LINC complex and nuclear envelope, including Lamin A/C, Lamin B1, Emerin, SUN1, and SUN2, as well as the nuclear membrane integrity and myogenic diﬀerentiation capacity, were negatively aﬀected. 15 of 21 Cells 2020, 9, 1286 The comparison of the features of the Lmna mutants described here with those previously reported by others resulted in coincidences and diﬀerences. For instance, the Emerin extranuclear localization, compatible with endoplasmic reticulum, detected in null, delta, and R249W clones, is similar to the previously reported for Lmna null, LmnaL530P/L530P, LmnadK32, and LmnaN195K mouse myoblasts, as well as LMNA p.Y259X patient cells [15,25–27,36]. Contradictory results have been described for Lamin B1 localization. It has been reported that the loss of A-type lamins allows the assembly of B-type lamins [25]. In addition, more recently, LMNB1 localization defects have been shown in induced pluripotent stem cells carrying LMNA mutations [17], which is in line with the results we obtained in the present work. On the other hand, the Lmna null, delta, and R249W analyzed in this work showed a nuclear, although low intensity, signal for SUN1 and SUN2 at the nuclear envelope. A regulation by Lamin A of SUN2 localization has been previously reported in mouse ﬁbroblasts [37]. In addition, SUN2 nuclear distribution has been dramatically altered in synaptic nuclei of Lmna−/−and LmnaH222P/H222P muscle [38]. On the other hand, the weak nuclear envelope localization characteristic for Lmna exon 4 mutant C2C12 cells diﬀers from initial reports showing that localization of SUN1 does not depend on A-type lamins in mouse ﬁbroblasts and HeLa cells [38–40]. However, other works in Lmna null or LmnaL530P/L530P mouse ﬁbroblasts showed an extranuclear signal or loss of nuclear envelope signal [27,41,42]. The diﬀerences observed between these reports and the present work might be due to the diﬀerent cell system used—ﬁbroblasts versus myoblasts. 4.1. Comparison of Lmna Exon 4 Mutants Generated in This Study with Previously Reported LMNA Mutants Importantly, it has been reported that LMNA mutations causing Emery–Dreifuss muscular dystrophy are associated with low SUN1 levels in human myoblasts [43], an outcome similar to the one obtained in the Lmna exon 4 mutants generated and characterized in this work. In summary, diﬀerent conclusions might be obtained for the SUN1–Lamin A/C connection that could be a consequence of diﬀerent cellular or molecular scenarios. Further studies are needed to better clarify the role that Lamin A/C proteins play on SUN1 localization and function. LMNA p.H222P, one of the most studied LMNA mutations, is also located in exon 4. However, some of its features diﬀer from the data obtained with the exon 4 mutants analyzed here. For instance, the alterations of MAPK and AKT signaling pathways described in the LmnaH222P–H222P model [10,32] were not detected in the Lmna exon 4 mutants analyzed in this work. Moreover, Lmna knockdown in C2C12 cells has been previously associated with a signiﬁcant increase in ERK1/2 activation [44], an abnormality that is not found in the null C2C12 clones studied here (Figure 6A). To note, the siRNA (small interfering Ribonucleic Acid) used in the former did not fully inhibit Lamin A/C expression, whereas the use CRISPR/Cas9 induced the complete elimination of Lamin A/C expression (Figure 2A). In our study, we observed a high heterogeneity both inter and intra clones for p-ERK and p-AKT levels. Because this was not the case for other proteins analyzed here (Figure S5), we conclude that none of these two pathways are homogeneously altered in the Lmna exon 4 mutants studied in this work, whereas Lmna p.H222P mutation is speciﬁcally associated with an increased activation of both pathways. Another possible explanation, would be the fact that the majority of the studies for Lmna p.H222P have been carried out using cardiomyocytes from a LmnaH222P knockin mouse model, whereas the cells used in the present study were mouse myoblasts. The detailed characterization of the same cell type from a mouse LmnaR249W model (currently under study) will help us to discern the correct explanation of these diﬀerences. 4.2. Mechanistic and Functional Defects Associated with Lmna Exon 4 Mutations Myogenic diﬀerentiation is the most important function of myoblasts. It has been previously shown that the expression of Lamin A mutants associated with diﬀerent laminopathies inhibits the in vitro diﬀerentiation of C2C12 myoblasts [28,29]. Abnormal myoblast diﬀerentiation has also been reported in LmnadK32 cells [15]. Moreover, Lmna null satellite cells are characterized by a defective myogenic diﬀerentiation [45]. The same defects were observed for the majority of the null, delta, and R249W mutants analyzed in this work, which conﬁrmed the fact that mutations and/or downregulation Cells 2020, 9, 1286 16 of 21 16 of 21 of Lamin A/C are associated with an impairment in myogenic diﬀerentiation that might explain a signiﬁcant part of the pathological phenotype of laminopathies. However, other scenarios cannot be ruled out. For instance, a premature diﬀerentiation associated to SMAD6 (SMAD Family Member 6) overexpression and BMP4 (Bone Morphogenetic Protein 4) downregulation has been reported in Lmna∆8–11 myoblasts [46]. Although this seems not to be the case for Lmna exon 4 mutant myoblasts (data not shown), further studies will help to fully understand the diﬀerentiation problems induced by these Lmna mutations. Other potential mechanisms underlying the abnormal diﬀerentiation of the Lmna exon 4 mutants is an abnormal function of the LINC complex. The deﬁcient subcellular localization observed for many of the LINC complex components (Emerin, Lamin A/C, Lamin B1, and SUN1/SUN2) in Lmna exon 4 mutant myoblasts could be the cause of the defects detected in nuclear membrane integrity and myogenic diﬀerentiation. Interestingly, the LINC complex, a molecular link between the nucleo- and cyto-skeleton, has been associated with muscular dystrophies [30]. Furthermore, the alteration of the nuclear transport due to a malfunction of the nuclear pores has already been associated with LINC complexes defects [47,48], and might be the functional cause of the pathological phenotype associated with Lmna exon 4 mutants. However, additional studies are needed to conﬁrm the causal relationship between Lmna exon 4 mutations and LINC complex in laminopathies. Other mechanisms and pathways previously associated with muscular dystrophies and cardiomyopathies might be associated with the Lmna exon 4 mutations. For instance, an altered WNT (Wingless and Int-1)/β-catenin signaling has been found in hearts of LmnaH222P mice [10,11], and pharmacological activation of this pathway improves heart dysfunction in this mouse model [49]. 4.3. DNA Damage: A Feature Speciﬁc for the R249W Mutation? The nuclear lamina is a critical structure for the stability and proper nuclear distribution of DNA [53]. A weak nuclear envelope aﬀects DNA integrity, inducing the accumulation of DNA damage, a feature linked to LMNA-associated diseases [19,54–56]. The majority of the mouse myoblasts studied in this work did not show elevated levels of DNA damage, which is consistent with the fact that we have explored this property only in undiﬀerentiated myoblasts, and increased levels DNA damage are mainly observed in diﬀerentiated cells and tissues of HGPS and LMNA-related, muscular dystrophies [19,55]. Surprisingly, even in this undiﬀerentiated myoblast state, the Lmna p.R249W mutant under study showed signiﬁcantly elevated levels of DNA damage. The fact that this feature was also observed in human LMNA p.R249W myoblasts (Figure 7) and in muscle biopsies from a 1 year old patient carrying a LMNA p.R249W mutation [19I] points to a speciﬁc association with DNA integrity and/or repair mechanisms. Further studies will help to validate this connection and explore therapeutic opportunities related to DNA damage and repair. 4.2. Mechanistic and Functional Defects Associated with Lmna Exon 4 Mutations On the other hand, disruption of the nucleocytoplasmic Ran gradient and consequent nuclear protein import defects have been reported in cells carrying LMNA mutations [5,50]. Moreover, alterations of the nuclear morphology, such as the ones associated with Lmna exon 4 mutants, are associated with an abnormal distribution of the chromatin and, subsequently, changes in the epigenetics marks [51,52]. Whether all these pathways are also causally connected with the Lmna exon 4 mutations we have described herein remains to be determined and will be part of further studies. 4.4. CRISPR/Cas9 Activity and Gene Therapy Implications The delta and R249W clones analyzed in this work showed a signiﬁcant reduction in the levels of Lamin A/C (Figure 2). This was probably due to the CRISPR-mediated deletion of one or more Lmna alleles (C2C12 cells used in this work were hypertriploid), as mRNA expression was signiﬁcantly reduced in all of them (data not shown). Therefore, the phenotypes observed in the Lmna exon 4 mutants could have been due not only to the mutations generated in Lmna, but also to a defective expression of lamin proteins. It is important to note that Lmna haploinsuﬃciency is associated with dilated cardiomyopathy [57]. On the other hand, all the clones analyzed in this work lost their Cells 2020, 9, 1286 17 of 21 17 of 21 wild type alleles and showed a single mutant allele. This genomic scenario strongly diﬀers from the predominantly autosomal dominant scenario of LMNA-associated diseases in which missense mutations could lead to Lamin or Prelamin A accumulation. To note, a completely diﬀerent situation is found in the null mutations generated by CRISPR/Cas9. All this evidence has important implications for the therapeutic potential that CRISPR technology could have for the treatment of laminopathies and the potential use as disease models of the clones we generated. One of the most interesting results we obtained in this work was the high percentage of delta clones isolated upon CRISPR/Cas9 activity on Lmna exon 4 (Figure S1). This indicated that these types of in-frame mutations might have some selective advantage over the rest of the Indels generated by CRISPR/Cas9 activity. Moreover, one of these clones, delta57, is characterized by the expression of a Lmna exon 4 null gene. Interestingly, although delta57 shows nuclear morphology and LINC complex abnormalities, it does retain some myogenic diﬀerentiation capacity. This could indicate that an exon 4 skipping strategy for the elimination of mutations located in this exon could be compatible with a partially functional myoblast. As far as we know, in LMNA-associated disease, successful exon skipping has only been shown for exon 5 in human cells [20]. The potential of this therapeutic strategy needs further experimentation in appropriate in vivo models. In conclusion, in this work, we characterized in detail the consequences associated with mutations in Lmna exon 4 in myoblasts. 4.4. CRISPR/Cas9 Activity and Gene Therapy Implications This new knowledge has opened the possibility of exploring new therapeutic approaches based on the DNA damage speciﬁcally induced by the Lmna R249W mutation and the use of exon skipping strategies for the elimination of Lmna exon 4 mutations. Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4409/9/5/1286/s1. Table S1: Sequencing information of Lmna exon 4 mutant clones and pools. Figure S1: Scheme of the Lmna gene and protein structure and the mutants generated in this work. Figure S2: Generation of 3D skeletal muscle models. Figure S3: Deletions/insertions detected in C2C12 pools and clones. Figure S4: Lamin A-C subcellular localization and expression in delta mutants. Figure S5: All the Western blots analyzed in this work for Lamin A/C, ERK1/2, AKT, SUN1, SUN2, and γH2AX. Figure S6: Lamin B1 subcellular localization in Lmna exon 4 mutants. Figure S7: Emerin subcellular localization in Lmna exon 4 mutants. Figure S8: Single color channels for Figure 5b panel. Figure S9: SUN1 sub-cellular localization in Lmna exon 4 mutants. Figure S10: SUN2 sub-cellular localization in Lmna exon 4 mutants. Author Contributions: Conceptualization, I.P.d.C.; Methodology, D.G.-D., C.E., F.d.M., A.G.C, J.R.-A., I.P.d.C.; Investigation, D.G.-D., C.E., F.d.M., A.G.C., B.V.-M., A.M., S.A.-Q., M.A.R.-M.; Resources, A.T.B., G.B.; Writing—Original Draft Preparation, D.G.-D., C.E., I.P.d.C.; Writing—Review and Editing, D.G.-D., C.E., F.d.M., A.G.C, B.V.-M., A.M., S.A.-Q., A.B., G.B., J.R.-A, M.A.R.-M, I.P.d.C.; Supervision, I.P.d.C.; Project Administration, I.P.d.C.; Funding Acquisition, F.d.M., I.P.d.C. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Ministerio de Ciencia e Innovación (Acción estratégica en Salud intramural PI16III/00017-TPY1348/16) and by Fundación Andrés Marcio, niños contra la laminopatía (TPY-259/19). F.M. was funded by a Miguel Servet II contract from Instituto de Salud Carlos III (ISCIII). Acknowledgments: We thank all the members of the Human Genetics Area of the Instituto de Investigación de Enfermedades Raras (IIER) for their continuous support and help. We are deeply grateful to Beatriz Martínez (IIER) for the logistic and scientiﬁc support at the beginning of the project. The authors thank Óscar Zaragoza for confocal microscopy assistance. We are very grateful to Eric Schirmer and. Colin Stewart for kindly providing SUN2 and SUN1 antibodies. Conﬂicts of Interest: The authors declare no conﬂict of interest. The 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. Appendix A Synthesis of Prepolymer Precursors References 1. De Leeuw, R.; Gruenbaum, Y.; Medalia, O. Nuclear Lamins: Thin Filaments with Major Functions. Trends Cell Biol. 2018, 28, 34–45. [CrossRef] [PubMed] 1. De Leeuw, R.; Gruenbaum, Y.; Medalia, O. Nuclear Lamins: Thin Filaments with Major Functions. Trends Cell Biol. 2018, 28, 34–45. [CrossRef] [PubMed] 2. Reddy, K.L.; Zullo, J.M.; Bertolino, E.; Singh, H. Transcriptional repression mediated by repositioning of genes to the nuclear lamina. Nature 2008, 452, 243–247. [CrossRef] [PubMed] 3. Dechat, T.; Pﬂeghaar, K.; Sengupta, K.; Shimi, T.; Shumaker, D.K.; Solimando, L.; Goldman, R.D. Nuclear lamins: Major factors in the structural organization and function of the nucleus and chromatin. Genes Dev. 2008, 22, 832–853. [CrossRef] [PubMed] 3. Dechat, T.; Pﬂeghaar, K.; Sengupta, K.; Shimi, T.; Shumaker, D.K.; Solimando, L.; Goldman, R.D. Nuclear lamins: Major factors in the structural organization and function of the nucleus and chromatin. Genes Dev. 2008, 22, 832–853. [CrossRef] [PubMed] 4. Swift, J.; Discher, D.E. The nuclear lamina is mechano-responsive to ECM elasticity in mature tissue. J. Cell Sci. 2014, 127, 3005–3015. [CrossRef] 4. Swift, J.; Discher, D.E. The nuclear lamina is mechano-responsive to ECM elasticity in mature tissue. J. Cell Sci. 2014, 127, 3005–3015. [CrossRef] 5. Busch, A.; Kiel, T.; Heupel, W.-M.; Wehnert, M.; Hübner, S. Nuclear protein import is reduced in cells expressing nuclear envelopathy-causing lamin A mutants. Exp. Cell Res. 2009, 315, 2373–2385. [CrossRef] 5. Busch, A.; Kiel, T.; Heupel, W.-M.; Wehnert, M.; Hübner, S. Nuclear protein import is reduced in cells expressing nuclear envelopathy-causing lamin A mutants. Exp. Cell Res. 2009, 315, 2373–2385. [CrossRef] 6 Worman H J Nuclear lamins and laminopathies J Pathol 2012 226 316–325 [CrossRef] 6. Worman, H.J. Nuclear lamins and laminopathies. J. Pathol. 2012, 226, 316–325. [CrossRef] 6. Worman, H.J. Nuclear lamins and laminopathies. J. Pathol. 2012, 226, 316–325. [CrossRef] 7. Gruenbaum, Y.; Foisner, R. Lamins: Nuclear Intermediate Filament Proteins with Fundamental Functions in Nuclear Mechanics and Genome Regulation. Annu. Rev. Biochem. 2015, 84, 131–164. [CrossRef] 8. Cenni, V.; Capanni, C.; Mattioli, E.; Columbaro, M.; Wehnert, M.; Ortolani, M.; Fini, M.; Novelli, G.; Bertacchini, J.; Maraldi, N.M.; et al. Rapamycin treatment of Mandibuloacral dysplasia cells rescues localization of chromatin-associated proteins and cell cycle dynamics. Aging 2014, 6, 755–770. [CrossRef] 9. Pellegrini, C.; Columbaro, M.; Capanni, C.; D’Apice, M.R.; Cavallo, C.; Murdocca, M.; Lattanzi, G.; Squarzoni, S. All-trans retinoic acid and rapamycin normalize Hutchinson Gilford progeria ﬁbroblast phenotype. Oncotarget 2015, 6, 29914–29928. [CrossRef] 10. Synthesis of Prepolymer Precursors Gelatin methacryloyl (GelMA) was synthesized with a 40% degree of methacrylation, as previously described (García-Lizarribar, Fernández-Garibay et al. 2018). The material was then dialyzed in Milli Q water with 6–8 kDa molecular weight cut-off (MWCO) membranes (Spectrumlabs, Pireas, Greece, 08-700-142) for 4 days. Sodium carboxymethylcellulose (CMC) (Sigma-Aldrich, 419273) was methacrylated 18 of 21 18 of 21 Cells 2020, 9, 1286 at a maximum degree of methacrylation, as previously described [58]. The methacrylation reaction was performed by mixing a solution of 1% (w/v) of the polymer in 50 mM MES (2-(N-morpholino)ethanesulfonic acid) buffer at pH 6.5 with 20 mM EDC (1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride), 10 mM N-hydroxysuccinimide (Sigma-Aldrich, 130672), and 10 mM 2 aminoethylmethacrylate (Sigma-Aldrich, 479659). The reaction was stopped after 24 h with the addition of acetone (Panreac, 161007, Barcelona, Spain) and filtered using a vacuum flask. The precipitate was dissolved in 10 mM PBS and dialyzed in Milli Q water with 3.5 kDa MWCO membranes (Thermofisher, 68035). Finally, the solutions of methacrylated polymers (GelMA and CMCMA) were lyophilized and stored at −20◦C. Preparation of Prepolymer Solutions A mixture of the prepolymer precursors (GelMA and CMCMA) was diluted in growth medium, containing the photoinitiator lithium phenyl(2,4,6-trimethylbenzoyl) phosphinate (LAP; TCI EUROPE N.V, L0290), at 65 ◦C for 3 h to obtain a homogeneous solution. The concentrations of GelMA, CMCMA, and LAP were ﬁxed to obtain ﬁnal concentrations of 5%, 1%, and 0.1% (w/v), respectively. References Muchir, A.; Pavlidis, P.; Decostre, V.; Herron, A.J.; Arimura, T.; Bonne, G.; Worman, H.J. Activation of MAPK pathways links LMNA mutations to cardiomyopathy in Emery-Dreifuss muscular dystrophy. J. Clin. Invest. 2007, 117, 1282–1293. [CrossRef] 11. Muchir, A.; Wu, W.; Choi, J.C.; Iwata, S.; Morrow, J.; Homma, S.; Worman, H.J. Abnormal p38 mitogen-activated protein kinase signaling in dilated cardiomyopathy caused by lamin A/C gene mutation. Hum. Mol. Genet. 2012, 21, 4325–4333. [CrossRef] [PubMed] 12. Larrieu, D.; Britton, S.; Demir, M.; Rodriguez, R.; Jackson, S.P. Chemical Inhibition of NAT10 Corrects Defects of Laminopathic Cells. Science 2014, 344, 527–532. [CrossRef] [PubMed] 13. Beyret, E.; Liao, H.-K.; Yamamoto, M.; Hernandez-Benitez, R.; Fu, Y.; Erikson, G.; Reddy, P.; Izpisua Belmonte, J.C. Single-dose CRISPR–Cas9 therapy extends lifespan of mice with Hutchinson–Gilford progeria syndrome. Nat. Med. 2019, 25, 419–422. [CrossRef] [PubMed] 14. Santiago-Fernández, O.; Osorio, F.G.; Quesada, V.; Rodríguez, F.; Basso, S.; Maeso, D.; Rolas, L.; Barkaway, A.; Nourshargh, S.; Folgueras, A.R.; et al. Development of a CRISPR/Cas9-based therapy for Hutchinson–Gilford progeria syndrome. Nat. Med. 2019, 25, 423–426. [CrossRef] 19 of 21 Cells 2020, 9, 1286 19 of 21 15. Bertrand, A.T.; Brull, A.; Azibani, F.; Benarroch, L.; Chikhaoui, K.; Stewart, C.L.; Medalia, O.; Ben Yaou, R.; Bonne, G. Lamin A/C Assembly Defects in LMNA-Congenital Muscular Dystrophy Is Responsible for the Increased Severity of the Disease Compared with Emery-Dreifuss Muscular Dystrophy. Cells 2020, 9, 844. [CrossRef] 16. Scharner, J.; Brown, C.A.; Bower, M.; Iannaccone, S.T.; Khatri, I.A.; Escolar, D.; Gordon, E.; Felice, K.; Crowe, C.A.; Grosmann, C.; et al. Novel LMNA mutations in patients with Emery-Dreifuss muscular dystrophy and functional characterization of four LMNA mutations. Hum. Mutat. 2011, 32, 152–167. [CrossRef] 17. Steele-Stallard, H.B.; Pinton, L.; Sarcar, S.; Ozdemir, T.; Maﬃoletti, S.M.; Zammit, P.S.; Tedesco, F.S. Modeling Skeletal Muscle Laminopathies Using Human Induced Pluripotent Stem Cells Carrying Pathogenic LMNA Mutations. Front. Physiol. 2018, 9, 1332. [CrossRef] 18. Bertrand, A.T.; Ziaei, S.; Ehret, C.; Duchemin, H.; Mamchaoui, K.; Bigot, A.; Mayer, M.; Quijano-Roy, S.; Desguerre, I.; Lainé, J.; et al. Cellular microenvironments reveal defective mechanosensing responses and elevated YAP signaling in LMNA-mutated muscle precursors. J. Cell Sci. 2014, 127, 2873–2884. [CrossRef] 19. Earle, A.J.; Kirby, T.J.; Fedorchak, G.R.; Isermann, P.; Patel, J.; Iruvanti, S.; Moore, S.A.; Bonne, G.; Wallrath, L.L.; Lammerding, J. Mutant lamins cause nuclear envelope rupture and DNA damage in skeletal muscle cells. Nat. Mater. 2019, 19, 464–473. [CrossRef] 20. References Scharner, J.; Figeac, N.; Ellis, J.A.; Zammit, P.S. Ameliorating pathogenesis by removing an exon containing a missense mutation: A potential exon-skipping therapy for laminopathies. Gene Ther. 2015, 22, 503–515. [CrossRef] 21. Yaﬀe, D.; Saxel, O. Serial passaging and diﬀerentiation of myogenic cells isolated from dystrophic mouse muscle. Nature 1977, 270, 725–727. [CrossRef] [PubMed] 22. Ortega, M.A.; Fernández-Garibay, X.; Castaño, A.G.; De Chiara, F.; Hernández-Albors, A.; Balaguer-Trias, J.; Ramón-Azcón, J. Muscle-on-a-chip with an on-site multiplexed biosensing system for in situ monitoring of secreted IL-6 and TNF-α. Lab. Chip 2019, 19, 2568–2580. [CrossRef] [PubMed] 23. Oliveros, J.C.; Franch, M.; Tabas-Madrid, D.; San-León, D.; Montoliu, L.; Cubas, P.; Pazos, F. Breaking-Cas—interactive design of guide RNAs for CRISPR-Cas experiments for ENSEMBL genomes. Nucleic Acids Res. 2016, 44, W267–W271. [CrossRef] [PubMed] 24. Vaughan, A.; Alvarez-Reyes, M.; Bridger, J.M.; Broers, J.L.; Ramaekers, F.C.; Wehnert, M.; Morris, G.E.; Whitﬁeld, W.G.F.; Hutchison, C.J. Both emerin and lamin C depend on lamin A for localization at the nuclear envelope. J. Cell Sci. 2001, 114, 2577–2590. p 25. Sullivan, T.; Escalante-Alcalde, D.; Bhatt, H.; Anver, M.; Bhat, N.; Nagashima, K.; Stewart, C.L.; Burke, B. Loss of A-type lamin expression compromises nuclear envelope integrity leading to muscular dystrophy. J. Cell Biol. 1999, 147, 913–920. [CrossRef] 26. Muchir, A.; van Engelen, B.G.; Lammens, M.; Mislow, J.M.; McNally, E.; Schwartz, K.; Bonne, G. Nuclear envelope alterations in ﬁbroblasts from LGMD1B patients carrying nonsense Y259X heterozygous or homozygous mutation in lamin A/C gene. Exp. Cell Res. 2003, 291, 352–362. [CrossRef] 27. Hale, C.M.; Shrestha, A.L.; Khatau, S.B.; Stewart-Hutchinson, P.J.; Hernandez, L.; Stewart, C.L.; Hodzic, D.; Wirtz, D. Dysfunctional Connections Between the Nucleus and the Actin and Microtubule Networks in Laminopathic Models. Biophys. J. 2008, 95, 5462–5475. [CrossRef] 28. Favreau, C.; Higuet, D.; Courvalin, J.-C.; Buendia, B. Expression of a Mutant Lamin A That Causes Emery-Dreifuss Muscular Dystrophy Inhibits In Vitro Diﬀerentiation of C2C12 Myoblasts. Mol. Cell. Biol. 2004, 24, 1481–1492. [CrossRef] 29. Markiewicz, E.; Ledran, M.; Hutchison, C.J. Remodelling of the nuclear lamina and nucleoskeleton is required for skeletal muscle diﬀerentiation in vitro. J. Cell Sci. 2005, 118, 409–420. [CrossRef] 30. Meinke, P.; Thuy, D.N.; Wehnert, M.S. The LINC complex and human disease. Biochem. Soc. Trans. 2011, 39, 1693–1697. [CrossRef] 31. Gerbino, A.; Procino, G.; Svelto, M.; Carmosino, M. Role of Lamin A/C Gene Mutations in the signaling defects leading to cardiomyopathies. Front. Physiol. 2018, 9, 1356. [CrossRef] [PubMed] 32. References Choi, J.C.; Wu, W.; Muchir, A.; Iwata, S.; Homma, S.; Worman, H.J. Dual speciﬁcity phosphatase 4 mediates cardiomyopathy caused by lamin A/C (LMNA) gene mutation. J. Biol. Chem. 2012, 287, 40513–40524. [CrossRef] [PubMed] 20 of 21 Cells 2020, 9, 1286 20 of 21 33. Burla, R.; La Torre, M.; Merigliano, C.; Vernì, F.; Saggio, I. Genomic instability and DNA replication defects in progeroid syndromes. Nucleus 2018, 9, 368–379. [CrossRef] [PubMed] 34. Graziano, S.; Kreienkamp, R.; Coll-Bonﬁll, N.; Gonzalo, S. Causes and consequences of genomic instability in laminopathies: Replication stress and interferon response. Nucleus 2018, 9, 289–306. [CrossRef] 35. Chen, S.N.; Lombardi, R.; Karmouch, J.; Tsai, J.Y.; Czernuszewicz, G.; Taylor, M.R.G.; Mestroni, L.; Coarfa, C.; Gurha, P.; Marian, A.J. DNA Damage Response/TP53 Pathway Is Activated and Contributes to the Pathogenesis of Dilated Cardiomyopathy Associated with LMNA (Lamin A/C) Mutations. Circ. Res. 2019, 124, 856–873. [CrossRef] 36. Ho, C.Y.; Jaalouk, D.E.; Vartiainen, M.K.; Lammerding, J. Lamin A/C and emerin regulate MKL1-SRF activity by modulating actin dynamics. Nature 2013, 497, 507–511. [CrossRef] 37. Liang, Y.; Chiu, P.H.; Yip, K.Y.; Chan, S.Y. Subcellular Localization of SUN2 Is Regulated by Lamin A and Rab5. PLoS ONE 2011, 6, e20507. [CrossRef] 38. Méjat, A.; Decostre, V.; Li, J.; Renou, L.; Kesari, A.; Hantaï, D.; Stewart, C.L.; Xiao, X.; Hoﬀman, E.; Bonne, G.; et al. Lamin A/C–mediated neuromuscular junction defects in Emery-Dreifuss muscular dystrophy. J. Cell Biol. 2009, 184, 31–44. [CrossRef] 39. Haque, F.; Lloyd, D.J.; Smallwood, D.T.; Dent, C.L.; Shanahan, C.M.; Fry, A.M.; Trembath, R.C.; Shackleton, S. SUN1 Interacts with Nuclear Lamin A and Cytoplasmic Nesprins To Provide a Physical Connection between the Nuclear Lamina and the Cytoskeleton. Mol. Cell. Biol. 2006, 26, 3738. [CrossRef] 40. Hasan, S.; Güttinger, S.; Mühlhäusser, P.; Anderegg, F.; Bürgler, S.; Kutay, U. Nuclear envelope localization of human UNC84A does not require nuclear lamins. FEBS Lett. 2006, 580, 1263–1268. [CrossRef] 41. Chen, C.-Y.; Chi, Y.-H.; Mutalif, R.A.; Starost, M.F.; Myers, T.G.; Anderson, S.A.; Stewart, C.L.; Jeang, K.-T. Accumulation of the inner nuclear envelope protein Sun1 is pathogenic in progeric and dystrophic laminopathies. Cell 2012, 149, 565–577. [CrossRef] [PubMed] 42. Ostlund, C.; Folker, E.S.; Choi, J.C.; Gomes, E.R.; Gundersen, G.G.; Worman, H.J. Dynamics and molecular interactions of linker of nucleoskeleton and cytoskeleton (LINC) complex proteins. J. Cell Sci. 2009, 122, 4099–4108. [CrossRef] [PubMed] 43. References Mattioli, E.; Columbaro, M.; Capanni, C.; Maraldi, N.M.; Cenni, V.; Scotlandi, K.; Marino, M.T.; Merlini, L.; Squarzoni, S.; Lattanzi, G. Prelamin A-mediated recruitment of SUN1 to the nuclear envelope directs nuclear positioning in human muscle. Cell Death Diﬀer. 2011, 18, 1305–1315. [CrossRef] [PubMed] 44. Muchir, A.; Wu, W.; Worman, H.J. Reduced expression of A-type lamins and emerin activates extracellular signal-regulated kinase in cultured cells. Biochim. Biophys. Acta 2009, 1792, 75–81. [CrossRef] [PubMed] 45. Frock, R.L.; Kudlow, B.A.; Evans, A.M.; Jameson, S.A.; Hauschka, S.D.; Kennedy, B.K. Lamin A/C and emerin are critical for skeletal muscle satellite cell diﬀerentiation. Genes Dev. 2006, 20, 486–500. [CrossRef] [PubMed] 46. Janin, A.; Bauer, D.; Ratti, F.; Valla, C.; Bertrand, A.; Christin, E.; Chopin, E.; Streichenberger, N.; Bonne, G.; Gache, V.; et al. SMAD6 overexpression leads to accelerated myogenic diﬀerentiation of LMNA mutated cells. Sci. Rep. 2018, 8, 5618. [CrossRef] 47. Liu, Q.; Pante, N.; Misteli, T.; Elsagga, M.; Crisp, M.; Hodzic, D.; Burke, B.; Roux, K.J. Functional association of Sun1 with nuclear pore complexes. J. Cell Biol. 2007, 178, 785–798. [CrossRef] 48. Jahed, Z.; Soheilypour, M.; Peyro, M.; Mofrad, M.R.K. The LINC and NPC relationship—It’s complicated! J. Cell Sci. 2016, 129, 3219–3229. [CrossRef] 49. Le Dour, C.; Macquart, C.; Sera, F.; Homma, S.; Bonne, G.; Morrow, J.P.; Worman, H.J.; Muchir, A. Decreased WNT/β-catenin signalling contributes to the pathogenesis of dilated cardiomyopathy caused by mutations in the lamin a/C gene. Hum. Mol. Genet. 2017, 26, 333–343. [CrossRef] 50. Chen, H.; Li, C.; Peng, X.; Zhou, Z.; Weinstein, J.N.; Liang, H.; Caesar-Johnson, S.J.; Demchok, J.A.; Felau, I.; Kasapi, M.; et al. A Pan-Cancer Analysis of Enhancer Expression in Nearly 9000 Patient Samples. Cell 2018, 173, 386–399. [CrossRef] 51. Goldman, R.D.; Shumaker, D.K.; Erdos, M.R.; Eriksson, M.; Goldman, A.E.; Gordon, L.B.; Gruenbaum, Y.; Khuon, S.; Mendez, M.; Varga, R.; et al. Accumulation of mutant lamin A causes progressive changes in nuclear architecture in Hutchinson–Gilford progeria syndrome. Proc. Natl. Acad. Sci. USA 2004, 101, 8963–8968. [CrossRef] [PubMed] 52. Camozzi, D.; Capanni, C.; Cenni, V.; Mattioli, E.; Columbaro, M.; Squarzoni, S.; Lattanzi, G. Diverse lamin-dependent mechanisms interact to control chromatin dynamics. Nucleus 2014, 5, 427–440. [CrossRef] [PubMed] 21 of 21 21 of 21 Cells 2020, 9, 1286 53. Guelen, L.; Pagie, L.; Brasset, E.; Meuleman, W.; Faza, M.B.; Talhout, W.; Eussen, B.H.; de Klein, A.; Wessels, L.; de Laat, W.; et al. References Domain organization of human chromosomes revealed by mapping of nuclear lamina interactions. Nature 2008, 453, 948–951. [CrossRef] [PubMed] 54. Liu, B.; Wang, J.; Chan, K.M.; Tjia, W.M.; Deng, W.; Guan, X.; Huang, J.; Li, K.M.; Chau, P.Y.; Chen, D.J.; et al. Genomic instability in laminopathy-based premature aging. Nat. Med. 2005, 11, 780–785. [CrossRef] [PubMed] 55. Musich, P.R.; Zou, Y. DNA-damage accumulation and replicative arrest in Hutchinson–Gilford progeria syndrome. Biochem. Soc. Trans. 2011, 39, 1764–1769. [CrossRef] [PubMed] 56. Saha, B.; Zitnik, G.; Johnson, S.; Nguyen, Q.; Risques, R.A.; Martin, G.M.; Oshima, J. DNA damage accumulation and TRF2 degradation in atypical Werner syndrome ﬁbroblasts with LMNA mutations. Front. Genet. 2013, 4, 129. [CrossRef] [PubMed] 57. Wolf, C.M.; Wang, L.; Alcalai, R.; Pizard, A.; Burgon, P.G.; Ahmad, F.; Sherwood, M.; Branco, D.M.; Wakimoto, H.; Fishman, G.I.; et al. Lamin A/C haploinsuﬃciency causes dilated cardiomyopathy and apoptosis-triggered cardiac conduction system disease. J. Mol. Cell. Cardiol. 2008, 44, 293–303. [CrossRef] 58. García-Lizarribar, A.; Fernández-Garibay, X.; Velasco-Mallorquí, F.; Castaño, A.G.; Samitier, J.; Ramon-Azcon, J. Composite Biomaterials as Long-Lasting Scaﬀolds for 3D Bioprinting of Highly Aligned Muscle Tissue. Macromol. Biosci. 2018, 18, 1–13. [CrossRef] © 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/).
W2145160339.txt	https://journals.plos.org/plosgenetics/article/file?id=10.1371/journal.pgen.1003567&type=printable	en		Breakage-fusion-bridge Cycles and Large Insertions Contribute to the Rapid Evolution of Accessory Chromosomes in a Fungal Pathogen	PLOS genetics	2,013	cc-by	13,180	Breakage-fusion-bridge Cycles and Large Insertions Contribute to the Rapid Evolution of Accessory Chromosomes in a Fungal Pathogen Daniel Croll, Marcello Zala, Bruce A. McDonald Plant Pathology, Institute of Integrative Biology, ETH Zurich, Zurich, Switzerland Abstract Chromosomal rearrangements are a major driver of eukaryotic genome evolution, affecting speciation, pathogenicity and cancer progression. Changes in chromosome structure are often initiated by mis-repair of double-strand breaks in the DNA. Mis-repair is particularly likely when telomeres are lost or when dispersed repeats misalign during crossing-over. Fungi carry highly polymorphic chromosomal complements showing substantial variation in chromosome length and number. The mechanisms driving chromosome polymorphism in fungi are poorly understood. We aimed to identify mechanisms of chromosomal rearrangements in the fungal wheat pathogen Zymoseptoria tritici. We combined population genomic resequencing and chromosomal segment PCR assays with electrophoretic karyotyping and resequencing of parents and offspring from experimental crosses to show that this pathogen harbors a highly diverse complement of accessory chromosomes that exhibits strong global geographic differentiation in numbers and lengths of chromosomes. Homologous chromosomes carried highly differentiated gene contents due to numerous insertions and deletions. The largest accessory chromosome recently doubled in length through insertions totaling 380 kb. Based on comparative genomics, we identified the precise breakpoint locations of these insertions. Nondisjunction during meiosis led to chromosome losses in progeny of three different crosses. We showed that a new accessory chromosome emerged in two viable offspring through a fusion between sister chromatids. Such chromosome fusion is likely to initiate a breakage-fusion-bridge (BFB) cycle that can rapidly degenerate chromosomal structure. We suggest that the accessory chromosomes of Z. tritici originated mainly from ancient core chromosomes through a degeneration process that included BFB cycles, nondisjunction and mutational decay of duplicated sequences. The rapidly evolving accessory chromosome complement may serve as a cradle for adaptive evolution in this and other fungal pathogens. Citation: Croll D, Zala M, McDonald BA (2013) Breakage-fusion-bridge Cycles and Large Insertions Contribute to the Rapid Evolution of Accessory Chromosomes in a Fungal Pathogen. PLoS Genet 9(6): e1003567. doi:10.1371/journal.pgen.1003567 Editor: Joseph Heitman, Duke University Medical Center, United States of America Received November 28, 2012; Accepted May 1, 2013; Published June 13, 2013 Copyright: ß 2013 Croll et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by an ETH Zurich grant [ETH-03 12; www.ethz.ch] to DC and BAM and by Swiss National Science Foundation grants to DC [PA00P3_145360; www.snf.ch] and BAM [31003A_134755; www.snf.ch]. 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: daniel.croll@usys.ethz.ch of dispersed repeats [14–16]. Telomeres play a major role in maintaining chromosome stability [17,18]. Although chromosomes lacking a telomere are particularly susceptible to chromosomal fusion, subtelomeric double-strand breaks may also cause chromosomal fusion [19]. McClintock’s classic cytogenetic work on maize in the 1930s and 1940s showed that mis-repair of damaged chromosomal ends could generate cycles of chromosomal degeneration termed breakage-fusion-bridge (BFB) cycles [20,21]. BFB cycles begin when a telomere breaks off a chromosome. When the damaged chromosome replicates, its sister chromatids fuse and form a bridge during anaphase, with the two centromeres of the fused sister chromatids pulled into opposite poles of the dividing cell. After the bridge breaks, the resulting daughter cells receive defective chromosomes that lack telomeres and can initiate new BFB cycles. BFB cycles have also been identified in animals [22,23] and yeast [24,25]. In humans, BFB cycles play a significant role in cancer progression [18,26,27]. Fungal chromosomes are generally too small for traditional cytogenetic analyses based on chromosome staining and microscopic examination. But fungi were found to show extensive Introduction Chromosomal rearrangements are major drivers of genome evolution. Dobzhansky [1] realized that chromosomal polymorphism would ‘‘supply the raw materials for evolution’’, providing some of the earliest support for Darwin’s theory of evolution. Since Dobzhansky’s work on Drosophila, cytogenetic studies have revealed a large number of chromosomal rearrangements in the genomes of plant and animal species [2], including humans [3]. Chromosomal rearrangements were shown to contribute to sex chromosome differentiation [4,5], reproductive isolation [6], speciation [7–10] and complex adaptive phenotypes [11]. Chromosomal rearrangements involve deletions, duplications, inversions and translocations within and among chromosomes. In most cases, the molecular mechanisms that generated the observed rearrangements are not known, but a common explanation is misrepair of double-stranded DNA breaks [12,13]. Repetitive DNA has been strongly associated with chromosome rearrangements in plant and animal genomes and is thought to promote non-allelic homologous recombination during meiosis due to the misalignment PLOS Genetics \| www.plosgenetics.org 1 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen series of genes important for virulence [43]. In the rice blast fungus Magnaporthe oryzae and related species, a major effector called AVRPita that confers virulence on rice was frequently translocated between subtelomeric regions of different chromosomes including accessory chromosomes [44]. Flanking retrotransposons likely contributed to the extreme mobility of the AVR-Pita gene within and among closely related species. The largest known complement of accessory chromosomes is found in the wheat pathogen Zymoseptoria tritici (syn. Mycosphaerella graminicola [45]). The eight smallest chromosomes of the reference genome of Z. tritici, ranging in size from 409–773 kb, were identified as accessory chromosomes [46]. The core chromosomes of the reference genome range in size from 1,186–6,089 kb [46]. In contrast to accessory chromosomes found in other pathogenic fungi, Z. tritici accessory chromosomes contain over six hundred annotated genes, however the function of these genes is poorly understood [46]. The fungus shows extensive chromosomal length and number polymorphisms within random mating field populations [30,47]. Some of the chromosomal diversity appears to be generated through meiosis because progeny populations exhibited frequent chromosome loss and disomy as a result of nondisjunction of accessory chromosomes [48]. The origin of the accessory chromosomes of Z. tritici is not known, though both horizontal chromosome transfer from an unknown donor and degeneration of core chromosomes have been proposed [46]. Comparative genomics of closely related species suggested that several accessory chromosomes originated prior to the emergence of Z. tritici [49]. Several lines of evidence suggest that accessory chromosomes may be important for virulence, including the finding that genes on accessory chromosomes are under accelerated evolution and that these are more likely to show a protein signature consistent with a role in pathogenicity [46,49]. The large set of accessory chromosomes in Z. tritici and its close relatives provides a powerful model system to elucidate the mechanisms underlying fungal chromosomal polymorphisms and the origins of accessory chromosomes. We combined population genomic resequencing and PCR-based chromosome segment genotyping to measure the diversity in chromosomal structure at a global scale. We then performed controlled sexual crosses to trace the fate of accessory chromosomes through meiosis and to identify structural rearrangements in chromosomes among the progeny. We confirmed the findings from our resequencing data with electrophoretic karyotyping that enabled chromosomal separation, isolation and visualization by Southern blotting. Our study provides the most comprehensive view to date of mechanisms underlying chromosomal polymorphisms in evolving fungal populations. Author Summary Chromosomal rearrangements are a hallmark of genetic differences between species. But changes in chromosome structure can also occur spontaneously within species, within populations, or even within individuals. The causes and consequences of chromosomal rearrangements affecting natural populations are poorly understood. We investigated a class of fungal chromosomes called accessory chromosomes that are not shared among all individuals within a species. Using a fungal pathogen possessing numerous accessory chromosomes as a model, we assessed chromosome diversity based on wholegenome sequencing and a PCR assay of chromosomal segments that included a global collection of isolates. We show that the accessory chromosomes are highly variable in their gene content and that geographic differences correlate with the number and the structure of the chromosomes. We applied the same approach to document chromosomal rearrangements occurring during sexual reproduction. We identified viable offspring carrying a novel chromosome that originated from a large duplication affecting the majority of the chromosome. Our study showed that chromosomal structure can evolve rapidly within a species to generate a highly diverse set of accessory chromosomes. This chromosomal diversity may contribute significantly to the adaptive potential of fungal pathogens. chromosomal polymorphisms following the invention of pulsedfield gel electrophoresis (PFGE). Application of PFGE revealed that many fungal species exhibit a high variability in chromosome number and size, even among individuals drawn from the same random mating population [28–30]. Mechanisms generating the differences in chromosome length and number remained largely elusive, although chromosome breakage and non-allelic homologous recombination among repetitive elements during meiosis were suggested to play a role [28,31]. High chromosomal variability in pathogenic fungi may play an important adaptive role [32]. For example, dramatic changes in copy numbers of an arsenite efflux transporter in Cryptococcus neoformans occurred during experimental evolution favoring arsenite tolerance [33]. Chromosomal disomy was associated with increased antifungal drug resistance in several human pathogens including C. neoformans and Candida albicans [34,35]. Copy-number variation and aneuploidy were frequently found in clinical and environmental isolates of the same species [36–39]. Some of the most polymorphic chromosomal complements were found in plant pathogenic fungi. Several species carry chromosomes that are not shared among all members of the species [40]. Chromosomes exhibiting a presence/absence polymorphism within a species have been referred to as B, dispensable, supernumerary or accessory chromosomes to differentiate them from the ‘‘core’’ chromosomes that are shared among all members of a species [29,32,40]. We refer to the chromosomes not shared among all individuals as accessory chromosomes because many of these chromosomes play an adaptive role in pathogen evolution, hence these chromosomes are not truly dispensable [32]. Nor do they fit the classic definition of B chromosomes, because they can carry many coding genes and may be necessary for survival in some environments. One of the best studied fungal accessory chromosomes was found in isolates of the pathogen Nectria haematococca and contains a gene cluster important for virulence on peas [41,42]. The tomato pathogen Fusarium oxysporum f. sp. lycopersici contains several accessory chromosomes that carry a PLOS Genetics \| www.plosgenetics.org Results Global populations are highly differentiated for presence or absence of chromosomal segments Z. tritici is distributed globally and exhibits high genetic diversity for neutral markers [50–52] as well as high phenotypic diversity for quantitative traits, including virulence and thermal adaptation [51–53]. To assess the composition and frequency of accessory chromosomes across global populations, we designed 57 PCR assays covering all 8 known accessory chromosomes found in the reference strain IPO323 (chromosomes 14–21; [46]). Amplicons ranging in size from 400–600 bp were targeted to coding regions and primer sites were chosen in conserved regions of each gene (Table S1). The genes comprised in the PCR assay were evenly distributed along the accessory chromosomes and were located mostly in GC-rich regions interspersed by regions of higher repeat 2 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 1. Global survey of diversity in accessory chromosomes of Zymoseptoria tritici. A) Presence or absence of chromosomal segments assayed by PCR in a global collection of four field populations located in Australia, Israel, United States and Switzerland (total n = 98). Green bars represent the number of chromosomal segments found within the populations. Core chromosomes 10 and 13 were included for comparison with the accessory chromosomes 14–21. B) Population differentiation based on the presence of chromosomal segments calculated by Wright’s FST. C) The physical location of each gene used for the PCR assays is shown on schematics drawn for each accessory chromosome. The variation of GC-content along the chromosomes is shown in red. D) Content of short direct repeats assessed in 20 kb segments. E) Location of coding regions according to PLOS Genetics \| www.plosgenetics.org 3 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen the reference genome [46]. The location of probes used for Southern hybridizations to CHEF gels are indicated in red. F) Gene ontology terms for genes comprised in the PCR assay. Only genes on accessory chromosomes 14 and 16 were described by gene ontology terms [46]. doi:10.1371/journal.pgen.1003567.g001 Table 1). We identified genomic divergence between the reference isolate IPO323 and the resequenced isolates by mapping all sequence reads to the finished reference genome. To avoid spurious read mapping in repetitive regions of the chromosomes, we restricted our comparison to the coding regions of the accessory chromosomes. Furthermore, we considered exons of multi-exon genes separately to avoid biases introduced by gene length. In summary, we mapped reads to 1763 exons corresponding to 654 unique genes. The average exon length on accessory chromosomes is 314 bp compared to 517 bp on core chromosomes. The read depth from the resequencing data of 9 Swiss isolates mapped against the reference genome did not suggest any disomic chromosomes (i.e. doubled read depth for a particular chromosome). However, the different isolates varied greatly in gene content on accessory chromosomes. Four isolates (3C4, 3D1, 3F5 and 1A5; Figure 3) showed a nearly complete set of coding sequences compared to the reference genome, with a substantial number of coding sequences present on all 8 accessory chromosomes. Isolate 3D7 contained the smallest complement of accessory chromosome genes, as only four chromosomes showed a substantial proportion of coding sequences to be present. The read mapping to coding regions indicated that accessory chromosomes 14, 19 and 21 likely differ in length among homologous chromosomes (Figure 3). Chromosome 16 was found in all isolates except one. However, chromosome 16 likely differs substantially among isolates due to a large number of deletions compared to the chromosome 16 of the reference genome. Nearly all surveyed 20 kb segments along chromosome 16 showed missing genes in at least some of the resequenced isolates. The strongest variation in coding sequence complements was found among variants of chromosome 14. Isolate 3D1 lacked 149 out of 292 coding sequences, while smaller segments of missing coding sequences were found in six isolates (9G4C, 3B8, 3C4, 3F5, 1E4 and 1A5). The number of missing coding sequences ranged from 18–45 among these six isolates. At one end of chromosome 19, isolate 3B8 showed 46 missing coding sequences out of 220 coding sequences. Similarly, isolates 1A5 and 1E4 showed 31 missing coding sequences out of 155 coding sequences on chromosome 21. content (Figure 1C and 1D). Gene density varies along accessory chromosomes and the PCR assays covered the entire range of known gene locations for each chromosome (Figure 1E). The function of most genes included in the PCR assay is unknown and only 7 out of 57 genes were characterized by gene ontology (Figure 1F; [46]). As a control we designed 15 additional PCR assays covering core chromosomes 10 and 13. Known microsatellite loci were included in each PCR as a positive control. In total, we surveyed 98 isolates sampled from a global collection of four wheat fields at 72 evenly spaced chromosome positions (Table S2): Oregon, United States (n = 19), Israel (n = 23), Australia (n = 30) and Switzerland (n = 26). The PCR assays on the core chromosomes 10 and 13 showed that 10 chromosomal segments were present in all 98 isolates (Figure 1A). Three chromosomal segments were missing in 1–3 isolates distributed at random across the populations. One segment on each chromosome was missing in a large fraction of the isolates, but was at approximately the same frequency across all populations (Figure 1A). None of the isolates was missing an entire core chromosome. In contrast, chromosomal segments on accessory chromosomes showed large frequency variations among populations and different accessory chromosomes showed different patterns of segmental presence/absence (Figure 1A). With the exception of chromosome 18, all accessory chromosomes were found at a frequency higher than 50% in the four field populations. Chromosome 16 was present at the highest frequency with several chromosomal segments being fixed within populations. Individual accessory chromosomes showed substantial differences compared to the chromosomes of the Dutch reference strain IPO323. Central chromosomal segments located on chromosome 14 were almost entirely missing in isolates from Australia, the United States and Israel. Swiss isolates showed a central chromosomal segment at approximately half the frequency as segments closer to the telomeric ends of the chromosome. The haplotypic diversity for the presence or absence of individual chromosomal segments was substantial among isolates (Figure S1). Nearly every isolate showed a unique combination of presence or absence of individual accessory chromosome segments. We assessed the population differentiation for presence or absence of chromosomal segments among populations using Wright’s FST statistic. Frequencies of several accessory chromosome segments were strongly differentiated among populations. The central segments of chromosome 14 showed FST ranging from 0.15– 0.55 (Figure 1B). High levels of differentiation were also found for the second segment of chromosome 15 and the first segment of chromosome 17. Chromosome 18 showed elevated levels of differentiation across the chromosome, largely because this chromosome was almost entirely missing from the Australian and USA populations (Figure 1A). In contrast, previous data on neutral genetic markers on core chromosomes showed little differentiation among these and other populations [50]. Chromosome 14 harbors a substantial length polymorphism gained through a recent insertion To investigate the nature of large missing chromosomal segments, we performed chromosome-length dotplots of the reference strain chromosome sequence against assemblies of the resequenced isolates. In particular we were interested in whether the large missing segments of chromosome 14 found in isolate 3D1 were due to a single deletion event. The comparison of the reference chromosome 14 of IPO323 with genomic scaffolds of resequenced isolates showed that both the Swiss isolate 3D1 and a previously sequenced Iranian isolate A26b carried one large deletion spanning nearly 400 kb (Figure 4). In addition, we identified two shorter deletions at homologous locations in both isolates (at 210–250 kb and 690–720 kb) compared to the reference chromosome 14. Interestingly, isolate 9G4C was lacking the large central deletion, however, this isolate shared the two peripheral deletions with isolates 3D1 and A26b (Figure 4). A fourth isolate (1E4) shared only the 690–720 kb deletion (Figure 4). In order to determine the sequence of events leading to the large length polymorphism of chromosome 14 segregating within Z. tritici populations, we performed dotplots with genomic assemblies Population resequencing revealed variation in gene content among homologous chromosomes We found substantial karyotypic diversity in accessory chromosomes among isolates from Switzerland (Figure 2). In order to obtain a fine-scale map of structural variation in accessory chromosomes among isolates, we performed Illumina resequencing on 9 of the Swiss isolates (mapping coverage 10–236; PLOS Genetics \| www.plosgenetics.org 4 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 2. Electrophoretic karyotype diversity in accessory chromosomes among eight resequenced Swiss Zymoseptoria tritici isolates compared to the reference isolate IPO323. Chromosomal bands were separated by CHEF gel electrophoresis. The size marker is Saccharomyces cerevisiae chromosomes (Sc). doi:10.1371/journal.pgen.1003567.g002 second set of breakpoints is located at 256,832 bp and 609,754 bp (breakpoints C and D on Figure 5D). of three closely related species. We identified significant matching scaffold sequences from isolates of the closest relative Z. pseudotritici spanning the central deletions found in 3D1 and A26b (Figure 4). In the more ancestral species Z. ardabiliae we did not identify any significant matches for chromosome 14. However, in the more distantly related species Z. passerinii, a genomic scaffold spanned the entire central region. The deletion matched the regions identified in 3D1 and A26b, as well as Z. pseudotritici (Figure 4). This suggests that the ancestral chromosome 14 was significantly shorter than the chromosome 14 found in the reference strain IPO323. Furthermore, this finding indicates that the missing sequences in 3D1 and A26b actually represent large insertions into chromosome 14 of the reference strain. We aimed to ascertain whether the predicted length variants of chromosome 14 are reflected in the karyotypic profiles of the different isolates. For this, we used chromosome-specific probes to identify chromosome 14 in different Z. tritici isolates and Z. passerinii. Hybridization with two chromosome-specific probes (see Table 2) located at opposite ends of the chromosomes showed that the reference isolate IPO323 carried a chromosome 14 in the size range of 780 kb (Figures 5A and 5B; data shown for probe 2) as expected for the isolate [46]. Isolates 3D1 and A26b both carried a substantially shorter chromosome 14 in the range of 400–450 kb, as predicted from the genomic scaffold alignments. The outgroup species Z. passerinii also carried a chromosome 14 that is substantially shorter than in IPO323 (Figures 5A and 5B). Isolate 9G4C was predicted to be of intermediate size between the variants found in IPO323 and 3D1 and A26b. Hybridization with chromosome-specific probes indeed identified a chromosome 14 variant of about 530 kb (Figures 5A and 5B; data shown for probe 2). To better understand mechanisms leading to the sequence insertions, we identified the precise locations of the breakpoints by performing multiple sequence alignments of the reference chromosome 14 and the scaffold sequences of 3D1, A26b, 9G4C and Z. passerinii. Interestingly, the four sequence breakpoints characterizing the central section of chromosome 14 are at exactly homologous positions in Z. tritici and Z. passerinii (Figure 5D). The first set of breakpoints is located at 213,639 bp and 250,917 bp (breakpoints A and B on Figure 5D) in the IPO323 genome. The PLOS Genetics \| www.plosgenetics.org Genomic characterization of the insertion in chromosome 14 We aimed to identify the nature of the novel sequences inserted into chromosome 14 of the reference strain. The overall GCcontent of chromosome 14 was 48.5% and corresponds to the lowest chromosomal GC content of the Z. tritici reference genome [46]. The two sequences located between breakpoints A-B and E-F showed a consistently lower GC-content than neighboring sequences (Figure 6). The largest sequence, located between breakpoints C and D, showed a heterogeneous GC-content. The density of repeat sequences increased sharply near the breakpoints of the shorter sequences located between breakpoints A-B and E-F (Figure 6). Furthermore, no genes were located between breakpoints A-B and only a single gene was found between breakpoints E-F (Figure 6). The large sequence inserted between breakpoints C-D contained several gene-poor regions. However, the overall gene density of this large sequence is similar to other regions of chromosome 14 (Figure 6). The large inserted sequence contained 16 genes with predicted functions related to a wide variety of metabolic, signaling and transcription factor activities (Figure 7B). By performing a self-alignment of the reference strain chromosome 14 sequence, we identified a substantial number of repeated sequences distributed along the chromosome. In particular, we found a large palindromic sequence located between 500–550 kb that showed high sequence similarity on both sequence strands (Figure 5C). Chromosome 14 of the reference strain contains a series of transposable element (TE) remnants distributed along the chromosome (Figure 7A). Several of the inserted sequences contain TE remnants near the flanking regions. In particular, a non-long terminal repeat (non-LTR) element is found near both flanking regions of the insertion between alignment breakpoints E and F. The same element is found at flanking regions of the two other insertions (alignment breakpoints A and D). The large palindromic sequence is flanked by outwards facing LTR Copia element remnants. 5 June 2013 \| Volume 9 \| Issue 6 \| e1003567 [SRS383009] Distorted segregation of chromosomes in sexual crosses A major contribution to polymorphisms in accessory chromosomes may arise through meiotic recombination [48]. We performed controlled crosses involving three pairs of isolates from the Swiss population and analyzed 48 progeny from each cross. We applied the same PCR assays targeting 15 chromosomal segments on two core chromosomes and 57 chromosomal segments on the accessory chromosomes. Chromosomal segments on core chromosomes that were missing in either of the two parents were found to be segregating in approximately equal proportions in all three progeny sets (Figure 8). Patterns of segregation were different for several accessory chromosomes. In Cross 1 (9B8B69G4C), we found a loss of chromosome 16 in one offspring despite the fact that both parental isolates were carrying a near full-length chromosome 16 (Figure 8E). In Cross 2 (1A561E4), we found that 8 progeny were missing all chromosome 14 segments, although both parental isolates carried the corresponding chromosome segments (Figure 8C). Similarly, chromosomes 16, 18, 20 and 21 were entirely missing from one offspring though both parents carried these chromosomes. Cross 3 (1A563D7) showed the strongest segregation distortions. Parental isolate 3D7 was missing four accessory chromosomes (Figure 8A; chromosomes 14, 15, 18 and 21). Two of these four chromosomes (15 and 21) were inherited in significantly higher proportion than expected under random segregation (X2 test, p,0.0007 multiple comparisons corrected, Figure 8B). Interestingly, in Cross 1 the parental strains similarly differed in their presence of chromosomes 15 and 21, however we did not detect any significant segregation distortion in this cross (Figure 8E). Furthermore, two progeny of Cross 3 lost accessory chromosomes 17, 19 and 20 entirely, although both parental strains carried these chromosomes. Meiosis generates novel electrophoretic karyotype profiles 945,872 We randomly selected 24 and 34 offspring from Cross 1 and Cross 2, respectively, in order to identify changes in electrophoretic karyotype profiles among progeny. Progeny of both crosses showed substantial karyotypic diversity. Through hybridization with chromosome-specific probes, we found that parental isolates of Cross 2 showed length variation for chromosome 19 of approximately 0.3 Mb (data not shown). Chromosomes 15 and 21 showed nearly identical chromosome lengths among the parental isolates. Progeny of Cross 2 segregated the two length variants of chromosome 19 in approximately equal proportions (data not shown). Larger chromosomes (1.0–3.0 Mb) of parents and progeny of Cross 2 showed similarly diverse electrophoretic karyotypes (Figure 9A). In Cross 2, we identified two progeny (A2.2 and A66.2) out of 34 tested with a chromosomal band estimated to be around 0.9 Mb. However, neither of the two parents were found to have a chromosomal band in the range of 0.7–1.2 Mb, as shown by different PFGE gels optimized to separate either the smallest (,1.0 Mb) or medium-sized chromosomal bands (1.0–3.0 Mb) (Figures 9A and B). 1 All samples were submitted under NCBI BioProject ID [PRJNA178194]. doi:10.1371/journal.pgen.1003567.t001 90 ST99CH9B8B 12,195,122 22.37 97,941 [SAMN01815815] [SRS383143] [SRS382972] 486,086 89 ST99CH9G4C 12,384,026 22.68 93,236 [SAMN01815816] [SRS383142] 499,110 88 PLOS Genetics \| www.plosgenetics.org ST99CH1E4 9,224,238 18.14 83,812 [SAMN01815817] [SRS383148] 665,994 [SAMN01815818] 558,175 91,984 89 15.79 89 8,114,322 ST99CH1A5 ST99CH3F5 5,781,183 10.56 96,887 [SAMN01815819] [SRS383147] 539,729 86 ST99CH3D7 5,814,117 10.96 136,993 [SAMN01815820] [SRS383145] [SRS383146] 88 ST99CH3D1 5,655,717 9.66 731,290 [SAMN01815821] [SRS383144] 109,158 [SAMN01815822] 493,063 98,927 [SAMN01815823] 475,584 109,481 88 10.17 87 59,709,810 ST99CH3C4 ST99CH3B8 6,145,362 11.11 Longest scaffold (bp) Scaffold N50 (bp) Mean coverage at mapped positions (X) Coverage of reference genome (%) Number of reads Isolate Table 1. Illumina resequencing and assembly statistics of Zymoseptoria tritici isolates from the Swiss population. BioSample ID1 SRA sample ID Rapid Chromosomal Evolution in a Fungal Pathogen Genome resequencing and chromosomal hybridization reveals aberrant fusion event during meiosis In order to elucidate the origin of the novel chromosome found in two offspring of isolates 1A5 and 1E4, we performed whole genome resequencing of these progeny. The sequencing reads were mapped to all coding sequences of the reference genome, identically to the procedure used for the resequencing of the Swiss population. The parental isolates 1A5 and 1E4 both carried an 6 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 3. Genome resequencing of nine Zymoseptoria tritici isolates from a Swiss population. Illumina sequencing reads were mapped to annotated coding regions of the reference genome IPO323. The height of the bars represents the number of coding regions found in 20 kb segments in the reference genome. Horizontal rows show the nine different resequenced isolates. For each resequenced isolate black segments of the bars indicate the sum of present coding sequences per 20 kb segments. White segments of the bars indicate absent coding sequences. doi:10.1371/journal.pgen.1003567.g003 to the reference genome (Figure 10B). The two progeny A2.2 and A66.2 also lacked the region between 1–85 kb compared to the reference genome (Figure 10B). The central region of chromosome 17 was divided into two sharply distinct regions based on read depth. A region of high read depth between 85–350 kb and a region of low read depth between 350–481 kb (Figure 10B). We tested whether the increased read depth on chromosome 17 was distinct from the read depth on other chromosomes of the progeny. We calculated the average read density on coding sequences across all 13 core chromosomes as a reference baseline. The average read densities of the parental isolate 1A5 and the two progeny A2.2 and A66.2 were respectively 11.96, 30.93 and 36.34 reads per base pair of coding sequence. We compared these average values to read densities on accessory chromosomes (Figure 10C). In order to mitigate biases introduced by large almost complete set of accessory chromosomes except that 1E4 lacked chromosome 17 (Figure 3). Progeny A2.2 and A66.2 both showed a complete set of accessory chromosomes. However, in contrast to parental isolate 1A5, we did not find any mapping reads for coding sequences spanning the terminal portion of chromosome 17 (ranging from 481–558 kb on the reference chromosome 17; Figure 10A). This missing chromosome segment would result in a reduced length of approximately 100 kb compared to the length of chromosome 17 in the reference strain (full length 584 kb; [46]). To test for potential duplication events, we used read depth on chromosome 17 as a proxy for duplicated sequences. The parental isolate 1A5 showed a homogeneous distribution of read depth along the chromosome. The parental isolate is suggested to be missing a large chromosomal segment between 1–85 kb compared PLOS Genetics \| www.plosgenetics.org 7 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen PLOS Genetics \| www.plosgenetics.org 8 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 4. Comparisons of synteny among different variants of accessory chromosome 14 in related species. Scaffolds from de novo assemblies of the Z. tritici isolates 1E4 (n = 17), 9G4C (n = 16), 3D1 (n = 11) and A26b (n = 7) aligning to chromosomes 14 of the reference isolate IPO323 (horizontal axis) are shown. Scaffolds from isolates of related species Z. pseudotritici (5.9.1; n = 9) and Z. passerinii (P63; n = 4) aligning to chromosomes 14 of the reference isolate IPO323 are shown below. Orange bars show conserved deletions shared among different isolates and species in comparison to the reference chromosome 14. The grey bar indicates a putative deletion in Z. passerinii not spanned by a scaffold. Scaffolds are differentiated by color. doi:10.1371/journal.pgen.1003567.g004 and large insertions. The chromosomal complement is highly plastic with substantial variation both in the number of accessory chromosomes carried by each isolate and variation in gene content among homologous chromosomes. We located the exact breakpoints of multiple insertions in chromosome 14 that led to a drastic chromosome length polymorphism within a population. Meiosis played a significant role in shaping accessory chromosome complements. Segregation of some accessory chromosomes was distorted, with nondisjunction during meiosis leading to frequent losses of accessory chromosomes. We found evidence for the fusion of sister chromatids of chromosome 17 in two offspring from the same cross. These progeny carried a nearly doubled chromosome 17 generated through a chromosomal fusion in a subtelomeric region that was likely initiating a breakage-fusion-bridge cycle. missing segments on the various accessory chromosomes, we calculated the average read density using only mapped positions for each isolate. Accessory chromosomes of the parental isolate 1A5 showed read densities ranging from 69.4–94.4% of the average read density on core chromosomes, with chromosome 17 showing the lowest read density. Both progeny showed on average slightly higher read densities ranging from 69.1–122% and 76.1– 133% for A2.2 and A66.2, respectively. Among all accessory chromosomes, chromosome 17 showed the largest increase (1.77– 1.92 fold) in relative read density compared to the parental isolate 1A5 in both progeny (Figure 10C). We hypothesized that this nearly two-fold increase in read density reflected a large duplication event occurring on chromosome 17. To determine the genomic content of the novel chromosomal band found in the two offspring, we excised the new chromosomal band found at 0.9 Mb from the PFGE gel of progeny A2.2. After purification and whole-genome amplification, we performed Illumina sequencing on the resulting amplified DNA. The sequencing reads were mapped to all coding sequences of the reference genome. The average read density per chromosome was highly variable, with most chromosomes showing an average read density of 0.78–17.4 reads per base pair (Figure 10D). By far the highest read density was found for coding sequences on chromosome 17 with 175 reads per base pair. We designed two genomic probes specific to chromosome 17 and hybridized the probes to chromosomal bands separated by PFGE. The probes showed that parental isolate 1A5 was carrying a chromosome 17 of the expected length as compared to the reference isolate IPO323 (Figure 9B). We found no hybridization signal on any chromosomal band for parental isolate 1E4. Both progeny A2.2 and A66.2 showed a specific hybridization signal for chromosome 17 on the novel chromosomal band at 0.9 Mb (Figures 9B and 9C; probe 4 see Table 2). A second chromosomespecific probe for chromosome 17 gave identical results (data not shown). Taken together, this strongly suggests that the novel chromosome band is either entirely or almost entirely constituted by sequences belonging to chromosome 17. Effects of population structure on global chromosomal diversity The global survey of chromosomal segments revealed highly diverse accessory chromosome complements. We found that isolates not only differed in the number of accessory chromosomes as expected, but that homologous chromosomes showed markedly different gene contents due to numerous insertions and deletions. Several accessory chromosomes such as chromosome 16 were found near fixation in some populations, such as Australia and Israel. In contrast we found that chromosome 18 was almost entirely missing from the sampled Australian population. The near fixation or losses of accessory chromosomes in some populations may be due to stochastic processes such as founder events during the establishment of the pathogen in previously unaffected geographical regions. Populations also differed strongly in the diversity of chromosomal haplotypes detected by the PCR assays. The Swiss population had a much higher number of unique haplotypes for chromosomes 14, 16 and 18 than the Australian population. Founder effects were hypothesized to explain the low genetic diversity found for neutral markers in Australian Z. tritici populations that were introduced along with wheat not later than ,200 years ago [50]. In agreement with this earlier finding, accessory chromosomal segments of the Australian population showed the strongest deviation from global frequencies. However, large variations in accessory chromosome complements were also found in other populations. Hence, the diversity in chromosomal complements reflects a previously uncharacterized form of genetic Discussion We showed that the accessory chromosomes of Z. tritici underwent significant structural rearrangements including fusions Table 2. Primer sequences used for the Southern hybridization probes specific for chromosome 14 and 17, respectively. Chromosome Probe GeneID1 Forward primer (59-39) Reverse primer (59-39) 14 probe 1 97558 TTT GGA ACA TCT GAC CAC GA GCG CTG TTC TTC CAA GTA ATC 14 probe 2 111740 CAA GCA CCC TCT CAC AAA CA AAG CCC GTA GGT TGG TAT TG 17 probe 3 97809 CTC TTG ACT TCC TCC ATT GTC AT GTG AAT TGT TCG GGG AAG AG 17 probe 4 97838 CCA ATC CCA AGA AAA CCG GAC CTT TTG TGA GCT TCT CAA GTA 1 GeneID refers to the reference genome annotation [46]. doi:10.1371/journal.pgen.1003567.t002 PLOS Genetics \| www.plosgenetics.org 9 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 5. A) Pulsed-field gel electrophoresis of accessory chromosomes found in the Zymoseptoria tritici isolates IPO323 (reference isolate), 3D1, 9G4C, 26b and the Z. passerinii isolate P63. The size marker (Sc) represents chromosomes of Saccharomyces cerevisiae. Asterisks represent chromosome 14 variants identified by Southern hybridization with a chromosome specific probe. B) Southern hybridization with a PLOS Genetics \| www.plosgenetics.org 10 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen chromosome 14 specific probe. C) Chromosomal dotplot of chromosome 14 in the reference genome of IPO323. Orange blocks indicate the identified deletions compared to other isolates of Z. tritici revealing a palindrome between 500 and 550 kb. D) Multiple sequence alignments of genomic scaffolds spanning breakpoints A–D are shown for Z. tritici isolates IPO323, A26b, 3D1 and 9G4C. In addition, we aligned the genomic scaffold found in the isolate P63 of Z. passerinii. Exact locations of the breakpoints A–D are shown in parentheses and refer to chromosome 14 of the reference genome IPO323. doi:10.1371/journal.pgen.1003567.g005 differentiation in this pathogen. Frequency differences in accessory chromosomes among populations may also result from selection operating on chromosomes carrying genes that confer a selective advantage or disadvantage in particular environments. For example, gene products such as effectors that contribute to host virulence in a gene-for-gene interaction may be strongly disfavored in some wheat fields due to the presence of matching resistance genes [54]. If virulence factors such as effectors are located on accessory chromosomes, this may enable rapid adaptation in an arms race to overcome detection by the host immune system. The rapid loss of non-essential virulence factors located on accessory chromosomes may provide a significant selective advantage to a fungal pathogen [32]. events generating a chromosomal length polymorphism. Accessory chromosome 16 showed numerous short deletions spanning only a few coding sequences in the Swiss population compared to the reference chromosome. Our resequencing analysis (Figure 3) suggests that several chromosomes may be missing chromosomal ends including telomeres. However, our resequencing data was not informative on the integrity of telomeric repeats and we could not be certain that telomeres were missing in these isolates. In the reference genome, one telomere sequence on chromosome 21 could not be sequenced and may be missing [46]. Intact telomeres play a crucial role in chromosomal stability by ensuring homologous chromosomal pairing and disjunction during meiosis [17]. Defective telomeres are thought to initiate the development of breakage-fusion-bridge cycles leading to major chromosomal anomalies [17,18]. If some accessory chromosomes are indeed defective for telomeres, this may play a major role in generating the observed chromosome polymorphisms. Accessory chromosomes showed extensive variation in gene content The resequencing of Swiss isolates revealed extensive variation in gene content among homologous accessory chromosomes. In comparison to the chromosome sequence of the reference strain, accessory chromosomes of the resequenced isolates carried deletions ranging from a few genes to large sections affecting several dozens of genes. Surprisingly, missing segments were rarely contiguous as would be expected from single deletion and insertion Large insertion led to visible chromosomal length polymorphism The most dramatic chromosomal length polymorphism segregating within a population was found for chromosome 14, with the Figure 6. Genomic characterization of chromosome 14 in the Zymoseptoria tritici reference isolate IPO323. Orange shades indicate the alignment breakpoints identified among isolates of Z. tritici. A) Repeat density (for repeats up to a period size of 50 bp). B) GC content is shown in sliding windows with a window length of 5 kb. C) Numbers of genes are reported for each 5 kb window. doi:10.1371/journal.pgen.1003567.g006 PLOS Genetics \| www.plosgenetics.org 11 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 7. Remnants of transposable elements and predicted gene functions on chromosome 14 insertions. A) Identification of remnant long terminal repeat (LTR) and non-LTR transposable elements. Orange regions on chromosome 14 show the three inserted sequences flanked by alignment breakpoints A–F (as on Figure 4 and 5). The palindrome (shaded in brown; see Figure 5C) identified on chromosome 14 is flanked by two outwards facing Copia-1 type LTRs. Similarly, the inserted sequence at 700 kb (E–F) is flanked by two outwards facing non-LTR transposable elements. Related non-LTR transposable elements were also found near the flanking regions (A and D) of the large central insertion and the shorter insertion at 230 kb. B) Enlarged view of the large inserted sequence (between C and D) in chromosome 14. Black segments show the location of annotated genes. Genes with a functional prediction by Gene Ontology terms are indicated according to Goodwin et al. [46]. doi:10.1371/journal.pgen.1003567.g007 predicted to encode transcription factors or other functions. All three inserted sequences are flanked on at least one end by remnants of the same class of transposable elements. The presence of these elements near the flanking regions suggests that non-allelic homologous recombination with an unknown chromosome may have played a role in the insertion of these sequences into chromosome 14. Interestingly, the two shorter insertions in chromosome 14 showed a markedly lower GC-content than surrounding regions and these inserts were virtually devoid of genes. These isochores may be regions of reduced recombination, as the inserted regions may lack homologous sequences necessary for meiotic crossing-over. The largest inserted sequence also contains a very large palindromic sequence similar in extent to palindromes on the human Y chromosome [56,57]. The palindrome is flanked by the remnants of two copies of a transposable element, similar to the inserted sequences. In yeast, palindromes were shown to mediate gene amplification and intrachromosomal recombination and may lead to genomic instability [58,59]. Goodwin et al. [46] hypothesized that accessory chromosomes originated through an ancient horizontal transfer from an unknown donor species. Our analyses show that the accessory chromosome 14 was maintained through multiple speciation events and hence may be a remnant of an ancient core chromosome. The large insertions observed in extant Z. tritici shortest identified chromosome variant approximately half the length of the longest known chromosome variant. In a related pathogen found on barley (Z. passerinii) that is ancestral to the more closely related pathogens found on wild grasses [55] we identified a homologous chromosome. The ancestral form of chromosome 14 is largely identical to the shortest variant found in Iranian and Swiss populations. Several lines of evidence suggest that the large insertions leading to the chromosome 14 variant found in the reference strain occurred recently. First, the longest chromosome variant is found almost exclusively in the Swiss population, which is closest to the location where the Dutch reference isolate IPO323 was isolated. The A26b isolate from Iran was sampled close to the center of origin of Z. tritici and this isolate carried the shortest known chromosomal variant. Second, sequences immediately adjacent to the insertion breakpoint locations showed only a single nucleotide polymorphism compared to the reference chromosome. Third, sequences near the breakpoint location were highly similar even when compared with the phylogenetically distant Z. passerinii. A major open question is the source of the inserted sequences. We did not find closely related sequences either at a different location in the reference genome or in any resequenced strain. The largest insertion in chromosome 14 contains several dozen genes and may have functional consequences for isolates carrying the large chromosome variant, because several of these genes were PLOS Genetics \| www.plosgenetics.org 12 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 8. Variability in accessory chromosomes among progeny of three crosses between Zymoseptoria tritici isolates. A) Progeny of a cross between isolates 3D7 and 1A5 (Cross 3). Chromosomal segments of core chromosomes 10 and 13 and accessory chromosomes 14–21 were assayed by PCR. The two top rows indicate the two parental genotypes. The green bars show the number of individual chromosomal segments among the 48 progeny. B), D) and F) Test for random segregation of chromosomal segments that are present in only one of the two parental isolates. The 2log10 transformed p values were corrected for non-independence and the horizontal bar represents the Bonferroni-corrected significance PLOS Genetics \| www.plosgenetics.org 13 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen threshold (p,0.0007). C) Chromosomal segment numbers among 48 progeny from a cross between isolates 1E4 and 1A5 (Cross 2). E) Chromosomal segment numbers among 48 progeny from a cross between isolates 9B8B and 9G4C (Cross 1). doi:10.1371/journal.pgen.1003567.g008 Figure 9. Pulsed-field gel electrophoresis of accessory chromosomes in the parental isolates 1E4 and 1A5, as well as their progeny A2.2 and A66.2. The isolate IPO323 was added as a reference. Both progeny showed a new chromosomal band at 0.9 Mb that is absent in either parental isolate and other screened progeny. Sc represents chromosomes of Saccharomyces cerevisiae added as a size marker. A) Pulsed-field gel electrophoresis of medium-sized chromosomes (up to approx. 3 Mb) of parental isolates 1E4 and 1A5 and 7 progeny. Progeny A2.2 and A66.2 showed a new chromosomal band at 0.9 Mb indicated by an asterisk. B) Pulsed-field gel electrophoresis of accessory chromosomes of parental isolates 1E4 and 1A5 and progeny A2.2 and A66.2. Asterisks identify chromosome 17 variants identified by hybridization. C) Southern hybridization of a chromosome 17 specific probe on chromosomes separated by pulsed-field gel electrophoresis of parental isolates, progeny and the reference isolate IPO323 as in B). doi:10.1371/journal.pgen.1003567.g009 PLOS Genetics \| www.plosgenetics.org 14 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Figure 10. Initiation of a breakage-fusion-bridge cycle of Zymoseptoria tritici chromosome 17 during meiosis. A) Genome resequencing of two progeny (A2.2 and A66.2) from a cross between the parental isolates 1E4 and 1A5. Parent 1A5 carries accessory chromosome 17 (parent 1E4 is missing chromosome 17). Illumina sequencing reads were mapped to known coding regions of chromosome 17 on the reference genome IPO323. PLOS Genetics \| www.plosgenetics.org 15 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Black and white segments of the bars represent presence and absence, respectively, of particular coding sequences in 20 kb sections along chromosomes 17. B) Variation in read density along chromosome 17 of parental isolate 1A5 and progeny A2.2 and A66.2. C) Variation in read density among accessory chromosomes in the offspring A2.2 and A66.2. Illumina sequencing reads were mapped to coding sequences of the reference genome IPO323. Read density is reported as fold-difference between the offspring and the parental isolate 1A5. As a reference, the mean folddifference on core chromosomes is reported as a horizontal line. Both offspring showed a near two-fold higher read density on chromosome 17 compared to other accessory chromosomes. D) Illumina sequencing of an excised chromosomal band at 0.9 Mb identified by PFGE in the offspring A2.2. Illumina sequencing reads were mapped to coding sequences of all chromosomes of the reference genome IPO323. E) Schematic illustration of the hypothesized non-allelic homologous recombination between inverted repeats that generated chromosome 17 in offspring A2.2 and A66.2. The resulting isodicentric chromosome can initiate a breakage-fusion-bridge cycle while the acentric chromosome will be lost during successive rounds of cell division. doi:10.1371/journal.pgen.1003567.g010 populations suggest that chromosome 14 is undergoing a degeneration process. The insertions do not seem to have a severe effect on the fitness of the organism, as many different length variants of chromosome 14 were found segregating within the Swiss population. The tolerance to large sequence rearrangements may be a hallmark of the degeneration process affecting accessory chromosomes. and non-duplicated regions, consistent with the striking difference in read depth observed across chromosome 17 (Figure 10B). If the fused chromosome contains two centromeres, it is expected to form a bridge at anaphase and undergo BFB cycles [20,21,26]. The rejoining of broken ends during new rounds of cell division will create new chromosomal arrangements including deletions and duplications. The lack of a homologous chromosome 17 during meiosis may have contributed to the initiation of a BFB cycle. The fate of the novel chromosome over subsequent generations is currently under investigation. The meiotic pairing of the large duplicated chromosome 17 with the parental chromosome variant is likely to generate further highly unstable chromosomal variants. Z. tritici possesses a genomic defense mechanism known as RIP [46] that is common to a large number of ascomycete fungi [63]. RIP rapidly degenerates highly similar genomic regions through the introduction of point mutations. We predict that the novel chromosome 17 variant generated by duplicating a large fraction of the original chromosome would be subjected to rapid degeneration as a result of RIP. We propose that BFB cycles coupled with RIP played a major role in creating the degenerated accessory chromosomes of Z. tritici. Segregation distortion and chromosomal loss during meiosis Meiosis is thought to play a major role in genomic instability in fungi [28,48,60]. Non-allelic homologous recombination among dispersed repeats was hypothesized to be the main source of chromosomal length polymorphism in fungi [28,60]. In Z. tritici, aberrations during meiosis were suggested to lead to the loss of accessory chromosomes and hence contribute to chromosomal number polymorphism among isolates [48]. Our analyses of progeny from three different crosses showed that chromosomal loss affected nearly all accessory chromosomes. We detected low levels of chromosomal losses for all accessory chromosomes except chromosome 15. The loss of a chromosome may be due to nondisjunction of sister chromatids during meiosis. This defect during meiosis would create progeny carrying a duplicated (i.e. disomic) chromosome. Frequent loss of accessory chromosomes during meiosis poses an apparent paradox in Z. tritici. Populations would be expected to gradually lose all accessory chromosomes over generations in the absence of mechanisms to maintain the accessory chromosomes. Interestingly, our data on inheritance of accessory chromosomes revealed a mechanism that may maintain accessory chromosome complements in populations. Analyses of segregation frequencies revealed that chromosomes 15 and 21, if present in only one of the two parental strains, were inherited significantly more frequently than expected under random segregation of the chromosomes. Distorted segregation was restricted to one cross and no distortion was detected in the second cross differing in the presence of chromosomes 15 and 21. Segregation distortion was found to be a key characteristic of numerous animal B chromosomes [61]. We hypothesize that segregation distortion is one of the mechanisms that maintains accessory chromosomes in Z. tritici populations. Rapid chromosomal structure evolution drives diversity in accessory chromosomes Our study revealed extensive yet viable chromosomal rearrangements generated by meiosis. Genomic instability and insertion of exogenous sequences led to highly diversified sets of homologous chromosomes affecting hundreds of genes. The large number of insertions and deletions found among accessory chromosomes suggests that these chromosomes underwent an extensive degeneration process. The chromosomal degeneration process may well have been initiated in an ancestor of Z. tritici. The shorter gene length and lower gene density on accessory chromosomes compared to core chromosomes suggests that degeneration processes affected accessory chromosomes over long evolutionary time scales. We identified large insertions and the initiation of breakage-fusion-bridge cycles as two major contributors to chromosomal abnormalities. Surprisingly, isolates of Z. tritici appear to be highly tolerant of these abnormalities, which may contribute to the maintenance of extensive karyotypic diversity in populations. The extensive degeneration, distorted segregation and frequent loss of accessory chromosomes highlight a central question surrounding fungal accessory chromosomes: How and when do these chromosomes originate? We showed that chromosome 14 is ancient, as its origin predates several speciation events prior to the emergence of Z. tritici in the Fertile Crescent [55]. We postulate that the accessory chromosomes found in extant Z. tritici populations likely originated from the core chromosomes through a degeneration process. The initiation of chromosome degeneration is particularly likely in isolates that carry disomic chromosomes due to nondisjunction. Disomy would provide redundancy in gene content and, hence, relax selection Chromosomal fusion can initiate breakage-fusion-bridge cycles The most striking example of chromosomal plasticity was the fusion of sister chromatids to generate a much longer chromosome 17 in two progeny of Cross 2 (Figure 10). This meiotic abnormality occurred in a cross between a strain carrying a chromosome 17 similar to the reference isolate and a strain lacking the entire chromosome. One mechanism for creating the new chromosome is non-allelic homologous recombination between inverted repeats on sister chromatids ([62], Figure 10E). The recombination event could create an isodicentric chromosome 17 carrying duplicated PLOS Genetics \| www.plosgenetics.org 16 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen pressure to maintain chromosomal integrity. Chromosomal degeneration may then proceed rapidly through BFB cycles, nondisjunction and RIP of duplicated regions. The emergence of a highly diverse and rapidly evolving set of accessory chromosomes in Z. tritici illustrates how an accessory genome can be created to serve as a cradle for adaptive evolution in this and other fungal pathogens. repeats with a repeat motif between 2–50 bp [71]. For each repeat, we calculated the total length of the repeat and subtracted the number of mismatches in the repeat motif, as a proxy for the extent and purity of the repeat element. Genome resequencing of Z. tritici isolates We used the previously published genome assemblies of two Iranian isolates of Z. tritici ST01IRA26b and ST01IRA48b. In addition, we included five genomes of Z. pseudotritici (STIR04_3.11.1, STIR04_2.2.1, STIR04_4.3.1, STIR04_5.3, STIR04_5.9.1) four genomes of Z. ardabiliae (STIR04_3.3.2, STIR04_3.13.1, STIR04_1.1.1, STIR04_1.1.2) and one genome of the outgroup species Z. passerinii (P63) [72,73]. All genome assemblies are available under the NCBI BioProject [PRJNA63131] on GenBank. We resequenced nine Z. tritici isolates from Switzerland (full isolate names: ST99CH1A5, ST99CH1E4, ST99CH3B8, ST99CH3C4, ST99CH3D1, ST99CH3D7, ST99CH3F5, ST99CH9B8B and ST99CH9G4C) and two progeny from Cross 2 (A2.2 and A66.2). We performed Illumina paired-end sequencing on 500–700 bp insert libraries to generate between 1–2 Gb of quality-trimmed sequence data per isolate (theoretical coverage of 25–506). The read length was either 82 bp or 90 bp. Illumina sequence data are available from the NCBI Short Read Archive (see Table 1 for accessions). Materials and Methods Populations and isolates included in the study We assessed the diversity in chromosomal structure in a global collection of Z. tritici. We included field populations from Israel (n = 23), Oregon, USA (n = 19), Switzerland (n = 26) and Australia (n = 30) (Table S2). These populations were previously assayed for neutral genetic diversity and variation in quantitative traits [53]. These isolates showed substantial variation for several quantitative characters, including virulence, fungicide resistance and thermal adaptation, among and within populations [51–53,64]. Establishment of sexual crosses Sexual crosses were performed between three pairs of isolates from the Swiss population (see Table S2) using the established protocol for Z. tritici [65]. The crosses were between isolates ST99CH9B8B and ST99CH9G4C (Cross 1), ST99CH1A5 and ST99CH1E4 (Cross 2) and ST99CH1A5 and ST99CH3D7 (Cross 3). Genome de novo assembly We used SOAPdenovo v. 1.5 [74] to generate de novo assemblies, including scaffolding and gap closing. De novo assemblies yielded a scaffold N50 ranging from 79,920– 121,161 bp depending on the resequenced isolate. Total assembly space (scaffolds and singletons) ranged from 35.57–38.33 Mb (see Table 1). All genome assemblies are available on GenBank under BioProject [PRJNA178194] (see Table 1). The comparison with the total finished genome size for the reference isolate IPO323 (39.7 Mb) shows that the genomic assemblies account for a very large proportion of the genome of the sequenced isolates. The assembly statistics were similar to the assemblies reported earlier for the same species [72]. Chromosomal segment PCR assay In order to survey presence-absence polymorphism among accessory chromosomes, we designed PCR assays to amplify approximately 500 bp of coding sequences at regular intervals of approximately 100 kb along the chromosomes of reference strain IPO323. For detailed information on the targeted genes and chromosomal locations see Table S1. Primers for PCR amplification were designed on conserved sections of the targeted coding sequence. Sequence conservation was assessed using the reference assembly of nine resequenced Swiss isolates and two resequenced Iranian isolates (for details see below). We used Primer 3.0 for primer design [66]. In order to control for successful PCR, we included a primer pair of a microsatellite locus in each PCR mix [67]. Successful PCRs produced a band at approximately 250 bp that was clearly distinguishable from the PCR product associated with each chromosomal segment. PCR reactions were performed in 20 ml volumes containing approximately 5–10 ng genomic DNA, 0.5 mM of each primer, 0.25 mM dNTP, 0.6 U Taq polymerase (DreamTaq, Thermo Fisher, Inc.) and the corresponding PCR buffer. PCR products were visualized on agarose gels. Illumina read assembly on the reference genome We mapped the Illumina reads of each resequenced isolate and offspring to the finished genome of IPO323 [46]. We used Bowtie 2.1.0 [75] to perform the mapping, allowing only reads that were mapped as paired-ends. We assessed the read coverage on the reference genome by filtering all reads based on their mapping quality (minimum mapping quality of 20) with GATK version 2.39-ge5ebf34 [76]. Coverage of coding sequences was extracted using the BEDtools utilities [77]. We scored the absence of coding sequences conservatively, requiring that less than 10 bp of a coding sequence should be covered and that the average read density on the coding sequence would be below 26. Plotting and analyses of chromosomal segment PCR assays We used the R graphics package ggplot2 to plot the raw datasets and analyses [68,69]. Measures of genetic differentiation among populations (FST) were calculated with the function var.comp in the R package hierfstat [70]. The presence-absence data generated by the PCR assays were considered as two possible alleles at haploid loci. We tested for segregation distortion of chromosomal segments among progeny by testing for deviations from the expected 1:1 segregation ratio of presence-absence polymorphism among progeny with a x2 contingency table. We accounted for non-independence of chromosomal segments and multiple testing with a conservative Bonferroni correction. We calculated the repeat content on accessory chromosomes by identifying direct PLOS Genetics \| www.plosgenetics.org Alignment of genomic scaffolds Structural changes among chromosomes of different isolates were analyzed using Nucmer [78]. We used the –mum option requiring unique anchor matches that are unique in both the query and the reference genome. Genome assemblies were compared in pairwise comparisons between the finished reference genome of IPO323 and the draft assemblies of the different isolates of Z. tritici, Z. pseudotritici, Z. ardabiliae and Z. passerinii. In order to visualize synteny among different variants of chromosome 14, we extracted all scaffolds matching the reference chromosome 14. We discarded scaffolds that were shorter than 10 kb and that showed a 17 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen match identity with the reference chromosome of less than 80%. Scaffold alignments were plotted with the R package ggplot2 [69]. Repetitive and palindromic sequences of the reference chromosome 14 of IPO323 were visualized by performing a self-alignment with LASTZ (http://www.bx.psu.edu/,rsharris/lastz). 28 h at 55uC. During the incubation the lysing solution was changed once after 18 h and gently mixed every 2–3 h. Chromosomal plugs were washed three times for 15–20 min in 5–6 ml of a 0.1 M EDTA (pH 9.0) solution and then stored in the same solution at 4uC until they were used. Characterization of chromosome 14 Pulsed-field gel electrophoresis The finished chromosome 14 sequences were analyzed for short and medium length tandem repeats with the software Tandem Repeat Finder v. 4.04 [79]. We set the matching weight to 2, the mismatching and indel penalty to 10 and the match and indel probability to 80 and 10, respectively. The minimum alignment score was required to be 10 and the maximum period size of repeats was set to 50 bp. The occurrence of repeats was visualized along a 5 kb sliding window (with increments of 1 kb). The gene density on each chromosome was reported as the occurrence of start codons according to the latest annotation [46]. GC content of each chromosome was reported in 5 kb sliding windows with increments of 1 kb. We identified transposable element remnants on chromosome 14 by querying the annotated repeat libraries provided by Repbase Update [80]. Pulsed-field gel electrophoresis (PFGE) was carried out using a BioRad CHEF II apparatus (BioRad Laboratories, Hercules CA, USA). Chromosomal plugs were inserted into the wells of a 1.2% and 1.0% (wt/vol) agarose gel (Invitrogen, Switzerland) to separate small chromosome (,1 Mb) and medium-sized chromosomes (1.0 Mb–3.0 Mb), respectively. Small chromosomes (i.e. accessory chromosomes) were separated at 13uC in 0.56 Tris-borate-EDTA Buffer (Sambrook & Russell 2001) at 200 V with a 60–120 s pulse time gradient for 24–26 h. Medium-sized chromosomes were separated at 100 V with a 250–900 s pulse time gradient for 48– 50 h using the same buffer and running temperature as above. Gels were stained in ethidium bromide (0.5 mg/ml) for 30 min immediately after the run. Destaining was performed in water for 5–10 min. Photographs were taken under ultraviolet light with a Molecular Imager (Gel Doc XR+, BioRad, Switzerland). As size standards, we used chromosome preparations of Saccharomyces cerevisiae (BioRad, Switzerland) and Hansenula wingei (BioRad, Switzerland). Preparation of fungal material for molecular karyotyping High molecular weight chromosomal DNA (Ch-DNA) was prepared by in situ digestion of cell walls of agarose-embedded conidia. We used a slightly modified non-protoplasting method according to McCluskey et al. [81]. The following Z. tritici isolates were used: ST01IRA26b, ST99CH9B8B (parental isolate of Cross 1), ST99CH9G4C (parental isolate of Cross 1), ST99CH1A5 (parental isolate of Cross 2 and 3), ST99CH1E4 (parental isolate of Cross 2), ST99CH3B8, ST99CH3C4, ST99CH3D1, ST99CH3D7, ST99CH3F5 and IPO323. In addition, we included the isolate P63 of Z. passerinii [55]. To screen progeny of sexual crosses, we randomly selected 24 and 34 confirmed progeny from Cross 1 and Cross 2, respectively. All isolates were transferred from stocks maintained in glycerol at 280uC to Yeast Malt Agar (YMA) plates and were grown for 3 to 4 days in the dark at 18uC. After incubation, conidia were washed off the plates with sterile water and 600–800 ml of suspended conidia were transferred to 2 to 3 fresh YMA plates. The plates were incubated for 2 to 3 days as described above. Conidia were harvested using sterile distilled water and filtered through sterile Miracloth (Calbiochem, La Jolla CA, USA) into 50 ml screw-cap Falcon tubes. The tubes were filled with distilled water up to 50 ml total volume. The suspension was centrifuged at 3750 rpm at room temperature for 15 min with a clinical centrifuge (Allegra X-12R, Beckman Coulter, Brea CA, USA). The resulting pellets were resuspended in 1–3 ml TE buffer (10 mM Tris-HCL, pH 7.5; 1 mM EDTA, pH 8.0) and gently vortexed. The spore concentration of the solution was determined using a Thoma haematocytometer cell counter. An aliquot of 1.5 ml spore suspension with a concentration between 86107 to 26108 spores/ml was transferred to a fresh 50 ml screw-cap tube and incubated at 55uC in a water bath for several minutes. To each tube, 1.5 ml pre-warmed (55uC) low-melting-point agarose prepared in TE Buffer was added (2% w/v; molecular biology grade, Biofinex, Switzerland). The solution was thoroughly mixed by gentle pipetting. An aliquot of 500 ml was solidified on ice for approximately 10 min in a precooled plug casting mold (BioRad Laboratories, Switzerland). A total of five agarose plugs per isolate were incubated in 15 ml screw-top tubes containing 5 ml of a lysing solution containing 0.25 M EDTA, pH 8.0, 1.5 mg/mL protease XIV (Sigma, St. Louis MO, USA), 1.0% sodium dodecyl sulfate (Fluka, Switzerland). The incubation was performed for PLOS Genetics \| www.plosgenetics.org Southern transfer and hybridization of pulsed-field gels Southern blotting and hybridization were performed according to standard protocols [82]. In summary, hydrolysis was performed in 0.25 M HCl for 30 min and DNA was blotted onto Amersham HybondTM-N+ membranes (GE Healthcare, Switzerland) overnight under alkaline conditions [82]. DNA was fixed onto the membranes at 80uC for 2 h. Membranes were prehybridized overnight with 25 ml of a buffer containing 20% (w/v) SDS, 10% BSA, 0.5 M EDTA (pH 8.0), 1 M sodium phosphate (pH 7.2) and 0.5 ml of sonicated fish sperm solution (Roche Diagnostics, Switzerland). Probes were labeled with 32P by nick translation (New England Biolabs, Inc.) following the manufacturer’s instructions. Hybridization was performed overnight at 65uC. Blots were subjected to stringent wash conditions with a first wash in 16SSC and 0.1% SDS and a second wash with 0.26SSC and 0.1% SDS. Both washes were performed at 60uC. Membranes were exposed to X-ray film (Kodak BioMax MS) for 2 to 3 days at 280uC. All hybridization probes used to identify specific chromosomes are listed in Table 2. Excision of chromosomal band and amplification Chromosomal DNA was separated with CHEF gel electrophoresis as previously described for the separation of small chromosomes except that a 1.0% agarose gel was used. The novel 0.9 Mb chromosomal band from isolate A2.2 was excised and DNA was recovered using the Wizard SV Gel and PCR Clean-up System kit (Promega, Switzerland) with the following modifications to the manufacturer’s recommendations: during the incubation at 65uC the gel slice was vortexed two times for 5 minutes, sonication was for 3 min and followed by a final incubation for 1 min. The resulting purified DNA was amplified using a whole genome amplification kit (REPLI-g Mini Kit, Qiagen, Germany). Amplified DNA was subjected to whole genome sequencing with an Illumina HiSeq 2000 as described above. Supporting Information Figure S1 Global survey of diversity in accessory chromosomes of Zymoseptoria tritici. The presence or absence of chromosomal 18 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen Table S2 Global collection of Zymoseptoria tritici isolates included in the study. (DOCX) segments were assayed by PCR in a global collection of four field populations located in Australia, Israel, United States and Switzerland (total n = 98). Horizontal rows indicate different isolates included in the study. Green and red rectangles indicate the presence and absence, respectively, of a chromosomal segment assayed by PCR. Locations of individual PCR assays are indicated in Figure 1 and Supplementary Table S1. Core chromosomes 10 and 13 were included for comparison with the accessory chromosomes 14–21. (EPS) Acknowledgments We are grateful to Jana Drabešová for assistance in the laboratory. Ed Louis, Christine Grossen, Beat Ruffner and four anonymous reviewers provided helpful feedback on earlier versions of this manuscript. We acknowledge the Genetic Diversity Centre of the ETH Zurich for use of their lab facilities. Chromosomal position, gene identifier and primer sequences for the PCR assay on core and accessory chromosomes. (DOCX) Table S1 Author Contributions Conceived and designed the experiments: DC BAM. Performed the experiments: DC MZ. Analyzed the data: DC. Wrote the paper: DC BAM. References 24. Rank GH, Xiao W, Kolenovsky A, Arndt G (1988) FLP recombinase induction of the breakage-fusion-bridge cycle and gene conversion in Saccharomyces cerevisiae. Curr Genet 13: 273–281. 25. Hackett JA, Feldser DM, Greider CW (2001) Telomere dysfunction increases mutation rate and genomic instability. Cell 106: 275–286. 26. Gisselsson D, Pettersson L, Höglund M, Heidenblad M, Gorunova L, et al. (2000) Chromosomal breakage-fusion-bridge events cause genetic intratumor heterogeneity. Proc Natl Acad Sci USA 97: 5357–5362. doi:10.1073/ pnas.090013497. 27. Thompson SL, Compton DA (2010) Chromosomes and cancer cells. Chromosome Res 19: 433–444. doi:10.1007/s10577-010-9179-y. 28. Zolan ME (1995) Chromosome-length polymorphism in fungi. Microbiol Rev 59: 686–698. 29. Kistler HC, Miao VPW (1992) New modes of genetic change in filamentous fungi. Annu Rev Phytopathol 30: 131–153. 30. McDonald BA, Martinez J (1991) Chromosome Length Polymorphisms in a Septoria tritici Population. Curr Genet 19: 265–271. 31. Fraser JA, Huang JC, Pukkila-Worley R, Alspaugh JA, Mitchell TG, et al. (2005) Chromosomal translocation and segmental duplication in Cryptococcus neoformans. Eukaryotic Cell 4: 401–406. doi:10.1128/EC.4.2.401-406.2005. 32. Croll D, McDonald BA (2012) The accessory genome as a cradle for adaptive evolution in pathogens. PLoS Pathog 8: e1002608. doi:10.1371/journal. ppat.1002608. 33. Chow EWL, Morrow CA, Djordjevic JT, Wood IA, Fraser JA (2012) Microevolution of Cryptococcus neoformans Driven by Massive Tandem Gene Amplification. Mol Biol Evol 29: 1987–2000. doi:10.1093/molbev/mss066. 34. Selmecki A, Forche A, Berman J (2006) Aneuploidy and isochromosome formation in drug-resistant Candida albicans. Science 313: 367–370. doi:10.1126/ science.1128242. 35. Sionov E, Lee H, Chang YC, Kwon-Chung KJ (2010) Cryptococcus neoformans overcomes stress of azole drugs by formation of disomy in specific multiple chromosomes. PLoS Pathog 6: e1000848. doi:10.1371/journal.ppat.1000848. 36. Selmecki A, Bergmann S, Berman J (2005) Comparative genome hybridization reveals widespread aneuploidy in Candida albicans laboratory strains. Mol Microbiol 55: 1553–1565. doi:10.1111/j.1365-2958.2005.04492.x. 37. Hu G, Liu I, Sham A, Stajich JE, Dietrich FS, et al. (2008) Comparative hybridization reveals extensive genome variation in the AIDS-associated pathogen Cryptococcus neoformans. Genome Biol 9: R41. doi:10.1186/gb-2008-92-r41. 38. Desnos-Ollivier M, Patel S, Spaulding AR, Charlier C, Garcia-Hermoso D, et al. (2010) Mixed infections and In Vivo evolution in the human fungal pathogen Cryptococcus neoformans. MBio 1: e00092-20. doi:10.1128/mBio.0009110. 39. Li W, Averette AF, Desnos-Ollivier M, Ni M, Dromer F, et al. (2012) Genetic Diversity and Genomic Plasticity of Cryptococcus neoformans AD Hybrid Strains. G3 (Bethesda) 2: 83–97. doi:10.1534/g3.111.001255. 40. Covert SF (1998) Supernumerary chromosomes in filamentous fungi. Curr Genet 33: 311–319. 41. Coleman JJ, Rounsley SD, Rodriguez-Carres M, Kuo A, Wasmann CC, et al. (2009) The genome of Nectria haematococca: contribution of supernumerary chromosomes to gene expansion. PLoS Genet 5: e1000618. doi:10.1371/ journal.pgen.1000618. 42. Miao VPW, Covert SF, VanEtten HD (1991) A fungal gene for antibiotic resistance on a dispensable (‘‘B’’) chromosome. Science 254: 1773–1776. 43. Ma L-J, van der Does HC, Borkovich KA, Coleman JJ, Daboussi M-J, et al. (2010) Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature 464: 367–373. doi:10.1038/nature08850. 44. Chuma I, Isobe C, Hotta Y, Ibaragi K, Futamata N, et al. (2011) Multiple translocation of the AVR-Pita effector gene among chromosomes of the rice blast fungus Magnaporthe oryzae and related species. PLoS Pathog 7: e1002147. doi:10.1371/journal.ppat.1002147. 1. Dobzhansky TG (1937) Genetics and the Origin of Species. New York City: Columbia University Press. 2. Coghlan A, Eichler EE, Oliver SG, Paterson AH, Stein L (2005) Chromosome evolution in eukaryotes: a multi-kingdom perspective. Trends Genet 21: 673– 682. doi:10.1016/j.tig.2005.09.009. 3. Sebat J, Lakshmi B, Troge J, Alexander J, Young J, et al. (2004) Large-scale copy number polymorphism in the human genome. Science 305: 525–528. doi:10.1126/science.1098918. 4. Ranz JM, Maurin D, Chan YS, Grotthuss von M, Hillier LW, et al. (2007) Principles of genome evolution in the Drosophila melanogaster species group. PLoS Biol 5: e152. doi:10.1371/journal.pbio.0050152. 5. Bachtrog D (2005) Sex chromosome evolution: molecular aspects of Ychromosome degeneration in Drosophila. Genome Res 15: 1393–1401. doi:10.1101/gr.3543605. 6. Noor MA, Grams KL, Bertucci LA, Reiland J (2001) Chromosomal inversions and the reproductive isolation of species. Proc Natl Acad Sci USA 98: 12084– 12088. doi:10.1073/pnas.221274498. 7. Lai Z, Nakazato T, Salmaso M, Burke JM, Tang S, et al. (2005) Extensive chromosomal repatterning and the evolution of sterility barriers in hybrid sunflower species. Genetics 171: 291–303. doi:10.1534/genetics.105.042242. 8. Rieseberg LH (2001) Chromosomal rearrangements and speciation. Trends Ecol Evol 16: 351–358. 9. Rieseberg LH, Willis JH (2007) Plant speciation. Science 317: 910–914. doi:10.1126/science.1137729. 10. Faria R, Navarro A (2010) Chromosomal speciation revisited: rearranging theory with pieces of evidence. Trends Ecol Evol 25: 660–669. doi:10.1016/ j.tree.2010.07.008. 11. Joron M, Frezal L, Jones RT, Chamberlain NL, Lee SF, et al. (2011) Chromosomal rearrangements maintain a polymorphic supergene controlling butterfly mimicry. Nature 477: 203–206. doi:doi:10.1038/nature10341. 12. Pastink A, Eeken JC, Lohman PH (2001) Genomic integrity and the repair of double-strand DNA breaks. Mutat Res 480–481: 37–50. 13. Schubert I (2007) Chromosome evolution. Curr Opin Plant Biol 10: 109–115. doi:10.1016/j.pbi.2007.01.001. 14. Argueso JL, Westmoreland J, Mieczkowski PA, Gawel M, Petes TD, et al. (2008) Double-strand breaks associated with repetitive DNA can reshape the genome. Proc Natl Acad Sci USA 105: 11845–11850. doi:10.1073/pnas.0804529105. 15. Bzymek M, Lovett ST (2001) Instability of repetitive DNA sequences: the role of replication in multiple mechanisms. Proc Natl Acad Sci USA 98: 8319–8325. doi:10.1073/pnas.111008398. 16. Raskina O, Barber JC, Nevo E, Belyayev A (2008) Repetitive DNA and chromosomal rearrangements: speciation-related events in plant genomes. Cytogenet Genome Res 120: 351–357. doi:10.1159/000121084. 17. Murnane JP (2006) Telomeres and chromosome instability. DNA Repair 5: 1082–1092. doi:10.1016/j.dnarep.2006.05.030. 18. Bailey SM, Murnane JP (2006) Telomeres, chromosome instability and cancer. Nucleic Acids Res 34: 2408–2417. doi:10.1093/nar/gkl303. 19. Pfeiffer P, Goedecke W, Obe G (2000) Mechanisms of DNA double-strand break repair and their potential to induce chromosomal aberrations. Mutagenesis 15: 289–302. 20. McClintock B (1938) The Production of Homozygous Deficient Tissues with Mutant Characteristics by Means of the Aberrant Mitotic Behavior of RingShaped Chromosomes. Genetics 23: 315–376. 21. McClintock B (1941) The Stability of Broken Ends of Chromosomes in Zea Mays. Genetics 26: 234–282. 22. Bi X, Wei S-CD, Rong YS (2004) Telomere protection without a telomerase; the role of ATM and Mre11 in Drosophila telomere maintenance. Curr Biol 14: 1348–1353. doi:10.1016/j.cub.2004.06.063. 23. Toledo F, Buttin G, Debatisse M (1993) The origin of chromosome rearrangements at early stages of AMPD2 gene amplification in Chinese hamster cells. Curr Biol 3: 255–264. PLOS Genetics \| www.plosgenetics.org 19 June 2013 \| Volume 9 \| Issue 6 \| e1003567 Rapid Chromosomal Evolution in a Fungal Pathogen 45. Quaedvlieg W, Kema GHJ, Groenewald JZ, Verkley GJM, Seifbarghi S, et al. (2011) Zymoseptoria gen. nov.: a new genus to accommodate Septoria-like species occurring on graminicolous hosts. Persoonia 26: 57–69. doi:10.3767/ 003158511X571841. 46. Goodwin SB, M’barek SB, Wittenberg AHJ, Crane CF, Hane JK, et al. (2011) Finished genome of the fungal wheat pathogen Mycosphaerella graminicola reveals dispensome structure, chromosome plasticity, and stealth pathogenesis. PLoS Genet 7: e1002070. doi:10.1371/journal.pgen.1002070. 47. Mehrabi R, Taga M, Kema GHJ (2007) Electrophoretic and cytological karyotyping of the foliar wheat pathogen Mycosphaerella graminicola reveals many chromosomes with a large size range. Mycologia 99: 868–876. 48. Wittenberg AHJ, van der Lee TAJ, Ben M’barek S, Ware SB, Goodwin SB, et al. (2009) Meiosis drives extraordinary genome plasticity in the haploid fungal plant pathogen Mycosphaerella graminicola. PLoS ONE 4: e5863. doi:10.1371/ journal.pone.0005863. 49. Stukenbrock EH, Jørgensen FG, Zala M, Hansen TT, McDonald BA, et al. (2010) Whole-genome and chromosome evolution associated with host adaptation and speciation of the wheat pathogen Mycosphaerella graminicola. PLoS Genet 6: e1001189. doi:10.1371/journal.pgen.1001189. 50. Zhan J, Pettway RE, McDonald BA (2003) The global genetic structure of the wheat pathogen Mycosphaerella graminicola is characterized by high nuclear diversity, low mitochondrial diversity, regular recombination, and gene flow. Fungal Genet Biol 38: 286–297. 51. Zhan J, Stefanato FL, McDonald BA (2006) Selection for increased cyproconazole tolerance in Mycosphaerella graminicola through local adaptation and in response to host resistance. Mol Plant Pathol 7: 259–268. doi:10.1111/ j.1364-3703.2006.00336.x. 52. Zhan J, McDonald BA (2011) Thermal adaptation in the fungal pathogen Mycosphaerella graminicola. Mol Ecol 20: 1689–1701. doi:10.1111/j.1365294X.2011.05023.x. 53. Zhan J, Linde CC, Jürgens T, Merz U, Steinebrunner F, et al. (2005) Variation for neutral markers is correlated with variation for quantitative traits in the plant pathogenic fungus Mycosphaerella graminicola. Mol Ecol 14: 2683–2693. doi:10.1111/j.1365-294X.2005.02638.x. 54. Stukenbrock EH, McDonald BA (2009) Population genetics of fungal and oomycete effectors involved in gene-for-gene interactions. Mol Plant Microbe Interact 22: 371–380. doi:10.1094/MPMI-22-4-0371. 55. Stukenbrock EH, Banke S, Javan-Nikkhah M, McDonald BA (2007) Origin and domestication of the fungal wheat pathogen Mycosphaerella graminicola via sympatric speciation. Mol Biol Evol 24: 398–411. 56. Rozen S, Skaletsky H, Marszalek JD, Minx PJ, Cordum HS, et al. (2003) Abundant gene conversion between arms of palindromes in human and ape Y chromosomes. Nature 423: 873–876. doi:10.1038/nature01723. 57. Skaletsky H, Kuroda-Kawaguchi T, Minx PJ, Cordum HS, Hillier L, et al. (2003) The male-specific region of the human Y chromosome is a mosaic of discrete sequence classes. Nature 423: 825–837. doi:10.1038/nature01722. 58. Butler DK, Yasuda LE, Yao MC (1996) Induction of large DNA palindrome formation in yeast: implications for gene amplification and genome stability in eukaryotes. Cell 87: 1115–1122. 59. Lisnić B, Svetec I-K, Štafa A, Zgaga Z (2009) Size-dependent palindromeinduced intrachromosomal recombination in yeast. DNA Repair 8: 383–389. doi:10.1016/j.dnarep.2008.11.017. 60. Fierro F, Martı́n JF (1999) Molecular mechanisms of chromosomal rearrangement in fungi. Crit Rev Microbiol 25: 1–17. doi:10.1080/10408419991299185. 61. Jones RN, Rees H (1982) B Chromosomes. New York: Academic Press. 62. Sasaki M, Lange J, Keeney S (2010) Genome destabilization by homologous recombination in the germ line. Nat Rev Mol Cell Biol 11: 182–195. doi:10.1038/nrm2849. PLOS Genetics \| www.plosgenetics.org 63. Selker EU (2002) Repeat-induced gene silencing in fungi. Adv Genet 46: 439– 450. 64. Zhan J, Torriani SFF, McDonald BA (2007) Significant difference in pathogenicity between MAT1-1 and MAT1-2 isolates in the wheat pathogen Mycosphaerella graminicola. Fungal Genet Biol 44: 339–346. doi:10.1016/ j.fgb.2006.10.008. 65. Kema GH, Verstappen EC, Todorova M, Waalwijk C (1996) Successful crosses and molecular tetrad and progeny analyses demonstrate heterothallism in Mycosphaerella graminicola. Curr Genet 30: 251–258. 66. Rozen S, Skaletsky HJ (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 132: 365–386. 67. Goodwin SB, van der Lee TAJ, Cavaletto JR, Lintel Hekkert Te B, Crane CF, et al. (2007) Identification and genetic mapping of highly polymorphic microsatellite loci from an EST database of the septoria tritici blotch pathogen Mycosphaerella graminicola. Fungal Genet Biol 44: 398–414. doi:10.1016/ j.fgb.2006.09.004. 68. R Development Core Team (2013) R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. 69. Wickham H (2009) ggplot2: elegant graphics for data analysis. New York: Springer. 70. Goudet J (2004) hierfstat, a package for r to compute and test hierarchical Fstatistics. Mol Ecol Notes 5: 184–186. doi:doi:10.1111/j.14718286.2004.00828.x. 71. Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 16: 276–277. 72. Stukenbrock EH, Bataillon T, Dutheil JY, Hansen TT, Li R, et al. (2011) The making of a new pathogen: insights from comparative population genomics of the domesticated wheat pathogen Mycosphaerella graminicola and its wild sister species. Genome Res 21: 2157–2166. doi:10.1101/gr.118851.110. 73. Stukenbrock EH, Quaedvlieg W, Javan-Nikhah M, Zala M, Crous PW, et al. (2012) Zymoseptoria ardabilia and Z. pseudotritici, two progenitor species of the septoria tritici leaf blotch fungus Z. tritici (synonym: Mycosphaerella graminicola). Mycologia 104: 1397–407. doi:10.3852/11-374. 74. Li R, Zhu H, Ruan J, Qian W, Fang X, et al. (2010) De novo assembly of human genomes with massively parallel short read sequencing. Genome Res 20: 265–272. doi:10.1101/gr.097261.109. 75. Langmead B, Salzberg SL (2012) Fast gapped-read alignment with Bowtie 2. Nat Methods 9: 357–359. doi:10.1038/nmeth.1923. 76. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, et al. (2011) A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nat Genet 43: 491–498. doi:10.1038/ng.806. 77. Quinlan AR, Hall IM (2010) BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics 26: 841–842. doi:10.1093/bioinformatics/ btq033. 78. Kurtz S, Phillippy A, Delcher AL, Smoot M, Shumway M, et al. (2004) Versatile and open software for comparing large genomes. Genome Biol 5: R12. doi:10.1186/gb-2004-5-2-r12. 79. Benson G (1999) Tandem repeats finder: A program to analyze DNA sequences. Nucleic Acids Res 27: 573–580. 80. Jurka J (2000) Repbase update: a database and an electronic journal of repetitive elements. Trends Genet 16: 418–420. 81. McCluskey K, Russell BW, Mills D (1990) Electrophoretic karyotyping without the need for generating protoplasts. Curr Genet 18: 385–386. doi:10.1007/ BF00318221. 82. Sambrook J, Russell DW (2001) Molecular Cloning. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. 20 June 2013 \| Volume 9 \| Issue 6 \| e1003567
https://openalex.org/W1980921371	https://europepmc.org/articles/pmc4373255?pdf=render	English	null	Perception of speech rhythm in second language: the case of rhythmically similar L1 and L2	Frontiers in psychology	2,015	cc-by	13,907	ORIGINAL RESEARCH published: 25 March 2015 doi: 10.3389/fpsyg.2015.00316 Perception of speech rhythm in second language: the case of rhythmically similar L1 and L2 Mikhail Ordin* and Leona Polyanskaya* Fakultät für Linguistik und Literaturwissenschaft, Universität Bielefeld, Bielefeld, Germany We investigated the perception of developmental changes in timing patterns that happen in the course of second language (L2) acquisition, provided that the native and the target languages of the learner are rhythmically similar (German and English). It was found that speech rhythm in L2 English produced by German learners becomes increasingly stress- timed as acquisition progresses. This development is captured by the tempo-normalized rhythm measures of durational variability. Advanced learners also deliver speech at a faster rate. However, when native speakers have to classify the timing patterns characteristic of L2 English of German learners at different proﬁciency levels, they attend to speech rate cues and ignore the differences in speech rhythm. Edited by: Judit Gervain Edited by: Judit Gervain, Université Paris Descartes, France Edited by: Judit Gervain, Université Paris Descartes, France Reviewed by: Juan M. Toro, Universitat Pompeu Fabra, Spain Pilar Prieto, Universitat Pompeu Fabra, Spain Correspondence: Mikhail Ordin and Leona Polyanskaya, Fakultät für Linguistik und Literaturwissenschaft, Universität Bielefeld, Bielefeld 33615, Germany mikhail.ordin@uni-bielefeld.de; leona.polyanskaya@uni-bielefeld.de Specialty section: Specialty section: This article was submitted to Language Sciences, a section of the journal Frontiers in Psychology Received: 08 November 2014 Accepted: 05 March 2015 Published: 25 March 2015 Keywords: speech rhythm, rhythm metrics, durational variability, rhythm acquisition, rhythm perception, timing patterns, rhythm development, second language Keywords: speech rhythm, rhythm metrics, durational variability, rhythm acquisition, rhythm perception, timing patterns, rhythm development, second language Introduction Reviewed by: Juan M. Toro, Universitat Pompeu Fabra, Spain Pilar Prieto, Universitat Pompeu Fabra, Spain Reviewed by: Juan M. Toro, Universitat Pompeu Fabra, Spain Pilar Prieto, Universitat Pompeu Fabra Spain The diﬀerences between languages and linguistic varieties are manifested in the acoustic compo- nents of the signal that are perceived by the auditory system and cognitively processed to extract linguistic structures. Some minute acoustic diﬀerences are perceived by the native speakers, while some gross acoustic changes in the speech stream may be ignored—either not perceived not attended to. In this study we concentrated on the perceptual relevance of the changes in speech rhythm in second language (L2) that happen in the course of L2 acquisition, provided that the native and the target languages of the learner are rhythmically similar. Correspondence: Mikhail Ordin and Leona Polyanskaya, Fakultät für Linguistik und Literaturwissenschaft, Universität Bielefeld, Bielefeld 33615, Germany mikhail.ordin@uni-bielefeld.de; leona.polyanskaya@uni-bielefeld.de g g g y y We start by introducing the notion of rhythm. Further we move on to discussing why perception of rhythmic patterns might be linguistically relevant. Then we report how speech rhythm develops in L2 English spoken by German learners, and why it is worth studying whether people are sensi- tive to the changes in L2 speech rhythm, when the target and native languages of the learner are rhythmically similar. A brief overview of empirical studies in German and English speech rhythm is provided to highlight rhythmic similarities between languages. Later, we report the results of the perception experiment aimed to answer the main question of the research: are the rhythmic changes that happen in the course of acquisition perceptually relevant, if the L1 and L2 of the learner are rhythmically similar? In the end, we show the theoretical implications of our ﬁndings. where Those languages that produce the eﬀect of stress-timing display vowel reduction, more complex C clusters, have more diﬀerent syllable types, opposition between phonologically long and short vowels, between geminate and non-geminate consonants, are less likely to exhibit vowel har- mony and ﬁxed stress. Rhythm metrics reﬂect these language- speciﬁc phonological properties. To name a few examples, 1C is thought to be indicative of the syllabic structure, syllable com- plexity and consonantal phonotactic constraints. 1V is supposed to be indicative of the degree of vowel reduction. VarcoV and VarcoC reﬂect the same properties 1C and 1V do, but Varco measures are supposed to neutralize the eﬀect of the tempo dif- ferences, and thus to reduce the eﬀect of idiosyncrasies in speech production. %V indicates the syllabic structure and inventory. Languages with more restricted syllabic inventory operate less complex syllables, usually of the CV structure. The more types of syllables there are in the language inventory, the more conso- nants are added to the onset or coda of the syllables. This reduces the proportion of vocalic intervals to the overall duration of the utterance, and %V decreases. Prieto et al. (2012) demonstrated that prosodic edges and heads are marked by manipulating dura- tional ratios in a language-speciﬁc way, and this may also account for small diﬀerences in rhythm measures between languages. The analysis of the surface durational variability of V and C intervals indeed allows spreading the languages on a continuous scale with respect to their rhythmic properties and to say that one language is more or less stress-timed than another, or that the two lan- guages are rhythmically similar. However, it did not yet provide an unambiguous support for the Rhythm Class Hypothesis that suggests that languages are split into distinct rhythm categories. Some of these metrics are inﬂuenced by the speech rate to a higher degree than the others (Dellwo and Wagner, 2003; Dellwo, 2006; Wiget et al., 2010). For example, 1V depends on the mean duration of vowels in an utterance. That is, if speech is deliv- ered at a faster rate, mean durations become smaller and 1V tends to decrease. Dellwo (2006) suggested Varco measure to normalize for the tempo diﬀerences and to capture the diﬀer- ences in durational variability irrespective of the diﬀerences in speech rate between the languages. where where m—number of interval in an utterance for which PVI is calculated, m—number of interval in an utterance for which PVI is calculated, d—duration of kth interval. Higher values of %V and lower values of the other metrics cor- respond to the languages that are traditionally deﬁned as syllable- timed (Ramus et al., 1999; Low et al., 2000; Grabe and Low, 2002; Dellwo and Wagner, 2003; White and Mattys, 2007), and pos- sibly provide the necessary cues to diﬀerentiate the durational patterns of the rhythmically contrastive languages (Ramus and Mehler, 1999; Ramus et al., 1999). A new concept of speech rhythm has been introduced in an attempt to ﬁnd the perceptually relevant acoustic correlates of rhythmic patterns (Ramus et al., 1999). It rests on the assumption that consonantal and vocalic intervals in the speech signal can exhibit language-speciﬁc patterns of durational variability. Lan- guages that are traditionally classiﬁed as “stress-timed” exhibit higher degree of durational variability compared to “syllable- timed” languages. That is, stress-timing is characterized by more substantial diﬀerences in duration of vowels and consonantal clusters within the same utterance produced by the same speaker. To capture the variability in duration of speech intervals, a num- ber of the so-called rhythm metrics have been proposed. Among the most commonly used interval-based rhythm metrics are the pairwise variability index (PVI) (Grabe and Low, 2002), the stan- dard deviation in duration of speech intervals (1) and the per- centage of vocalic material in an utterance (%V) (Ramus et al., 1999), the coeﬃcient of variation in duration of speech inter- vals (Varco) (Dellwo and Wagner, 2003). Conventionally these metrics are applied to vocalic (V) and consonantal (C) inter- vals, i.e., sequences of consecutive vowels or consonantal clus- ters that can straddle the syllabic and word boundaries within an utterance). Yet the metrics have also been applied to capture the durational variability of other speech intervals in order to inves- tigate the multiple rhythms on multiple timescale, e.g., on the timescale of feet (Nolan and Asu, 2009) or syllables (Ordin et al., 2011). , ; , ) Dauer (1983, 1987) and Schiering (2007) analyzed the phono- logical structure of languages which give the impression of stress- timing or syllable-timing. where Grabe and Low (2002) also suggested a normalized version of PVI in an attempt to neutralize the inﬂuence of the speech rate on the measures of the local dura- tional variability. Formulas 1 and 2 show how the raw (rPVI) and the normalized (nPVI) versions are calculated. White and Mat- tys (2007) and Wiget et al. (2010) reported Varco measures, %V and nPVI are more robust to the ﬂuctuations of the speech rate compared to non-normalized metrics. Citation: Ordin M and Polyanskaya L (2015) Perception of speech rhythm in second language: the case of rhythmically similar L1 and L2. Front. Psychol. 6:316. doi: 10.3389/fpsyg.2015.00316 The word rhythm implies the idea of periodicity. Based on the auditory impression that certain events or certain speech constituents reoccur periodically in the speech stream, the languages were classiﬁed into stressed-timed (in which stressed syllables were perceived to be distributed at roughly equal intervals, e.g., German, English, Dutch, Russian) and syllable-timed (in which all syllables were perceived to be of roughly equal duration, e.g., French, Italian, Spanish). Later, a new March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 1 Perception of speech rhythm in L2 Ordin and Polyanskaya r PVI = "m −1 X k = 1 dk −dk + 1 /(m −1) # (2) rhythmic class of mora-timed languages (in which moras are sup- posedly perceived as roughly equal in duration, e.g., Japanese, West Greenlandic) was added. Experimental studies, however, failed to ﬁnd empirical evidence to support this impression (Roach, 1982; Dauer, 1983; Pamies Bertran, 1999). However, adults (Ramus et al., 1999) and even infants (Nazzi et al., 1998; Ramus and Mehler, 1999; Nazzi and Ramus, 2003) are able to diﬀerentiate between rhythmic patterns of languages that are traditionally classiﬁed as stress- and syllable-timed. Therefore, researchers continued looking for the acoustic correlates of audi- torily perceived diﬀerences in speech rhythm. (2) Frontiers in Psychology \| www.frontiersin.org Rhythm Changes in Second Language Acquisition Papers focussed on acquisition of speech rhythm in L2 are rare. Most of these studies concentrate on comparing rhythm in L2 speech with the target represented by an adult native speaker. Examined L2 speech is usually produced by rather advanced learners. The results showed that the rhythm scores in L2 speech are intermediate between those in the native and the target lan- guage of the learners (White and Mattys, 2007). This is usually interpreted as the inﬂuence of the native language of the learner on his speech production in the L2. Low et al. (2000) showed that nPVI-V in L2 Singaporean English is inﬂuenced by the L1 Chi- nese language. Rhythm in L2 English was shown to be aﬀected by L1 Chinese, French, Spanish, Romanian and Italian (White and Mattys, 2007; Gut, 2009; Mok, 2013, etc.). These important ﬁndings regarding sensitivity of listeners to rhythmic diﬀerences have been done using discrimination tests. We know that people can discriminate between utterances even with small diﬀerences in durational variability. However, certain functions attributed to speech rhythm are not based on discrim- ination, but rather on classiﬁcation (segmentation, evaluation of accentedness, detection of linguistic origin of the speaker, etc.). Classiﬁcation is diﬀerent from discrimination. The listener may be able to perceive some acoustic diﬀerences when attending to them, but nevertheless ignore these diﬀerences when attributing an acoustic signal to a certain group, or when making a decision whether an acoustic signal is a representative of a certain class. In this particular study we focused not merely on whether the diﬀerences in L2 rhythm between utterances delivered by learn- ers at diﬀerent proﬁciency levels are detected. We were rather interested in whether listeners are able to reliably classify the utterances of L2 learners into distinct classes based on timing diﬀerences between utterances, and if so, which timing patterns listeners use to form the classes. The studies with the emphasis on development of rhythmic patterns in the course of L2 acquisition are even rarer. One of the few exceptions is the study by Ordin and Polyanskaya (2014) who compared how speech rhythm develops in L1 and in L2 acquisition. They found that speech rhythm develops from more syllable-timed toward more stress-timed patterns both in child L1 and in adult L2 speech. Research Question Previous studies have showed that rhythmic patterns change as language acquisition progresses even when the native and the target languages of the learners are rhythmically similar (Ordin et al., 2011). In our study, we were interested whether these developmental changes in speech rhythm are perceptually rel- evant. It is already known that the listeners are sensitive to the rhythmic diﬀerences between rhythmically contrastive lan- guages (Ramus and Mehler, 1999) as well as between German and English, i.e., between rhythmically similar languages (Vicenik and Sundara, 2013). Listeners are also able to distinguish rhyth- mic patterns of the utterances from the same language (White et al., 2012; Arvaniti and Rodriquez, 2013). Therefore, we think that the ﬁne distinctions between rhythmic patterns typical of L2 English of adult learners at diﬀerent proﬁciency levels might be detected. Importance of Rhythmic Patterns for Speech Processing This led to the hypoth- esis that alongside with constructing the ﬁrst representation of their native language, babies use rhythmic patterns to bootstrap on the syntactic properties of the language and on lexicon (Christophe and Dupoux, 1996; Mazuka, 1996; Mehler et al., 1996, 2004; Nespor et al., 1996). Rhythmic patterns are also used to develop strategies for segmentation of continuous speech and consequent word extraction and learning (Christophe et al., 2003; Thiessen and Saﬀran, 2007). In light of these considera- tions, we could suggest that the ability to recognize the durational cues pertaining to the speech rhythm is of the utmost importance for language acquisition and speech processing (e.g., for devel- opment and implementation of language-speciﬁc segmentation strategies). Therefore, sensitivity to timing diﬀerences, which is already observed in infancy, also persists in adulthood (Ramus and Mehler, 1999; White et al., 2012). Adults also use rhythmic cues to recognize the foreign accent in L2 speech and to detect the linguistic origin of the speaker (Kolly and Dellwo, 2014), to evaluate the degree of accentedness in L2 speech (Polyanskaya et al., 2013), to extract discrete linguistic units from continuous speech (Christophe et al., 2003). German share phonological parameters that are known to aﬀect the rhythm metrics. Both of these languages are classiﬁed as stress-timed in terms phonetic timing patterns captured by met- ric scores (Grabe and Low, 2002) and exhibit the phonological characteristics typical of stress-timed languages (Dauer, 1987; Schiering, 2007). Therefore German learners of English do not have to acquire phonological characteristics like production of complex syllables and complex consonantal clusters, opposition of long and short vowels, etc. Table 1 provides the metric scores in monolingual adult speech delivered by adult native speak- ers of either German or English, as reported in various studies. No unambiguous tendency is evident as for in which of these languages the durational variability is higher. %V seems a bit lower in German, which can be explained by a slightly higher syllabic complexity and a higher number of C clusters in Ger- man than in English (Delattre, 1965 cited in Gut, 2009). Com- parison of the metric scores for German and English with those reported for traditional syllable-timed languages (Ramus et al., 1999; Grabe and Low, 2002; White and Mattys, 2007) shows that both German and English exhibit higher duration variability and lower %V. Importance of Rhythmic Patterns for Speech Processing People are sensitive to the timing patterns which are cap- tured by rhythm metrics. Mehler et al. (1996) hypothesized that pre-linguistic infants perceive incoming continuous speech as a succession of vocalic and consonantal segments, vocalic segments are processed as informative harmonic signals of variable duration and intensity, which are alter- nating with unanalyzed noise (consonantal intervals). This n PVI = 100 × "m −1 X k = 1 dk −dk + 1 (dk + dk + 1)/2 /(m −1) # (1) March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 2 Perception of speech rhythm in L2 Ordin and Polyanskaya Time-Intensity-Grid-Representation of incoming continuous speech is based on innate perception mechanisms that help them to construct the ﬁrst representation of their language. Ramus et al. (1999) observed that languages with similar rhythmic properties tend to share more typological characteristics of grammatical and phonological structure. This led to the hypoth- esis that alongside with constructing the ﬁrst representation of their native language, babies use rhythmic patterns to bootstrap on the syntactic properties of the language and on lexicon (Christophe and Dupoux, 1996; Mazuka, 1996; Mehler et al., 1996, 2004; Nespor et al., 1996). Rhythmic patterns are also used to develop strategies for segmentation of continuous speech and consequent word extraction and learning (Christophe et al., 2003; Thiessen and Saﬀran, 2007). In light of these considera- tions, we could suggest that the ability to recognize the durational cues pertaining to the speech rhythm is of the utmost importance for language acquisition and speech processing (e.g., for devel- opment and implementation of language-speciﬁc segmentation strategies). Therefore, sensitivity to timing diﬀerences, which is already observed in infancy, also persists in adulthood (Ramus and Mehler, 1999; White et al., 2012). Adults also use rhythmic cues to recognize the foreign accent in L2 speech and to detect the linguistic origin of the speaker (Kolly and Dellwo, 2014), to evaluate the degree of accentedness in L2 speech (Polyanskaya et al., 2013), to extract discrete linguistic units from continuous speech (Christophe et al., 2003). Time-Intensity-Grid-Representation of incoming continuous speech is based on innate perception mechanisms that help them to construct the ﬁrst representation of their language. Ramus et al. (1999) observed that languages with similar rhythmic properties tend to share more typological characteristics of grammatical and phonological structure. Frontiers in Psychology \| www.frontiersin.org Participants Piske et al. (2001) analyzed a range of factors that inﬂuence pro- nunciation of L2 learners. These factors, among others, included the age and the length of exposure to the L2, amount of L2 use, language learning aptitude and motivation, learning mode. In our study we controlled these factors by collecting the rel- evant information in a detailed language-background question- naire (see Appendix 1 in online Supplementary Materials). Based on the questionnaire, we selected only those speakers who formed a homogeneous group and varied only in the degree of L2 mas- tery. The relevant information gleaned from the questionnaire was further veriﬁed in an informal interview during the recording sessions. Speech Material Germany at the time of the recordings. However, they reported to have little to no command of German, lived in close English- speaking community at the UK military bases in Nord-Rhein Westphalia, worked in only English-speaking environment, had English as their home and neighborhood language, came from monolingual English-speaking families and were raised in mono- lingual environment. Rhythm Changes in Second Language Acquisition The authors showed that both vocalic and consonantal variability in duration in L2 English increases as a function of the length of residence in the UK in adult speech when the target (English) and the native (Italian or Punjabi) languages of the learners are rhythmically contrastive. Ordin et al. (2011) showed that durational variability in speech of L2 learners also increases with proﬁciency growth when the target (English) and the native (German) languages of the learners exhibit similar rhythmic properties. English and March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 3 Perception of speech rhythm in L2 Ordin and Polyanskaya TABLE 1 \| Metric scores for German and English as reported in various studies. Language Rhythm metrics References %V VarcoV n-PVI-V rPVI-C 1C VarcoC German 41.7 52.5 68.7 65.0 Russo and Barry, 2008 42.8 71.7 Dellwo and Wagner, 2003 46.4 59.7 55.3 52.6 Grabe and Low, 2002 39.8 51.5 53.6 67.0 62.0 54.0 Arvaniti (2012)—overall score 36 44 55 73 62 51 Arvaniti (2012)—scores obtained on read sentences that were deliberately designed to enhance durational variability 41 52 56 60 54 50 Arvaniti (2012)—scores obtained on read sentences that were deliberately designed to inhibit durational variability 41 52 53 56 55 50 Arvaniti (2012)—scores obtained on sentences uncontrolled for phonotactics 42 55 52 72 55 50 Arvaniti (2012)—scores obtained on spontaneous speech English 38.0 64.0 73.0 70.0 59.0 White and Mattys, 2007 42.0 55.7 Dellwo and Wagner, 2003 41.1 57.2 64.1 56.7 Grabe and Low, 2002 40.1 53.5 Ramus et al., 1999 45.7 54.8 59.9 68.9 60.0 Arvaniti (2012)—overall score 41 48 55 83 68 57 Arvaniti (2012)—scores obtained on read sentences that were deliberately designed to enhance durational variability 50 46 51 57 49 53 Arvaniti (2012)—scores obtained on read sentences that were deliberately designed to inhibit durational variability 44 50 56 61 55 55 Arvaniti (2012)—scores obtained on sentences uncontrolled for phonotactics 48 66 66 77 68 59 Arvaniti (2012)—scores obtained on spontaneous speech TABLE 1 \| Metric scores for German and English as reported in various studies. Frontiers in Psychology \| www.frontiersin.org Elicitation Procedure The selected learners of English ﬁrst underwent a pronuncia- tion test so that we could assess the learners’ mastery of pro- nunciation. The test was devised by the authors and consisted of two parts: Perception and production. The perception part was compiled from Vaughan-Rees (2002) and included phoneme recognition, emotion recognition, intention recognition tasks. The production part included sentence reading. The sentences for production were composed to evaluate segmental realizations and prosodic control of the participants in the second language. The test ran for approximately 20 min. The test and the details on the controlled pronunciation features and assessment criteria can be found in Appendix 2 in online Supplementary Materials. We have recorded 51 German learners of L2 English (17–35 years old, M = 21; 27 females). We selected for participation only those people who grew up in or near the city of Bielefeld in North-Rhein Westphalia. The variety of German spoken in that region closely resembles what is understood as a Northern standard variety of German (Hochdeutsch). The selected partic- ipants did not exhibit features of regional varieties of German. All the participants were monolingual native speakers of German without speech or hearing disorders. Further on, a 5-min phonetic aptitude test (PAT) was admin- istered. The authors devised this test based on the oral mimicry tests described by Pike (1959), Suter (1976), and Thompson (1991). The test is aimed to predict the general phonetic ability by asking the participant to imitate novel sounds that do not exist in their native or target language and to mimic novel prosodic phe- nomena (e.g., lexical tones, tonal contours with accents which are We have also recorded 10 native speakers of English (south- ern British variety, 25–40 years, M = 30, 6 females) to compare the metric scores of the L2 learners of English with those of the L1 English speakers. The English speakers were residents in March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 4 Perception of speech rhythm in L2 Ordin and Polyanskaya not aligned according to the convention of the learner’s target or native language, etc.). The test and the details on the assessment criteria can be found in Appendix 3 in online Supplementary Materials. The sounds to imitate were presented by the holder of the IPA certiﬁcate conﬁrming his proﬁciency in producing and perception of sounds existing in world languages. 1According to the evaluators’ opinion, we did not have true beginners, and our participants better correspond to the lower-intermediate (B1.1 according to the Common European Framework for Languages), upper-intermediate (B2.1) and advanced (C1.1 and C1.2) levels. CEFL speciﬁes skills the learner should achieve at each of the six levels: A1, A2, B1, B2, C1, C2. However, as each level is usu- ally covered in language schools during two intensive courses, teachers split each level into two sublevels, e.g., C1.1 and C1.2. However, oﬃcial CEFL guidelines do not split six levels into sublevels, and division into C1.1 and C1.2 is done—rather arbitrarily—by the teachers. We did not want to resort to commonly used place- ment tests to evaluate the learners’ proﬁciency level because standard placement tests are designed to make an initial assessment to place the student into the course that ﬁts his level. Placement tests are not designed to estimate the proﬁciency level for certiﬁcation of the achieved proﬁciency level. Therefore, using several human evaluators (to avoid human bias) is the best methodological option, in our opinion. Elicitation Procedure The perfor- mance of participants in PAT did not correlate with their L2 pro- ﬁciency (we had both high and low proﬁciency learners with both high and low phonetic aptitude). Neither did the performance of the L2 learners in the PAT correlate with their performance in the English pronunciation test with any of the metrics calculated on their speech. This shows that the ability to imitate rhythmic pat- terns of the target language is not related to the general phonetic aptitude and we can eliminate a potential alternative explanation that the diﬀerences in rhythmic patterns between learners at dif- ferent proﬁciency levels are pertaining to the phonetic aptitude rather than to the overall proﬁciency. ratings between the teachers, we used Cronbach alpha, which is 0.90 for vocabulary, 0.89 for ﬂuency and 0.92 for grammati- cal accuracy. This shows high agreement between the raters and conﬁrms the reliability of their assessments. We averaged three ratings across the parameters for each rater and each interview, and thus got three mean ratings per learner. The teachers’ assessments and the results of the pronunciation tests were used to place the learner into one of the following proﬁ- ciency groups: beginners (12 speakers with ratings between 4 and 6), intermediate (9 speakers with mean ratings between 6 and 8), and advanced learners (22 speakers with ratings above 8)1. We used the results of the pronunciation test to assess the pronun- ciation skills of the learners. Eight speakers were not attributed to any group, either because the teachers did not agree with each other in their assessments (2 speakers were excluded for this reason) or because of the discrepancy between the results of the pronunciation tests and the teachers’ assessment of accuracy, ﬂuency and vocabulary resources. Pronunciation skills do not always agree with the general assessment of the learner’s reading, writing, listening and speaking skills, vocabulary size, grammar accuracy, etc. That is why we deemed it necessary to combine the tutors’ assessment of ﬂuency, accuracy and vocabulary on the one hand and the mastery of pronunciation on the other hand. In case when pronunciation lags far behind the general L2 mastery or exceeds the expected level, the learner was not attributed to any of the proﬁciency groups. At the next stage, an informal interview was conducted by the ﬁrst author. Segmentation Thirty three elicited sentences per speaker were annotated in Praat (Boersma and Weenink, 2010). Annotation was performed by the second author. Each sentence was divided into V and C intervals. The segmentation was carried out manually by the sec- ond author based on the criteria outlined in Peterson and Lehiste (1960) and Stevens (2002) for V and C intervals. The burst of energy corresponding to the release of the closure was taken as the starting point of a consonantal interval with the initial voiceless plosive sound after a pause and at the beginning of a sentence. Either the stop release, or apparent beginning of a voice bar, or other cues indicating apparent vibration of the vocal folds (whatever came ﬁrst) were considered as the beginning of a consonantal interval with the initial voiced plosive. The mark- ers of the turbulent noise were taken as the beginning of fricative consonants. The beginning of the ﬁrst formant was taken as the beginning of a sonorant consonant. Consonantal intervals in the The tests and recordings were made individually with every participant in a sound-treated booth of the audio-visual studio at the Bielefeld University in Germany. The recordings were made in WAV PCM at 44 kHz, 16 bit, mono. Elicitation Procedure General questions about preferences in read- ing and music, lifestyle, career choice, biography, and childhood were asked (Appendix 4 in online Supplementary Materials). The interviews were recorded and lasted approximately 12 min long with each participant. Following the interview, we ran a sentence elicitation task, similar to one used by Bunta and Ingram (2007). Thirty three sentences were elicited from each speaker. We used 33 pic- ture prompts for the elicitation procedure. The participants viewed picture slides in PowerPoint presentation. Each slide was accompanied with a descriptive sentence. The participants were instructed to remember the sentences. The participants could move to the next image or to go back to the previous slide at their own pace. When they had viewed all the slides, they were asked to look at the images again, without the accompanying text, and to recall and say the sentences that they had been asked to remember. In a very rare case (<5%) when the speaker could not remember the sentence or retrieved a modiﬁed sentence from memory, verbal prompts were used to help the speaker to pro- duce the correct sentence. For example, the participant said “The dog is running after the cat,” and the expected sentence was “The dog is chasing the cat.” The researcher responded to the partici- pant: “Yes, it is. You could also say chasing, which means running after. Can you say what you see at this picture once again?” One verbal prompt was suﬃcient to elicit the expected sentence when there was a mismatch in the ﬁrst trial. The recording ran con- tinuously throughout the sentence elicitation procedure. The list of elicited sentences and the examples of picture prompts can be found in Appendix 5 in online Supplementary Materials. Assessment of Learners’ Proﬁciency Three experienced teachers of English as a foreign language lis- tened to the recorded interviews and evaluated learners’ ﬂuency, grammatical accuracy, and vocabulary resources. They used a 10- point scale for each parameter. To estimate the consistency of March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 5 Perception of speech rhythm in L2 Ordin and Polyanskaya middle of a sentence were considered to start after the vowel ﬁn- ishes, and to stretch until the onset of the following vowel. The end of the consonantal interval in the ﬁnal position was marked at the end of the acoustic energy. The consonantal intervals in the ﬁnal positions were considered to start immediately after the vowel and ﬁnish at the end of the fricative noise (for obstruents) or at the end of the ﬁrst formant (for sonorants). Conventional procedure based on the analysis of the waveform and the spec- tral characteristics of the speech signal was based to identify the boundaries of the vocalic intervals. The end of the vowel was identiﬁed by the abrupt change in the vowel formant structure or by termination of the formants, and by the signiﬁcant drop in the waveform amplitude. The onset of the vowel was marked at the beginning of the voicing identiﬁed as the start of the regular vertical stripes on the spectrogram in the region of the second and higher formats. The marker indicating the vowel onset or oﬀset was placed at the point closest to the zero crossing on the waveform. sentence to account for possible developmental changes in speech tempo in the course of L2 acquisition, and for the interaction of speech rhythm and speech tempo. Although some rhythm metrics were claimed to be better than others at quantifying rhythm, there is no consensus on which metrics have more discriminative power. White and Mat- tys (2007), for example, advocated for pairwise metrics, while Ramus et al. (1999) favored 1C and %V. Loukina et al. (2011) performed the analysis of 15 rhythm metrics and in experiments separating pairs of languages by rhythmic properties showed that a rhythm measure that is successful at separating one pair often performs poorly at separating another pair. Considering the lack of consensus on the optimal set of metrics, we decided not to limit our investigation to the metrics which were found more useful in certain studies. Assessment of Learners’ Proﬁciency Metric Description %V Percentage of vocalic intervals 1V Standard deviation of vocalic intervals duration 1C Standard deviation of consonantal intervals duration nPVI-V Averaged of the mean differences between successive vocalic intervals nPVI-C Averaged of the mean differences between successive consonantal intervals rPVI-V Averaged difference in duration of successive vocalic intervals rPVI-C Averaged difference in duration of successive consonantal intervals VarcoV Coefﬁcient of variation of vocalic intervals, i.e., standard deviation divided by the mean VarcoC Coefﬁcient of variation of consonantal intervals, i.e., standard deviation divided by the mean MeanV Mean duration of vocalic intervals MeanC Mean duration of consonantal intervals TABLE 2 \| Metrics used in this study. Pauses and hesitations were not included into V or C intervals and were discarded. If the same type of the interval was annotated prior and following the pause, we treated them as two separate intervals because they are likely to be perceived as such. Final syllables were included into analysis. March 2015 \| Volume 6 \| Article 316 Assessment of Learners’ Proﬁciency Instead we tested all the metrics in order to see which ones better capture the diﬀerences in rhythm between sentences produced by L2 learners at diﬀerent proﬁciency levels. A series of by-sentence ANOVA tests (Table 3) with the values of the metrics as the dependent variables and proﬁciency level as the factor shows that non-normalized rhythm metrics (1V, 1C, rPVI-v, rPVI-c) and %V do not diﬀer between the proﬁciency levels. As the raw metrics do not diﬀer between the proﬁciency levels, we are not including them into further statistical tests. In diﬃcult cases where it was necessary to place the boundary between the consonantal interval represented by a sonorant con- sonant with a clear formant structure and a vowel, the decision was based on the amplitude of the ﬁrst format. Such diﬃcult cases were associated with the boundaries or categorizing allophones of /l/ (e.g., in the words girl, ball, table). We based our segmen- tation on purely phonetic criteria, therefore /l/ was sometimes marked as a vowel (in case of a vocalized [l]), and sometimes as a consonant. The decision was based on (1) auditory analy- sis by an experienced phonetician, and (2) amplitude of the ﬁrst formant. If the amplitude did not drop after the preceding vowel and the segment was perceived by a phonetician as a vocalized [l], then the segment was segmented as a vocalic interval. We did not want to pre-deﬁne certain types of segments either as consonan- tal or vocalic. We adopted a phontic approach to speech rhythm. Within the adopted framework, speech rhythm is represented by the surface timing patterns, which are purely phonetic, and pho- netic properties are not discrete and cannot be pre-assigned to a certain phonological category a-priory. TABLE 2 \| Metrics used in this study. Calculating the Rhythm Metrics The sentences elicited using the picture prompts were used to calculate the rhythm metrics. The sentence elicitation procedure helped us to avoid the reading mode and made speech material more similar to natural spontaneous speech. Besides, we obtained lexically identical sentences from every participant, which is nec- essary to analyze the development of speech rhythm per se, not aﬀected by the diﬀerences between the sentences in phonotactics, number of syllables in polysyllabic words, syntactic structures and phrasing. TABLE 3 \| Non-signiﬁcant ANOVA tests for the rhythm metrics between proﬁciency groups. Metric Signiﬁcance of Signiﬁcance of welch test levene’s test (if Levene’s test is signiﬁcant) or F statistics of the analysis of variance rPVI-v 0.513 0.4 rPVI-c 0.007 0.267 1V 0.748 0.692 1C < 0.0005 0.154 %V 0.691 0.068 March 2015 \| Volume 6 \| Article 316 TABLE 3 \| Non-signiﬁcant ANOVA tests for the rhythm metrics between proﬁciency groups. g Traditional rhythm metrics were calculated on each sentence. The overview of the selected metrics was given in the Table 2. We also calculated the mean duration of V and C intervals for each Frontiers in Psychology \| www.frontiersin.org Perception of speech rhythm in L2 Ordin and Polyanskaya ANOVAs on the rate-normalized rhythm measures revealed signiﬁcant diﬀerence between proﬁciency levels at p < 0.0005 for each metric. These metrics were included into multivari- ate model. The MANOVA test with nPVI metrics, Varco met- rics and mean durations of V and C intervals as the dependent variables and proﬁciency level as the factor revealed a signiﬁcant eﬀect of proﬁciency level on the rhythm measures, 3 = 0.856, F(12, 2822) = 19.06, p < 0.0005, µ2 = 0.075. Figures 1–3 show that the metric scores increase as L2 acquisition progresses, which indicates that German learners of English deliver L2 speech at a higher rate and with higher degree of stress-timing as their L2 mastery grows. The diﬀerences between the proﬁciency lev- els pairwise for each metric are mostly signiﬁcant (signiﬁcance values are given in Table 4). the pairwise durational variability and speech rate discriminate between the proﬁciency levels much better than utterance-wise variability. We also wanted to see how close the advanced German learn- ers of English are to their target in regard to acquisition of rhyth- mic patterns. Calculating the Rhythm Metrics For this, we compared the metric scores calculated on the sentences produced by the advanced learners of English with those calculated on the sentences spoken by native English speakers. T-tests (Table 7) reveal that the metric scores do not FIGURE 2 \| nPVI-v and nPVI-r in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. The MANOVA was followed up with the discriminant anal- ysis. We used only those metrics that were found to diﬀer sig- niﬁcantly between proﬁciency levels in our previous tests. The analysis revealed two discriminant functions. The ﬁrst function explained 96.9% of variance, canonical R2 = 0.14, and the sec- ond explained only 3.1% of variance, R2 = 0.005. In combination these functions signiﬁcantly diﬀerentiated the proﬁciency levels, 3 = 0.856, χ2 (12) = 220.318, p < 0.005. The second function alone did not signiﬁcantly diﬀerentiate between the proﬁciency levels, 3 = 0.995, χ2 (5) = 7.232, p = 0.204. This can also be seen on the discriminant function plot (Figure 4). Classiﬁcation results (Table 5) show that the model classiﬁes correctly 57% of cases (chance is 33%). The correlations between the outcomes and discriminant functions revealed that the measures of local—pairwise— variability and of speech rate loaded on the ﬁrst function, and global measures of variability loaded more highly on the second function (see Table 6). As the ﬁrst function explains substantially more variance that the second function, we can conclude that FIGURE 2 \| nPVI-v and nPVI-r in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. FIGURE 1 \| VarcoV and VarcoC in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. FIGURE 3 \| meanV and meanC in the sentences produced by native FIGURE 3 \| meanV and meanC in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. FIGURE 3 \| meanV and meanC in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. Calculating the Rhythm Metrics FIGURE 1 \| VarcoV and VarcoC in the sentences produced by native English speakers and by German learners of English at beginning, intermediate and advanced proﬁciency levels. Error bar shows 95% conﬁdence interval. March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 7 Perception of speech rhythm in L2 Ordin and Polyanskaya TABLE 4 \| Signiﬁcance for comparisons of rhythm metrics between proﬁciency levels pairwise (with Hochberg’s correction). Comparison VarcoV VarcoC nPVI-V nPVI-C meanV meanC Beginner—Intermediate <0.0005 0.855 <0.0005 0.17 0.004 0.315 Intermediate—Advanced 0.328 <0.0005 0.096 <0.0005 0.004 0.011 TABLE 6 \| Structure matrix of the discriminant function coefﬁcients. Function 1 2 meanV −0.5∗∗ 0.07 nPVI-C −0.472∗∗ −0.404 nPVI-V −0.469∗∗ −0.444 meanC −0.347∗∗ −0.224 VarcoC 0.436 0.79∗∗ VarcoV 0.409 −0.552∗∗ The stars indicate larger correlation between each variable and one of the discriminant functions. TABLE 4 \| Signiﬁcance for comparisons of rhythm metrics between proﬁciency levels pairwise (with Hochberg’s correction). 4 \| Signiﬁcance for comparisons of rhythm metrics between proﬁciency levels pairwise (with Hochberg’s correction). TABLE 6 \| Structure matrix of the discriminant function coefﬁcients. TABLE 6 \| Structure matrix of the discriminant function coefﬁcients. Function 1 2 meanV −0.5∗∗ 0.07 nPVI-C −0.472∗∗ −0.404 nPVI-V −0.469∗∗ −0.444 meanC −0.347∗∗ −0.224 VarcoC 0.436 0.79∗∗ VarcoV 0.409 −0.552∗∗ The stars indicate larger correlation between each variable and one of the discriminant functions. FIGURE 4 \| Discriminant function plot. purposes of marking edges and heads of prosodic constituents (see Prieto et al., 2012). FIGURE 4 \| Discriminant function plot. The analysis shows that speech rate and the degree of stress-timing increase as a function of proﬁciency growth. This tendency, however, can only be captured by normalized rhythm metrics. Raw metrics do not diﬀer between the proﬁciency lev- els. The values of the raw metrics are inﬂuenced by the speech tempo, i.e., by the mean durations of speech intervals: The faster one talks, the shorter the V and S intervals become; the shorter speech intervals result in smaller durational diﬀerences in pairs of consecutive intervals and in smaller standard deviation in duration of speech intervals. As the mean durations of speech intervals signiﬁcantly diﬀer between the proﬁciency levels, we should also expect signiﬁcant diﬀerences in the values of the raw metrics. However, this was not conﬁrmed. Calculating the Rhythm Metrics We believe that the values of the raw metrics are inﬂuenced by two conﬂicting forces: The tendency to deliver speech at a faster rate and with higher durational variability at high proﬁciency levels. This con- ﬂict prevents the emergence of signiﬁcant diﬀerences in the val- ues of raw metrics between proﬁciency levels. Normalization— removing the inﬂuence of speech tempo—allows us to notice the trend to enhance durational variability in L2 speech with proﬁ- ciency. The lack of signiﬁcant diﬀerences in %V between the pro- ﬁciency levels presents an interesting case. In earlier studies %V has been reported to be robust to ﬂuctuations in speech tempo (Wiget et al., 2010) and to discriminate between L1 and L2 speech (White and Mattys, 2007). However, in our experiment %V was not informative. %V is the proportion of the vocalic material in a sentence, and that is determined by phonotatic diﬀerences. The proportion of vocalic material will be lower in the languages that allow complex consonantal clusters and reduction of vowels in unstressed positions (e.g., German, Russian, English). These lan- guages are traditionally classiﬁed as stress-timed (Dauer, 1983). The languages on the opposite end of the spectrum impose strong phonotactic constraints, prefer simple CV syllables and feature TABLE 5 \| Classiﬁcation Results based on the Discriminant Analysis. Predicted Group Membership (in %) Beginners Intermediate Advanced Original Group Beginners 38.4 0 55.1 Membership (in%) Intermediate 26.6 0.3 73.1 Advanced 9.8 0 90.2 TABLE 5 \| Classiﬁcation Results based on the Discriminant Analysis. diﬀer between sentences spoken by advanced German learners and native speakers of English, with the exception of meanC (overall shorter C intervals in the utterances of L2 speakers) and rPVI-C (raw pairwise variability of consonantal intervals is higher in speech of learners of English). The diﬀerence in 1C is on the verge of signiﬁcance (p = 0.069), and the scores are again higher in sentences produced by L2 learners. Signiﬁcant and marginally signiﬁcant diﬀerence in consonantal variability is easily accounted for the diﬀerences in articulation rate of C intervals: longer C intervals in L2 speech result in larger standard deviations and pairwise durational diﬀerences. What is impor- tant is that pairwise durational variability of consonantal inter- vals per se, i.e., when the diﬀerences in speech rate are normalized, is also signiﬁcantly higher in speech of advanced L2 learners. Frontiers in Psychology \| www.frontiersin.org Calculating the Rhythm Metrics In our study, we used the same set of sentences elicited from diﬀerent speakers, thus the lexical diﬀerences that could potentially inﬂuence %V were elim- inated. The target and the native languages of the L2 learners were similar in terms of phonotactic and phonological proper- ties, and the learners did not have problems with producing the clusters of consonants in English sentences. %V captures phono- tactic and phonological diﬀerences, but the sentences spoken by learners at diﬀerent proﬁciency levels in our study manifested only phonetic diﬀerences in timing patterns, phonotactics and phonological characteristics were the same. Therefore, it is not surprising that %V was not found to diﬀer between sentences produced by L2 learners at diﬀerent proﬁciency levels. To conclude, the analysis conﬁrms signiﬁcant diﬀerences in rhythmic patterns between proﬁciency levels in L2. Rhythm mea- sures are more consistently stress-timed at higher proﬁciency levels. Raw metrics are inﬂuenced by conﬂicting tendencies to deliver speech at a faster rate and with higher durational vari- ability at higher proﬁciency levels, and thus do not increase with proﬁciency. The developmental tendency to increase the degree of stress-timing in L2 speech has been observed even when both the native and the target languages of the learner are rhythmi- cally similar. The main research question of our study was to investigate the perceptual relevance of the rhythmic diﬀerences between proﬁciency levels. Based on the literature review, we assumed that listeners are suﬃciently sensitive to the durational variability of C and V intervals to discriminate timing patterns of L2 utterances delivered by learners at diﬀerent proﬁciency levels. We wanted to ﬁnd out whether the detected diﬀerences in timing patterns between proﬁciency levels are used to classify utterances into discrete categories. To address this question, we set up the perception experiment. The discriminant analysis also reveals that the advanced learn- ers are more consistent in realization of timing patterns com- pared to lower-proﬁcient learners. Inspection of the discriminant function plot (Figure 7) reveals that the variate scores for the advanced learners are more compact, while the variate scores for the beginners are spread more evenly along the ﬁrst discrimi- nant function. The discriminant function plot also showed that the variate scores for diﬀerent groups of acquirers overlap (see overlapping circles on Figure 7). This means that beginners pro- duced sentences sometimes with high degree of durational vari- ability, and sometimes with lower degree of durational variability. Calculating the Rhythm Metrics Advanced learners constantly produced the sentences with high degree of durational variability. In other words, the productions of beginners varied greatly between stress-timed and syllable- timed rhythm patterns, but productions of advanced learners were more consistently stress-timed. Methods Participants We have recruited 25 native English speakers to act as listeners in the perception study (age range—21–24 years, M = 22; 13 females). Care was taken to form a socially homogeneous group of listeners with the same language background. All participants were students of Ulster University, monolingual English speak- ers (see our criteria for monolinguality in the description of the participants for Experiment 1). All listeners grew up in or around Belfast and were speaking the same regional variety of English (veriﬁed by a native speaker of English, phonetician and Belfast resident). We ensured that the participants did not diﬀer in age, educational level, social status, language background, experience with foreign languages, and all had equal exposure to educated standard British English. y We can draw the same conclusion if we look at Table 5. Rhythm and tempo measures correctly predict the speaker’s pro- ﬁciency level for 57% of sentences. The overall accuracy is sig- niﬁcantly above chance (33%), but the accuracy for the sentences produced by speakers on diﬀerent proﬁciency levels varies sub- stantially. Sentences produced by advanced speakers were clas- siﬁed correctly in 90.2% of cases, while sentences produced by beginners were classiﬁed correctly only in 38.4% of cases. This means that the half of the sentences spoken by beginners exhibit higher degree of variability that is typical of stress-timed rhythm in 90.2% sentences spoken by advanced learners. On the other hand, only 9.8% of sentences spoken by advanced learners exhibit lower durational variability overlapping with 38.4% of sentences from beginners. The analysis of the discriminant function plot and the classiﬁcation accuracy indicates that the timing patterns Calculating the Rhythm Metrics Advanced learners overshoot with increasing durational variabil- ity of C intervals, although they successfully acquire variability of V intervals. This can be explained by less assimilation of conso- nants in clusters within syllables in L2 speech (i.e., tendency to clearly produce all the consonants in the clusters) and by incom- plete mastery of ﬁne modiﬁcations in prosodic timing for the March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 8 Perception of speech rhythm in L2 Ordin and Polyanskaya TABLE 7 \| t-tests comparing metric scores in English speech of native English speakers and advanced L2 learner of English. meanV meanC %V 1V 1C VarcoV VarcoC rPVI-V rPVI-C nPVI-V nPVI-C t(1054) 1.216 3.371 −1.314 −0.064 1.818 −1.433 0.131 1.014 4.506 1.412 4.506 p = 0.224 = 0.001 = 0.189 = 0.949 = 0.069 = 0.152 = 0.896 = 0.311 < 0.0005 = 0.158 < 0.0005 less vowel shortening (e.g., French, Japanese). These factors increase the proportion of vocalic material in speech. Therefore, we assume that %V is a powerful predictor to discriminate between rhythmically contrastive languages. %V can also reﬂect the diﬀerences in lexical material, i.e., whether the utterances per se diﬀer in phonotactic characteristics. In our study, we used the same set of sentences elicited from diﬀerent speakers, thus the lexical diﬀerences that could potentially inﬂuence %V were elim- inated. The target and the native languages of the L2 learners were similar in terms of phonotactic and phonological proper- ties, and the learners did not have problems with producing the clusters of consonants in English sentences. %V captures phono- tactic and phonological diﬀerences, but the sentences spoken by learners at diﬀerent proﬁciency levels in our study manifested only phonetic diﬀerences in timing patterns, phonotactics and phonological characteristics were the same. Therefore, it is not surprising that %V was not found to diﬀer between sentences produced by L2 learners at diﬀerent proﬁciency levels. become more stable and consistent as a result of the acquisition progress. less vowel shortening (e.g., French, Japanese). These factors increase the proportion of vocalic material in speech. Therefore, we assume that %V is a powerful predictor to discriminate between rhythmically contrastive languages. %V can also reﬂect the diﬀerences in lexical material, i.e., whether the utterances per se diﬀer in phonotactic characteristics. Stimuli Supposedly, during the training session the participants formed new percep- tion categories for further discrimination between the stimuli from diﬀerent groups. Then the testing session began. Eighteen out of thirty tree elicited sentences per speaker were selected for stimuli preparation. Six sentences had three stressed syllables (e.g., the ‘dog is ‘ eating the ‘bone), six sentences included two stressed syllables (e.g., the ‘book is on the ‘table) and six sen- tences had only one stressed syllable (e.g., it’s ‘raining outside). The selected sentences produced by the selected speakers were listened to in order to make sure that the sentences were indeed pronounced with the expected number of stressed syllables. The selected sentences are marked with asterisk in Appendix 5 in online Supplementary Materials. We selected 378 sentences in total for the perception experiment (21 speakers ∗18 sentences). For the testing session, we prepared 270 stimuli (diﬀerent from those used in the training session, 5 speakers per proﬁciency group, 18 sentences per speaker). The procedure was the same as in the training session, but the listeners received no feedback, and all the stimuli were played only once. We used the speech resynthesis technique (Ramus and Mehler, 1999) to prepare the stimuli. We replaced all consonantal intervals in the selected sentences with “s” and all vocalic intervals with “a” and resynthesizing sentences with constant fundamen- tal frequency in MBROLA. The durations of “s” and “a” intervals were equal to the duration of C and V intervals in the origi- nal sentences. This technique degraded segmental and most of the prosodic information from the sentences. The only preserved diﬀerences between the identical sentences spoken by learners at diﬀerent proﬁciency levels were the diﬀerences in durational ratios of C and V intervals. Regardless of the recent criticism of this technique (Arvaniti and Rodriquez, 2013), its usefulness has been demonstrated in a number of studies (Ramus et al., 1999; Ramus and Mehler, 1999; Vicenik and Sundara, 2013; Kolly and Dellwo, 2014, etc.), and we found this delexicalization method to be optimal for the purposes of our study. The duration of the experiment varied between participants and usually exceeded 90 min. The participants could take a short break and have a rest pause during the training session and between the training and the testing session, but not during the testing session. Stimuli We selected sentences elicited from seven speakers per proﬁ- ciency group in the ﬁrst experiment to prepare the stimuli. The selected speakers from the advanced group had the highest mean ratings given by the evaluators (see description of the ﬁrst exper- iment, Section Procedure). The selected speakers from the begin- ners had the lowest mean ratings from the evaluators. We also randomly selected seven speakers from the group of intermediate learners. March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 9 Perception of speech rhythm in L2 Ordin and Polyanskaya Eighteen out of thirty tree elicited sentences per speaker were selected for stimuli preparation. Six sentences had three stressed syllables (e.g., the ‘dog is ‘ eating the ‘bone), six sentences included two stressed syllables (e.g., the ‘book is on the ‘table) and six sen- tences had only one stressed syllable (e.g., it’s ‘raining outside). The selected sentences produced by the selected speakers were listened to in order to make sure that the sentences were indeed pronounced with the expected number of stressed syllables. The selected sentences are marked with asterisk in Appendix 5 in online Supplementary Materials. We selected 378 sentences in total for the perception experiment (21 speakers ∗18 sentences). We used the speech resynthesis technique (Ramus and Mehler, 1999) to prepare the stimuli. We replaced all consonantal intervals in the selected sentences with “s” and all vocalic intervals with “a” and resynthesizing sentences with constant fundamen- tal frequency in MBROLA. The durations of “s” and “a” intervals were equal to the duration of C and V intervals in the origi- nal sentences. This technique degraded segmental and most of the prosodic information from the sentences. The only preserved diﬀerences between the identical sentences spoken by learners at diﬀerent proﬁciency levels were the diﬀerences in durational ratios of C and V intervals. Regardless of the recent criticism of this technique (Arvaniti and Rodriquez, 2013), its usefulness has been demonstrated in a number of studies (Ramus et al., 1999; Ramus and Mehler, 1999; Vicenik and Sundara, 2013; Kolly and Dellwo, 2014, etc.), and we found this delexicalization method to be optimal for the purposes of our study. was played. When all 108 stimuli were presented, the partici- pant had a 2-min break before the stimuli were played again. The training procedure was repeated three times. Results We calculated rhythm metrics on the stimuli that were classiﬁed by the majority of listeners as Burabah, Losto, and Mahutu. The metrics were calculated on V and C intervals. We performed the discriminant analysis to test whether rhythm metrics statistically discriminate between the stimuli classiﬁed into three groups. The analysis revealed two discriminant functions. The ﬁrst function explained 94.8% of variance, R2 = 0.52, and the second function explains 5.2% of variance, R2 = 0.05. These functions in combi- nation signiﬁcantly diﬀerentiate between the groups, λ = 0.457, χ2 (20) = 149.9, p < 0.0005. The second function alone is not sig- niﬁcant, λ = 0.945, χ2 (9) = 11.101, p = 0.282. The overall accu- racy of the model is 69% (chance is 33.3%), accuracy of Burabah is 91%, Losto—52%, Mahutu—58.5% (chance level is 33.3% for each category). See Table 8 for the details on the classiﬁcation accuracy. Stimuli During the experiment the participants were oﬀered hot and cold drinks and sweet snacks to help them cope with possible fatigue. The participants could have their drinks and snacks during the rest pauses as well as during the train- ing session. The order of stimuli presentation was randomized using the internal Praat algorithm in attempt to counterbalance for possible fatigue eﬀect. March 2015 \| Volume 6 \| Article 316 Procedure The experiment was carried out with each participant individ- ually in the phonetic laboratory of Ulster University. The stim- uli were presented to the listeners in two sessions: Training and testing. The listeners were not informed that the stimuli were derived from L2 English speech because we did not want the listeners use linguistic expectations regarding what the stim- uli in L2 English might sound like. This might have created a bias that would be diﬃcult to control. Instead, the listeners were told that the stimuli were derived from three rare exotic African languages. We coined these languages Burabah (sen- tences of the advanced L2 learners converted into “sasasa” stim- uli), Losto (stimuli based on durations in sentences of intermedi- ate learners of English), and Mahutu (resynthesized sentences of beginners). The structure matrix (Table 9) reveals that the ﬁrst function is loaded with the raw metrics and mean durations of V and C intervals, while the second function is loaded with the normalized metrics. This means that the normalized metrics cannot discrim- inate between the groups, but mean durations and raw metrics discriminate between the stimuli identiﬁed as Burabah, Losto, and Mahutu with probability signiﬁcantly above chance. g We chose 108 stimuli for the training session (18 stimuli per speaker, 2 speakers per proﬁciency group). Before the session, each listener was exposed to nine stimuli, randomly selected from those used later in the training session, 3 stimuli per proﬁciency group, i.e., per “exotic language.” The listener had 1 min to listen to these stimuli by clicking with a mouse on nine buttons on the computer screen. Each button had a caption with the “language” name. After 1-min familiarization, the stimuli were presented to listener one by one. The listener had to identify from which lan- guage (Mahutu, Losto, or Burabah) it originates. The listener was expected to click one of the three buttons on the computer screen with a mouse pointer. Each button had a caption with the “lan- guage” name. On response, the listener was provided with the feedback which “language” it really was, and the next stimulus TABLE 8 \| Classiﬁcation Results (prior probabilities: all groups equal). Predicted Group membership Mahutu(%) Losto(%) Burabah(%) Original Mahutu 58.5 26.4 15.1 Losto 28.4 52.2 19.4 Burabah 1.3 7.6 91.1 March 2015 \| Volume 6 \| Article 316 TABLE 8 \| Classiﬁcation Results (prior probabilities: all groups equal). Frontiers in Psychology \| www.frontiersin.org Procedure Frontiers in Psychology \| www.frontiersin.org 10 Perception of speech rhythm in L2 Ordin and Polyanskaya FIGURE 6 \| 1V and 1C for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 7 \| meanV and meanC for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. durational variability for the listeners performing the classiﬁ TABLE 9 \| Structure matrix of the discriminant function coefﬁcients. Metrics Function I II meanV 0.719∗ 0.404 meanC 0.607∗ −0.181 rPVI-v 0.458∗ 0.351 rPVI-c 0.371∗ −0.097 1C 0.352∗ −0.064 1V 0.421 0.559 VarcoV −0.002 0.341∗ nPVI-c 0.102 0.208∗ nPVI-v 0.002 0.203∗ VarcoC 0.035 0.127∗ The stars indicate larger correlation between each variable and one of the discriminant functions. TABLE 9 \| Structure matrix of the discriminant function coefﬁcients. FIGURE 6 \| 1V and 1C for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 6 \| 1V and 1C for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 5 \| rPVI-V and rPVI-C for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 7 \| meanV and meanC for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 7 \| meanV and meanC for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 5 \| rPVI-V and rPVI-C for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. FIGURE 7 \| meanV and meanC for the stimuli identiﬁed as Mahutu, Losto, or Burabah. Error bar shows 95% conﬁdence interval. Figures 5–7 show the diﬀerences in the rhythm metrics that signiﬁcantly diﬀer between stimuli classiﬁed into three groups. Only mean durations and non-normalized metrics (rPVI and the standard deviation) diﬀer signiﬁcantly between the stimuli iden- tifyed as Burabah, Mahutu, and Losto and statistically discrimi- nate between the groups. Rhythm metrics normalized for tempo and %V do not diﬀer between the stimuli classiﬁed into three dif- ferent groups, and do not discriminate between stimuli attributed to diﬀerent classes. durational variability for the listeners performing the classiﬁ- cation task. However, we still do not know the relative con- tribution of the rhythm compared to tempo in classiﬁcation. Procedure Step Metrics β T B p R2 R2 change Signiﬁcance of R2 change 1 meanV 0.525 10.1 61.108 <0.0005 0.276 0.276 <0.0005 2 meanV 0.377 7.163 43.814 <0.0005 0.386 0.110 <0.0005 meanC 0.363 6.914 46.28 <0.0005 TABLE 11 \| Coefﬁcients and parameters of the regression model with Frequency_Mahutu as the dependent variable. Step Metrics β T B p R2 R2 change Signiﬁcance of R2 change 1 meanV 0.525 10.1 61.108 <0.0005 0.276 0.276 <0.0005 2 meanV 0.377 7.163 43.814 <0.0005 0.386 0.110 <0.0005 meanC 0.363 6.914 46.28 <0.0005 Coefﬁcients and parameters of the regression model with Frequency_Mahutu as the dependent variable. tempo equals 5.62 syl/s. for the stimuli identiﬁed as Burabah, 4.41 syl/s. for the stimuli identiﬁed as Losto, and 4.4 syl/s. for the stimuli identiﬁed as Mahutu. ANOVA analysis showed that the diﬀerence in tempo between the groups is signiﬁcant, F(2, 196) = 64.077, p < 0.0005. Pairwise comparisons (with the Bonfer- roni correction) reveal that the diﬀerence lies between “Losto” and “Burabah” stimuli, while the diﬀerence between “Losto” and “Mahutu” groups is not signiﬁcant. Speech tempo in the stimuli identiﬁed as Burabah is 25.7% higher than in the stim- uli identiﬁed as Losto. This increase is above the threshold for just noticeable tempo diﬀerence (Quene, 2007; Thomas, 2007). Speech tempo in the stimuli classiﬁed as Mahutu is 6.6% slower than in the stimuli identiﬁed as Losto, and this diﬀerence is below the just noticeable threshold. added, and the model was signiﬁcantly improved. Adding raw rhythm metrics as predictors did not improve the model fur- ther. Table 10 summarized the main details of the regression model. The results show that the most important predictors are mean durations of V and C intervals, which are negatively correlated with the frequency of “Burabah” response. This means that the shorter the speech intervals (i.e., the faster the tempo), the more likely the listener will classify the stimulus as Burabah. We also performed stepwise multiple regressions with Fre- quency_Mahutu and with Frequency_Losto as dependent variable (details of the regression models are in Tables 11, 12 respec- tively). The analyses show that the most inﬂuential predictors for both Frequency_Mahutu and Frequency_Losto are meanV and meanC. The predictors are positively correlated with the fre- quency of “Losto” and “Mahutu” responses, which means that the stimuli with longer C and V intervals (i.e., slower speech rate) are more likely to be identiﬁed as Losto or Mahutu. Procedure Listeners’ sensitivity to speech tempo can be explained by a number of studies in physiology of hearing. Schreiner and Urbas (1986, 1988) showed that auditory neurons ﬁre in response to a sharp increase in intensity that usually coincides with the vowel onset. Consequently, the rate at which “s” and “a” alternate in the stimuli determines the rate at which the neurons ﬁre. Moreover, some studies suggest a direct relation between a syllable-length unit (“sa” unit in our stimuli) and the neural response in the auditory cortex (Viemeister, 1988; Greenberg, 1997; Wong and Schreiner, 2003; Greenberg and Ainsworth, 2004). Besides, the auditory system imposes certain limitations on the speech tempo. If the assumptions to the speech rate and to the length of the syllable-like units are violated, speech processing and decoding of speech at the cortical level is compromised (Ghitza and Green- berg, 2009; Ghitza, 2011). Therefore, there is a physiological basis for discriminating fast and slow stimuli, or stimuli with longer and shorter syllable-like units. Frontiers in Psychology \| www.frontiersin.org Procedure To address this issue, we calculated the frequency for Burabah, Losto, and Mahutu response for each stimulus, i.e., how many listeners out of 25 identiﬁed each stimulus as Burabah (Fre- quency_Burabah), Losto (Frequency_Losto), or Mahutu (Fre- quency_Burabah). After that we performed stepwise multiple regression to assess the ability of meanV, meanC, rPVI-C, rPVI- V, 1V, and 1C to predict Frequency_Burabah. Stepwise regres- sion was chosen because we wanted to evaluate whether both the mean durations and the variability measures were necessary to predict Frequency_Burabah. The constructed model included only two steps. At the ﬁrst step, meanV was entered into equation as the most powerful predictor. At the second step, meanC was Figures 5–7 show that the stimuli identiﬁed as Burabah exhibit shorter mean durations and smaller standard deviations in duration of speech intervals and smaller durational diﬀer- ences in pairs of consecutive intervals. As the raw metrics are inﬂuenced by the speech rate, we cannot say what the partici- pants were listening for to make their judgments—speech tempo, durational variability, or both. As meanV and meanC display the highest correlations in the structure matrix (Table 9), we could conclude that tempo is probably more important than Figures 5–7 show that the stimuli identiﬁed as Burabah exhibit shorter mean durations and smaller standard deviations in duration of speech intervals and smaller durational diﬀer- ences in pairs of consecutive intervals. As the raw metrics are inﬂuenced by the speech rate, we cannot say what the partici- pants were listening for to make their judgments—speech tempo, durational variability, or both. As meanV and meanC display the highest correlations in the structure matrix (Table 9), we could conclude that tempo is probably more important than March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 11 Perception of speech rhythm in L2 Ordin and Polyanskaya TABLE 10 \| Coefﬁcients and parameters of the regression model with Frequency_Burabah as the dependent variable. parameters of the regression model with Frequency_Burabah as the dependent variable. 1 meanV −0.638 −13.574 −102.93 <0.0005 0.407 0.407 <0.0005 2 meanV −0.489 −10.506 −78.832 <0.0005 0.519 0.111 <0.0005 meanC −0.365 −7.853 −64.49 <0.0005 TABLE 11 \| Coefﬁcients and parameters of the regression model with Frequency_Mahutu as the dependent variable. Conclusion diﬀerently by listeners with diﬀerent native languages for pur- poses of speech processing (Christophe et al., 2003; Murty et al., 2007; Thiessen and Saﬀran, 2007; Kim et al., 2008). However, their importance in processing non-linguistic stimuli when cog- nitive mechanisms are less intensely employed might be low. When the presented stimuli are not processed as speech-like, rhythmic diﬀerences between stimuli are not used for discrim- ination or classiﬁcation (Ramus et al., 2000). Thus, in our experiment listeners rely more on those patterns in acoustic signal that have direct physiological correlates rather than on the patterns that are processed though the relay of cognitive ﬁlters. We have shown signiﬁcant diﬀerences in speech timing between the sentences produced by the German learners of English at diﬀerent proﬁciency levels. As L2 acquisition progresses, L2 English is delivered at a faster rate and with a higher degree of stress-timing. Further analysis revealed that realization of tim- ing is more stable in L2 speech produced by advanced L2 learn- ers. Advanced learners tend to speak consistently with a higher degree of stress-timing. Lower proﬁciency speakers randomly vary the degree of durational variability in their speech, some- times delivering L2 speech with high durational variability, and sometimes with a more syllable-timed rhythm. We suggest that timing control in L2 speech production improves as acquisition progresses, and rhythm becomes more stable. We would like to emphasize that our results do not indicate the inability of the participants to hear the rhythmic diﬀerences. To test the ability to detect the diﬀerences in L2 speech rhythm between proﬁciency levels, discrimination test is to be carried out, and a number of studies showed that such small diﬀer- ences are detected. Using classiﬁcation task, we can determine which timing patterns are used to classify the utterances into groups. It is possible, that larger diﬀerences in durational vari- ability (e.g., between rhythmically contrastive languages that are traditionally deﬁned as stress-timed and syllable-timed) might become more linguistically relevant and used in classiﬁcation. Smaller diﬀerences as those revealed between L2 varieties are not suﬃciently diﬀerent to be processed as linguistically rele- vant, and timing patterns are classiﬁed based on direct phys- iological correlates. Further research is necessary to address which rhythmic diﬀerences could be processed as linguistically relevant. Although rhythmic changes in L2 acquisition can be easily proﬁled with normalized rhythm metrics, raw metrics do not exhibit a clear uni-directional development. Discussion The results show that listeners classify the stimuli based on speech tempo and ignore the diﬀerences in the durational variability between the “sasasa” sequences. The Figures 5–7 also show that there is no diﬀerence between the stimuli identiﬁed as Losto and Mahutu for 1V, 1C, rPVI-V, rPVI-C, meanC, and meanV measures. Faster stimuli with both low and high variation in duration of V and C intervals were classiﬁed as Burabah, and slower stimuli were almost randomly attributed to either Losto or Mahutu. We conclude that the listeners formed only two cate- gories: one for faster stimuli that were classiﬁed as Burabah, and the other for slower stimuli that were randomly identiﬁed either as Mahutu or Losto. We are not aware of any evidence of direct physiologi- cal correlates for the ability to diﬀerentiate ﬁne distinctions in durational variability. Thus, we assume that diﬀerentiation of ﬁne distinctions in rhythmic patterns involves cognitive pro- cessing. Peculiarities of predominant rhythmic patterns in a certain language correlate with grammatical, morphological and other structural characteristics. Rhythmic patterns guide the way the language is acquired. They inﬂuence the strategies of segmentation of continuous speech. Rhythmic cues are exploited This result agrees with psychoacoustic data in tempo percep- tion. Quene (2007) and Thomas (2007) studied just-noticeable diﬀerences in tempo and found that 5–8% change in tempo (expressed as beats per minute for non-speech stimuli and syllables-per-minute for speech stimuli) is easily detected by the subjects. We analyzed the tempo diﬀerences between the stimuli which were classiﬁed as Losto, Mahutu, and Burabah. Average March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 12 Perception of speech rhythm in L2 Ordin and Polyanskaya TABLE 12 \| Coefﬁcients and parameters of the regression model with Frequency_Losto as the dependent variable. Step Metrics β t B p R2 R2 change Signiﬁcance of R2 change 1 meanV 0.427 7.741 41.821 <0.0005 0.183 0.183 <0.0005 2 meanV 0.358 5.993 35.018 <0.0005 0.207 0.024 =0.005 meanC 0.170 2.847 18.207 =0.005 FIGURE 8 \| Splitting the “sasasa” stimuli into two categories based on durational variability of vocalic and consonantal intervals and speech rate. TABLE 12 \| Coefﬁcients and parameters of the regression model with Frequency_Losto as the dependent variable. References Gut, U. (2009). Non-native Speech. A Corpus-Based Analysis of the Phonetic and Phonological Properties of L2 English and L2 German. Frankfurt: Peter Lang. Arvaniti, A. (2012). The usefulness of metrics in the quantiﬁcation of speech rhythm. J. Phonet. 40, 351–373. doi: 10.1016/j.wocn.2012.02.003 Kim, J., Davis, C., and Cutler, A. (2008). Perceptual tests of rhythmic similarity: II. Syllable rhythm. Langu. Speech 51, 343–359. doi: 10.1177/0023830908099069 Kolly, M.-J., and Dellwo, V. (2014). Cues to linguistic origin: the contribution of speech temporal information in foreign accent recognition. J. Phonet. 42, 12–23. doi: 10.1016/j.wocn.2013.11.004 Arvaniti, A., and Rodriquez, T. (2013). The role of rhythm class, speaking rate and F0 in language discrimination. Lab. Phonol. 4, 7–38. doi: 10.1515/lp-2013-0002 Boersma, P., and Weenink, D. (2010). Praat: Doing Phonetics by Com- puter (Version 5.1.22). Retrieved: December 15, 2010. Available online at: http://www.praat.org/ Loukina, A., Kochanski, G., Rosner, B., Shih, C., and Keane, E. (2011). Rhythm measures and dimensions of durational variation in speech. J. Acoust. Soc. Am. 129, 3258–3270. doi: 10.1121/1.3559709 Bunta, F., and Ingram, D. (2007). The acquisition of speech rhythm by bilingual Spanish- and English-speaking four-and ﬁve-year-old children. J. Speech Lang. Hear. Res. 50, 999–1014. doi: 10.1044/1092-4388(2007/070) Low, L., Grabe, E., and Nolan, F. (2000). Quantitative characterizations of speech rhythm: syllable-timing in Singapore English. Langu. Speech 43, 377–401. doi: 10.1177/00238309000430040301 Christophe, A., and Dupoux, E. (1996). Bootstrapping lexical acquisition: the role of prosodic structure. Linguist. Rev. 13, 383–412. doi: 10.1515/tlir.1996.13.3- 4.383 Mazuka, R. (1996). “How can a grammatical parameter be set before the ﬁrst word?,” in Signal to Syntax: Bootstrapping from Speech to Grammar in Early Acquisition, eds J. L. Morgan and K. Demuth (Mahwah, NJ: Lawrence Erlbaum Associates Inc), 313–330. Christophe, A., Gout, A., Peperkamp, S., and Morgan, J. L. (2003). Discovering words in the continuous speech stream: the role of prosody. J. Phonet. 31, 585–598. doi: 10.1016/S0095-4470(03)00040-8 Mehler, J., Dupoux, E., Nazzi, T., and Dehaene-Lambertz, G. (1996). ˇDCoping with linguistic diversity: the infant’s viewpoint,” in From Signal to Syntax: Bootstrap- ping from Speech to Grammar in Early Acquisition, eds J. Morgan and K. D. Demuth (Hillsdale, NJ: Erlbaum), 101–116. Dauer, R. (1983). Stress-timing and syllable-timing reanalyzed. J. Phonet. 11, 51–62. Dauer, R. (1987). “Phonetic and phonological components of language rhythm,” in Proceedings of the 11th International Congress of Phonetic Sciences, (Tallinn, Estonia), 447–450. Mehler, J., Sebastian-Galles, N., and Nespor, M. (2004). Conclusion Faster speech rate at higher proﬁciency levels lowers the values of raw metrics, while the need to enhance durational variability pushes the met- rics up. These conﬂicting forces did not allow raw metrics to reveal a clear developmental change as a function of acquisition progress. Perception experiment was set up to investigate whether monolingual English use the diﬀerences in L2 speech timing between proﬁciency levels to group the utterances with diﬀer- ent timing patterns into the same class. Although L2 speech indeed becomes increasingly more stress-timed with proﬁciency, native speakers of English, when asked to classify diﬀerent timing March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 13 Perception of speech rhythm in L2 Ordin and Polyanskaya between the stimuli are not suﬃciently large to be linguistically relevant. patterns into separate groups, paid attention to the diﬀerences in speech rate and ignored the diﬀerences in speech rhythm between the utterances produced by the L2 learners at diﬀer- ent proﬁciency levels. Faster utterances were grouped separately from slower utterances. Both groups included utterances with high and low durational variability of speech intervals. This trend is schematically illustrated on Figure 8. The sensitivity of the listeners to speech tempo is physiologically determined. The fact that listeners ignore rhythmic diﬀerences in classiﬁcation can be explained by non-linguistic nature of the stimuli. Process- ing of “sasasa” stimuli in our experiment, assumingly, does not involve cognitive mechanisms that are employed in processing of linguistic material, and listeners pay attention to those features of the acoustic signal that have direct physiological correlates. Further research is necessary to understand whether the cogni- tive ﬁlter is not applied to processing these stimuli because they are not perceived as speech, or because the diﬀerences in rhythm References “Biological foundations of language: language acquisition, cues for parameter setting and the bilingual infant,” in The New Cognitive Neuroscience, ed M. Gazzaniga (Cambridge, MA: MIT Press), 825–836. Dellwo, V. (2006). “Rhythm and speech rate: a variation coeﬃcient for deltaC,” in Language and Language-Processing, eds P. Karnowski and I. Szigeti (Frankfurt am Main: Peter Lang), 231–241. Mok, P. (2013). Speech rhythm of monolingual and bilingual children at 2;06: Can- tonese and English. Bilingualism: Language and Cognition 16, 693–703. doi: 10.1017/S1366728910000453 Dellwo, V., and Wagner, P. (2003). “Relations between language rhythm and speech rate,” in Proceedings of the 15th International Congress of Phonetics Sciences (Barcelona), 471–474. Murty, L., Otake, T., and Cutler, A. (2007). Perceptual tests of rhythmic similarity: I. Mora Rhythm. Langu. Speech 50, 77–99. doi: 10.1177/00238309070500010401 Ghitza, O. (2011). Linking speech perception and neurophysiology: speech decod- ing guided by cascaded oscillators locked to the input rhythm. Front. Psychol. 2:130. doi: 10.3389/fpsyg.2011.00130 Nazzi, T., Bertoncini, J., and Mehler, J. (1998). Language discrimination by new- borns: toward an understanding of the role of rhythm. J. Exp. Psychol. Hum. Percept. Perform. 24, 756–766. doi: 10.1037/0096-1523.24.3.756 Ghitza, O., and Greenberg, S. (2009). On the possible role of brain rhythms in speech perception: intelligibility of time-compressed speech with periodic and aperiodic insertions of silence. Phonetica 66, 113–126 doi: 10.1159/000208934 Nazzi, T., and Ramus, F. (2003). Perception and acquisition of linguistic rhythm by infants. Speech Commun. 41, 233–243. doi: 10.1016/S0167-6393(02)00106-1 Grabe, E., and Low, L. (2002). “Durational variability in speech and the rhythm class hypothesis,” in Papers in Laboratory Phonology, Vol. 7, eds C. Gussenhoven and N. Warner (New York, NY: Mouton de Gruyter), 515–546. Nespor, M., Guasti, M., and Christophe, A. (1996). “Selecting word order: the rhythmic activation principle,” in Interfaces in Phonology, ed U. Kleinhenz (Berlin: Akademie Verlag), 1–26. Greenberg, S. (1997). “Auditory function,” in Encyclopedia of Acoustics, ed M. Crocker (New York, NY: Wiley), 1301–1323. Nolan, F., and Asu, E. (2009). The pairwise variability index and coexisting rhythms in language. Phonetica 66, 64–77. doi: 10.1159/000208931 Ordin, M., and Polyanskaya, L. (2014). Development of timing patterns in ﬁrst and second languages. System 42, 244–257. doi: 10.1016/j.system.2013.12.004 Greenberg, S., and Ainsworth, W. (2004). “Speech processing in the auditory sys- tem: An Overview,” in Speech Processing in the Auditory System, eds S. Green- berg, W. Ainsworth, A. Popper, and R. Fay (New York, NY: Springer-Verlag), 1–62. Ordin, M., Polyanskaya, L., and Ulbrich, C. (2011). Supplementary Material The Supplementary Material for this article can be found online at: http://www.frontiersin.org/journal/10.3389/fpsyg. 2015.00316/abstract Acknowledgments We acknowledge the ﬁnancial support of the German Research Foundation (DFG) and the Open Access Publication Fund of Bielefeld University for the article processing charge. This work was supported by the Alexander von Humboldt foundation. We are also thankful to Ferenc Bunta and David Ingram for sharing the picture prompts that we used for the sentence elicitation task. References “Acquisition of timing patterns in second language,” in Proceedings of Interspeech 2011 (Florence), 1129–1132. March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 14 Perception of speech rhythm in L2 Ordin and Polyanskaya Pamies Bertran, A. (1999). Prosodic typology: on the dichotomy between stress- timed and syllable-timed languages. Lang. Design 2, 103–130. Suter, R. W. (1976). Predictors of pronunciation accuracy in second lan- guage learning. Lang. Learn. 26, 233–253. doi: 10.1111/j.1467-1770.1976.tb 00275.x Peterson, G., and Lehiste, I. (1960). Duration of syllable nuclei in English. J. Acoust. Soc. Am. 32, 693–703. doi: 10.1121/1.1908183 Thiessen, E., and Saﬀran, J. (2007). Learning to learn: infants’ acquisition of stress- based strategies for word segmentation. Lang. Learn. Dev. 3, 73–100. doi: 10.1080/15475440709337001 Pike, E. V. (1959). A test for predicting phonetic ability. Lang. Learn. 9, 35–41. doi: 10.1111/j.1467-1770.1959.tb01127.x Piske, T., McKay, I., and Flege, J. (2001). Factors aﬀecting degree of foreign accent in an L2: a review. J. Phonet. 29, 191–215. doi: 10.1006/jpho.2001.0134 Thomas, K. (2007). Just noticeable diﬀerences and tempo change. J. Sci. Psychol. 14–20. Polyanskaya, L., Ordin, M., and Ulbrich, C. (2013). “Contribution of timing pat- terns into perceived foreign accent,” in Elektronische Sprachsignalverarbeitung 2013, ed P. Wagner (Dresden: TUDpress), 71–79. Thompson, I. (1991). Foreign accents revisited: the English pronunciation of Russian immigrants. Lang. Learn. 41, 177–204. doi: 10.1111/j.1467- 1770.1991.tb00683.x Prieto, P., del Mar Vanrell, M., Astruc, L., Payne, E., and Post, B. (2012). Phonotac- tic and phrasal properties of speech rhythm. Evidence from Catalan, English, and Spanish. Speech Commun. 54, 681–702. doi: 10.1016/j.specom.2011.12.001 Vaughan-Rees, M. (2002). Test Your Pronunciation: Book With Audio CD. London: Longman. Vicenik, C., and Sundara, M. (2013). The role of intonation in lan- guage and dialect discrimination by adults. J. Phonet. 41, 297–306. doi: 10.1016/j.wocn.2013.03.003 Quene, H. (2007). On the just noticeable diﬀerence for tempo in speech. J. Phonet. 35, 353–362. doi: 10.1016/j.wocn.2006.09.001 Viemeister, N. (1988). “Psychophysical aspects of auditory intensity coding,” in Auditory Function, eds G. Edelman, W. Gall, and W. Cowan (New York, NY: Wiley), 213–241. Ramus, F., Hauser, M., Miller, C., Morris, D., and Mehler, J. (2000). Language dis- crimination by human newborns and by cotton-top tamarin monkeys. Science 288, 349–351 doi: 10.1126/science.288.5464.349 Ramus, F., and Mehler, J. (1999). Language identiﬁcation with suprasegmental cues: a study based on speech resynthesis. J. Acoust. Soc. Am. 105, 512–521. doi: 10.1121/1.424522 White, L., and Mattys, S. (2007). References Calibrating rhythm: ﬁrst language and sec- ond language studies. J. Phonet. 35, 501–522. doi: 10.1016/j.wocn.2x007. 02.003 White, L., Mattys, S., and Wigit, L. (2012). Language categorization by adults is based on sensitivity to durational cues, not rhythmic class. J. Mem. Lang. 66, 665–679. doi: 10.1016/j.jml.2011.12.010 Ramus, F., Nespor, M., and Mehler, J. (1999). Correlates of linguistic rhythm in the speech signal. Cognition 73, 265–292. doi: 10.1016/S0010-0277(99)00058-X Roach, P. (1982). “On the distinction between ‘stress-timed’ and ‘syllable-timed’ languages,” in Linguistic Controversies, ed D. Crystal (London: Edward Arnold), 73–79. Wiget, L., White, L., Schuppler, B., Grenon, I., Rauch, O., and Mattys, S. (2010). How stable are acoustic metrics of contrastive speech rhythm? J. Acoust. Soc. Am. 127, 1559–1569. doi: 10.1121/1.3293004 Russo, R., and Barry, W. J. (2008). “Isochrony reconsidered. Objectifying relations between rhythm mesures and speech tempo,” in Proceedings of Speech Prosody 2008 (Campinas, Brazil), 419–422. Wong, S., and Schreiner, C. (2003). Representation of stop-consonants in cat pri- mary auditory cortex: intensity dependence. Speech Commun. 41, 93–106. doi: 10.1016/S0167-6393(02)00096-1 Schiering, R. (2007). The phonological basis of linguistic rhythm. Cross-linguistic data and diachronic interpretation. Sprachtypologie Universalienforschung 60, 337–359. doi: 10.1524/stuf.2007.60.4.337 Conﬂict of Interest Statement: The authors declare that the research was con- ducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Conﬂict of Interest Statement: The authors declare that the research was con- ducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. Schreiner, C., and Urbas, J. (1986). Representation of amplitude modulation in the auditory cortex of the cat. I. The anterior auditory ﬁeld (AAF). Hear. Res. 21, 227–241. doi: 10.1016/0378-5955(86)90221-2 Schreiner, C., and Urbas, J. (1988). Representation of amplitude modulation in the auditory cortex of the cat. II. Comparison between cortical ﬁelds. Hear. Res. 32, 49–64. doi: 10.1016/0378-5955(88)90146-3 Copyright © 2015 Ordin and Polyanskaya. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this jour- nal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Stevens, K. (2002). Toward a model for lexical access based on acoustic land- marks and distinctive features. J. Acoust. Soc. Frontiers in Psychology \| www.frontiersin.org References Am. 111, 1872–1891. doi: 10.1121/1.1458026 March 2015 \| Volume 6 \| Article 316 Frontiers in Psychology \| www.frontiersin.org 15
https://openalex.org/W4375863324	https://setjournal.com/SET/article/download/68/42	Latin	null	Inclined MHD Effects in Tapered Asymmetric Porous Channel with Peristalsis: Applications in Biomedicine	Science, engineering and technology	2,023	cc-by	5,775	Abstract This paper emphasized an inclined magnetohydrodynamics (MHD) effects in tapered asymmetric porous channel with peristalsis in the presence of slip boundary conditions. Here we considered the two-dimensional channel with a porous medium. The fundamental assumptions of long wavelength and low Reynolds number are applied in the relevant nonlinear equations for momentum, heat, and mass transfer as part of mathematical modeling. The equations subjected to slip boundary conditions have been solved numerically by the Mathematica software. Various essential physical characteristics of velocity, temperature, concentration, and heat transfer rate are captured graphically in the end. The velocity profile is found parabolic for various involved parameters. It is observed that the embedded parameters behave in the exact opposite manner when compared with temperature and concentration distributions. The sinusoidal behavior of the heat transfer rate is also displayed. The unique aspect of this effort is specifically to relate the Joule heating, Darcy resistance, and inclined magnetic field effects in peristaltic flow for a non-Newtonian Jeffrey fluid in an asymmetric tapered channel under the influence of slip boundary conditions. Such preferences have a wide range of applications in engineering, biology, and industry. The outcomes of the presented work are also proficient in the medical field for the treatment of cancer using MHD. The MHD also aids in controlling blood pressure during systolic and diastolic pressure conditions by regulating the blood flow stream. non-Newtonian fluids in a variety of geometries is vast. Some helpful studies on MHD are covered in the references [1-9]. Corresponding author: Anum Tanveer (anum.maths@must.edu.pk) Received: 3 February 2023; Revised: 2 May 2023; Accepted: 3 May 2023; Published: 9 May 2023 © 2023 The Author(s). This work is licensed under a Creative Commons Attribution 4.0 International License Inclined MHD Effects in Tapered Asymmetric Porous Channel with Peristalsis: Applications in Biomedicine Anum Tanveer, Sharak Jarral thematics, Mirpur University of Science and Technology (MUST), Mirpur 10250, AJK, Pakistan Anum Tanveer, Sharak Jarral Department of Mathematics, Mirpur University of Science and Technology (MUST), Mirpur 10250, AJ Vol. 3, No. 1, pp. 36-46 https://doi.org/10.54327/set2023/v3.i2.68 Vol. 3, No. 1, pp. 36-46 https://doi.org/10.54327/set2023/v3.i2.68 Science, Engineering and Technology Science, Engineering and Technology www.setjournal.com https://doi.org/10.54327/set2023/v3.i2.68 www.setjournal.com 1. Introduction Magnetohydrodynamics is the study of the magnetic properties and behavior of electrically conducting fluids. Medical sciences and advanced biomedical engineering are progressively studying magnetohydrodynamics (MHD) that has frequent applications in both engineering and medical fields. The MHD is predominantly utilized in the cancer treatment, hypothermia and open wounds in the medical industry. MHD effects are also employed to study electrolytes, plasma, hot metals and saltwater. Physiological applications of magnetic devices and magnetic particles include drug transfers and magnetic resonance imaging. Today, the brochure on MHD peristaltic transport of The Greek term "Peristaltikos" originates the word "peristalsis" meaning contraction and expansion. The topic of peristalsis for viscous fluids is first executed by Latham [10]. Under long wavelength and low Reynolds number the peristalsis of viscous fluid has been examined by Shapiro et al. [11]. Due to its numerous uses in the fields of engineering and medicine, peristaltic transport is now the most exquisite topic in the scientific world. In biological systems, peristalsis is used for the movement of lymphatic vessels, capillary vasomotion and blood circulation in the arteries from 36 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 heart pumping. The references [12-19] discuss a few insightful studies on peristalsis. This study intends to look into how velocity, temperature and concentration are affected by magnetic field and gravity in porous media. In order to simplify the resulting mathematical problem, the long- wavelength and low Reynolds number approximation are employed. The physical properties of emergent components are discussed by plotting their graphs in terms of velocity, temperature, concentration and heat transfer rate. Non-Newtonian fluid plays a significant role in engineering and technology such as chemical industries, biological sciences and geophysics. Various types of non-Newtonian fluids including oils, ketchup, honey, paints and toothpaste are used in everyday activities. Due to numerous uses of non-Newtonian fluids in industries the peristaltic flow of these fluids cannot be neglected. Non-Newtonian fluids perform better computation when examining the rheology of naturally occurring processes. One of the non-Newtonian fluids is Jeffrey fluid that is found to give better relevance for physiological fluids. Such as describing blood flow in arteries. It is revealed that Jeffrey model is one of the significant non-Newtonian model which precise the best representation of viscoelastic fluids. The references [20-31] carry out some investigations regarding the use of Jeffrey liquid. 2. Mathematical Formulation 36-46 𝐵ത= (0, 0, 𝐵଴𝑠𝑖𝑛𝜂), (2) by generalized Ohm’s law = (0, 0, 𝐵଴𝑠𝑖𝑛𝜂), (2) neralized Ohm’s law ത ത ത ଵ ̅ ത 𝜌௙ቂ డ௎ഥ డ௧̅ + 𝑈ഥడ௎ഥ డ௑ത+ 𝑉തడ௎ഥ డ௒തቃ= − డ௉ത డ௑ത+ డ డ௑ത(𝑆̅௑௑ തതതത) + డ డ௒ത(𝑆̅௑௒ തതതത) + (1 −𝐶଴)𝜌௙𝑔𝛽௧(𝑇ത−𝑇଴) + 𝜌௙ቂ డ௎ഥ డ௧̅ + 𝑈ഥడ௎ഥ డ௑ത+ 𝑉തడ௎ഥ డ௒തቃ= − డ௉ത డ௑ത+ డ డ௑ത(𝑆̅௑௑ തതതത) + డ డ௒ത(𝑆̅௑௒ തതതത) + (1 −𝐶଴)𝜌௙𝑔𝛽௧(𝑇ത−𝑇଴) + 𝐵ത= (0, 0, 𝐵଴𝑠𝑖𝑛𝜂), (2) by generalized Ohm’s law డ డ௒ത(𝑆̅௑௒ തതതത) + (1 −𝐶଴)𝜌௙𝑔𝛽௧(𝑇ത−𝑇଴) + 𝐽̅ = 𝜎ቂ(𝐸ത+ 𝑉ത× 𝐵ത) − ଵ ௘௡೐(𝐽̅ × 𝐵ത)ቃ, (3) (3) ൬ ఘ೛ିఘ೑ ఘ೑൰𝑔𝛽௖(𝐶̅ −𝐶଴) − ఙ୆బ మୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗(𝑈ഥ−𝑚𝑉തsin 𝜂)− where 𝑉ത is fluid velocity, 𝐽̅ the current density,𝜎 the electrical conductivity and we disregarded the impact of the electric force, 𝐸ത = 0. By using Eq. (2) and (3), we obtain: ఓ ୩భ(ଵା஛భ) ቀ𝑈ഥ+ λଶ ୢ௎ഥ ୢ୲ቁ+ 𝜌௙𝑔sin 𝜂 , (10) (10) 𝜌௙ቂ డ௏ഥ డ௧̅ + 𝑈ഥడ௏ഥ డ௑ത + 𝑉തడ௏ഥ డ௒തቃ= − డ௉ത డ௒ത+ డ డ௑ത(𝑆̅௑௒ തതതത) + 𝐽̅ × 𝐵ത= ఙ୆బ మୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗[(−𝑈ഥ+ 𝑚𝑉തsin 𝜂)𝑖−(𝑉ത+ 𝑚𝑈ഥsin 𝜂)𝑗+ 0𝑘]. (4) డ డ௒(𝑆̅௒௒ തതതത) −ఙ୆బ మୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗(𝑉ത+ 𝑚𝑈ഥsin 𝜂) − ఓ ୩భ(ଵା஛భ) ቀ𝑉ത+ (4) 𝑚𝑈ഥsin 𝜂)𝑗+ 0𝑘]. 𝑚𝑈ഥsin 𝜂)𝑗+ 0𝑘]. λଶ ୢ௏ഥ ୢ୲ቁ −𝜌௙𝑔cos 𝜂, (11) where 𝑚= ఙ୆బ ௘௡೐ represents the Hall current parameter. λଶ ୢ௏ഥ ୢ୲ቁ −𝜌௙𝑔cos 𝜂, (11) (11) ቂడ்ത డ௧̅ + 𝑈ഥడ்ത డ௑ത+ 𝑉തడ்ത డ௒തቃ= α ቀ డమ்ത డ௑തమ + డమ்ത డ௒തమቁ+ ଵ ఘ೑಴೑ ቂ𝑆̅௑௑ തതതത డ௎ഥ డ௑ത+ 𝑆̅௑௒ തതതതቀ డ௎ഥ డ௒ത+ డ௏ഥ డ௑തቁ+ 𝑆̅௒௒ തതതത డ௏ഥ డ௒തቃ+ Joule heating is listed below [34]: Joule heating is listed below [34]: ௃̅.௃̅ ఙ= ఙ஻బమ௦௜௡మఎ ଵା௠మ௦௜௡మఎ[𝑈ഥ+ 𝑉ത]. (5) (5) ଵ ఘ೑಴೑ ቂ𝑆̅௑௑ തതതത డ௎ഥ డ௑ത+ 𝑆̅௑௒ തതതതቀ డ௎ഥ డ௒ത+ డ௏ഥ డ௑തቁ+ 𝑆̅௒௒ തതതത డ௏ഥ డ௒തቃ+ The following equation describes the Cauchy stress tensor and extra stress tensor T and S for an incompressible Jeffrey material [36]. τ ൤𝐷஻ቀ డ஼̅ డ௑ത డ்ത డ௑ത+ డ஼̅ డ௒ത డ்ത డ௒തቁ+ ஽೅ ்೘൬ቀ డ்ത డ௑തቁ ଶ + ቀ డ்ത డ௒തቁ ଶ ൰൨+ ଵ ఘ೑௖೑ ఙ୆బ మୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗[𝑈ഥଶ+ 𝑉തଶ] , (12) 𝑇= −𝑃𝐼+ 𝑆, (6) 𝑆= ఓ ଵା஛భቀ𝛾̇ + λଶ ୢఊ̇ ୢ୲ቁ . (7) (12) 𝑇= −𝑃𝐼+ 𝑆, (6) (7) ቂ డ஼̅ డ௧̅ + 𝑈ഥడ஼̅ డ௑ത+ 𝑉തడ஼̅ డ௒തቃ= 𝐷஻ቀ డమ஼̅ డ௑തమ + డమ஼̅ డ௒തమቁ+ ஽೅ ்೘ቀ డమ்ത డ௑തమ+ The Rivilin-Ericksen tensor 𝛾̇ is listed below [19] డమ்ത డ௒തమቁ. (13) డమ்ത డ௒തమቁ. డమ்ത డ௒തమቁ. (13) 𝛾̇ = (∇𝑉ത) + (∇𝑉ത)௧. 𝛾̇ = (∇𝑉ത) + (∇𝑉ത)௧. 2. Mathematical Formulation Consider a two dimensional inclined tapered asymmetric porous channel through which the Jeffrey nanofluid is flowing under peristaltic wave in the presence of magnetic field. The lower and upper walls are located at positions 𝑌= 𝐻ଵ and 𝑌= 𝐻ଶ with 𝑇ଵ,𝐶ଵ 𝑎𝑛𝑑 𝑇଴, 𝐶଴ the temperature and concentration of upper and lower walls respectively (see Figure 1) as follows. The impact of heat and mass transport has a variety of uses in the biomedical field. Numerous technological applications observed mass transfer in action in addition to heat transfer. In chemical engineering mass transfer is vital notably in chemical reactors and industrial operations. Convection is the most effective way to describe how heat is moved from one point to another during peristaltic flow through porous media in cryotherapy. Conduction, convection and radiation are the basic three ways that transfer heat in multiple systems. Evaluation of skin irritation, hypertension, the eradication of harmful cancer cells, the dilution technique for assessing blood flow and paper manufacturing is just a few of the plethora of processes that are involved in heat transmission. The references [32-43] cover some useful studies in heat and mass transfer. Figure 1. Geometrical demonstration of problem. Figure 1. Geometrical demonstration of problem. The geometry of walls is characterized by the subsequent form [31]. The objective of this research is to construct a mathematical model to evaluate the consequences of inclined MHD along peristaltic channel. Jeffrey nanofluid in a tapered asymmetric channel with porous media has been considered in flow analysis. In our research the combined investigation of inclined asymmetric porous channel, slip boundary conditions, Darcy resistance, sloped MHD with peristaltic flow of non-Newtonian Jeffrey fluid is taken into account. In the realm of research, our paper is extremely innovative. 𝐻ଵ= −𝑑−𝐾ଵ തതത𝑥̅ −𝑎ଵ തതതsin ቂ ଶగ ఒ(𝑥̅ −𝑐𝑡̅) + 𝜙ቃ, (1a) 𝐻ଶ= 𝑑+ 𝐾ଵ തതത𝑥̅ + 𝑎ଶ തതതsin ቂ ଶగ ఒ(𝑥̅ −𝑐𝑡̅)ቃ. (1b) (1a) (1b) Here d shows half width of the channel, λ is wavelength, 𝑎തଵ and 𝑎തଶis wave amplitude of lower and upper wall respectively,𝑡̅ denotes the timeand 𝜙shows the phase angle. We enforce uniform magnetic field which is [44-45]. 37 Science, Engineering and Technology Vol. 3, No. 2, pp. 2. Mathematical Formulation The frame (𝑥̅, 𝑦ത) in which the flow is thought to be steady is flowing at a velocity of c. The transformations between the laboratory and wave frame are [6]: In term of Jeffrey fluid model, the Darcy resistance is [13] 𝑥̅ = 𝑋ത+ 𝑐𝑡̅, 𝑦ത= 𝑌ത, 𝑢ത= 𝑈ഥ−𝑐, 𝑣̅ = 𝑉ത, 𝑝̅(𝑥̅, 𝑦ത) = 𝑃ത(𝑋ത, 𝑌ത, 𝑡̅), 𝑇= 𝑇ത. (14) 𝑥̅ = 𝑋ത+ 𝑐𝑡̅, 𝑦ത= 𝑌ത, 𝑢ത= 𝑈ഥ−𝑐, 𝑣̅ = 𝑉ത, 𝑝̅(𝑥̅, 𝑦ത) = 𝑅= − ఓ ୩భ(ଵା஛భ) ቀ1 + λଶ ୢ ୢ୲ቁ𝑉ത. (8) (8) 𝑃ത(𝑋ത, 𝑌ത, 𝑡̅), 𝑇= 𝑇ത. (14) (14) 𝑃ത(𝑋ത, 𝑌ത, 𝑡̅), 𝑇= 𝑇ത. where 𝑉ത represent the fluid’s velocity and kଵ the permeability parameter. where 𝑉ത represent the fluid’s velocity and kଵ the permeability parameter. 𝑥= ௫̅ ఒ, 𝑦= ௬ത ௗ, 𝑢= ௨ഥ ௖, 𝑣= ௩ത ௖, 𝑡= ௖௧̅ ఒ, 𝑎ଵ= ௔భ തതതത ௗ, 𝑎ଶ= ௔మ തതതത ௗ, 𝛿= ௗ ఒ, ℎଵ= ுభ തതതത ௗ, ℎଶ= ுమ തതതത ௗ, 𝑣= ఓ ఘ೑, 𝑘ଵ= ఒ௞భ തതതത ௗ, 𝑝= ௗమ௣̅ ௖ఒఓ, 𝑀= ට ఙ ఓ𝐵଴𝑑, 𝜆ଶ തതതത= ఒమ ௖ ௗ, 𝑃𝑟= ௩ ఈ, 𝑅𝑒= The flow of an incompressible nanofluid in a fixed frame is represented by the following equations [6]: డ௎ഥ డ௑ത+ డ௏ഥ డ௒ത= 0, (9) డ௎ഥ డ௑ത+ డ௏ഥ డ௒ത= 0, (9) 38 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 ௖ௗ ௩, 𝑆= ௌ̅ௗ ఓ௖ , 𝐺𝑚= ൫ఘ೎ିఘ೑൯௚ఉ೎(஼భି஼బ)ௗమ ఓ௖ , 𝑁𝑡= ఛ஽೅(்భି்బ) ௩்೘ , 𝑁𝑏= ఛ஽ಳ(஼భି஼బ) ௩ , 𝐵𝑟= 𝑃𝑟× 𝐸𝑐, 𝐹𝑟= ௖మ ௚ௗ, 𝐷𝑎= ௞భ ௗమ, 𝐺𝑟= ( ଵି஼బ)ఘ೑௚ఉ೟(்భି்బ)ௗమ ఓ௖ , 𝐸𝑐= ௖మ ௖೑(்భି்బ) , 𝜃= ்തି்బ ்భି்బ, 𝜁= ஼̅ି஼బ ஼భି஼బ (15) 𝑅𝑒𝛿ቂ(𝑢+ 1) డ఍ డ௫+ ௩ ఋ డ఍ డ௬ቃ= ቀ𝛿ଶడమ఍ డ௫మ+ డమ఍ డ௬మቁ+ ே௧ ே௕ቀ𝛿ଶడమఏ డ௫మ+ డమఏ డ௬మቁ, (19) (19) ௖మ ௚ௗ, 𝐷𝑎= ௞భ ௗమ, 𝐺𝑟= ( ଵି஼బ)ఘ೑௚ఉ೟(்భି்బ)ௗమ ఓ௖ , 𝐸𝑐= The stream function and designated velocity fields are now being inserted. ௖మ ௖೑(்భି்బ) , 𝜃= ்തି்బ ்భି்బ, 𝜁= ஼̅ି஼బ ஼భି஼బ (15) 𝑢= డట డ௬, 𝑣= −𝛿 డట డ௫ . (20) (15) (20) where the stress elements are represented by the following form: 3. Solution to Problem 𝑆௫௫= ଶఋ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ𝜓௫௬, (21) 𝑆௫௬= ଵ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ൫𝜓௬௬− 𝛿ଶ𝜓௫௫൯, (22) 𝑆௬௬= − ଶఋ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ𝜓௫௬. (23) 𝑆௫௫= ଶఋ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ𝜓௫௬, (21) (21) After using dimensionless variables (15) and falling bars, the Eqs. (10-13) become: 𝑅𝑒𝛿ቂ(𝑢+ 1) డ௨ డ௫+ ௩ ఋ డ௨ డ௬ቃ= − డ௣ డ௫+ 𝛿 డ డ௫(𝑆௫௫) + డ డ௬൫𝑆௫௬൯+ 𝐺𝑟𝜃+ 𝐺𝑚𝜁− ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗(𝑢+ 1) + ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗(𝑚𝑣sin 𝜂) − ଵ ஽ೌ(ଵା஛భ) ቆ1 + ௖஛మ ௗቀ𝛿 డ௨ డ௫+ 𝑆௫௬= ଵ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ൫𝜓௬௬− 𝛿ଶ𝜓௫௫൯, (22) (22) 𝑆௬௬= − ଶఋ (ଵାఒభ) ቂ1 + ఒమ௖ఋ ௗቀ𝜓௬ డ డ௫−𝜓௫ డ డ௬ቁቃ𝜓௫௬. (23) (23) Using the preceding non-dimensional governing flow Eqs. (16)-(19) with long wavelengths and low Reynolds number approximations one gets డ௩ డ௬ቁቇ× (𝑢+ 1) + ோ௘ ி௥sin 𝜂 , (16) (16) 𝑅𝑒𝛿ଶቂ(𝑢+ 1) డ௩ డ௫+ ௩ ఋ డ௩ డ௬ቃ= − డ௣ డ௬+ 𝛿ଶడ డ௫൫𝑆௫௬൯+ 𝛿 డ డ௬൫𝑆௬௬൯−𝛿𝑣ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቁ− 0 = − డ௣ డ௫+ ቀ ଵ ଵାఒభቁ డయట డ௬య+ 𝐺𝑟𝜃+ 𝐺𝑚𝜁−ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗+ ଵ ஽ೌ(ଵାఒభ)ቁቂ డట డ௬+ 1ቃ+ ோ௘ ி௥sin 𝜂, (24) 0 = − డ௣ డ௫+ ቀ ଵ ଵାఒభቁ డయట డ௬య+ 𝐺𝑟𝜃+ 𝐺𝑚𝜁−ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗+ 𝑅𝑒𝛿ଶቂ(𝑢+ 1) డ௩ డ௫+ ௩ ఋ డ௩ డ௬ቃ= − డ௣ డ௬+ 𝛿ଶడ డ௫൫𝑆௫௬൯+ ଵ ஽ೌ(ଵାఒభ)ቁቂ డట డ௬+ 1ቃ+ ோ௘ ி௥sin 𝜂, (24) (24) 𝛿 డ డ௬൫𝑆௬௬൯−𝛿𝑣ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቁ− 0 = − డ௣ డ௬ , (25) 0 = − డ௣ డ௬ , (25) 𝛿ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቁ𝑚sin 𝜂× (𝑢+ 1) − ఋ௩ ஽ೌ(ଵା஛భ)ቆ1 + 𝛿ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቁ𝑚sin 𝜂× (𝑢+ 1) − ఋ௩ ஽ೌ(ଵା஛భ)ቆ1 + 0 = డమఏ డ௬మ+ 𝐵𝑟൤ቀ ଵ ଵାఒభቁቀ డమట డ௬మቁ ଶ + ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቀ డట డ௬+ ௖஛మ ௗቀ𝛿డ௨ డ௫+ డ௩ డ௬ቁቇ−𝛿ோ௘ ி௥cos 𝜂, (17) ቂ(𝑢+ 1) డఏ డ௫+ ௩ ఋ డఏ డ௬ቃ= 𝐸𝑐ቂδ𝑆௫௫ డ௨ డ௫+ 𝑆௫௬ቀ డ௨ డ௬+ 0 = డమఏ డ௬మ+ 𝐵𝑟൤ቀ ଵ ଵାఒభቁቀ డమట డ௬మቁ ଶ + ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗ቀ డట డ௬+ 1ቁ ଶ ൨+ 𝑃𝑟𝑁𝑏 డఏ డ௬ డ఍ డ௬+ 𝑃𝑟𝑁𝑡ቀ డఏ డ௬ቁ ଶ , (2 0 = డమ఍ డ௬మ+ ே௧ ே௕ డమఏ డ௬మ . (2 ௖஛మ ௗቀ𝛿డ௨ డ௫+ డ௩ డ௬ቁቇ−𝛿ோ௘ ி௥cos 𝜂, (17) ቂ(𝑢+ 1) డఏ డ௫+ ௩ ఋ డఏ డ௬ቃ= 𝐸𝑐ቂδ𝑆௫௫ డ௨ డ௫+ 𝑆௫௬ቀ డ௨ డ௬+ 0 = డ௬మ+ 𝐵𝑟൤ቀଵ 1ቁ ଶ ൨+ 𝑃𝑟𝑁𝑏 డఏ డ௬ డ డ௬ 0 = డమ఍ డ௬మ+ ே௧ ே௕ డమఏ డ௬మ . 5. Result and Discussion 𝜓= ௤ ଶ , డట డ௬+ 𝛼ଵቀ ଵ ଵାఒభቁ డమట డ௬మ= 0, 𝜃+ 𝛼ଶ డఏ డ௬= 1, 𝜁+ 𝛼ଷ డ఍ డ௬= 1 𝑎𝑡 𝑦= ℎଶ = 1 + 𝑘ଵ𝑥+ 𝑎ଶsin[2𝜋(𝑥−𝑡)], In this section, the effects of pertinent parameters on common profiles (concentration, temperature, and velocity) are discussed. In-depth discussion is held regarding the factors that affect the peristaltic transport of a nanofluid, including Grashof number Gr(0.5 ≤ 𝐺𝑟≤2.0 ), Hartmann number M( 0.0 ≤𝑀≤3.0 ), nanoparticle Grashof number Gm( 0.5 ≤𝐺𝑚≤2.0), flow rates Q(4.0 ≤𝑄≤5.5),the ratio of relaxation to retardation times λଵ( 1.0 ≤λଵ≤4.0 ), Darcy number Da( 0.3 ≤𝐷𝑎≤1.2 ), Brinkman number Br( 0.2 ≤ Br ≤1.4 ), Brownian motion parameter Nb( 0.2 ≤ 𝑁𝑏≤0.8), thermophoresis parameter Nt(0.2 ≤𝑁𝑡≤ 0.8 ), and Prandtl number Pr( 0.5 ≤𝑃𝑟≤2.0 ). The numerical computation is performed using the Mathematica built-in numerical ND-Solve method. 𝑎ଶsin[2𝜋(𝑥−𝑡)], (29) (29) 𝜓= − ௤ ଶ , డట డ௬−𝛼ଵቀ ଵ ଵାఒభቁ డమట డ௬మ= 0, 𝜃−𝛼ଶ డఏ డ௬= 0, 𝜁−𝛼ଷ డ఍ డ௬= 0 𝑎𝑡 𝑦= ℎଵ = −1 − 𝑘ଵ𝑥− 𝑎ଵsin[2𝜋(𝑥−𝑐𝑡) + 𝜙]. (3 (30) The wave and fixed frame's flow rates are related by [6]: 𝑄= 𝑞+ 1 + 𝑑. (31) (31) The outlook of velocity profiles for variation in various involved parameters is parabolic. The velocity profile at the centre of the channel decays and the walls of the channel experiencing an opposite behaviour when M is increased (see Figure 2(a)). Dual responce of velocity is captured for rising values of Gm and Gr (see Figure 2(b) and Figure 2(c)). It depicted in Figure 2(d) as the value of Da increases, then the velocity distribution increases at central portion and dropped at the lower part of channel. Figure 2(e) corresponds to dual impact of 𝜆ଵ on velocity profile. As in Figure 2(f) the flow rate Q is elevated, the velocity distribution increases. Moreover, it is observed that the velocity profiles satisfy the boundary requirements. Our results for Figure 2(a) and Figure 2(d) are analogous to the outcomes reported in ref. [46]. Dimensionless form of Heat transfer rate in is listed below [18]. 𝑍= ℎଶ ᇱ(𝑥) ቚ డఏ డ௬ቚ ௬ୀ௛మ . (32) (32) 3. Solution to Problem ௖஛మ ௗቀ𝛿డ௨ డ௫+ డ௩ డ௬ቁቇ−𝛿ோ௘ ி௥cos 𝜂, (17) ቂ( + 1) డఏ+ ௩డఏቃ 𝐸ቂδ𝑆 డ௨+ 𝑆 ቀ డ௨+ 0 1ቁ ଶ ௖஛మ ௗቀ𝛿డ௨ డ௫+ డ௩ డ௬ቁቇ−𝛿ோ௘ ி௥cos 𝜂, (17) 0 డ௬మ+ 𝐵𝑟൤ቀଵାఒ 1ቁ ଶ ൨+ 𝑃𝑟𝑁𝑏 డఏ డ௬ డ఍ డ௬+ ௖஛మ ௗቀ𝛿డ௨ డ௫+ డ௩ డ௬ቁቇ−𝛿ோ௘ ி௥cos 𝜂, (17) (17) 1ቁ ଶ ൨+ 𝑃𝑟𝑁𝑏 డఏ డ௬ డ఍ డ௬+ 𝑃𝑟𝑁𝑡ቀ డఏ డ௬ቁ ଶ , (26) (26) ቂ(𝑢+ 1) డఏ డ௫+ ௩ ఋ డఏ డ௬ቃ= 𝐸𝑐ቂδ𝑆௫௫ డ௨ డ௫+ 𝑆௫௬ቀ డ௨ డ௬+ 𝛿ଶడ௩ డ௫ቁ+ 𝑆௬௬ డ௩ డ௬+ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗((𝑢+ 1)ଶ+ 𝑣ଶ)ቃ+ ଵ ௉௥ቂ𝛿ଶడమఏ డ௫మ+ డమఏ డ௬మቃ+ 𝑁𝑏ቂ𝛿ଶడఏ డ௫ డ఍ డ௫+ డఏ డ௬ డ఍ డ௬ቃ+ డఏ ଶ డఏ ଶ ቂ(𝑢+ 1) డఏ డ௫+ ௩ ఋ డఏ డ௬ቃ= 𝐸𝑐ቂδ𝑆௫௫ డ௨ డ௫+ 𝑆௫௬ቀ డ௨ డ௬+ 𝛿ଶడ௩ డ௫ቁ+ 𝑆௬௬ డ௩ డ௬+ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗((𝑢+ 1)ଶ+ 𝑣ଶ)ቃ+ 0 = డమ఍ డ௬మ+ ே௧ ே௕ డమఏ డ௬మ . (27) (27) After abolishing the pressure gradient term Eqs. (24- 25) reduce to subsequent form. ଵ ௉௥ቂ𝛿ଶడమఏ డ௫మ+ డమఏ డ௬మቃ+ 𝑁𝑏ቂ𝛿ଶడఏ డ௫ డ఍ డ௫+ డఏ డ௬ డ఍ డ௬ቃ+ ൤𝛿ଶቀ డఏ డ௫ቁ ଶ + ቀ డఏ డ௬ቁ ଶ ൨, (18) ቀ ଵ ଵାఒభቁ డరట డ௬ర+ 𝐺𝑟 డఏ డ௬+ 𝐺𝑚 డ఍ డ௬−ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗+ ଵ ஽ೌ(ଵାఒభ)ቁ డమట డ௬మ= 0, (28) ቀ ଵ ଵାఒభቁ డరట డ௬ర+ 𝐺𝑟 డఏ డ௬+ 𝐺𝑚 డ఍ డ௬−ቀ ெమୱ୧୬మ஗ ଵା௠మୱ୧୬మ஗+ ൤𝛿ଶቀ డఏ డ௫ቁ ଶ + ቀ డఏ డ௬ቁ ଶ ൨, (18) ଵ ஽ೌ(ଵାఒభ)ቁ డమట డ௬మ= 0, ൤𝛿ଶቀ డఏ డ௫ቁ ଶ + ቀ డఏ డ௬ቁ ଶ ൨, (18) ଵ ஽ೌ(ଵାఒభ)ቁ డమట డ௬మ= 0, (28) (18) ଵ ஽ೌ(ଵାఒభ)ቁ డమట డ௬మ= 0, (28) (28) 39 Vol. 3, No. 2, pp. 36-46 Science, Engineering and Technology The boundary conditions are defined as follows [24]: The boundary conditions are defined as follows [24]: 4. Numerical Method By using the Mathematica ND solve technique, the modified dimensionless equations are numerically processed. The benefit of this approach is that it picks the right algorithm and detects any possible errors automatically. Furthermore, this technique delivers excellent computational output with just three to four minutes of CPU time required for each evaluation. In reality, this approach avoids complex solution expressions and directly displays graphical depictions. The Eqs, (26)-(28) subjected to boundary conditions (29) and (30) have been solved numerically by the ND_solve method. The commercial software Mathematica includes a built-in feature for validating the results. The governing equations (26-28) are coupled and have a significant degree of nonlinearity. It is impossible to find the exact solution. Consequently, the numerical solution has been obtained. The Eqs, (26)-(28) subjected to boundary conditions (29) and (30) have been solved numerically by the ND_solve method. The commercial software Mathematica includes a built-in feature for validating the results. The governing equations (26-28) are coupled and have a significant degree of nonlinearity. It is impossible to find the exact solution. Consequently, the numerical solution has been obtained. In Figure 3(a) Temperature is seen to grow when the values ofpertinent parameters M boost.It is noticed that in Figure 3(b) the escalating values of the Brikmann number Br enhances the temperature profile owing to the effects of viscous dissipation. We noticed that Temperature rises up with the increament in the value of Nb and Nt (see Figure 3(c) and Figure 3(d)). These outcomes for temperature profile are appropriately matched with the results gained by Abd-Alla et al. [6]. The temperature distribution also appears to fulfill the boundary condition. 40 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 Science, Engineering and Technology Vol. 3, No. 2, pp. 36 46 (a) (b) (c) (d) (e) (f) Figure 2. Velocity u versus y for: (a). M, (b). Gr, (c). Gm, (d). Da, (e). 𝜆ଵ and (f). Q when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1.0 , αଵ= 0.1, αଶ= 0.1, αଷ= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0. (a) (b) (b) (a) (d) (c) (c) (d) ( ) (f) (e) (f) (e) Figure 2. Velocity u versus y for: (a). M, (b). Gr, (c). Gm, (d). Da, (e). 𝜆ଵ and (f). 4. Numerical Method Q when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1.0 , αଵ= 0.1, αଶ= 0.1, αଷ= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0. 41 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 (a) (b) (c) (d) Figure 3. Temperature 𝜃 versus y for: (a). M, (b). Br, (c). Nb, (d). Nt when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1, αଵ= 0.1, αଶ= 0.1, αଷ= 0.1 (b) (a) (a) (b) (c) (d) (d) (c) Figure 3. Temperature 𝜃 versus y for: (a). M, (b). Br, (c). Nb, (d). Nt when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1, αଵ= 0.1, αଶ= 0.1, αଷ= 0.1 For various values of the M, Br, Nt the heat transfer rate Z is displayed in Figure 5. We noticed that heat transfer coefficient Z rises with the increament in the value of M, Br. so, results are shown in figure 5(a) and figure 5(b). Similarly, heat transfer rate Z falls with the increase of Nt which is exposed in figure 5(c).The sinusoidal behaviour of heat transfer rate is observed.It is noticed that outlook of heat transfere rate is oscillatary. Our outcome for Br parameter is analogous to the upshot reported in ref [18]. Concentration 𝜁 falls when the values of pertinent parameters M, Nt and Pr boost (see Figure 4(a), Figure 4(c) and Figure (4d)) and reverse pattern is seen for Nb (see Figure 4(b)). However, the concentration is not significantly affected by the Brownian motion parameter. The distribution of concentration also filfil the boundaryconditions. It is observed that, the embedded parameters behave in the exact opposite manner when we compared the temperature profile with regard to concentration distribution. Our findings for concentration profile are matched to those results for pertinent perameters reported in ref. [3]. 42 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 (a) (b) (c) (d) Figure 4. 6. Conclusion In this research, we addressed on the inclined Magnetohydrodynamics (MHD) effects in a tapered asymmetric porous channel with peristalsis under the effects of slip boundary conditions. From present study, we concluded that fluid velocity declines in the channel's central portion whilst boosted up near the walls under a stronger magnetic field parameter. Because greater Hartman number strengthen theresistive force (Lorentz force) that opposes fluid velocity. For escalating levels of Gm and Gr the fluid viscosity decays and in view of this fact the dual response of velocity is captured.Temperature goes up whilst concentration decays for increament in values of M. It happens due to the increament in magnetic field. Growing value of Br enhances the temperature profile owing to the effects of viscous dissipation. A rise in Nt and Nb represents an increament in temperature which is consistent with the nanoparticles' effective transfer from the wall to the fluid, causing a notable boost in the temperature profile.It is revealed that outlook of heat transfer rate Z is oscillatory because of the contraction and expansion of the walls of the channel. The absolute value of heat transfer rate Z enhances when M and Br goes up and contrary trend is demonstrated for Nt. [3] M. Rafiq, M. Sajid, S.E. Alhazmi, M.I. Khan, E.R. El- Zahar, “MHD electroosmotic peristaltic flow of Jeffrey nanofluid with slip conditions and chemical reaction,’’ Alexandria Engineering Journal, 61(12), pp. 9977-9992, 2022. [4] B.B. Divya, G. Manjunatha, C. Rajashekhar, H. Vaidya, K.V. Prasad, “Effects of inclined magnetic field and porous medium on peristaltic flow of a Bingham fluid with heat transfer,’’ Journal of Applied and Computational Mechanics, 7(4), pp. 1892-1906, 2021. [5] A.M. Abd-Alla, S. M. Abo-Dahab, E. N. Thabet, M. A. Abdelhafez, “Impact of inclined magnetic field on peristaltic flow of blood fluid in an inclined asymmetric channel in the presence of heat and mass transfer,’’ Waves in Random and Complex Media, pp. 1-25, 2022. [6] A.M. Abd-Alla, E.N. Thabet, F.S. Bayones, “Numerical solution for MHD peristaltic transport in an inclined nanofluid symmetric channel with porous medium,’’ Scientific Reports, 12(1), pp. 1-11, 2022. [7] N.B. Naduvinamani, A.S. Guttedar, L. Devindrappa, “On the Magnetohydrodynamic (MHD) Peristaltic Flow of a Hyperbolic Tangent Fluid in a Tapered Asymmetric Channel,’’ Journal of Nanofluids, 11(5), pp. 737-744, 2022. [8] M. Devakar, K. Ramesh, K. References [1] M. Nazeer, M.I. Khan, M.U. Rafiq, N.B. Khan, “Numerical and scale analysis of Eyring-Powell nanofluid towards a magnetized stretched Riga surface with entropy generation and internal resistance,’’ International Communications in Heat and Mass Transfer, 119, 104968, 2020. Figure 5 (c). Heat transfer rate Z versus x for Nt [2] Y.M. Chu, F. Ahmad, M.I. Khan, M. Nazeer, F. Hussain, N.B. Khan, L, Mei, “Numerical and scale analysis of non-Newtonian fluid (Eyring-Powell) through pseudo- spectral collocation method (PSCM) towards a magnetized stretchable Riga surface,’’ Alexandria Engineering Journal, 60(2), pp. 2127-2137, 2021. 6. Conclusion Vajravelu, “Magnetohydrodynamic effects on the peristaltic flow of couple stress fluid in an inclined tube with endoscope,’’ Journal of Computational Mathematics and Data Science, 2, 100025, 2022 Data and Materials Accessibility No data or additional materials were utilized for the research described in the article. 4. Numerical Method Concentration 𝜁versus y for: (a). M, (b). Nb, (c). Nt and (d). Pr when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1, αଵ= 0.1,αଶ= 0.1, αଷ= 0.1. (a) (b) (b) (a) (c) (d) (d) (c) Figure 4. Concentration 𝜁versus y for: (a). M, (b). Nb, (c). Nt and (d). Pr when 𝑥= 0.3, 𝑡= 0.1, 𝑎ଵ= 0.1, 𝑎ଶ= 0.1, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 0.1, 𝑃𝑟= 0.7, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.2, 𝑁𝑏= 0.2, 𝐷𝑎= 0.3, 𝑀= 1.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 1, αଵ= 0.1,αଶ= 0.1, αଷ= 0.1. (a) (b) Figure 5. Heat transfer rate Z versus x for: (a) M, (b) Br, when 𝑥= 0.3, 𝑡= 0.5, 𝑎ଵ= 0.08, 𝑎ଶ= 0.08, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 1, 𝑃𝑟= 0.8, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.5, 𝑁𝑏= 0.5, 𝐷𝑎= 0.5, 𝑀= 2.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 0.1, αଵ= 0.1, αଶ= 0.1, αଷ= 0.1 (a) (b) (b) (a) Figure 5. Heat transfer rate Z versus x for: (a) M, (b) Br, when 𝑥= 0.3, 𝑡= 0.5, 𝑎ଵ= 0.08, 𝑎ଶ= 0.08, 𝐺𝑟= 0.5, 𝐺𝑚= 0.5, 𝐵𝑟= 1, 𝑃𝑟= 0.8, 𝑑= 0.5, 𝑄= 3.0, 𝑁𝑡= 0.5, 𝑁𝑏= 0.5, 𝐷𝑎= 0.5, 𝑀= 2.0, 𝑚= 0.3, η = ஠ ଺, 𝜆ଵ= 0.1, 𝑘ଵ = 0.1, αଵ= 0.1, αଶ= 0.1, αଷ= 0.1 43 Vol. 3, No. 2, pp. 36-46 Science, Engineering and Technology Figure 5 (c). Heat transfer rate Z versus x for Nt Competing Interest Statement The authors declare no known competing financial interests or personal relationships that could have influenced the work reported in this paper. [9] F.S. Bayones, A.M. Abd-Alla, E.N. Thabet, “Magnetized dissipative Soret effect on nonlinear radiative Maxwell nanofluid flow with porosity, chemical reaction and 44 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 Joule heating,’’ Waves in Random and Complex Media, pp. 1-19, 2022. roughness,” International Journal of Nonlinear Analysis and Applications, 13(2), pp. 117-130, 2022. [10] T.W. Latham, “Fluid motions in a peristaltic pump (Doctoral dissertation,’’ Massachusetts Institute of Technology). 1966. [20] M. Kahshan, D. Lu, A.M. Siddiqui, “A Jeffrey fluid model for a porous-walled channel: Application to flat plate dialyzer,’’ Scientific reports, 9(1), pp. 1-18, 2019. [11] A.H. Shapiro, M.Y. Jaffrin, S.L. Weinberg, “Peristaltic pumping with long wavelengths at low Reynolds number,’’ Journal of fluid mechanics, 37(4), pp. 799- 825, 1969. [21] F. Ali, S. Murtaza, I. Khan, N.A. Sheikh, and K. S, “Atangana–Baleanu fractional model for the flow of Jeffrey nanofluid with diffusion-thermo effects: applications in engine oil, Advances in Difference Equations,” pp. 1-21, 2019. [12] Z. Hussain, k. Al-Khaled, U. Ashrif, A. Abbasi, S.U. Khan, W. Farooq, M.Y. Malik, “A mathematical model for radiative peristaltic flow of Jeffrey fluid in curved channel with Joule heating and different walls: Shooting technique analysis,’’ Ain Shams Engineering Journal, 13(5), 101685, 2022 [22] A.S. Kotnurkar, J. Beleri, I.A. Badruddin, N.A. Ahammad, “Effect of Thermal Radiation and Double- Diffusion Convective Peristaltic Flow of a Magneto- Jeffrey Nanofluid through a Flexible Channel,” Mathematics, 10(10), 170, 2022. .[13] T. Hayat, S. Farooq, B. Ahmad, A. Alsaedi, “Effectiveness of entropy generation and energy transfer on peristaltic flow of Jeffrey material with Darcy resistance,’’ International Journal of Heat and Mass Transfer, 106, pp. 244-252. 2017. [23] A.S. Kotnurkar, V.T. Talawar, “Influence of thermal jump and inclined magnetic field on peristaltic transport of Jeffrey fluid with silver nanoparticle in the eccentric annulus. Heliyon,’’ 8(9), e10543, 2022. [24] M. Rafiq, M. Sajid, S.E. Alhazmi, M.I. Khan, E.R. El- Zahar, “MHD electroosmotic peristaltic flow of Jeffrey nanofluid with slip conditions and chemical reaction,’’ Alexandria Engineering Journal, 61(12), pp. 9977- 9992, 2022. [14] A.M. Abd-Alla, E.N. Thabet, F.S. Bayones, A.M. Alsharif, “Heat transfer in a non-uniform channel on MHD peristaltic flow of a fractional Jeffrey model via porous medium,’’ Indian Journal of Physics, pp. 1-11, 2022. [25] P.P. Singh, G. Kumar, S. Competing Interest Statement peristaltic flow in a micropolar nanofluid: mathematical model with thermophysical features,” Scientific Reports, 12(1), 21540, 2022. [32] J. Wang, H. Deng, Z. Tao, Y. Li, J. Zhu, “Heat transfer in a rotating rectangular channel with impingement jet and film holes,’’ International Journal of Thermal Sciences, 163, 106832, 2021 [44] M.V. Krishna, B.V. Swarnalathamma, “Heat and mass transfer on the Peristaltic flow of an incompressible electrically conducting Williamson fluid through a porous medium in a symmetric channel with Hall current effects and inclined magnetic field,” International Journal of Scientific Engineering and Research, 6(9), pp. 941-957, 2019, [33] R. Vinoth, B. Sachuthananthan, “Flow and heat transfer behavior of hybrid nanofluid through microchannel with two different channels,’’ International Communications in Heat and Mass Transfer, 123, 105194, 2021. [44] H. Yasmin, N. Iqbal, A. Tanveer,“Engineering applications of peristaltic fluid flow with hall current, thermal deposition and convective conditions,” Mathematics, 8(10), 1710, 2020. [34] W. Abbas, N. Eldabe, S. Marzouk,” Soret and Dufour effects with Hall currents on peristaltic flow of Casson fluid with heat and mass transfer through non-Darcy porous medium inside vertical channel,’’ Egyptian Journal of Chemistry, 64(9), pp. 5215-5225, 2021. [45] Z. Abbas, M.Y. Rafiq, S. Khaliq, A. Ali, “Dynamics of the thermally radiative and chemically reactive flow of Sisko fluid in a tapered channel,” Advances in Mechanical Engineering, 14(10), 16878132221129735, 2022. [35] S. Nazir, M. Kashif, A. Zeeshan, H. Alsulami, M. Ghamkhar, “A study of heat and mass transfer of non- Newtonian fluid with surface chemical reaction, “Journal of the Indian Chemical Society, 99(5), 100434, 2022. [36] A. M. Abd-Alla, S. M. Abo-Dahab, E. N. Thabet, M.A. Abdelhafez, “Heat and mass transfer for MHD peristaltic flow in a micropolar nanofluid: mathematical model with thermophysical features,” Scientific Reports, 12(1), 21540, 2022. [37] F. Hussain, ‘’Heat and Mass Transfer Analysis in MHD Peristaltic Flow of Newtonian and Non-Newtonian Fluids bounded with Porous Tilted Channel,” 2022. [38] A.M. Abd-Alla, S.M. Abo-Dahab, E.N. Thabet, M.A. (Abdelhafez), “Peristaltic pump with heat and mass transfer of a fractional second grade fluid through porous medium inside a tube. Scientific reports,” 12(1), pp. 1-14, 2022. [39] A.M. Abd-Alla, S.M. Abo-Dahab, E.N. Thabet, M.A. (Abdelhafez), F.S. Bayones, “Heat and mass transfer in a peristaltic rotating frame Jeffrey fluid via porous medium with chemical reaction and wall properties,” Alexandria Engineering Journal, 2022. [40] A. Salmi, H.A. Madkhali, B. Ali, M. Nawaz, S.O. Alharbi, A.S. Competing Interest Statement Raghav, “Peristaltic Flow of Jeffrey Fluid through a slanted cylinder,’’ JOURNAL OF ALGEBRAIC STATISTICS, 13(2), pp. 3689-3695, 2022. [15] M.I. Khan, M. Lashin, N.B. Khedher, B. Ahmed, S.U. Khan, M. Oreijah, A.M. Galal, “Peristaltic Phenomenon in an Asymmetric Channel Subject to Inclined Magnetic Force and Porous Space,’’ Bioengineering, 9(10), 588, 2022. [26] S. Bajwa, S. Ullah, A.S. Al-Johani, I. Khan, M. Andualem, “Effects of MHD and Porosity on Jeffrey Fluid Flow with Wall Transpiration,’’ Mathematical Problems in Engineering, 2022 [16] A.M. Abd-Alla, R.A. Mohamed, S.M. Abo-Dahab, M. S. Soliman, “Rotation, initial stress effect on MHD peristaltic flow of reacting radiating fourth-grade nanofluid with viscous dissipation and Joule heating,’’ Waves in Random and Complex Media, pp. 1- 35, 2022. [27] M. Aleem, M.I. Asjad, A. Ahmadian, M. Salimi, M. Ferrara, “Heat transfer analysis of channel flow of MHD Jeffrey fluid subject to generalized boundary conditions,’’ The European physical journal plus, 135(1), pp. 1-15, 2020. [17] Z. Asghar, M.W. Saeed Khan, M.A. Gondal and A. Ghaffari,” Channel flow of non-Newtonian fluid due to peristalsis under external electric and magnetic field,’’ Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering, 09544089221097693, 2022. [28] Z. Hussain, K. Al-Khaled, U. Ashrif, A. Abbasi, S.U. Khan, W. Farooq, and M. Y. Malik, “A mathematical model for radiative peristaltic flow of Jeffrey fluid in curved channel with Joule heating and different walls: Shooting technique analysis,’’ Ain Shams Engineering Journal, 13(5), 101685, 2022 [18] A. Tanveer, S. Mahmood, T. Hayat, A. Alsaedi, “On electroosmosis in peristaltic activity of MHD non- Newtonian fluid,” Alexandria Engineering Journal, 60(3), pp. 3369-3377, 2021. [29] A.S. Kotnurkar, V.T. Talawar, “Influence of thermal jump and inclined magnetic field on peristaltic transport of Jeffrey fluid with silver nanoparticle in the eccentric annulus,’’ Heliyon, 8(9), e10543, 2022 [19] N. Ammar, H.A. Ali, “Mathematical modeling for peristaltic transport of non-Newtonian fluid through inclined non-uniform channel under the effect of surface [30] A.M. Abd-Alla, S.M. Abo-Dahab, M.A. Abdelhafez, E.N. Thabet, “Effects of heat transfer and the endoscope on Jeffrey fluid peristaltic flow in tubes,’’ Multidiscipline Modeling in Materials and Structures, 17(5), pp. 895-914, 2021. [31] H. Vaidya, C. Rajashekhar, B.B. Divya, G. Manjunatha, K.V. Prasad, I.L. Animasaun, “Influence of transport 45 Science, Engineering and Technology Vol. 3, No. 2, pp. 36-46 properties on the peristaltic MHD Jeffrey fluid flow through a porous asymmetric tapered channel,” Results in Physics, 18, 103295, 2020. Competing Interest Statement Alqahtani, “Numerical study of heat and mass transfer enhancement in Prandtl fluid MHD flow using Cattaneo-Christov heat flux theory,” Case Studies in Thermal Engineering, 33, 101949, 2022. [41] A. Tanveer, S. Mahmood, T. Hayat, A. Alsaedi, “On electroosmosis in peristaltic activity of MHD non- Newtonian fluid,” Alexandria Engineering Journal, 60(3), pp. 3369-3377, 2021. [42] F.S. Bayones, A. M. Abd-Alla, E. N. Thabet, “Effect of heat and mass transfer and magnetic field on peristaltic flow of a fractional Maxwell fluid in a tube,” Complexity, pp. 1-12, 2021. [43] A. M. Abd-Alla, S.M. Abo-Dahab, E.N. Thabet, M.A. Abdelhafez, “Heat and mass transfer for MHD 46
https://openalex.org/W4319655392	https://zenodo.org/records/7143056/files/348-Prandi-2023.pdf	English	null	Computer-assisted simultaneous interpreting: A cognitive-experimental study on terminology	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	124,500	Translation and Multilingual Natural Language Processing Editors: Oliver Czulo (Universität Leipzig), Silvia Hansen-Schirra (Johannes Gutenberg-Universität Mainz), Reinhard Rapp (Hochschule Magdeburg-Stendal), Mario Bisiada (Universitat Pompeu Fabra) Editors: Oliver Czulo (Universität Leipzig), Silvia Hansen-Schirra (Johannes Gutenberg-Universität Mainz), Reinhard Rapp (Hochschule Magdeburg-Stendal), Mario Bisiada (Universitat Pompeu Fabra) Editors: Oliver Czulo (Universität Leipzig), Silvia Hansen-Schirra (Johannes Gutenberg-Universität Mainz), Reinhard Rapp (Hochschule Magdeburg-Stendal), Mario Bisiada (Universitat Pompeu Fabra) In this series (see the complete series history at https://langsci-press.org/catalog/series/tmnlp) this series (see the complete series history at https://langsci-press.org/catalog/series/tmnlp): 11. Fantinuoli, Claudio (ed.). Interpreting and technology. 11. Fantinuoli, Claudio (ed.). Interpreting and technology. 12. Nitzke, Jean. Problem solving activities in post-editing and translation from scratch: A multi-method study. 12. Nitzke, Jean. Problem solving activities in post-editing and translation from scratch: A multi-method study. 13. Vandevoorde, Lore. Semantic differences in translation. 13. Vandevoorde, Lore. Semantic differences in translation. 14. Bisiada, Mario (ed.). Empirical studies in translation and discourse. 15. Tra&Co Group (ed.). Translation, interpreting, cognition: The way out of the box. 16. Nitzke, Jean & Silvia Hansen-Schirra. A short guide to post-editing. 17. Hoberg, Felix. Informationsintegration in mehrsprachigen Textchats: Der Skype Translator im Sprachenpaar Katalanisch-Deutsch. 18. Kenny, Dorothy (ed.). Machine translation for everyone: Empowering users in the age of artificial intelligence. 19. Kajzer-Wietrzny, Marta, Adriano Ferraresi, Ilmari Ivaska & Silvia Bernardini. Mediated discourse at the European Parliament: Empirical investigations. 20. Marzouk, Shaimaa. Sprachkontrolle im Spiegel der Maschinellen Übersetzung: Untersuchung zur Wechselwirkung ausgewählter Regeln der Kontrollierten Sprache mit verschiedenen Ansätzen der Maschinellen Übersetzung. 21. Frittella, Francesca Maria. Usability research for interpreter-centred technology: The case study of SmarTerp. 22. Prandi, Bianca. Computer-assisted simultaneous interpreting: A cognitive-experimental study on terminology. ISSN: 2364-8899 Acknowledgements Abbreviations Introduction Abbreviations vii Introduction 1 1 Terminology work in interpreting 7 1.1 Translating and interpreting for LSP . . . . . . . . . . . . . . . . 7 1.1.1 Features of LSP . . . . . . . . . . . . . . . . . . . . . . . 8 1.1.2 Terminology as a parameter of quality . . . . . . . . . . 9 1.2 Terminology work in translation and interpreting . . . . . . . . 13 1.2.1 Terminology and translation . . . . . . . . . . . . . . . . 13 1.2.2 Terminology work in translation . . . . . . . . . . . . . 14 1.2.3 Terminology work in interpreting . . . . . . . . . . . . . 16 1.3 Terminology tools for conference interpreters . . . . . . . . . . 21 1.3.1 Requirements for a support tool for conference interpreters . . . . . . . . . . . . . . . . . . . . . . . . . 21 1.3.2 Usefulness and limitations of non-interpreter-specific tools 22 2 Information technologies and interpreting 25 2.1 Overview and classification of technologies . . . . . . . . . . . . 25 2.1.1 Technologies mediating interpreting . . . . . . . . . . . 26 2.1.2 Technologies generating interpreting . . . . . . . . . . . 28 2.1.3 Supporting technologies for interpreting . . . . . . . . . 31 2.2 CAI tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.2.1 Definition . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.2.2 Overview of CAI tools . . . . . . . . . . . . . . . . . . . 35 2.2.3 InterpretBank . . . . . . . . . . . . . . . . . . . . . . . A cognitive-experimental study on terminology A cognitive-experimental study on terminology Bianca Prandi language science press Bianca Prandi. 2023. Computer-assisted simultaneous interpreting: A cognitive-experimental study on terminology (Translation and Multilingual Natural Language Processing 22). Berlin: Language Science Press. Bianca Prandi. 2023. Computer-assisted simultaneous interpreting: A cognitive-experimental study on terminology (Translation and Multilingual Natural Language Processing 22). Berlin: Language Science Press. This title can be downloaded at: http://langsci-press.org/catalog/book/348 © 2023, Bianca Prandi Published under the Creative Commons Attribution 4.0 Licence (CC BY 4.0): http://creativecommons.org/licenses/by/4.0/ ISBN: 978-3-96110-397-3 (Digital) 978-3-98554-055-6 (Hardcover) ISSN: 2364-8899 DOI: 10.5281/zenodo.7143056 Source code available from www.github.com/langsci/348 Errata: paperhive.org/documents/remote?type=langsci&id=348 Cover and concept of design: Ulrike Harbort Proofreading: Amir Ghorbanpour, Andreas Hölzl, Christopher Straughn, Elen Le Foll, Jean Nitzke, Jeroen van de Weijer, Marten Stelling, Neneng Sri, Robin Lemke Fonts: Libertinus, Arimo, DejaVu Sans Mono Typesetting software: XƎLATEX Language Science Press xHain Grünberger Str. 16 10243 Berlin, Germany http://langsci-press.org Storage and cataloguing done by FU Berlin This title can be downloaded at: http://langsci-press.org/catalog/book/348 © 2023, Bianca Prandi Published under the Creative Commons Attribution 4.0 Licence (CC BY 4.0): http://creativecommons.org/licenses/by/4.0/ ISBN: 978-3-96110-397-3 (Digital) 978-3-98554-055-6 (Hardcover) This title can be downloaded at: http://langsci-press.org/catalog/book/348 © 2023, Bianca Prandi Published under the Creative Commons Attribution 4.0 Licence (CC BY 4.0): http://creativecommons.org/licenses/by/4.0/ ISBN: 978-3-96110-397-3 (Digital) 978-3-98554-055-6 (Hardcover) ISSN: 2364-8899 DOI: 10.5281/zenodo.7143056 Source code available from www.github.com/langsci/348 Errata: paperhive.org/documents/remote?type=langsci&id= ISSN: 2364-8899 DOI: 10.5281/zenodo.7143056 Source code available from www.github.com/langsci/348 Errata: paperhive.org/documents/remote?type=langsci&id=34 Cover and concept of design: Ulrike Harbort Proofreading: Amir Ghorbanpour, Andreas Hölzl, Christopher Straughn, Elen Le Foll, Jean Nitzke, Jeroen van de Weijer, Marten Stelling, Neneng Sri, Robin Lemke Fonts: Libertinus, Arimo, DejaVu Sans Mono Typesetting software XƎLATEX Storage and cataloguing done by FU Berlin Acknowledgements Abbreviations Introduction . 38 2.3 Conference interpreters’ use of supporting technologies . . . . . 42 2.3.1 Conference interpreters’ level of computerisation . . . . 42 2.3.2 Computer and software use for assignment preparation 43 2.3.3 Computer and software use peri- and in-process . . . . 46 1 Terminology work in interpreting Contents 2.3.4 Attitudes towards CAI tools . . . . . . . . . . . . . . . . 48 2.4 Research on CAI tools . . . . . . . . . . . . . . . . . . . . . . . . 51 3 Simultaneous interpreting as a complex cognitive activity 63 3.1 Working memory and information processing . . . . . . . . . . 64 3.2 The role of attention in processing visual stimuli . . . . . . . . . 69 3.3 Attention sharing and task interference . . . . . . . . . . . . . . 70 3.4 The construct of cognitive load . . . . . . . . . . . . . . . . . . . 74 3.4.1 Cognitive load in interpreter-computer interaction . . . 75 3.4.2 Attention splitting and audio-visual integration . . . . . 77 3.4.3 Cognitive load as time-based resource sharing . . . . . . 79 3.5 Cognitive implications in interpreting . . . . . . . . . . . . . . . 80 3.5.1 Gile’s Effort Model of simultaneous interpreting . . . . . 82 3.5.2 Seeber’s Cognitive Load Model of Simultaneous Interpreting . . . . . . . . . . . . . . . . . . . . . . . . . 87 3.5.3 Discussion and choice of a model for the present work . 95 3.6 A model of SI with digital terminology support . . . . . . . . . . 97 3.6.1 Application of Seeber’s CLM to SI with terminological support . . . . . . . . . . . . . . . . . . . . . . . . . . . 98 3.6.2 Manual terminology look-up . . . . . . . . . . . . . . . 99 3.6.3 Automatic terminology look-up . . . Acknowledgements Abbreviations Introduction . . . . . . . . . . . 102 4 Measuring cognitive load 105 4.1 Theoretical approaches . . . . . . . . . . . . . . . . . . . . . . . 106 4.2 Subjective approaches . . . . . . . . . . . . . . . . . . . . . . . . 106 4.3 Performance measures . . . . . . . . . . . . . . . . . . . . . . . 108 4.3.1 Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . 108 4.3.2 Dysfluencies and pauses . . . . . . . . . . . . . . . . . . 109 4.4 Behavioural approaches . . . . . . . . . . . . . . . . . . . . . . . 111 4.4.1 Time-related measures . . . . . . . . . . . . . . . . . . . 112 4.4.2 Eye movement measures and eyetracking . . . . . . . . 115 4.5 Physiological measures . . . . . . . . . . . . . . . . . . . . . . . 121 4.5.1 Brain measures . . . . . . . . . . . . . . . . . . . . . . . 122 4.5.2 Heart rate measures . . . . . . . . . . . . . . . . . . . . 123 4.5.3 Pupillary measures . . . . . . . . . . . . . . . . . . . . . 124 5 Method 127 5.1 Research gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 5.2 Hypotheses on cognitive load in SI with terminology look-up . 128 ii Contents Contents 5.3 Research approach . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements Abbreviations Introduction 129 5.4 Pilot study: Method and results . . . . . . . . . . . . . . . . . . . 131 5.4.1 Participants . . . . . . . . . . . . . . . . . . . . . . . . . 132 5.4.2 Test subjects’ training . . . . . . . . . . . . . . . . . . . 132 5.4.3 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . 133 5.4.4 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 139 5.4.5 Stimulus validation . . . . . . . . . . . . . . . . . . . . . 141 5.4.6 Preliminary results . . . . . . . . . . . . . . . . . . . . . 142 5.4.7 Discussion of pilot study results . . . . . . . . . . . . . . 145 5.5 Main study: Rationale and method . . . . . . . . . . . . . . . . . 147 5.5.1 Participants . . . . . . . . . . . . . . . . . . . . . . . . . 147 5.5.2 Participants’ training . . . . . . . . . . . . . . . . . . . . 154 5.5.3 Materials . . . . . . . . . . . . . . . . . . . . . . . . . . 155 5.5.4 Apparatus used for data collection . . . . . . . . . . . . 158 5.5.5 Measures . . . . . . . . . . . . . . . . . . . . . . . . . . 158 5.5.6 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 164 5.5.7 Data preparation . . . . . . . Acknowledgements Abbreviations Introduction . . . . . . . . . . . . . . . 168 6 Results and discussion 173 6.1 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 6.1.1 Performance measures: Analysis . . . . . . . . . . . . . 173 6.1.2 Behavioural measures: Analysis . . . . . . . . . . . . . . 186 6.1.3 Subjective measure: The debriefing questionnaire . . . . 203 6.2 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 6.2.1 Performance measures . . . . . . . . . . . . . . . . . . . 208 6.2.2 Behavioural measures . . . . . . . . . . . . . . . . . . . 215 6.2.3 Subjective measure: The debriefing questionnaire . . . . 225 6.2.4 Validation of the CLM applied to computer-assisted SI . 229 6.3 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 6.3.1 Participants and choice of experimental design . . . . . 232 6.3.2 Use of simulated ASR . . . . . . . . . . . . . . . . . . . . 236 6.3.3 Priming . . . . . . . . . . . . . . . . . . . . . . . . . . . 236 6.3.4 Eyetracking methodology . . . . . . . . . . . . . . . . . 237 7 Conclusion 239 7.1 Methodological implications . . . . . . . . . . . . . . . . . . . . 240 7.2 Didactic implications . . . . . . . . . . . . . . . . . . . . . . . . 242 7.3 Practical implications . . . . . . . . . . . . . . Acknowledgements Abbreviations Introduction . . . . . . . . . . 244 7.4 Future work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 iii Contents Appendix A: Speech A 247 Appendix B: Speech B 253 Appendix C: Speech C 259 References 265 Index 305 Name index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305 iv iv Acknowledgements For her invaluable advice, patience and feedback, I would like to express my deepest gratitude to my supervisor, Univ-Prof. Dr. Silvia Hansen-Schirra. This endeavour would not have been possible without the students who par- ticipated in the experiments, my speaker, and the helping hand of the Tra&Co research team. I would like to extend my sincere thanks to my mentor, Dr. Claudio Fantinuoli, for inspiring and supporting me throughout my research journey. Finally, my heartfelt thanks goes to my parents Anna and Claudio, my sister Francesca, my partner Marek and all my friends, in particular Francesca and Elisa, for their unwavering support over the years. Abbreviations AI Artificial Intelligence AIIC International Association of Conference Interpreters ANOVA Analysis of Variance AOI Area of Interest API Application Programming Interface APR Average Pause Ration ASR Automatic Speech Recognition ATE Automatic Term Extraction BLEU Bilingual Evaluation Understudy CACI Canadian Association of Conference Interpreters CAI Computer-Assisted Interpreting CAIT Computer-Assisted Interpreter Training CASI Computer-Assisted Simultaneous Interpreting CAT Computer-Assisted Translation CDIP Corpus-Driven Interpreter Preparation CIUTI International Standing Conference of University Institutes of Translators and Interpreters CE Cognitive Effort CL Cognitive Load CLM Cognitive Load Model CLT Cognitive Load Theory CM Conflict Matrix COVID-19 Coronavirus Disease 2019 CRF Cognitive Resource Footprint CTS Cognitive Translation Studies DOT Dolmetschorientierte Terminologiearbeit EABM Ergonomics for the Artificial Boothmate EB Episodic Buffer EEG Electroencephalography EKS Eye-Key Span EM Effort Model ERP Event-Related Potential Abbreviations EU European Union EVS Ear-Voice Span fMRI Functional Magnetic Resonance Imaging HIAT Halbinterpretative Arbeitstranskriptionen (Semi-Interpretative Working Transcriptions) HMI Human-Machine Interaction HTER Human-targeted Translation Edit Rate IATE Interactive Terminology for Europe ICL Intrinsic Cognitive Load ICPD Inter-Cluster Pause Duration ICT Information and Communication Technologies ILO International Labor Organization IS Interpreting Studies KIT Karlsruhe Institute of Technology LSP Language for Special Purpose LTM Long-Term Memory LTS Long-Term Store MEG Magnetoencephalography MI Machine Interpreting MQM Multidimensional Quality Metrics MS Microsoft MT Machine Translation MTPE Machine Translation Post-Editing NASA-TLX NASA Task-Load Index NIR Near-Infrared Spectroscopy NLP Natural Language Processing NMT Neural Machine Translation PET Positron Emission Tomography PL Phonological Loop PWR Pause to Word Ratio RSI Remote Simultaneous Interpreting RTF Real-Time Factor RWL Reading while Listening SD Standard Deviation SDI Sprachen & Dolmetscher Institut SI Simultaneous Interpreting SIMTXT Simultaneous Interpreting with Text SR Speech Recognition ST Sight Translation viii ST Source Text STM Short-Term Memory STS Short-Term Storage STT Speech to Text SYNLE Syntactic Left Embeddedness TAP Think-Aloud Protocol TAUS-DQF TAUS Dynamic Quality Framework TE Terminology Extraction TET Terminology Extraction Tool TICQ Translation and Interpreting Competence Questionnaire TIS Total Interference Score TIS Translation and Interpreting Studies TKC Terminological Knowledge Constellation TKE Terminological Knowledge Entity TM Translation Memory TMS Terminology Management System TPR Translation Process Research TT Target Text UI User Interface UN United Nations VR Virtual Reality VSSP Visual-Spatial Sketch Pad WER Word Error Rate WM Working Memory WPM Words per Minute WPT Word Production Time ix ix ix Introduction When conference interpreters interpret a speech simultaneously, they are often faced with the need to quickly and precisely render specialised terminology in the target language. Hence, pre-assignment preparation is fundamental to ac- quire specialised terminology to aptly express domain-specific knowledge in the target language. In the preparation phase, interpreters usually compile glossaries containing specialised terms likely to be used by the speaker (e.g. Rütten 2007, Fantinuoli 2017a, Will 2007, Gile 2009). Despite learning the terminological equiv- alents ahead of the event, interpreters may not always be able to retrieve the target-language equivalent from memory during interpreting. To cope with this difficulty, among other tactics (Gile 2009: 14), they may choose to look up the required term in their glossaries. Traditionally, specialised glossaries have been compiled on paper, or prepared in digital format and printed out for the booth (e.g. Jiang 2013, 2015). With the increasing permeation of the profession by technology over the past couple of decades, the booth is now increasingly paper-less (Rütten 2017). On a laptop or a tablet, interpreters can now conduct glossary queries in the digital medium. As for the software employed to create digital glossaries for interpreting as- signments, traditionally this has consisted in text processing programmes or database applications aimed at the general public. For lack of dedicated tools, some interpreters have resorted to computer-assisted translation (CAT) tools. The use of tools for corpus-based terminology work, terminology extraction, and of speech recognition, to name but a few examples, is therefore not the exclusive preserve of translators and terminologists but can also be found in interpreters’ terminology work. Despite the potential usefulness of these technologies for interpreters, scholar- ship remarked that the nature of interpreting imposes specific demands, both cog- nitive and related to interpreters’ workflow (e.g. Rütten 2004, Will 2000, 2007). Thus, around the same time when CAT tools started to appear, applications geared towards the specific needs of interpreters were created and later increas- ingly refined which fall under the name of computer-assisted interpreting (CAI) Introduction tools.1 Their aim is to support interpreters along several phases of their work- flow, especially during preparation, but also for terminology retrieval during the interpreting task. The recent advances in automatic speech recognition (ASR) technology have motivated its integration into CAI tools, which may now of- fer live support for terminology and other units of information without physical interaction between the interpreter and the machine. 1For a terminological clarification of the term “CAI tool” and its use in the present work, see §2.2.1. 2Here intended in the broader sense of the term and referring not only to the conference setting. 3Pöchhacker derives the notion of memes and supermemes from Chesterman (2016): as in trans- lation, these “socio-biological concepts” have arisen as metaphors to illustrate particular views of interpreting as an object of study. From this perspective, the interpreter is seen as a “human processor” performing several “cognitive skills […] the combination of which would account for the complex task of interpreting” (Pöchhacker 2004: 53). Introduction The first prototypes of ASR- enhanced CAI tools are already emerging (e.g. Fantinuoli 2017a). y g g ( g ) In light of these developments, several studies have been conducted on the topic of CAI tools over the past few years. They have mainly explored the tools’ potential to improve terminological accuracy during simultaneous interpreting (e.g. Prandi 2015a,b), the extent to which ASR improves the rendition of number words and specialized terms in the target language (e.g. Defrancq & Fantinuoli 2021), or their potential for offering support to interpreters in the consecutive mode (e.g. Wang & Wang 2019). With a few exceptions (Biagini 2015, Frittella 2022), the focus of such studies has been rather narrow, using specific perfor- mance indicators such as the accuracy of interpreted terms and numerals to as- sess the tools’ impact on the overall quality of the interpretation (e.g. Pisani & Fantinuoli 2021), without taking stock of the interpretation beyond these individ- ual items. Many findings have emerged from small-scale experiments conducted on students in the framework of master’s theses (e.g. Canali 2018, Van Cauwen- berghe 2020). Despite the emphasis on the postulated difficulty of integrating CAI tools into the interpreting process and, on the other hand, the widespread enthusiasm for the potential of ASR to alleviate cognitive load during SI, the impact of such solutions on the cognitive subprocesses underlying SI has so far remained largely unexplored. This represents an evident lacuna compared to the large body of research conducted on cognition in the translator-machine inter- action. Such interaction has been addressed by numerous empirical studies in the area of Translation Process Research (TPR) from multiple perspectives and with a variety of methods (Tardel 2021: xvii). Indeed, as “a research tradition within cognitive translation studies (CTS) [...] exploring factors that determine human translation behavior” (ibid.), TPR may constitute a valuable reference point and provide useful tools to the analysis of computer-assisted simultaneous interpreting (CASI), especially from a methodological standpoint. However, no empirically-validated methodology for the combined collection of product- and process-oriented data with a markedly cognitive focus has yet been developed to explore the phenomenon of technology-supported SI. 2 The present doctoral thesis seeks to address this limitation by developing and testing an empirical methodology for a cognitive exploration of CASI. Introduction In par- ticular, the present work derives its methods from TPR to analyse the impact on cognitive load of different forms of digital terminological support for interpreters through a within-subject experimental study. To the best of my knowledge, this is the first study to address computer-assist- ed SI from a cognitive perspective. As such, it presents an exploratory character which aims to provide first findings and, at the same time, to identify open ques- tions and formulate hypotheses for further investigations of the phenomenon. The following section establishes the paradigm for the present research work and provides a conceptual framework for the cognitive inquiry into the phe- nomenon of CASI. The rest of the present chapter illustrates how the present work is organised and briefly describes the content of each chapter. Choice of paradigm for the present study Especially in the simultaneous mode, interpreting has often been described as a complex cognitive activity, involving concurrent information processing and temporary storage tasks competing for attentional resources. This view of inter- preting2 as cognitive information processing represents one of the “supermemes” of interpreting, as observed by Pöchhacker (2004: 51)3.i Yet, even though this “internal” perspective has been particularly prolific in the academic inquiry into the phenomenon, interpreting may also be viewed as a socially embedded human activity, situated in a real communicative context. The supermeme of communicative activity elucidates interpreting as a combina- tion of listening and speaking aimed at facilitating communication beyond the linguistic barrier. For the scope of the present study, I chose to conduct my inquiry into inter- preting from an explicitly cognitive perspective. Selecting one perspective does not, however, mean discarding or denying the other, but rather focusing the spot- light on one aspect of this multi-faceted activity, and is necessary to establish the theoretical framework guiding investigation. The next step lies into the defini- 3 3 Introduction tion of my research paradigm4. The two supermemes of interpreting as cognitive information processing and as communicative activity are at the core of a num- ber of paradigms that can be divided into social, psycholinguistic, and cognitive approaches following Setton’s classification (Setton 2003). The social approach reflects a tradition which looks at interpreting within a broader framework including social and behavioural factors. Prominent issues deriving from this view of interpreting are the interaction between the actors involved in the communicative event, the role of the interpreter and the long- standing issue of neutrality, the view of quality assessment as a question of pragmatics (and not only of identity between the source text and the interpreted message), but also the issue of interpreting strategies. To this category may be ascribe the ”target-text oriented, translation-theoretical” (Pöchhacker 2004: 77) paradigm exemplified by Salevsky (1987), Schjoldager (1995), Pöchhacker (1994) and Kalina (1998), as well as the “dialogic discourse-based interaction” paradigm (Pöchhacker 2004: 79) of which Roy (1996, 2000) and Wadensjö (1993, 1998) are the most prominent representatives. Psycholinguistic approaches are grounded in theories of communication and focus on features of discourse rather than on the cognitive processing of inter- preting. This perspective is at the core of the interpretive paradigm pioneered by Seleskovitch (1976) and Lederer (1981) with their théorie du sens. 4See Pöchhacker (2004, 2016) for a detailed account of research paradigms in Interpreting Stud- ies. Choice of paradigm for the present study The models by Chernov (1979, 1994), Déjean Le Féal (1980, 1981), Donovan (1990), Laplace (1994) and Setton (1999) himself may also be included in this category. Finally, interpreting has been studied from a cognitive perspective and viewed as a matter of information processing. The scholars belonging to this tradition focus on the exploration of the cognitive underpinnings of interpreting. It is not by chance that they derive their methods of investigation from the cognitive sciences. This third approach has generated the cognitive processing paradigm (Pöchhacker 2004: 73) initiated by Gerver (1976) and further exemplified by Lam- bert (1988), Massaro (1978), Moser (1978) and Moser-Mercer (1997), Kurz (1996), Shlesinger (2000) and Gile (1988, 1997, 1999) and Seeber (2007, 2011, 2017). To this perspective one can also ascribe the “neurophysiological/neurolinguistic para- digm” (Pöchhacker 2004: 75) exemplified by Fabbro & Gran (1994), Kurz (1994, 1996), Darò (1994, 1997), Petsche et al. (1993), Rinne et al. (2000), and Tommola & Hyönä (1990). The focus of the present doctoral thesis lies on the way the use of digital ter- minological support tools during simultaneous interpreting affects the cognitive 4 4 processes involved in simultaneous interpreting. Hence, I situate my inquiry in the cognitive processing paradigm, while adopting an interdisciplinary approach to the exploration of simultaneous interpreting with digital terminology support. Methodologically, the study draws heavily on research methods developed and validated in the framework of empirical Translation Process Research (TPR). As translation and interpreting are rather similar activities in terms of their under- lying cognitive processes and cognitive control functions, the approaches devel- oped in TPR for the exploration of the translation process are expected to provide a valuable methodological reference point. Organisation The present thesis is organised as follows. Chapter 1 introduces the topic of ter- minology in interpreting. Specifically, it addresses terminology work in confer- ence interpreting highlighting its commonalities and differences with terminol- ogy work in translation. Additionally, it underlines the role of terminology as an important quality factor in translation and interpreting, which motivates the fo- cus on terminological support in the present study. The last section closes Chap- ter 1 with a discussion of the requirements for CAI tools and of the potential and limitations of non-bespoke terminology tools for interpreters. Against this back- ground, Chapter 2 frames the technology object of inquiry, i.e. CAI tools, within the larger framework of technology applied to interpreting. After an overview of the available technologies for interpreting (§2.1), CAI tools are discussed in detail (§2.2.2), and InterpretBank, the CAI tool chosen for the experiment, in par- ticular (§2.2.3). The chapter offers a review of interpreters’ practices in compiling terminological resources ahead of and during the interpreting assignment, specif- ically in terms of their level of computerisation and choice of tools. This section closes with a review of how CAI tools have been studied in interpreting research thus far (§2.4) and illustrates current attitudes towards CAI tools (§2.3.4), which motivates the present work. Hence, in Chapter 3 I proceed to discuss simulta- neous interpreting as a complex cognitive activity, specifically as a question of attention allocation and resource sharing between co-occurring subtasks (§3.1), a key issue in the inquiry into technology-supported (simultaneous) interpret- ing. §3.4 illustrates cognitive load as a fundamental construct often encountered in academic discourse around CAI tools, but not yet explored experimentally. These two sections pave the way for the discussion of interpreting as an issue of multi-tasking within the area of Interpreting Studies (§3.5). Here, I illustrate in detail two models of SI which address this activity from the perspective of the 5 5 Introduction concurrent performance and coordination of cognitive sub-tasks. On this basis, I motivate the choice of framework to operationalise hypotheses on SI with dig- ital terminological support (§3.5.3). Chapter 4 describes the methods adopted in TPR and neighbouring disciplines (e.g. cognitive psychology) to measure cogni- tive load. Chapter 5 presents and discusses the research approach (§5.3) and the methodology deployed to test the hypotheses formulated for the present study (§5.2). 1 Terminology work in interpreting The present chapter discusses the role of terminology in simultaneous interpret- ing at specialised conferences, highlighting the commonalities with translation while at the same time foregrounding the specific requirements of terminology work in interpreting. Against this background, the ideal features of a bespoke tool for terminology work in the different phases of interpreting are discussed. To conclude, the chapter highlights the limitations of terminology tools available to translators, thus motivating the need for interpreter-specific tools, which are discussed in the following chapter. Organisation §5.4 presents the methods and results of the pilot study conducted to test the research methodology and validate the stimuli to be used for data collection. §5.5 goes into the details of the experimental design adopted in the main study, describing the adaptations conducted in light of the results of the pilot test. In Chapter 6, the results of the experiment are presented and discussed against the background of the hypotheses formulated in §5.2 and of relevant publications in the area of TPR and CAI research. §6.2.4 discusses and validates the application of Seeber (2007, 2011, 2017) CLM of SI to illustrate task interference and cognitive load in SI with the support of traditional digital glossaries, CAI tools with manual look-up, and ASR-enhanced CAI tools. The limitations of the present study are addressed in §6.3. Chapter 7 presents the methodological, didactic and practical implications of the present study and concludes this work with final remarks on potential avenues for future CAI research. concurrent performance and coordination of cognitive sub-tasks. On this basis, I motivate the choice of framework to operationalise hypotheses on SI with dig- ital terminological support (§3.5.3). Chapter 4 describes the methods adopted in TPR and neighbouring disciplines (e.g. cognitive psychology) to measure cogni- tive load. Chapter 5 presents and discusses the research approach (§5.3) and the methodology deployed to test the hypotheses formulated for the present study (§5.2). §5.4 presents the methods and results of the pilot study conducted to test the research methodology and validate the stimuli to be used for data collection. §5.5 goes into the details of the experimental design adopted in the main study, describing the adaptations conducted in light of the results of the pilot test. In Chapter 6, the results of the experiment are presented and discussed against the background of the hypotheses formulated in §5.2 and of relevant publications in the area of TPR and CAI research. §6.2.4 discusses and validates the application of Seeber (2007, 2011, 2017) CLM of SI to illustrate task interference and cognitive load in SI with the support of traditional digital glossaries, CAI tools with manual look-up, and ASR-enhanced CAI tools. The limitations of the present study are addressed in §6.3. Chapter 7 presents the methodological, didactic and practical implications of the present study and concludes this work with final remarks on potential avenues for future CAI research. 6 6 1.1 Translating and interpreting for LSP Translation and interpreting share a large set of common features, as both activ- ities concern the interlingual and intercultural transmission of a message. The most apparent element of distinction is perhaps the channel through which the target text is produced: written, in the case of translation, and oral or visual, in the case of interpreting and signed language interpreting, respectively. This distinction is usually adopted for ease of explanation, for instance in communi- cation with laypeople. However, translation and interpreting are in numerous respects quite similar. For instance, hybrid forms of translation and interpreting are possible: one such instance is sight translation (ST), which consists in the spoken translation of a written source text. With the advent of new possibilities offered by technology, additional hybrid forms of translation and interpretation are emerging, such as live subtitling (or interlingual respeaking), in which the interpreter produces a live translation of the speaker’s words that is, however, rendered in written form. Second, TPR has elucidated that cognitive control func- tions are very similar in translation and interpreting. Both translation and inter- preting involve the performance of parallel sub-tasks (e.g. Vardaro et al. 2019) and the first draft of a translation does not require significantly more time to be produced than a target text (TT) in interpreting. Translation and interpreting also share another characteristic: a large amount of texts translated or interpreted are specialised texts. The profile of translators 1 Terminology work in interpreting is thus increasingly becoming that of a specialised translator (Campo 2005); spe- cialised conferences, technical meetings, seminars or workshops represent the most common type of assignment for professional conference interpreters, both for those working in the private market and those employed by public institu- tions (e.g. Rütten 2008: 22). When translating or interpreting in specialised fields, translators and inter- preters face a common challenge: they are called upon to enable expert commu- nication across languages despite not being subject-matter experts themselves. This is particularly true for interpreters, who often work “in very different the- matic scenarios, usually for a public of experts, and are given the task of trans- mitting highly specialised knowledge” (Rodríguez & Schnell 2009: 21). Despite the increasing expertise acquired by translators and interpreters during their ca- reers, specialised translation and interpretation in technical settings remains a complex task for both groups of professionals. 1.1 Translating and interpreting for LSP This complexity is related to the demands imposed on translators and interpreters by specialised language (see §1.1.1) and to the expectations of expert readers or listeners (see §1.1.2). 1.1 Translating and interpreting for LSP 1.1 Translating and interpreting for LSP This type of influence combined with similar etymology, especially for certain language pairs, may favour the use of cognates, i.e. “those translation words that have similar orthographic-phonological forms” (Costa et al. 2000: 1285), in the translation of specialised discourse. On a cognitive level, cognates have been found to be easier to process. This is referred to as the cognate facilitation effect (Costa et al. 2000). The use of terms with etymological roots shared by different languages may therefore to some extent facilitate access to scientific discourse. i Another feature of LSP is a certain preference for nominalisation and espe- cially for complex noun phrases or compounds (Olohan 2008: 247), especially in English and other Germanic languages. These linguistic devices are functional to the economy of expression which is essential to specialised language (Pig- nataro 2012: 135). With reference to Halliday & Martin (2003), Olohan (2013: 428) remarks in this respect that: This shift from verb to noun, they argue, is significant because the meaning construed by the nominalization is a new one – an abstract theoretical en- tity which forms part of a scientific theory. These and other features, like expanded nominal groups, privilege experts and exclude others from access- ing scientific discourse. In translation and interpreting, these linguistic features of LSP pose an addi- tional layer of difficulty as elliptical, multi-word expressions must be re-coded in the target language. This may prove particularly problematic if the target- language equivalent is not readily available (see §1.2.3.1).i Finally, LSP is characterised by specific phraseology and syntax (Olohan 2013: 426), as well as specialised terminology, which translators and interpreters must use “as a means to achieve the interlinguistic transfer of specialized knowledge units” (Velásquez 2002: 447). 1.1.1 Features of LSP As specialised communication is considered primarily informative (Olohan 2013: 427), the focus is on the content, which must be conveyed with the highest ac- curacy possible. A key characteristic of specialised communication is thus the use of languages for special purposes (LSP, Desblache 2001, Scarpa 2010), the in- house jargon (Kalina 2006) adopted by experts in a specific subject matter to con- vey specialised knowledge to other experts or laypeople. LSP thus “has a precise purpose and function, that is, to specifically communicate concrete information clearly to a target set of users so that it can be productively used” (Folaron 2019: 207). LSP can therefore be seen as the way specialised knowledge manifests itself and is constituted in discourse. As compared to ordinary language, LSP presents several specific features concerning its lexicon as well as its phraseology and syn- tax. LSPs may therefore be described as “contextual-functional varieties of the or- dinary language (Garzone 2006), […] characterized by specific morpho-syntactic forms and by some discursive and pragmatic features” (Pignataro 2012: 134).i Especially in scientific contexts, one such characteristic is the widespread use of English as a lingua franca (Ammon 2011). As remarked for instance by Hansen- Schirra et al. (2017), this might result in shining-through effects in translation, es- pecially if the language is particularly susceptible to Anglicism (as is the case for German). Additionally, Braun et al. (2014) evidence how academic texts present a high number of Latinisms and Grecisms in addition to Anglicisms. 8 1.1 Translating and interpreting for LSP 1.1.2 Terminology as a parameter of quality In light of the status of terminology for effective specialised communication, it is not surprising that terminology is acknowledged as an important parameter of quality in translation and interpreting. It may be affirmed that “the adequacy of the terminology in a text as well as its suitability for the level of specialization determines to a great extent the quality of a translation” (Martínez & Faber 2009: 91). While terminology is not the only benchmark against which the quality of a translation may be assessed, its relevance as an important element of qual- ity emerges for instance through its inclusion in many evaluation frameworks, 9 9 1 Terminology work in interpreting both for human and for machine translation (MT). For instance, Mertin (2006) developed a typology of translation errors which includes terminology as one of the categories of the evaluation matrix. The Multidimensional Quality Metrics framework (MQM, see Lommel, Uszkoreit, et al. 2014, Lommel, Burchardt, et al. 2014) for the assessment of translation quality includes terminology as one of the higher-level issue types. Terminology is one of the 20 “core” issue types in- cluded in MQM Core, a simplified version of MQM, and it is included as a key issue type in the TAUS DQF Error Typology (Görög 2014a,b), a subset of MQM. The SAE J2450 Translation Quality Metric for Language Translation of Service Information (SAE International 2001) also includes the issue type “wrong term”, which can be mapped to the “terminology” issue type in MQM. Similarly, Hjer- son (Popovic 2011: 59), a framework for automatic classification of errors in ma- chine translation output, includes “incorrect lexical choice” as one of the error classes. These are but a few examples, but they highlight the role attributed to terminology for the assessment of translation quality. By the same token, especially in the context of interpreting settings where LSP is used, e.g. specialised conferences and technical meetings or workshops, termi- nology emerges as an important element of quality, although it is but one param- eter (García de Quesada 2011: 231). The weight attributed to terminology for the evaluation of interpreting quality may also be said to reflect the overall approach to interpreting and its evaluation. 1.1 Translating and interpreting for LSP 1.1 Translating and interpreting for LSP In the context of specialised conferences, the adequate use of specialised ter- minology is particularly relevant in terms of the clients’ and end-users’ expec- tations, because precision, economy of expression, and accuracy are considered key aspects of specialised discourse. It is not surprising, then, that several sur- veys on quality in simultaneous interpreting have highlighted the use of correct terminology as one of the most important benchmarks of perceived quality (e.g. Gile 1990, Pöchhacker 1994), both in terms of the expectations on and of the eval- uation of the interpreting service, and both among end-users and interpreters themselves. Many such user surveys have been conducted over the years, pre- dominantly through questionnaires sometimes combined with other methods, such as interviews (Mack & Cattaruzza 1995, Vuorikoski 1993, 1998, see García Becerra 2016 for a discussion on administration methods). For instance, a survey by Kurz (1989, 1993) regarding user expectations on in- terpreting quality found that correct terminology was the third out of eight fac- tors in order of importance, after “sense consistency with the original message” and “logical cohesion of the utterance”. Meak (1990) conducted a small-scale survey on 10 Italian doctors. The results on the role of correct terminology are less conclusive in this case, but this is to be expected considering the limited sample. Nonetheless, even the most lenient respondents stressed that too frequent imprecise use of vocabulary may prove distracting. Marrone’s (1993) survey on a sample of 87 conference attendees combined ex- pectations and evaluation and found that “inaccurate terminological usage” (p. 37) was considered as a shortcoming by most respondents, more serious than un- pleasant delivery. Quality of style and correct terminology were ranked second in order of importance after information completeness and before intonation and delivery. Kopczynski (1994) explored the expectations of 57 Polish speakers and recep- tors of interpreting and found that although conference attendees may in some cases show a certain degree of leniency towards the use of incorrect terminology, this may be perceived as one of the main irritants. Terminological precision was identified as the second most important parameter of quality, independent of the respondents’ role and professions. By the same token, in a survey by Pöchhacker (1994) on quality evaluation, “mastery of technical language” was ranked second in order of importance after the quality of verbal expression. Mack & Cattaruzza (1995) surveyed user expectations in Italy. 1.1.2 Terminology as a parameter of quality As discussed by Pöchhacker (2001, 2004), inter- preting has been addressed essentially from two perspectives: on the one hand, it has been viewed as a primarily linguistic task of text reception and production; on the other, as a chiefly communicative, socially-embedded task. Terminology as a quality parameter tends to play an important role in evaluation frameworks belonging to the first approach, as here the focus is on the product. This approach reflects the expectation of equivalence between the interpreted and the source speech, and is reflected in the notions of accuracy and fidelity (Gile 1991a), re- sulting in what Déjean Le Féal (1990: 155) defined as “equivalent effect”. This approach is exemplified by Barik’s (1971) taxonomy of error and omission types in simultaneous interpreting, seen primarily as linguistic production. The second approach focuses on interpreting as interaction. Here, the evaluation of quality focuses on the communicative aim of interpreting as situated in a real commu- nicative event. Rather than equivalence, what is important for scholars choosing this approach is the efficacy in achieving a pragmatic communication goal. In this sense, important contributions stem from research on community interpreting, which by its nature promotes the involvement of all players in the communica- tion triad, i.e. the interpreter and the communication parties. An example of this approach to the evaluation of interpretation is Wadensjö (1998, 2005) evaluation framework. 10 1.1 Translating and interpreting for LSP 1 Terminology work in interpreting 1 Terminology work in interpreting An AIIC survey presented in Moser (1995, 1996) involved 94 users of interpret- ing who were interviewed based on a structured questionnaire. Terminological accuracy was deemed important especially by experienced attendees of technical meetings. It was even considered more important than completeness. For general meetings, the opposite was true. In sum, end-users attending technical conferences tend to rank the use of cor- rect terminology among the top quality factors for interpreting. Completeness of rendition, knowledge of the subject matter and precise terminology tend to be assigned higher values than delivery-related aspects such as native accent or pleasant voice. Interpreters also recognise the role of terminology as a key quality parameter. The first survey on interpreters’ expectations was conducted by Bühler (1986) on 41 AIIC members and 6 members of the association’s Admissions Commit- tee. Both linguistic and extra-linguistic criteria were included. Among linguistic criteria, “use of correct terminology” (p. 232) was ranked particularly high (83% highly important) by the members of the Admissions Committee, similarly to the ranking for “sense consistency with the original message”, which received the highest relative and absolute ranking by all participants. As Bühler suggests, this might indicate that using accurate terminology can promote a faithful rendi- tion of the message. Indeed, the use of adequate terminology may contribute to improve performance as it favours cohesion and the correct transmission of the message (García de Quesada 2011: 219). A larger number of respondents were reached through online questionnaires in the 2000s. In a survey by Chiaro & Nocella (2004), involving 286 respondents, interpreters ranked correct terminology usage as one of the second-most impor- tant factors in addition to fluency of delivery and correct grammatical usage. Zwischenberger & Pöchhacker (2010) collected survey data on interpreters’ ex- pectations and self-perceptions, which the authors see as “inherently linked with the issue of quality” (p. 11). They partly replicated Bühler’s survey by adopting the same quality criteria, although their questionnaire also involved an evalua- tion of recorded interpretations. As for the role of correct terminology, the re- sults were similar to Bühler’s. Zwischenberger and Pöchhacker, however, also explored how the relative weight assigned to the individual criteria is affected by the type of interpreted event. They found that correct terminology was listed as a top priority by 38% of the respondents who cited seminars and workshops as a setting with specific quality requirements (ibid., p. 1.1 Translating and interpreting for LSP In their research, correct terminology was identified as the most important quality factor. The find- ings of Weller & Yanez’s (1998) user survey are also in line with Mack and Cat- taruzza’s. 11 11 1.2 Terminology work in translation and interpreting The contribution of the terminological discipline to the field of specialised trans- lation has been fundamental. For this reason, translation scholars and practition- ers tend to consider terminology as an integral part of the professional transla- tion practice, as testified by its inclusion in training programmes. However, ter- minology work in translation and interpreting presents specific features which are related to the nature of translation and interpreting and to the purpose served by specialised terminology in these professions. For these reasons, while the ter- minological discipline has laid the foundations for terminology work in transla- tion and interpreting, the approach to terminology in these fields has evolved to adapt to their specific requirements, giving rise to dedicated models of transla- tors’ and interpreters’ terminology work. 1 Terminology work in interpreting 16). In light of these survey results, it appears clear that both interpreting ser- vice providers and users acknowledge terminology as an important quality cri- terion. Therefore, interpreters may more easily meet the quality expectations 12 1.2 Terminology work in translation and interpreting of their audience by using adequate specialised terminology. This goal can be achieved through effective domain knowledge acquisition and proper termino- logical preparation. Against this background, the following section will discuss the role of terminology work, with a focus on interpreting as compared to trans- lation. 1Although, as Drewer & Pulitano (2019) rightly observe, “the differences between these schools [are] too small to be able to talk about different schools at all” [my translation from the German original]. 1 Terminology work in interpreting While Wüster’s model lays the ground for a structured terminological disci- pline, it presents several limitations which determined its later expansion by loy- alists or its firm critique and rejection by competing schools of thought (Cabré Castellví 2003, Drewer & Pulitano 2019).1 One such limitation is the exclusion of the individual contexts in which termi- nology is inevitably embedded, and which determines a certain degree of vari- ability (Will 2007). Therefore, translators and interpreters should not rely on par- allel texts as reliable sources for the mining of terminology and the creation of ontologies, since in Wüster’s theory terminologies are seen as universal, super- ordinate structures, unaffected by cultural differences. To overcome the limi- tations of Wüster’s approach, Gerzymisch-Arbogast (1996) proposed a context- specific term model which allows for a comparison of term meaning across texts, as used by the individual authors. The establishment of knowledge systems of this type requires static textual environments, as in translation. 1.2.1 Terminology and translation Wüster (1931, 1979) may be considered as the nestor of terminology, for which he and the “Viennese school” (Drewer & Pulitano 2019) claimed the status of inde- pendent discipline. His normative intent, formulated in his “General Theory of Terminology”, pursued the goal of ensuring unequivocal communication in tech- nical fields. He saw terminology “as a tool for disambiguating scientific and tech- nical documentation and communication” (Cabré Castellví 1998: 17). Specialised communication requires brevity and clarity, which is only achieved through ad- equate linguistic tools, i.e. terms. Terminology thus emerges as essential both for intralingual and interlingual specialised communication (Arntz & Picht 1982), be it in the form of translation or interpreting. In Wüster’s model, which expands de Saussure’s (1959) triangle, the term is described as an entity made up of a denomination, the “symbol” in Ogden & Richards’s (1923) semiotic triangle, or de Saussure’s (1959) “signifier”, and a concept, the signified. Wüster includes the distinction between langue and parole in the model itself (Mikkelsen 1991: 163), a distinction which appears both in the signified (content) and in the signifier (expression). 13 13 1 Terminology work in interpreting 1.2 Terminology work in translation and interpreting tion process (Martínez & Faber 2009: 104) and is strictly intertwined with the individual context and co-text in which specialised terms are embedded. This further denotes how a prescriptive, onomasiological approach to terminology and terminography is not reflective of the way translators work on specialised texts. In their terminology work, translators can make use of a variety of tools and resources that “can contribute to facilitating and accelerating the identifica- tion, description, consultation and reuse of terminology in a translation context” (Kageura & Marshman 2019: 74). These tools and resources comprise terminol- ogy databases, terminology extraction tools (TET) from monolingual and bilin- gual comparable and parallel corpora (especially previous translations), concor- dancers, tools for terminology research, text-alignment software, and terminol- ogy management systems (TMS). A translator’s workbench typically integrates all or most of these technologies (Hansen-Schirra 2012: 211). While translating a text, translators can consult multilingual glossaries and terminology databases to look up target-language equivalents, find definitions, etc. (Blancafort et al. 2011: 4). A translation project often starts with monolingual term extraction from the source text or other documents. The source text can be pre-processed to identify and extract candidate terms; previously translated and aligned texts can be used to create terminology databases as parallel corpora. Terminology extraction can be performed both by project managers and in-house terminologists and made available to translators, either for entire domains or for individual translation projects and documents. In principle, terminology can be extracted manually, for instance by reading and annotating the text. However, due to the increasing volumes of texts to be translated and the increasingly shorter time to production, manual terminology extraction is often excessively time-consuming. Here, auto- matic terminology extraction (ATE) can provide valuable support in combination with subsequent refinement to exclude pseudo-terminological units (Pavel & No- let 2001: 88). Bilingual ATE can be performed on parallel corpora (Vintar 2001), often in the form of translation memories (TM), to create termbases containing terminological pairs. Especially for non-standardised and emerging domains, par- allel corpora (Blancafort et al. 2010: 263) or authoritative databases or handbooks may not be available (Heid & Gojun 2012: 586). This lack can be addressed by the use of comparable corpora, i.e. of “texts of the same domain (and possibly genre) in different languages which need not be translations of each other” (Blancafort et al. 2011: 1). 1.2.2 Terminology work in translation At the beginning of the 1980s, terminology positioned itself as a necessity for translation (Mayer 2019: 84). Perhaps the most influential contribution to ter- minology theory and practice applied to translation can be identified in Arntz & Picht’s (1982) “Einführung in die übersetzungsbezogene Terminologiearbeit”. The authors offer a detailed discussion of Wüster’s model and provide indica- tions for terminology work in translation. Unlike for terminologists, translators’ terminology work is always descriptive rather than prescriptive, thus aiming to pin down the knowledge system of a particular field and its expression through terminology (intended both as terms and as syntagma and collocations). While ideal terminology work proceeds onomasiologically, i.e. from the signified to the signifier, translators (and especially interpreters, as discussed in §1.2.3) often pro- ceed semasiologically, i.e. from the expression to the meaning. As Mayer (2019: 103) observes, the onomasiological approach is only rarely used in terminology work by translators because of the considerable effort it entails. f As remarked for interpreters (§1.1), it is not required of translators to be expert in a specific specialised field. Rather, like ad-hoc terminologists and terminogra- phers (Wright & Wright 1997), they must be able to quickly acquire the elements of expert knowledge necessary to fill their knowledge gaps and deliver a high- quality translation. Thus, terminology work is conducted as part of the transla- 14 1.2 Terminology work in translation and interpreting 2https://www.linguee.com (Accessed: 10.09.2021) 3https://context.reverso.net/translation/ (Accessed: 10.09.2021) 4http://www.sketchengine.eu/ (Accessed: 06.09.2021) 1.2 Terminology work in translation and interpreting Incidentally, the results of terminology extraction are not only di- rectly useful to translators but can be fed to further tools which translators can use to support their translation workflow, such as computer-assisted translation 15 1 Terminology work in interpreting (CAT) tools or MT systems (Heid & Gojun 2012: 588). Most CAT tools include TETs, such as MultiTerm Extract in SDL Trados, which require parallel texts. Corpora can be further analysed through concordancers to explore how terms are used in context to derive the most relevant phraseologisms (Pavel & Nolet 2001: 89). Some freely available examples are Linguee2 or Reverso Context3 for the exploration of parallel corpora. Sketch Engine4 (Kilgarriff et al. 2014) is also a popular concordancer which offers monolingual, bilingual, or multilingual cor- pora analysis. Terminological resources can be used for the immediate translation task, but they can also be further stored in dedicated terminology management tools to be consulted and reused in future translation tasks. Although dedicated tools are available to this aim (e.g. termbases in TMSs), the use of spreadsheet software or other general-purpose solutions to store and exchange terminology is also rather widespread (e.g. SDL 2008, Blancafort et al. 2011). 1.2.3 Terminology work in interpreting Even though interpreting shares many similarities with translation, it presents some distinctive features that affect the way terminology work is conducted to ensure effective interlingual oral communication. 1.2.3.1 Distinctive features of terminology work in interpreting Translators work with a source text fixed in writing, which they can revisit as often as needed during the task. They can scan the text to clarify doubts by analysing the co-text, conduct terminological and content-related queries to fill their knowledge gaps during the task, and in theory pace the translation process as they require. The final product is therefore seldom the first draft produced, but rather the result of a process of increasing refinement. In interpreting, the source text is seldom available. Even when interpreters have access to the script of the speech before the assignment, they encounter the final speech as pronounced by the speaker only in the moment in which it is actually delivered, i.e. during the assignment. Interpreters are therefore often faced with the task of foreseeing which topics will be addressed in the speech and, consequently, which termi- nology may be used. Therefore, they must strive to fill their domain-knowledge 16 1.2 Terminology work in translation and interpreting and terminological gaps ahead of the task (Rütten 2008: 22). Additionally, inter- preting is performed under severe time constraints and cannot be interrupted. It would be unthinkable for interpreters to stop their rendition of the speaker’s message because they need to acquire additional knowledge to correctly under- stand and transfer the meaning. The possibilities of revision and monitoring are also more limited in interpreting, as the target text produced by interpreters can- not be further refined or modified (Will 2020: 38). Indeed, self-corrections or reformulations are often regarded as undesirable in interpreting. For interpret- ing assignments, the process of knowledge and terminological acquisition is thus mainly relegated to the pre-process phase and can only be integrated to some ex- tent into interpreting either peri- or in-process (Kalina 2005: 778), i.e. during the assignment. g Terminology work in interpreting thus aims to avoid knowledge gaps (Rüt- ten 2008: 25). Such knowledge gaps are, however, not only related to an inter- preter’s overall general knowledge or knowledge of the subject field, but are also situation-related. While conferences and their interpretations can be recorded and made available even after the event has ended, the communication taking place during the conference is mostly meant to serve an immediate purpose and aimed at a specific audience. The situatedness of speeches and presentations also determines the situatedness of their interpretations. 1.2.3.1 Distinctive features of terminology work in interpreting What matters in an inter- preter’s preparation is its suitability for the assignment at hand; terms are consid- ered adequate if they help communication. Thus, terminological preparation is also more pragmatic and situation-oriented than the terminology work of trans- lators (Rütten 2012b), because the product is usually meant to remain available for a longer period of time, and especially of terminologists, whose work must remain valid also beyond the individual text (Rütten 2013: 49). The pragmatism and situatedness of interpreters’ terminology work emerge in the type of infor- mation contained in interpreters’ glossaries: the terms chosen may be even very simple and general if the interpreter deems them more suitable to ensure effec- tive multilingual communication. As remarked by Rütten (2012a: 43), “translators and terminologists are far less free to generalize (and interpreters, in a way, are meant to generalize if the situation calls for it)”. Another difference between terminology work in translation and interpreting is due to the interpreting mode used in specialised conferences. In this setting, interpreters mostly work in the simultaneous mode, while consecutive interpret- ing is reserved to short presentations or other settings characterised by a higher degree of interaction between the speaker and the audience, e.g. technical work- shops. Unlike in the consecutive mode, in the simultaneous mode the interpreter 17 17 1 Terminology work in interpreting does not deliver the speech after the delegate, but rather almost at the same time as the speaker (for a discussion of interpreting modes, see e.g. Pöchhacker 2015). as the speaker (for a discussion of interpreting modes, see e.g. Pöchhacker 2015). The immediacy which characterises both modes of interpreting (Pöchhacker 2011: 10) is particularly apparent in simultaneous interpreting; the external pac- ing of the task profoundly affects the cognitive and strategic processes under- lying interpretation. Due to the limited amount of time and cognitive resources available during SI (Rütten 2013: 49), being able to quickly retrieve target-lan- guage equivalents from long-term memory to properly render specialised dis- course is paramount. This foregrounds the key role of terminology work pre-, peri- and in-process in interpretation for specialised conferences. Conducting preparatory work ahead of the interpreting process allows for the internalisa- tion of domain knowledge and specialised terminology, which can later be more easily recalled by interpreters, thus facilitating both their understanding of the source speech and their rendition in the target language. 1.2.3.1 Distinctive features of terminology work in interpreting Gile (2009) describes the role of language availability in his gravitational model. When units of linguistic knowledge are readily available because they have been activated, for example through preparation, or because they are often used, they can be easily accessed by the interpreter, as they gravitate to the core of available lexicon. When this does not happen, i.e. when the units of linguistic knowledge (terms) “drift out- wards”, a series of issues can occur both at the comprehension and the produc- tion level. Activating terminology before the event is therefore paramount to prevent such problems. More specifically, shifting cognition upstream from the simultaneous phase (Stoll 2010: 3) allows for easier anticipation and faster under- standing of the speaker and, in turn, for faster processing and a more accurate rendition. Finally, favouring the use of specific terminology over the adoption of alternative strategies such as paraphrasing, abstracting, or using hypernyms also helps interpreters avoid cognitive overload by freeing up cognitive resources thanks to a faster rendition. However, not all and not only domain-relevant terms are included in interpreters’ glossaries: only those terms are noted which the in- terpreter fears not being able to retrieve from long-term memory, even though they are trivial words (Rütten 2011: 42). This shows that terminology work in in- terpreting is also highly personal, corresponding to the individual interpreter’s preferences, working and memory structures (Rütten 2012a: 48, Wagener 2012: 9, Rütten 2018: 147). 1.2.3.2 Modelling terminology work in interpreting For the reasons outlined thus far, interpreting does not allow for a direct appli- cation of Gerzymisch-Arbogast’s (1996) model. In interpreting, the context is in 18 1.2 Terminology work in translation and interpreting fieri and co-created by the communication parties, albeit to different extents ac- cording to the individual interpreting modes. Additionally, the time constraints inherent to interpreting, especially in the simultaneous mode, determine an even clearer preference for the semasiological approach, which saves time when com- pared to the ideal onomasiological approach (Mikkelsen 1991: 168). It should be noted that expeditious terminology work is essential not only in terms of con- sultation, but increasingly also during the preparation phase, which has become progressively shorter (Rütten 2008). Speeding up preparation also seems conve- nient in terms of the profitability of interpreting, as preparing for an assignment seems to require a considerable amount of time.5 At the same time, interpreters need to constitute knowledge systems by rely- ing on “external textual structures” (Will 2007: 67) to be able to understand the texts they are interpreting. Will’s (2007) model of terminology work in SI closed the gap left open by previous models by considering the particular nature of in- terpretation and the conditions under which texts are produced and translated during SI. Will’s model represents an expansion and integration of his own model of knowledge management during a simultaneous interpreting assignment (Will 2000). In this first model, Will separated the interpreting assignment into three phases of knowledge acquisition: the preparation phase (Stage I), the confer- ence itself (Stage II) and the de-briefing phase post-assignment (Stage III). Or, in Kalina’s (2005: 778) terms, into the pre-process, peri- and in-process, and post- process stages, respectively. As Will argues, new knowledge in the form of single terminological units can only be constituted during Stage I and III and during the non-interpreting phases of Stage II (i.e. right before and right after the interpret- ing session). Additionally, Will draws on Gerzymisch-Arbogast’s model as well as on Mudersbach’s (1999) and Floros’s (2003) models. Mudersbach’s model inte- grates terminological units into their relative knowledge systems necessary for text comprehension, while Floros describes how relevant knowledge is selected. As Will (2007: 65) observes, “it would be impossible [for conference inter- preters] to acquire the same amount of knowledge as their specialised public”. 5One or more working days according to AIIC’s (2002) workload study. 1.2.3.2 Modelling terminology work in interpreting They “have thus to be able to constitute and to use relevant information in a very effective and specific way”. With some rare exceptions, conference inter- preters working with LSP do not possess the same level of domain knowledge as the speakers and audiences they are called to interpret for. They are therefore presented with an arduous task: that of compiling terminological resources and acquiring sufficient information on the subject matter at hand to be able to pre- cisely convey the speaker’s message despite not being experts themselves. Quite 19 1 Terminology work in interpreting often, this is further complicated by the limited time available for preparation. How, then, can they solve this conundrum? According to Will’s model of terminology work in SI, when preparing for a specialised conference, the interpreter must first of all attribute an individual term (“Texterm”) to a specific knowledge system, by comparing it with a refer- ence definition (“Systerm”). This process leads to the creation of a “Terminolog- ical Knowledge Entity” (TKE), “the smallest complete knowledge unit for under- standing and producing technical texts” (Will 2007: 69). To obtain a picture of the general knowledge structures in a speech, TKEs are grouped to form what Will defines as “Terminological Knowledge Constellations” (TKC). The most relevant knowledge systems for the assignment in question should be further investigated, while the less relevant ones can be treated in a more “economic” way, which allows for a more effective approach to terminology work. According to Will, conference interpreters’ terminology work can thus be equated to a detective’s investigative work: rather than following an onomasiological approach, which would involve a top-down approach, i.e. starting from acquiring the knowledge and then identifying the terminology used to express the concepts, conference interpreters tend to follow a semasiological approach. They explore the corpus of preparation material looking for the relevant terminology, the “explicit units” that refer to a knowledge system, using terms as evidence to identify the units of knowledge they convey. Semantic links between terms reflect logical connec- tions between the individual knowledge items. This process allows interpreters to acquire the necessary terminological knowledge and at the same time to link such knowledge to conceptual reference systems.fi This “detective work” presupposes the availability of the “evidence”, of suffi- cient preparation material to ensure effective terminology extraction and knowl- edge constitution. 1.3 Terminology tools for conference interpreters The need for supporting terminology tools capable of optimising and automat- ing the interpreting workflow clearly emerges against the background of what was discussed in §1.2.3. Will’s model of the terminology workflow in interpret- ing highlights that dedicated tools should address several requirements specific to interpretation: they should lighten interpreters’ preparation load by promot- ing targeted preparation and easy access to glossaries; be as little obtrusive and complex as possible, especially during the in-process phase, in order not to in- terfere with the concurrent cognitive processes involved in interpreting; easily adapt to the individual preparation styles of conference interpreters; facilitate the goal of moving cognition upstream of the interpreting phase (Stoll 2009). 1.2.3.2 Modelling terminology work in interpreting In order to face these limitations, Fantinuoli (2006) proposes the adoption of a corpus-driven preparation methodology. After creating cor- pora of specialised texts with corpora creation tools, conference interpreters can explore the topic with a concordancer starting from an automatically-extracted terminology list. This also facilitates the identification of “phraseological knowl- edge” (Fantinuoli 2017a: 29), primarily in the form of collocations, which in ad- dition to the individual terms also characterise specialised terminology and the client’s jargon. Thus, following a Corpus-Driven Interpreter Preparation (CDIP), “interpreters preparing for a conference can obtain a list of relevant terms and texts within minutes, even when targeted preparatory materials have not been made available by the conference organisers (as is often the case in professional settings)” (Fantinuoli 2006: 188). 20 1.3 Terminology tools for conference interpreters 1.3.1 Requirements for a support tool for conference interpreters To define the ideal architecture and features of a bespoke tool for interpreters, Rütten (2000, 2004, 2007) laid down a model of the ideal interpreter-specific tool which mirrors the workflow defined by Will (2000, 2007).i Rütten envisaged a five-tier structure organised around a “central starting page” (Rütten 2004: 173, see Figure 1.1) from which the different modules can be accessed. Each module serves a specific purpose and represents one phase of an interpreter’s workflow. According to Rütten, interpreters would benefit from: • a modality allowing for online and offline research of documents and in- formation; • a modality allowing for online and offline research of documents and in- formation; • a module for document management, connected to • a function for terminology extraction, also from parallel texts when avail- able, for the compiling of multilingual dictionaries; • a terminology management module synchronised to the previous one which would allow for the identification of duplicates or similar entries; • a “trainer” to systematise vocabulary memorisation. In addition to these interdependent modules, the author argues in favour of an “overall quick-search key that can be used blindly and independently of the module or function” (Rütten 2004: 175) for querying the whole database and all the modules. For students, a dedicated training function with exercises to fine- tune the sub-skills involved in interpreting could also prove beneficial. Rütten’s 21 1 Terminology work in interpreting 1 – Online + Offline research 2 – Document management 3 – Terminology extraction 4 – Terminology management 5 – Trainer Central starting page Figure 1.1: Rütten’s software model (Rütten 2004: 173) 1 – Online + Offline research Central starting page Figure 1.1: Rütten’s software model (Rütten 2004: 173) suggestions map the circularity of interpreters’ terminology work, from prepara- tion material to glossaries before and during interpreting, and back to glossaries after the assignment. The document management and terminology extraction modules facilitate the collection of textual material to constitute the necessary reference knowledge systems. The vocabulary trainer addresses another key ne- cessity for interpreters: the acquisition and activation of relevant specialised vo- cabulary ahead of the interpreting task, as terminological resources, unlike for translators, can also be consulted to some extent during interpreting. 1.3 Terminology tools for conference interpreters 1.3 Terminology tools for conference interpreters may be useful also during interpreting, for instance if texts are read out loud by speakers and the translation is already available. Translation memory systems which save all terminological entries in a single database can promote managing the data as a whole (Rütten 2012a: 44). Additionally, some TM tools and TMSs usually allow embedding a picture to illustrate a term, which can be useful to quickly grasp the term’s meaning (ibid.) and support memorisation. Tools for terminology extraction, either stand-alone or integrated into CAT tools, can save interpreters time by automatically identifying candidate terms and promote knowledge acquisition by allowing interpreters to focus on the meaning of the text. These tools can also represent useful emergency solutions when long documents are made available last-minute during the assignment (Rütten 2012a: 48). Finally, tools for terminology search on the web or in online databases and dic- tionaries can prove useful to identify or check the validity of a target-language equivalent during glossary compilation. The application of TMs, TMSs, CAT tools, etc., to interpreting presents, however, some limitations, which emerge especially in relation to their in-process use. To start, while non-bespoke tools do address some operations and subprocesses, they do not cover the entire inter- preting workflow satisfactorily (Rütten 2013: 49).it Non-interpreter-specific tools are often criticised by interpreters because they are perceived as too complex and not intuitive. TMSs, for instance, usually offer multiple data levels, which promote a systematisation of entries, but are often seen as too cumbersome or excessively complex (Rütten 2012a: 50, Stoll 2009: 141). The host of functions offered by non-bespoke tools does not necessarily have to be exploited completely. However, interpreters not familiar with these tools may perceive them as overwhelming and unnecessarily complex. Additionally, an interface offering too many functions or unoptimised entry structures may be difficult to process, especially for in-process terminology look-up when attention is already allocated to several subprocesses. None of the available tools offers “a mouse-free, blind and intuitive search function” (Rütten 2012a: 46). The limitations of non-bespoke tools outlined above have led to the devel- opment of increasingly sophisticated tools dedicated to conference interpreters, known as CAI tools. The following chapter discusses them in detail, highlighting the postulated advantages they offer, as well as their unique features as compared to CAT tools and the other resources discussed thus far (see also §1.2.2). 1.3.2 Usefulness and limitations of non-interpreter-specific tools Against the background of Will’s and Rütten’s models, the question arises whether and to what degree the tools already available to translators may al- ready satisfy the specific requirements of interpreters’ terminology work. y y pi q p gy As remarked by Rütten (2011: 43), CAT tools and other technologies such as TETs, TMSs, concordancers, etc., already cover several subphases of an inter- preter’s workflow. These tools have the potential to provide useful support to in- terpreters, although they have not been explicitly targeted to these professionals. Useful functions offered by translation memory tools and terminology manage- ment tools are for instance sorting and filtering functions or differentiated search functions (e.g. fuzzy search, see Rütten 2011: 43). It should be noted that complex searches or filtering are mainly feasible in the pre-, peri-, or post-process phase (Rütten 2013: 50, Wagener 2012: 8). Translation memories can represent useful tools especially for those interpreters also working as translators, who may reuse their terminological databases and save precious preparation time (Rütten 2013: 49). TMs can also be useful to show aligned parallel texts on screen. This function 22 1.3 Terminology tools for conference interpreters The chap- ter further explores whether and to what degree interpreters prefer CAI tools to other solutions for their terminology work and reviews the research conducted thus far on the topic of CAI. 23 2 Information technologies and interpreting In view of the complexity and the necessity of terminology work in interpreting for LSP settings, interpreters have looked for ways to facilitate their terminology work through technology. While interpreters can use tools that are also available to translators to pre- pare for their assignments and to access their terminological resources while interpreting, they can also adopt software developed specifically for interpreters. Known as CAI tools, these solutions aim to address interpreters’ unique needs and overcome the limitations of traditional tools identified in §1.3.2. With the aim to provide an in-depth discussion of CAI tools, the present chap- ter introduces the topic of technologies applied to interpreting and illustrates CAI tools within this larger framework. Special attention is dedicated to Inter- pretBank1 (Fantinuoli 2012, 2017a), the CAI tool adopted in the present study. To determine whether and to what degree interpreters include CAI tools in their workflows, the terminographic practices of conference interpreters are presented. The potential and limitations of CAI tools in addressing interpreters’ needs are discussed against the background of current CAI research, which motivates the contribution provided by the present work. 1https://interpretbank.com/ (Accessed: 24.07.2022) 2This chapter focuses on applications to professional conference interpreting. Therefore, tech- nologies which support training, known as Computer-Assisted Interpreter Training (CAIT) tools are not discussed here. 2 Information technologies and interpreting 2 Information technologies and interpreting Technologies can both mediate, generate and support interpretation (see Braun 2019). Supporting technologies include CAI tools. CAI tools can, however, also integrate technologies used to generate interpretation (e.g. ASR and MT), and may, in turn, be integrated into technologies mediating interpreting (see Will 2020: 65). To introduce the topic of CAI and define it in relation to other information and communication technologies (ICTs), the following sections pro- vide an overview and a classification of interpreting technologies.2 2.1 Overview and classification of technologies Like translation and numerous other areas of human life and work, the inter- preting profession has not remained indifferent to technological advances. Some developments in the field of technology have led to considerable progress in the interpreting practice, while other technologies have been perceived as a threat to its very existence since their inception. Something similar can be said of the trans- lation profession, which has experienced radical changes that have optimised translators’ workflows, but also threatened the status of human translation. 2.1 Overview and classification of technologies 2.1 Overview and classification of technologies of “striking a balance between cost-effectiveness and quality of interpretation” (Baigorri-Jalón 1999: 36). Not only financial, but also organisational and logisti- cal issues must be addressed, without sacrificing the quality of audio and video streams. Although some perceived and factual issues can be overcome through proper set-ups, a certain degree of resistance towards the “new technologies” remains among some conference interpreters. This extends to the other two cat- egories of interpreting technologies, as discussed in the following sections. The successful implementation of distance interpreting in several settings has, how- ever, shown that this technology is here to stay. Indeed, it could lay the foun- dations for the emergence of new conference interpreter profiles who fully em- brace the home-office, location-independent alternative to traditional conference settings, as the COVID-19 pandemic has shown (Fantinuoli 2019). Additionally, distance interpreting will require adjustments not only in attitudes, but also in training, for instance through the development of “training modules designed specifically for this modality, addressing cognitive, communicative and techni- cal aspects” (Ziegler & Gigliobianco 2018: 137). Furthermore, the integration of augmented reality technologies, such as VR glasses, could represent the next step in distance interpreting by helping address the perceived feeling of isola- tion through distance and the need for “self-control of the direction of sight” (ibid., p. 136). In translation, a comparable level of innovation was introduced by the advent of personal computers, which allowed for the execution of this intellectual activ- ity on a new medium. At the same time, this innovation brought about evident advantages in terms of productivity and quality, especially thanks to CAT tech- nologies (see §2.1.3). Technology has also modified how translation is performed. For instance, moving translation to the cloud has led to the creation of collabora- tive translation platforms, which allow for an unprecedented level of interaction between translators cooperating despite geographical limitations. The resulting translation is thus the product of a joint intellectual effort. ASR has allowed trans- lators to dictate their interpretation, rather than typing it, with evident gains in productivity and potentially also in quality (Carl, Lacruz, et al. 2016a). The use of machine translation has, to some extent, replaced translators’ work, but it has also offered language service providers and individual translators ways to speed up human translators’ work, e.g. through human post-editing of machine- translated texts. 2.1.1 Technologies mediating interpreting The first category of technologies applied to interpreting comprises solutions that have led to the emergence of new interpreting modes, or that allow for the performance of interpreting in previously inaccessible settings. The most promi- nent example of the first is simultaneous interpreting technology. Up until the 1920s, interpreting was only performed in the consecutive mode, or in the form of whispered interpretation, i.e. synchronously but without any technological aid. The first (successful attempt) at using simultaneous interpreting at a conference was at the 1928 ILO conference, where real SI was used during entire meetings, involving up to seven different languages (Baigorri-Jalón 1999: 33). After this first experiment, SI reached its “coming of age” (ibid., p. 34) at the Nuremberg Trials almost 20 years later. The simultaneous mode faced initial resistance by consecutive interpreters, who feared a decay in the quality of interpretation and, as a result, of communication, and perceived headphones and SI equipment as foreign objects. Nonetheless, SI quickly imposed itself for its time-saving charac- ter and cost-effectiveness, and remains the most widespread interpreting mode in the conference setting to date.i If simultaneous interpreting can be perceived as the first revolution brought about by technology in the field of interpreting, distance interpreting can be viewed as the second. This innovation made it possible to perform both modes of interpreting in new settings and with multiple possible configurations in the speaker-listener-interpreter triad. Distance interpreting “covers a whole range of technologically different setups” (Ziegler & Gigliobianco 2018: 121), which vary according to the constellation in the communication, the interpreting mode (e.g. in terms of the range of audio frequencies required), but also according to whether only the audio signal or both audio and video data are to be captured and transmitted. Since its inception, distance interpreting has posed the conundrum 26 2.1 Overview and classification of technologies For this reason, machine-translation post-editing (MTPE) can be placed at the intersection of technology-generated and technology-mediated translation. The combination of ASR and MTPE has even made post-editing of ASR output in audio-visual translation possible, as explored in the CompASs (Computer-Assisted Subtitling) project, a joint EU-financed effort of the German 27 2 Information technologies and interpreting TV channel ZDF and of the Translation & Cognition Center (TRA&CO) of the University of Mainz (see Tardel et al. 2019, 2021). TV channel ZDF and of the Translation & Cognition Center (TRA&CO) of the University of Mainz (see Tardel et al. 2019, 2021). 3https://www.skype.com/it/features/skype-translator/ (Accessed: 11.09.2021) 4https://azure.microsoft.com/en-us/services/cognitive-services/speech-translation/ (Accessed: 15.09.2021) 2.1.2 Technologies generating interpreting When considering technologies for the provision of intercultural communication services without the involvement of human interpreters, an inevitable parallel emerges between translation and interpreting. Over the years, although with dif- ferent levels of perceived pressure, both translators and interpreters have increas- ingly been faced with what they consider a potential threat to their livelihood: machine translation, on the one hand, and machine interpreting (MI), or speech translation, on the other. Just like MT, MI also aims to generate interpreting without the contribution of human interpreters. Despite its having long been discarded as a merely theoretical threat, the con- siderable advances brought about by neural networks and artificial intelligence (AI) have led to a resurfacing of machine interpreting. Like MT, MI requires a combination of several technologies. The essential steps involve ASR or speech-to-text technology (STT) to turn the speaker’s words into written text, a machine translation system to translate such text, and speech synthesis, or text-to-speech, to turn the translation into its oral form in the target language. The spoken output can, however, also be replaced by the creation of subtitles, without the use of speech synthesis. At present, the two main models are cascading and end-to-end systems. In cas- cading systems, the above-mentioned steps are performed consecutively and the source speech must be segmented. End-to-end systems do not require the inter- mediate step of ASR. This last possibility has only recently started to be explored, but it is increasingly reaching the quality produced by cascading systems. End- to-end systems are exemplified by Google’s Translatotron (Jia et al. 2019) and Translatotron 2 (Jia et al. 2022). The further development of the “neural para- digm” (Braun 2019: 296) could lead to rapid advances in the quality of MI output: “especially neural networks which can learn from previous tasks and shift atten- tion according to the relevance of an element in the source speech may have the potential to make machine interpreting more human-like” (ibid.). Most systems developed thus far work consecutively, i.e. the interpretation in written or oral form is made available after the speaker has produced an utter- ance. However, simultaneous machine interpreting has also been explored (e.g. Cho et al. 2013) and is attracting growing interest.i The first MI systems were of the consecutive type and relegated to limited domain applications. One example is the German Verbmobil project (Wahlster 28 2.1 Overview and classification of technologies Nor are they capable of inference and context anticipation: as Fantinuoli (2019: 342) observes, “they still lack background and 29 2 Information technologies and interpreting context knowledge”. There is, additionally, the issue of the suitability of input for MI processing: unlike MT, which allows for a quick test to verify whether a certain document can effectively be machine translated, MI does not allow for a re-do if things go south – if “a[n] MI system [...] fails to deliver a usable trans- lation, the communication simply breaks down” (ibid.). The interplay of these factors, coupled with the immediacy of simultaneous interpreting, is one of the reasons why fully automated machine interpreting will probably require further effort and time. As Braun (2019: 295) points out, the application of MI systems “to situations in which highly accurate professional language mediation is required remains a nontrivial challenge”. In order for MI to enter the third phase of human- machine interaction, i.e. that in which interpreters will go from machine-assisted to machine-assisting professionals, not only would the MI output need to be of a quality comparable to human interpreting, but it would also need to become more convenient in economic terms. Despite some scholars’ assertions that a to- tal replacement of human interpreters by machines will never be possible, “the real question is if AI will ever be able to tackle [the above-mentioned] issues at some point in the future” (Fantinuoli 2019: 345). There is no reason to affirm with absolute certitude that this will never be possible. At present, the question remains unanswered. There is, however, no doubt that machine interpreting will impact the inter- preting market in some way or another, at the very least in terms of the percep- tion of conference interpreters’ role and professionalism in the eyes of the public. Fantinuoli (2019) envisages a near future in which MI will start entering the low- end segment of the market, following its short-term entering of the recreational sector, where professional interpreting services were not used before. It is plau- sible to imagine that high-quality and highly professional human interpreting will be relegated to the higher end of the market, “at least until the advent of real human-like MI.” (p. 345). The use of machine interpreting in formal and high-level contexts should, how- ever, not be excluded. 5For instance, Cisco Webex offers “real time translation” (Webex Help Center 2022). The Euro- pean Parliament’s innovation partnership also aims at the development of a “Live Speech to Text and Machine Translation Tool for 24 Languages” (European Parliament 2019) 2.1 Overview and classification of technologies 2.1 Overview and classification of technologies 2000), a consecutive speech translation system for the domain of appointment- making. It was one of the first projects combining ASR and speech synthesis. An- other domain-specific system is VoiceTRAN Communicator, a cascading speech- to-speech system for the language pair Slovenian-English (Žganec-Gros et al. 2005). These first systems had limited applications. However, in recent years, ma- chine interpreting has become increasingly known to the general public, thanks to the implementation of general-purpose solutions capable of handling con- versations on a variety of topics. A well-known example is Microsoft’s Skype Translator3, which offers “real-time machine translation in 10 languages in voice and video chats and 60 languages in text chats” (Hoberg 2021: 21). Microsoft Azure also offers a speech translation API.4 Currently, there is growing inter- est for simultaneous machine interpreting systems capable of handling continu- ous speech. The Karlsruhe Institute of Technology (KIT) has developed a lecture translator to provide automatic simultaneous interpreting for foreign students who do not understand German. The output is made available in the form of written subtitles (see Dessloch et al. 2018 for a detailed description of the system). In addition to fostering accessibility in the university environment, the lecture translator further supports the students’ learning as the recordings of the lec- tures and their translations are stored in a repository. Another key advantage of using the lecture translator is the greater affordability of the system compared to human interpreters, despite the lower quality (Dessloch et al. 2018: 89). User tests involving students suggest that the lecture translator is well received and considered useful, especially by foreign students. A similar application of a system combining ASR with MT developed in the field of audiovisual translation is the CompAsS project (see also §2.1.1). The CompAsS tool aims at automating the subtitling process wherever possible, sup- porting the workflow of subtitlers. Initial studies suggest that the tool reduces the technical effort experienced by subtitlers (Tardel et al. 2019, 2021). What could seem like a relatively easy feat considering the sizeable improve- ments made by machine translation is, however, complicated by the very nature of spoken discourse. Unlike written communication, speech presents a certain degree of spontaneity and ambiguity, which machines are still unable to deal with without human assistance. 2.1 Overview and classification of technologies 2.1 Overview and classification of technologies discourse-oriented perspective (see e.g. Apfelbaum & Wadensjö 1997 for Verb- mobil, Hoberg 2022 for Skype Translator). This type of evaluation is, however, still in its infancy for continuous systems in conference settings (see Fantinuoli & Prandi 2021). A valuable step in this direction is the QEMI-C corpus (Mauri 2021). QEMI-C is a manually-compiled and -annotated, trilingual parallel corpus com- prising 40 authentic speeches. It can serve as a basis for the comparative quality evaluation of MI and human interpretation (ibid., p. 95) in conference settings, which research has yet to address to a large extent. Understanding the effects of MI use on communication appears particularly urgent, especially considering the overall negative perception of this technol- ogy by interpreters. In translation, where MT has long entered the profession, translators’ resistance towards this technology and a certain dislike of related translation tasks (e.g. MTPE) is well known, surfacing as negative attitudes for instance in surveys (e.g. Moorkens & O’Brien 2015) and in social media discourse (e.g. Läubli & Orrego-Carmona 2017), especially among experienced profession- als. Nonetheless, MT can also be used as a support, for instance to speed up the translation process for certain types of texts. Similarly, MI, especially with writ- ten output, may also be integrated into CAI tools to alleviate part of interpreters’ cognitive efforts (see §2.2 and e.g. Wang & Wang 2019 for consecutive interpret- ing). A deeper understanding of these technologies may help interpreters em- brace them as valuable support tools. 2.1.3 Supporting technologies for interpreting If technology-mediated and technology-generated interpreting may, for various reasons, encounter the adversity of practitioners, supporting technologies are the group of solutions which conference interpreters may be expected to embrace. This subset of technologies encompasses different types of applications, both hardware and software. The most remarkable examples are tablet interpreting, the sim-consec pen, supporting ICTs for terminology work also used by transla- tors, and bespoke tools for interpreters (CAI). 2.1 Overview and classification of technologies There is growing interest for machine interpreting from business and institutional players,5 and the quality of MI continues to improve. What remains to be determined is how the use of MI affects communication. Fur- ther insight into this aspect of MI-mediated communication may reasonably be expected to impact its adoption by end-users and its perception by interpreters. This has been a concern of interpreting scholars from the beginning, as con- secutive systems applied to dialogic conversations have been explored from a 30 2.1.3.1 Tablet interpreting Tablets are a rather new addition to the interpreter’s toolkit. The still rather scarce amount of research published on this subject has highlighted how tablets are applied in a variety of settings and can be used not only for consecutive in- terpreting (Rosado 2013, Behl 2013a,b, Goldsmith & Holley 2015, Goldsmith 2017, 31 31 2 Information technologies and interpreting Drechsel & Goldsmith 2016), which has been the main focus of the analyses con- ducted so far, but also for assignment preparation and during simultaneous inter- preting (Paone 2016). As for tablet usage during consecutive interpreting, a pilot study conducted in 2015 by Goldsmith and Holley offers a comparative user eval- uation of tools suitable for note-taking during consecutive interpreting, tablet features, and styluses. It highlights how tablet interpreters mostly seek reliability, durability and a good user experience when selecting tablets and tools. Research conducted thus far has identified several advantages provided by tablets over pen and paper for consecutive interpreting and over laptops in the interpreting booth. Tablets offer a “simplified user experience” (Drechsel & Goldsmith 2016: 11) and longer-lasting batteries, can easily fit in the small perimeter of a booth and are quieter than paper and most laptops. These aspects also favour their use for terminology look-up during interpreting. Drechsel & Goldsmith (2016: 17) even go so far as hypothesising that using such a “streamlined device may decrease cognitive load by allowing interpreters to focus on the necessary complements to the task at hand”, although they recognise that research is needed to test such a hypothesis. As with all tools, it is reasonable to assume that training may im- prove users’ command, and tablet interpreting performance as a consequence (Goldsmith 2017: 43). Goldsmith & Holley (2015) identify potential disadvantages: using tablets when moving can be cumbersome, tablets may generate mistrust in clients who are not familiar with this technology applied to interpreting, and they do come with additional stress, costs and a certain learning curve. As for the perceived professionalism, this largely depends on clients’ knowledge of and attitudes towards this technology applied to interpreting. 2.1 Overview and classification of technologies 2.1 Overview and classification of technologies and looking at her notes at the same time. The digital pen also allows for playback speed adjustment, which facilitates rendition and promotes precision. Overall, the advantages of the sim-consec pen have been identified by scholars as “better interpreting performance, which was seen in ‘more fluid delivery, closer source- target correspondence’ (Hamidi & Pöchhacker 2007: 14), greater accuracy, fewer ‘disfluencies’ (hesitation phenomena), greater interpreter confidence, and a more complete rendition” (Orlando 2014). Orlando (2010: 78) adds that “tablet PCs are more expensive and less portable than a pen and a notepad”, which speaks in favour of a digital pen as an alternative to traditional pen and paper support for consecutive interpreting.f Digital pens also offer interesting advantages for the training of future inter- preters. As Orlando (2010: 72) observes, trainers are often faced with the issue of “encouraging their students to develop their own personal [note-taking] sys- tems freely”, while at the same time being unable to “observe these systems in the process of being developed throughout the training”. As a consequence, they are unable to intervene with effective feedback, or to address the issues underly- ing a poor rendition or an ineffective note-taking technique. The possibility of observing the notes while in the making and to link the audio and the video of the notes taken does not only support trainers, but could additionally prove ben- eficial in terms of students’ self-awareness, self-evaluation and self-regulation, and encourage more objective, evidence-based assessment of the learners’ notes. 2.1.3.2 Sim-consecutive and digital pen The digital pen is another technology that can provide interesting advantages both for training and for practitioners. It belongs to the category of mobile com- puting platforms and consists of “a microphone, a built-in speaker, 3D recording headsets, and an infrared camera” (Orlando 2010: 77). Users can write as they would with any other pen, but must use special micro chipped paper onto which data is captured. Since the notes and the recording are synchronised, the inter- preter can then play back the recording from the notes. This technology opens up new possibilities for the profession, giving birth to a new hybrid mode of interpreting known as simultaneous consecutive inter- preting, sim-consec, or consec-simul with notes. In this interpreting mode, the interpreter takes notes as usual using the digital pen, and can then deliver the speech simultaneously while playing back the recording from the pen’s ear set 32 2.2 CAI tools Among the technologies aimed to support the interpreting process, CAI tools have received an increasingly wider share of attention. Unlike technologies which mediate the provision of the interpreting service (see §2.1.1) and can be described as setting-oriented technologies, CAI tools fall into the category of process-oriented technologies (Fantinuoli 2018a). The rationale behind their cre- ation is namely that of providing targeted support for each of the interdependent sub-processes involved in interpreting, thanks to a dedicated architecture and to features optimised for the interpreting workflow. 2 Information technologies and interpreting 2 Information technologies and interpreting CAI tools have been developed with the goal of addressing said limitations, targeting the specific needs of interpreters’ terminology work. During the prepa- ration phase, for instance, CAI tools can help “moving cognition upstream” (Stoll 2009) by helping interpreters extract terminology from ad-hoc corpora cre- ated through the automatic collection of texts on a particular subject, automat- ically annotating speech transcripts provided by clients, and looking up trans- lation equivalents, definitions and other linguistic and extra-linguistic informa- tion from several sources simultaneously. They can help interpreters increase their productivity by speeding up glossary creation and terminology memorisa- tion, facilitate information sharing through dedicated import and export tools, and provide seamless access to the terminology databases compiled during the preparation phase. Due to their relevance for the present work, CAI tools are discussed in detail in the following sections. 2.1.3.3 Supporting ICTs for terminology work Interpreters can make use of a host of technologies to support their terminology work before, during, and after the assignment. These supporting technologies include general-purpose tools, tools also used by translators, and bespoke tools for interpreters, i.e. CAI tools. For instance, interpreters can use text-processing or database software (e.g. MS Word or Excel) to compile their glossaries. While not developed specifically for interpreters, these solutions are very flexible and can easily be adjusted to inter- preters’ individual needs. CAT tools and other resources and solutions commonly used in translation such as TETs and TMSs can also be applied to interpreters’ terminology work. These tools can be used to process preparation documents and extract terminology to compile domain-specific glossaries. Online databases and electronic dictionaries can be used to look up terminology, while concor- dancers can be used to explore the use of terms in context. As pointed out in §1.3.2, however, these tools are not optimised for interpreters, and present a se- ries of limitations. 33 2.2.1 Definition Due to the multitude of tools and resources which interpreters can use as support, there has been some inconsistency in the use of the term CAI (Will 2020: 46). As remarked by Will (2020), the term CAI tool can in principle be used broadly to indicate any kind of supporting technology applied to interpreting. Some authors (e.g. Fantinuoli 2018a, Prandi 2017, 2018) have discussed as CAI tools only those bespoke tools explicitly developed to address interpreters’ needs. Other authors, on the other hand, use the term CAI tool in broader terms. For instance, Costa et al. (2014a) also include unit converters; Cavallo & Ortiz (2018) also mention “note-taking software, audio and video conference systems or learning platforms” (see also Will 2020: 47). Other authors have even used the term to indicate other categories of technologies applied to interpreting, including “terminology aids, such as laptops, notebooks, small handheld PDAs (Personal Digital Assistants) 34 2.2 CAI tools 2.2 CAI tools or similar instruments with Internet accessibility that may facilitate interpreters’ work” (Tripepi Winteringham 2010: 90). To provide some clarity in the use of the term, Will (2020: 47) proposes to de- fine non-bespoke tools targeting the pre- and post-process phases as “secondary CAI tools”. This term thus indicates “any computer-based applications to search, compile and record terminologically relevant structures for a subsequent inter- pretation” (Will 2020: 47). Secondary CAI tools are therefore mainly derived from computer linguistics and can also comprise corpus managers and corpus analy- sis tools, semantic networks (e.g. Babelnet, see Navigli & Ponzetto 2012, Navigli et al. 2021), terminology extraction tools, etc, i.e. the technologies discussed in §1.3. In addition to secondary CAI tools, “primary CAI tools”, i.e. tools explicitly targeted to interpreters, address the specific ergonomic and cognitive require- ments of the in-process phase. Among primary CAI tools, Will (2020) identifies a subcategory defined as “integrated” CAI tools, i.e. those tools which represent a complete workstation for interpreters, integrating both primary and secondary CAI functions, similarly to what CAT tools represent for translators (although in principle, interpreters could also use CAT tools to cover many subphases of their workflow). For the purpose of the present study, I henceforth use the term CAI tools to indicate only those tools which Will (2020) defines as primary CAI tools. 2.2.2 Overview of CAI tools Rütten’s (2004) model of a CAI tool (see §1.2.3.2) envisages a software programme capable of providing support for each phase of the workflow and of optimising time-consuming operations. Not all tools currently available on the market, how- ever, represent a complete technical implementation of the model. In their rela- tively short history, CAI tools have gone through an evolution from simple ter- minology management tools to all-around solutions for conference preparation, even going beyond what had been envisioned by the first scholarly speculations on the topic. In providing an overview of the tools available, and of those no longer sup- ported, I follow Fantinuoli’s (2016: 44) classification of (primary) CAI tools ac- cording to their “architecture and functionality spectrum”. This overview will serve as the basis to motivate the selection of the tool employed in the present study. In addition to a zero generation of tools, comprising non-bespoke-tools appli- cable to interpreting (i.e. secondary CAI tools), CAI software can be subdivided 35 2 Information technologies and interpreting into two generations (Fantinuoli 2016, 2018a). This categorisation should, how- ever, not be interpreted in strictly chronological terms. Rather, it mirrors the degree of sophistication of the individual applications. The first generation of tools consists of programs which offer little more than terminology management for interpreters (Hansen-Schirra 2012: 212). This cat- egory includes Terminus6 (Wintringham 2009), Glossarmanager,7 Glossary As- sistant (Martin 2014), Interplex8 (Sand 2003, 2010) and flashterm Interpreter 9 (Eisenrieth Dokumentations GmbH 2010). These tools can be used to organise terminological entries by topic, client, event and other criteria inside multilingual glossaries, the basic components of these tools’ architecture. The only exception is flashterm Interpreter. This tool is not term-based, but rather concept-based: the terminological entries can be assigned to categories and tagged individually, and filtered to create ad-hoc glossaries. Some tools, such as Interplex, allow for the creation of multiple databases, while in most CAI tools (also of the second gener- ation) all terminological entries are contained in a single database. Even though the basic functions offered by first-generation tools are similar (glossary creation and storage, basic import and export functions, a feature for glossary printing), some of the more recent tools provide additional functions that expand the range of elaboration by prospective users, such as the creation of ad-hoc term lists for quick terminology memorisation in flashterm Interpreter, or the inclusion of im- ages in the terminological entries. 6http://www.wintringham.ch/cgi/ayawp.pl/t/terminus (Accessed: 01.11.2021) 7https://www.glossarmanager.de (Accessed: 01.11.2021) 8http://www.fourwillows.com/interplex.html# (Accessed: 01.11.2021) 9https://www.flashterm.eu/en/ (Accessed: 01.11.2021) 2.2.2 Overview of CAI tools These tools usually also provide a feature for quick terminology look-up, both within single glossaries and/or in the whole database. To the second category of CAI tools can be ascribed applications that go be- yond providing a structured way of organising terminological resources for in- terpreters. These solutions aim to offer a complete workstation for conference preparation and to support interpreters along the different phases of their work- flow. Second-generation CAI tools typically provide the terminology manage- ment and glossary creation features present in first-generation solutions, but ad- ditionally offer further functions for workflow optimisation. Some examples are document integration and terminology extraction, quick search for translation equivalents and other linguistic and extra-linguistic information across multiple sources (such as online dictionaries and terminological databases), memorisa- tion of relevant terminology and terminology look-up both during the prepara- tion phase and on assignment thanks to the integration of search algorithms to 36 2.2 CAI tools 2.2 CAI tools speed up the querying process and effectively deal with time constraints. Second- generation CAI tools thus represent, to a large extent, the actualisation of Rüt- ten’s (2004) model. To this category one can ascribe Intragloss (discontinued), and Interpreter’s Help10 with its offline companion tool BoothMate.11 Intragloss was only available for the Mac operating system, while Interpreter’s Help is web- based and provides multi-platform offline applications. Intragloss allowed for quick terminology look-up across multiple online sources within the application itself, thus eliminating the need to switch between the tool and, for example, a search engine. Interpreter’s Help is a community-oriented tool where glossaries can be accessed both through the online platform or offline from the compan- ion app BoothMate. In addition to the above-mentioned features common to all second-generation tools, Interpreter’s Help also provides functions for managing assignments, sharing glossaries and assignment details online with colleagues, creating ad-hoc flashcards that allow users to memorise translation equivalents and other data, but also features that promote collaboration within the interpret- ing community: a public glossary database (Glossary Farm), a feature for upload- ing practice speeches, asking and providing feedback, and a community section with useful links and resources. BoothMate enables users to query their database thanks to a progressive search function that simply requires users to start typing the term they are looking for, and incrementally reduces the number of results presented. 10https://interpretershelp.com (Accessed: 01.11.2021) 11https://interpretershelp.com/boothmate (Accessed: 01.11.2021) 12For a detailed and up-to-date description of InterpretBank’s modules and functions, the reader may refer to https://interpretbank.com/site/index.html#features and to the user guide at: https: //www.interpretbank.com/site/docs/index.html (Accessed: 01.11.2021). 2.2.2 Overview of CAI tools While Fantinuoli distinguishes between two generations of CAI tools, the tech- nological advances provided by artificial intelligence hold great potential for fur- ther development of such tools towards fully-fledged virtual boothmates. This third generation of CAI tools, exemplified by InterpretBank (Fantinuoli 2012, 2017a), could represent the next step in human-machine interaction in the field of CAI. While first-generation tools mainly only provide a more rational and streamlined infrastructure for terminology work in interpreting with the bulk of the information processing still done by users, and second-generation tools take a step further by automatising and speeding up individual operations tradi- tionally carried out by the interpreter, third-generation, AI-enhanced CAI tools could provide a framework for optimising each step of the interpreting workflow through technology. When working with these tools, interpreters can focus on refining the work done by the tool, with the bulk of conference preparation hav- ing been pre-processed by the software. Even though this goal has not been fully achieved yet, and much will depend on the progress made by machine learning 37 37 2 Information technologies and interpreting and AI, the CAI tool InterpretBank already presents some features that go be- yond what is currently offered by second-generation tools and exemplifies third- generation CAI tools. For its innovativeness, I chose InterpretBank as the tool used in the study. In the following section, I therefore describe this CAI tool in detail. 2.2.3 InterpretBank InterpretBank combines a central Edit Modality – Rütten’s (2007) central start- ing page (see §1.3.1) – with three modules each dedicated to a stage of conference preparation: the Document Modality, the Memo Modality, and the Booth Modal- ity.12 From the Edit Modality, users can access the database, update old glossaries and create new ones, merge glossaries and move terms to other glossaries, auto- matically search for translation equivalents and definitions both offline, in the integrated resources, and online in pre-selected web pages. InterpretBank is cur- rently the only tool that offers solutions for creating glossaries starting from a single term thanks to the integration of corpus creation tools, and for auto- matically translating the whole glossary. The user’s intervention is still required to assess the results of automatic translation, but such a feature can help opti- mise interpreters’ preparation under time pressure by enabling them to focus on higher-level processing instead of devoting most of their time to manually compiling glossaries ahead of the event. These functions might prove particu- larly useful when conference interpreters are faced with a scarcity of preparation materials to be used as a basis for terminology extraction and key topic identifi- cation. While glossaries are organised within a single database, it is possible to create subglossaries and group glossaries together with tags, which provides an additional layer of structure and customisation. When interpreters do receive preparation documents from their clients or have found relevant resources on the internet, the Document Modality can assist them in making the most of the available resources. InterpretBank is one of the few tools that support users in extracting terminology from preparation documents, allowing for an application of the corpus-driven interpreter preparation (CDIP, see §1.2.3.2) put forward by Fantinuoli (2006). Like Interpreter’s Help and Intra- gloss, InterpretBank makes it possible for users to select relevant terms while reading through conference materials or other sources and add them to the glos- sary. When working with parallel texts, users can mark term pairs in parallel 38 2.2 CAI tools Figure 2.1: The Edit Modality in InterpretBank Figure 2.1: The Edit Modality in InterpretBank in order to quickly add them to their glossary, a feature also available in the two second-generation CAI tools previously mentioned. 13https://eur-lex.europa.eu/ (Accessed: 24.07.2022) 2.2.3 InterpretBank A unique feature of- fered by InterpretBank is the automatic terminology extraction feature, which can be used to mine terminology either from the whole set or from subsets of available sources, or from single files. The SMART terminology extraction func- tion learns from the user’s behaviour to improve terminology extraction results in subsequent TE tasks. The tool helps effectively process speech transcripts by allowing for automatic document annotation, a function which can represent a useful emergency strategy when interpreters receive written speeches to be read aloud shortly before the conference starts. An additional feature which can prove useful for interpreters dealing with EU-related topics is the possibility of downloading the texts of EU legislation directly from EUR-Lex13 into Interpret- Bank. Interpreters can mine these documents for terminology as with any other document added to the current glossary. The Edit and the Document Modality are therefore strictly intertwined: users can add entries to the glossary that they are currently working on from the Documents modality, and at a later time ex- plore how terms are used in the context of their preparation documents from in order to quickly add them to their glossary, a feature also available in the two second-generation CAI tools previously mentioned. A unique feature of- fered by InterpretBank is the automatic terminology extraction feature, which can be used to mine terminology either from the whole set or from subsets of available sources, or from single files. The SMART terminology extraction func- tion learns from the user’s behaviour to improve terminology extraction results in subsequent TE tasks. The tool helps effectively process speech transcripts by allowing for automatic document annotation, a function which can represent a useful emergency strategy when interpreters receive written speeches to be read aloud shortly before the conference starts. An additional feature which can prove useful for interpreters dealing with EU-related topics is the possibility of downloading the texts of EU legislation directly from EUR-Lex13 into Interpret- Bank. Interpreters can mine these documents for terminology as with any other document added to the current glossary. The Edit and the Document Modality are therefore strictly intertwined: users can add entries to the glossary that they are currently working on from the Documents modality, and at a later time ex- plore how terms are used in the context of their preparation documents from 39 2 Information technologies and interpreting the glossary entries. 2.2.3 InterpretBank The synergy between the two modalities also emerges from their presentation on the user’s interface: the Documents modality appears as an additional window juxtaposed to the glossary currently being processed by the user. After compiling terminological databases, interpreters can receive support in memorising event-related terminology in the Memory Modality, which creates virtual flashcards from the current glossary. Users can choose between manual and automatic presentation. The manual mode can prove particularly useful in the early stages of terminology memorisation, during which it can be useful for the user to mark the terms as “known” or “forgotten”. Following the principle of spaced repetition, the tool presents the user with the forgotten terms at the end of the practice session, so that said terms can be practised once again. The automatic flashcard presentation can be used to test one’s own reaction time to the term/stimulus, since presentation speed can be varied, but can also repre- sent an easy way to quickly review the terminology right before the assignment begins or during breaks. A similar option is also present in Interpreter’s Help, which provides the additional option of creating ad-hoc flashcard decks. Inter- pretBank offers two additional features: an option to view examples of the terms in the context of preparation documents, and the possibility of having the terms read aloud. This feature, which makes use of speech synthesis technology, is par- ticularly suited to the needs of interpreters, who process oral stimuli (with the exception of sight and signed language translation). Additionally, it can prove particularly useful for highly non-phonemic languages such as English. The phase of interpretation proper is covered by the Booth Modality, which completes the architecture of the tool. A similar function is also offered by In- terpreter’s Help companion app BoothMate. Like BoothMate, InterpretBank’s Booth Modality can also be used offline. In this modality, interpreters can quickly search for terminology in their glossaries. Unlike in the Edit Modality, in which users can only work on one glossary at a time, in the Booth Modality it is possi- ble to activate multiple glossaries which can all be queried simultaneously (the “active glossary”). Additionally, users can also choose to have InterpretBank per- form the search inside the whole database or even exploit external resources as an emergency strategy. The default search mode in the Booth Modality is the dynamic search, which progressively reduces the number of results with increasing input. 2.2.3 InterpretBank It is, however, also possible to manually start the query. A host of additional options, such as the fuzzy search and the “case and access insensitive” search, facilitate the lookup process during interpreting. 40 2.2 CAI tools Figure 2.2: The Booth Modality in InterpretBank Figure 2.2: The Booth Modality in InterpretBank Perhaps the most advanced feature offered by InterpretBank is the integra- tion of ASR technology. While still marketed as an experimental feature in the freelance version, the integrated ASR option represents the next step in support- ing interpreters through technology. The goal of ASR integration into the tool is that of presenting interpreters with elements typically deemed difficult to inter- pret, “problem triggers” (Gile 2009) such as numbers, specialised terminology, and named entities. This represents a step towards the creation of CAI tools serving as artificial boothmates and providing support without requiring exten- sive human-machine interaction. At the moment of writing, InterpretBank offers ASR support for specialised terminology and numbers. A similar feature is also offered for consecutive interpreting: users can create an “artificial notepad” al- lowing them to visualise the automatically-generated transcription of the source speech, looking up term equivalents with a click, automatically convert units, and share the ASR session with a colleague. Like Interpreter’s Help, InterpretBank also offers browser-based features that can be accessed through a private Cloud account. In the cloud, users can upload, edit, share and download glossaries, and access the memorisation and look-up features from the internet, an option particularly appreciated by interpreters who 41 2 Information technologies and interpreting prefer to be device-independent. The cloud options can also represent a practical alternative to using traditional import and export features to share glossaries as text or Excel files. 2.3 Conference interpreters’ use of supporting technologies Given the role of terminology acquisition for the interpreting process, the dis- tinctive features of terminology work in interpreting, and the availability of ded- icated software for interpreters, scholars have taken interest in exploring in- terpreters’ practices when compiling terminological databases for assignment preparation and consulting them during interpreting. The surveys conducted over the years paint a picture of conference inter- preters’ level of computerisation, use of technological support pre-, peri-, in- and post-process (both in terms of hardware and software), and of their needs in terms of dedicated solutions for terminology work. These surveys provide valu- able information on the extent to which CAI tools and other supporting ICTs are actually employed by interpreters in their professional workflows. 2.3.1 Conference interpreters’ level of computerisation When discussing the individual solutions chosen by interpreters as a support for their terminology work, their overall level of computerisation and their adoption of computer support in the various phases of interpretation is worth considering, as it may help put their practices into perspective.i The first inquiry into conference interpreters’ preparation practices dates back to 1992. Moser-Mercer (1992) distributed a questionnaire among 260 interpreters, all AIIC active members with English among their working languages. Among the 130 respondents, the level of “computerisation” (Moser-Mercer 1992: 510), i.e. the percentage of respondents who own or have access to a computer, was al- ready quite high (62.3%), and seemed to correlate positively with the years of professional experience and the number of interpreting days per year. Moser- Mercer speculated that this trend could be due to a higher level of specialisa- tion among more experienced professionals, which would in turn promote com- puter use for terminology work. A survey conducted by Drechsel (2004) supports Moser-Mercer’s (1992) findings, with only two respondents not using the PC in any of the phases of an interpreting assignment. It should be noted, however, that the sample was quite small and only representative of the German market (46 professional interpreters). A more nuanced picture of the use of computers as 42 2.3 Conference interpreters’ use of supporting technologies technological support for terminology emerges when considering the individual phases of the interpreting assignment. 14Berber-Irabien (2010) looked at search engines, online dictionaries and encyclopedias, terminol- ogy databases, electronic norms manuals, parallel texts, do-it-yourself corpora; technologies now considered obsolete – pocket electronic dictionaries, CD-ROM dictionaries, paper ency- clopedias – and terminology databases, as well as setting-oriented technologies (Fantinuoli 2018a: 155) such as remote interpreting, videoconferencing and telephone interpreting. 2.3.2 Computer and software use for assignment preparation Moser-Mercer’s (1992) survey did not venture into a detailed analysis of the tools used for preparation. However, it provides valuable insight into which types of solutions interpreters used at a time when both CAT and CAI tools were starting to be developed. The survey reports that 51.3% of those who used a computer medium chose terminology databanks (40%), databases (35.5%) and spreadsheet software (31%) to compile and organise terminological information and docu- mentation. The main sources for documentation ahead of the event were event documents and bilingual dictionaries, and a widespread use was made of termi- nology lists compiled personally or supplied by the clients. External databases were used only to a limited extent. These findings support the notion that termi- nology work in interpreting (as in translation) is performed ad-hoc and strictly interlinked with the individual assignment. The survey does, however, provide us with two interesting findings. At the time of her study, 67% of respondents showed an interest in terminology management software, while 72.7% expressed an interest in exchanging terminology in electronic form. This seems to speak in favour of a high degree of computerisation for preparation work. About a decade later, Valentini (2001, 2002) conducted a similar survey on a sample of 130 pro- fessional conference interpreters, both freelancers and staff, 40% of which with more than 20 years in the profession, working mostly (94.5%) in the simultane- ous mode. In the preparation phase, a third of her sample used electronic termi- nology databases (37% used paper glossaries), and 11.3% conducted terminology extraction (TE) automatically. The most widespread programs for terminology work were Microsoft Word or similar (67.7%), followed by Excel or similar (11.4% of respondents). Users of these solutions generally showed a higher degree of satisfaction than users of dedicated terminology software.t A similar picture of the software used by conference interpreters emerges from the t-survey by Zielinski & Ramírez-Safar (2005): of the 81 interpreters involved, 63% performed terminology work; 66.1% used Word or Excel tables for termi- nology management, while 53.2% also used terminology management systems (TMS). In this survey, no mention of tools developed specifically for interpreters (see Chapter 3) is made, even though some of them were already available at the time (e.g. Interplex, LookUp/DolTerm, and TermDB). As for TE, the trend is sim- ilar to the study by Valentini (2001, 2002) (13.9% informants used terminology extraction tools). 43 2 Information technologies and interpreting 2 Information technologies and interpreting Bilgen (2009) conducted a small-scale survey on a sample of experienced CACI members, the Canadian branch of AIIC. Bilgen’s (2009) findings are along the lines of the surveys previously mentioned: 70% of respondents used word pro- cessors or spreadsheets for terminology management, while 85% recorded terms on paper (multiple answers were possible). On-screen work was preferred before and after conferences. Interestingly, 77% of the sample never used a TMS, and no mention was made of interpreter-specific software. This contrasts with Zielin- ski & Ramírez-Safar’s (2005) findings, but in their sample most interpreters also worked as translators. Bilgen (2009) does not specify whether this is the case in her survey. Only one respondent reported using a TMS; another interpreter a database management programme. None of the surveyed interpreters mentioned interpreter-specific software. In a survey conducted in 2010 on a sample of 222 conference interpreters (mostly AIIC members and/or working for the EU or the UN), Berber-Irabien (2010) found that ICTs were mainly used in the preparation phase, and only scarcely (15%) for post-assignment debriefing. Terminology databases (such as IATE) were used by 88% of her sample. Disappointingly, terminology manage- ment software, and in particular programs for interpreters such as Interplex or LookUp, were mentioned in the overview of tools for interpreters, but were not included in the questionnaire categories. Of all ICTs considered14, search engines were those used with more frequency (in 38% of cases). While these findings overall support previously identified trends, no distinction was made between setting-oriented and process-oriented technologies, a limitation which hinders the direct comparison with previous studies. In 2012, Wagener conducted a survey on 102 interpreters, mainly German. In line with previous survey work, she found that most interpreters in her sample used general-purpose tools such as MS Word and even paper, while MS Excel was less popular (less than 30% of the sample). Only 26% of her sample used interpreter-specific tools. Regrettably, it was not mandatory for her respondents to include the names of the tools used. An even smaller percentage (13%) used terminology software. Jiang (2013) and Corpas Pastor & Fern (2016) conducted the two most recent surveys available on interpreters’ terminology practices. Jiang’s (2013) survey fo- cused on conference interpreters’ glossaries for SI. 2 Information technologies and interpreting The response rate was very 44 2.3 Conference interpreters’ use of supporting technologies high at 21%, and the large sample (𝑛= 476) mostly consisted of AIIC members, two thirds of whom had more than 15 years of professional experience. Jiang (2013) analysed several aspects of glossary creation for SI, such as the items in- cluded in the glossary, the sources used and the frequency of glossary creation. A very interesting finding was that the preferred medium remained loose paper (57.6%), followed by Word glossaries (55.7%). Most interpreters seemed to com- bine digital glossaries with a printout, so it can be assumed that both were taken into the booth. A smaller percentage of the surveyed interpreters (15%) used “glos- sary software”. An even smaller number of respondents use other tools such as Trados Multiterm and Interplex. An interpreter’s glossary consists mainly of a table or a list with glossary items (93.7% of responses). Given the high response rate and large sample, Jiang’s (2013) findings can be considered quite represen- tative of the glossary creation practices of experienced, professional conference interpreters working in the simultaneous mode until some years ago. Finally, Corpas Pastor & Fern (2016) aimed to explore interpreters’ use of “tech- nology tools” in the various interpreting modes and settings. For the purpose of this study, I will take into consideration the responses regarding simultaneous interpreting (a total of 92 out of 133). The respondents used mainly bilingual dictionaries, digital glossaries and thesauri, but also databases and term banks during preparation. Despite the rather small sample and the – in my opinion – somewhat arbitrary interpretation of data by the authors, their findings align with those of similar surveys. The surveys presented above span over 25 years. The sample sizes vary, as does the specific focus of each survey, although they all relate to technology use during preparation. Even though these surveys are only comparable to some ex- tent, they provide a picture of technology use for terminology work prior to the (simultaneous) interpreting task. On average, at least until 2016, conference in- terpreters tended to use computers for preparation, consult digital terminology databases and compile digital glossaries, mostly in the form of terminology lists and tables. General software prevailed, with a preference for Word or similar text processing programs, followed by Excel or other spreadsheet applications. 2 Information technologies and interpreting The interest in software solutions for preparation identified by Moser-Mercer (1992) seems to have led to a widespread use of various technologies for assign- ment preparation. Despite these developments, which – there is reason to be- lieve – will continue in the years to come, paper has not yet disappeared as an additional support medium for glossary work. On the contrary, paper seems to remain an integral component of interpreters’ terminology work and is often combined with the afore-mentioned text-processing programs. To a limited ex- tent, terminology management systems such as Trados Multiterm are also used, 45 2 Information technologies and interpreting although it seems that especially interpreters who also work as translators use these kinds of tools. As emerges from the t-survey (Zielinski & Ramírez-Safar 2005), automatic terminology extraction with dedicated TET seems to be a rare practice among interpreters, who prefer manual TE. Considering the speed at which technology progresses, and the acceleration, also due to the COVID-19 pandemic, of what has been defined as “the technological turn” in interpreting (Fantinuoli 2019), these results are in all probability no longer representative of the situation in 2022. It can be reasonably assumed that the ratio of interpreter- specific technologies for assignment preparation compared to more “traditional” solutions has now shifted more towards the first. Regrettably, to the best of my knowledge, no survey on interpreters’ use of technology pre-process has been conducted since 2016. 2.3.3 Computer and software use peri- and in-process If the paper medium has not yet been entirely supplanted by the electronic one in the preparation of glossaries (Bilgen 2009, Jiang 2013), it is nonetheless unde- niable that there has been a progressive penetration of technological support in the interpreting profession. As emerged from the surveys, however, at least until some years ago, the technological support inside the booth has remained limited. In Valentini’s (2001) survey, dating back almost 20 years, only 27.6% of survey participants stated that they use a computer in the booth. The largest percentage of booth computer users in her sample worked for the European institutions. At EU institutions, “the greater availability of infrastructure [...] facilitates and en- courages the penetration of information technology compared to other national and international institutions” (p. 162). A similar result emerged from Bilgen’s (2009) and Berber-Irabien’s (2010) surveys, which found that computer and soft- ware use was more popular before and after the assignment, while limited inside the booth. An encouraging piece of data that emerges from the survey conducted by Valentini (2001, 2002) is that 66.1% of respondents considered ICT support as potentially useful for SI. This is, however, a rather general claim and tells lit- tle about which ICTs are being referred to, nor which sub-processes they could support. Interestingly, 43.4% of the sample deemed terminology look-up with technological support “very useful”. Bilgen’s (2009) sample was also largely in favour of computer use in the booth, and half of the sample used one. In Valentini’s (2001, 2002) survey, no distinction was made between peri-pro- cess and in-process with reference to technology use in the booth. Drechsel (2004) did make this distinction. In the small sample he considered, 16 inter- preters used a PC peri-process, 16 to help their boothmates and 13 to look up 46 2.3 Conference interpreters’ use of supporting technologies terminology themselves; 14 respondents used laptops to work on documents and 16 for follow-up work during the conference. As for the medium, 11 respondents out of 46 used paper glossaries in addition to digital glossaries. The most recent inquiries in the terminology management practices of con- ference interpreters also show that, although with some limitations, the techno- logical support has increasingly made its way into the interpreting booth. For instance, Berber-Irabien (2010) found that terminology databases are the pre- ferred technology tool for accessing terminology during simultaneous interpret- ing. 2 Information technologies and interpreting 2 Information technologies and interpreting Because of the specificity of the sample, these findings cannot be generalised. Furthermore, the survey dates back to 2007 and is in all probability no longer representative. There is a need for an up-to-date inquiry into interpreters’ level of computerisation and especially of their use of software in all phases of an inter- preting assignment. Except for this survey, CAI tools are yet to be systematically integrated into such inquiries. Because of the specificity of the sample, these findings cannot be generalised. Furthermore, the survey dates back to 2007 and is in all probability no longer representative. There is a need for an up-to-date inquiry into interpreters’ level of computerisation and especially of their use of software in all phases of an inter- preting assignment. Except for this survey, CAI tools are yet to be systematically integrated into such inquiries. Despite the limitations of these surveys, they suggest a few trends in confer- ence interpreters’ use of terminology management software and other ICTs for terminology work, as well as in their level of computerisation. The average mod- ern conference interpreter seems to make extensive use of ICTs in the pre-process phase, bringing a laptop into the booth, often together with some information on paper or a printed glossary, and conducts post-conference work on a laptop, mostly in the form of terminology update. The most popular technologies used to compile glossaries are general-purpose text-processing or database software, such as MS Word or Excel. Terminology extraction and terminology manage- ment tools are also used, especially by interpreters who also work as translators. However, the use of these tools is quite limited among interpreters, as is the use of CAI tools, although they seem to be more popular. This may suggest that there is still room for further optimisation and tailoring of these tools to interpreters’ needs. The widespread use of paper, despite the possibilities offered by modern tech- nology for a paperless booth, may also point to a limitation of current CAI tools. It is possible, as Wagener (2012: 77) observes, that digital glossaries and paper are used for different purposes. The first support a kind of terminology work which aims at creating terminology resources that can be helpful also beyond the individual event. Having long-lasting terminological resources saves prepa- ration time and makes terminology work more sustainable in the long run. 2 Information technologies and interpreting Paper, on the other hand, may represent the ideal support for short-term terminologi- cal needs, i.e. to provide interpreters with the necessary aide-memoires for those terms which tend to be forgotten. CAI tools currently do not seem to address this specific need. 2.3.3 Computer and software use peri- and in-process Corpas Pastor & Fern (2016) also came to a similar conclusion, identifying bilingual dictionaries and personal glossaries as the preferred digital tool used by interpreters to search for terminology during the task. Nonetheless, most confer- ence interpreters seem not to rely on tools created specifically for interpreters. Corpas Pastor & Fern (2016) observe that “regarding technology tools during an interpreting task, most of our simultaneous interpreters do not count on help from technology resources. The only tools used in specific circumstances during an assignment are bilingual dictionaries, glossaries and, in some cases, web-based resources” (Corpas Pastor & Fern 2016: 26). Wagener (2012) had also found a widespread use of online dictionaries, search engines and other search tools in the booth, pointing to terminology work also happening peri- and in-process and not only pre-process. On the other hand, she had found a very limited use of terminology management systems in the booth, similarly to what had been postulated by Stoll (2009) and Rütten (2011) (see Wa- gener 2012). However, the vast majority of her respondents bring a technological support into the booth, be it a laptop, a netbook, a smartphone or a tablet, almost always coupled with handwritten or printed documents (90% of surveyed inter- preters). To date, the only inquiry into conference interpreters’ use of CAI tools for simultaneous interpreting is the survey conducted by the Sprachen und Dol- metscher Institut (SDI Munich) in 2007. The sample was made up of 135 AIIC members and conference interpreters members of other associations. Unfortu- nately, the survey was conducted in German, limiting the representativity of the sample. As for computer use, 68% of the sample brought a PC in the booth and 41% used it during interpreting, while 26% preferred booth-specific software. The participants who did not use CAI tools mainly used other tools, a good percent- age also “during interpreting” (59%). It is unclear whether “during interpreting” (“während des Dolmetschens”, SDI München 2007: 28) should be intended as during delivery itself, or whether this could also refer, for instance, to helping the boothmate with terminology look-up during their turn. A surprising finding, which would deserve further exploration, is that 41% used self-developed tools. 47 2.3.4 Attitudes towards CAI tools Having considered the current state of CAI tool development, the specific inter- preter needs and the potential of ASR for an even more efficient interpreter-tool interaction in-process, it is somewhat surprising that CAI tool usage is still rel- atively limited, as emerges from the surveys presented in §2.3.3. While provid- ing a snapshot of interpreters’ terminological practices at the moment of data 48 2.3 Conference interpreters’ use of supporting technologies collection, these surveys also bring to light some of the motivations behind the limited inclusion of CAI tools in interpreters’ workflows. While some negative attitudes have subsided over time, when asked to comment on their tool choice or to provide their opinion on the adoption of computers and CAI tools, conference interpreters seem to share some common points of criticism. Moser-Mercer (1992: 512) observed how the only tools that “met almost all re- quirements stipulated by interpreters” were Term-PC and SDL Multiterm. Both programs allowed for multilingual glossary creation, quick terminology retrieval, free definition of entry structures and options for glossary import, export and printing. Since then, several CAI tools were developed. Their use in the booth is, however, still considered problematic both by practitioners and researchers. In discussing the results of her survey on computer use in the interpreting booth, Valentini observed that terminology look-up while “on air” would only be pos- sible for redundant terminology, i.e. when terms are repeated more than once within a speech. In her opinion, interpreters would prefer other strategies such as simplification through paraphrasing or the use of hypernyms to the more dis- tracting search in their terminological resources, since clear and close rendition of the source message is a more prominent quality factor than terminological accuracy. The distraction potential of CAI tools seems, indeed, to be a common point of criticism on CAI tools and, more generally, of terminology look-up during interpreting, as emerges from the SDI (2007) questionnaire and Bilgen’s (2009) survey. In the first, the distracting effect (“ablenkende Wirkung”, SDI München 2007: 29) of terminology software is mentioned by 6% of the 82 interpreters not using such solutions. The main reasons for the exclusion of such software from interpreters’ toolkit are however the lack of necessity (57%), because more tradi- tional solutions are preferred, the lack of knowledge about such tools (35%) and the fact that these tools were considered unconvincing or not yet sophisticated enough (22%). 2 Information technologies and interpreting 2 Information technologies and interpreting To explain the continued presence of paper as additional support in the booth, even considered the preferred medium used for interpreters’ glossaries in her survey, Jiang (2013) advanced the hypothesis that interpreters’ may jot down essential glossary items on loose paper sheets, which would make terminology access immediate. This is a necessity, since “accurate and just-in-time retrieval of glossary items is vital, especially at technical meetings” (Jiang 2013: 90). While this is true, Jiang seems not to take into consideration the quick search function in Word or Excel files and the progressive search function available in some CAI tools when stating that “physically turning a page is probably simpler and faster than having to access a given item or page in a sizeable computer file or database” (Jiang 2013: 91). Rather, the choice to include paper may be more of a strategic nature (see §2.3.3). The potential disadvantages of CAI tools for terminology look-up have also been discussed in several publications on the application of technologies to con- ference interpreting. While CAI tools should theoretically “represent the most effective information interface when interpreting” (Tripepi Winteringham 2010: 90), the feasibility of their use is often questioned. These tools are described as potentially “time-consuming and distracting” (ibid.), or leading to a “loss of con- centration” (Tripepi Winteringham 2010: 91). Berber-Irabien (2010) similarly ob- serves that their use might interfere with listening and concentration. Given the complexity of the interpreting task, while using CAI tools “the interpreter [...] may not have the time or the cognitive ability to look up a word [...] or detect and choose the correct translation [...]” (Fantinuoli 2016: 49), because “typing an unknown word [...] requires an additional time-consuming effort which would affect the already existing efforts that interpreters support during their work” (Tripepi Winteringham 2010: 91). Additionally, “should the right word be found it may not be possible to incorporate it smoothly in speech” (Veisbergs 2007: 80). The surveys and publications discussed suggest a certain interest in technolo- gies for conference preparation and terminology work specifically developed to meet interpreters’ requirements and the need for dedicated tools. While CAI tools’ potential is overall acknowledged, their hypothesised and perceived lim- itations should not be ignored. The disadvantages that these analyses point to mainly concern the application of such tools on the simultaneous, in-process phase. 2.3.4 Attitudes towards CAI tools In view of the recent developments in CAI technology, this last point may not be as relevant as before. Other reasons mentioned are, in descend- ing order of importance, lack of space in the booth, poor customisation and resis- tance to technology because of excessive complexity of use. Similarly, Bilgen’s (2009) survey respondents observed that using the computer in the booth might be distracting for the interpreter and noisy, since the sound of typing could be picked up by the microphone, thus distracting the audience, too. This argument is, however, increasingly losing strength, as many modern laptops come with a silent keyboard, and some interpreters have already successfully replaced their laptops with tablets. 49 2.4 Research on CAI tools 2.4 Research on CAI tools 2.4 Research on CAI tools Moorkens & O’Brien 2016, Steurs et al. 2017), many translators express dissatis- faction towards their tools or frustration in their use. In translation, this feeling of frustration may be even more pronounced than in interpreting, as the use of specific tools is somehow imposed by the client, e.g. the translation agency. Translators’ “cognitive friction” (Ehrensberger-Dow & O’Brien 2015) in using CAT tools manifests in the irritation and negative attitudes expressed in a num- ber of surveys (e.g. O’Brien et al. 2017). Translators’ dissatisfaction concerns com- plex user interfaces, a lack of intuitiveness and user-friendliness and suboptimal navigation within the tool which slows down the workflow. This contrasts with the identification of high working speed and intuitive user interfaces as essential features of translator tools (see O’Brien et al. 2017). In interpreting, a certain amount of frustration may be avoided by simply choosing not to use tools perceived as useless or unsatisfactory, as the choice of supporting technologies is often left to the individual interpreter. This may, in part, be mirrored in the lower percentages of translation tools and CAI tools users among interpreters and the resulting digital divide with translators. How- ever, if other players become involved in the development of CAI tools, e.g. RSI providers including CAI as support in their platforms, it is possible that inter- preters may soon have to use tools they have not personally selected. As re- marked by Hansen-Schirra (2012), most tools for translators and interpreters are yet to exploit computational linguistics and speech technologies to their fullest extent, which results in tools not entirely optimised for their end-users. As for translation tools, greater involvement of interpreters in the development of sup- porting tools may help to mitigate some of their limitations and perceived draw- backs. At the same time, research on the use of CAI tools can help define how they impact the product of interpretation and to what extent they may support or impair the cognitive processes underlying interpreting. 2 Information technologies and interpreting The negative attitudes and the indifference encountered among interpreters towards non-bespoke supporting technologies and CAI tools may be compared to translators’ attitudes towards CAT tools and MT, a technology often integrated into CAT tools. While a considerable number of translators seem to have adopted CAT tools as a staple in their workflows, as suggested by recent surveys (e.g. 50 2 Information technologies and interpreting 2 Information technologies and interpreting Among the first publications which appeared on the topic, several contribu- tions aim to provide prospective users with an overview of the CAI tools to choose from and a set of criteria for their selection. Several articles by Costa et al. (2014a,b, 2015) detail the tools available on the market and highlight the fea- tures addressing conference interpreters’ needs. The authors offer “a tentative catalogue of current language technologies for interpreters, divided into termi- nology tools for interpreters, note-taking applications for consecutive interpret- ing, applications for voice recording and training tools” (Costa et al. 2015: 68). CAI and related terminology management tools, as well as other technologies relevant to interpreting, are discussed together. In Costa et al. (2015), the authors focus on CAI tools, analysing both three of the best-known CAI tools (Intragloss, InterpretBank and Interplex), and other lesser-known or non interpreter-specific terminology management software (i.e. SDL MultiTerm, AnyLexic, Lingo, Unilex and The Interpreter’s Wizard). In their analysis, the authors evaluate the tools chosen for the comparison against a set of 15 features identified as priorities in relation to conference interpreters’ needs. Of the 15 criteria, five are classified as fundamental and ten as secondary. The authors do not clarify the rationale be- hind this distinction, nor do they specify which features are deemed essential and which secondary. Additionally, they seem to base their evaluation on the rather unjustified goal of identifying the most complete solution. While this might be a useful piece of information, the tool so identified may not coincide with the one most suited to the needs of interpreters. In the authors’ analysis, the programme which received the highest score is SDL MultiTerm. Overall, the evaluation ap- pears rather arbitrary, as it excludes features which are particularly relevant to the interpreter’s workflow such as quick terminology look-up or terminology extraction from preparation documents (Fantinuoli 2016). Will (2015) proposes a similar evaluation of CAI tools and lays down a set of criteria for their assessment, with the goal of identifying deficits and proposing potential solutions to be implemented. 2.4 Research on CAI tools The recent developments in the field of CAI have renewed and multiplied the in- terest of the interpreting community for this subject. This is demonstrated by the increasing number of publications on the topic. In addition to the foundational works by Rütten (2000, 2004, 2007), Stoll (2009) and Will (2000, 2007) providing a model of software programs for interpreters (see §1.2.3.2), the publications de- voted to CAI tools address a number of topics and issues inherent to the tools, ranging from a simple presentation and comparison of the available solutions to the first empirical tests conducted on one or more tools. 51 2.4 Research on CAI tools 2.4 Research on CAI tools help functions (1), ergonomics (2) and user-friendliness (3). Unlike Costa et al. (2014a,b, 2015), the author compares Interplex, Terminus, LookUp professional and InterpretBank 3, i.e. CAI tools as they are intended in the present work. The tool evaluation presents the limitation of being subjective, although based on clearly motivated criteria. The author does not involve additional raters. The in- volvement of a large sample of conference interpreters in the evaluation would have been welcome. Some of his suggestions (such as terminology extraction, au- tomatic glossary population with definitions and translation equivalents) have, however, already been implemented into some tools, a sign of the rapid evolution of technology in this field. In addition to these comparisons and evaluation attempts of several tools by scholars, a number of bachelor’s and master’s theses have been devoted to the subject. This suggests a growing interest for these tools among the newer gen- erations of conference interpreters. In particular, De Merulis (2013) focused his analysis on the CAI tool InterpretBank, which he describes in detail. His was one of the first attempts at investigating the impact of CAI tool use on the qual- ity of simultaneous interpreting in terms of terminological accuracy. Similarly, Gacek (2015) reviews several CAI tools and offers a detailed description of Inter- pretBank, defined as one of the most user-friendly tools available to interpreters. He reports the results of two experiments carried out with the tool, for which qualitative data was collected through questionnaires. Users were asked to rate the usability of InterpretBank during a simultaneous interpreting session (first experiment) and after two months of usage (second experiment) and to compare it with printed glossaries prepared in Word and Excel. Data analysis, although limited to a sample of 12 questionnaires, seems to indicate that InterpretBank pro- vides an advantage in terms of efficiency, thanks to features such as the dynamic search function or the Memory Modality. Drawing on trainee interpreters’ interest into new technologies, interpreting research has looked at how CAI tools, in addition to more established technolo- gies such as RSI, could be integrated into the interpreting curriculum. Prandi (2015a,b) carried out an exploratory study on the integration of CAI tools into the Master’s degree in Interpreting at the University of Bologna/Forlì. 2 Information technologies and interpreting Will’s contribution has the merit of bas- ing the definition of the evaluation criteria on a theoretical model of interpreters’ terminological work, which he defines as “dolmetschorientierte Terminologiear- beit (DOT)” [interpreter-oriented terminology work] (Will 2015: 181), and is at the intersection of the terminology models by Wüster (1979) and Gerzymisch- Arbogast (1996), of Moser (1978) and Moser-Mercer’s (1997) process model of si- multaneous interpreting, and of Gile’s (1995) effort model (see §3.5.1). The three principles thus identified as fundamental for the implementation of the model (adequacy and pattern building, simultaneity, and phase-specific usage) in CAI tools are used as a basis for the definition of three key criteria against which the tools should be evaluated: flexible visualisation, comprehensive database and 52 2.4 Research on CAI tools The 12 study participants were divided into two groups: one group attended an intro- ductory class on InterpretBank and then received practical training in the booth (three sessions), while the other attended three introductory classes on the topic and trained with the tool only once. Students worked in pairs and were free to establish how to interact with the software and with the boothmate while us- ing the tool during SI. While the study was of qualitative nature and the sam- ple relatively small, results seem to indicate that CAI tools may help achieve 53 2 Information technologies and interpreting high terminological accuracy. InterpretBank was overall judged as user-friendly. Nonetheless, the study showed that hands-on training with the tool is also neces- sary: the group which enjoyed more booth time with InterpretBank before data collection showed a greater deal of independence in working with the tool dur- ing interpreting. The participants in this group had started to develop their own strategies for looking up terminology within the tool, for instance deciding be- forehand whether terms should be searched by the active interpreter or by the boothmate. They also showed greater agility in coordinating CAI tool usage with more traditional methods for suggesting equivalents for specialised terms, such as writing down terms. While these results cannot be generalised, they seem to provide an argument for the introduction of these increasingly relevant technolo- gies into the interpreting curriculum. On the background of the increasing interest for CAI tools, it is interesting to consider their level of integration into the interpreting curriculum. In 2017, Prandi (2020) conducted a survey involving 25 higher education institution mem- bers of the CIUTI15 network from 15 countries offering training programmes in conference interpreting. Survey results show that CAI tools are mostly seen as a secondary technology to be taught to prospective interpreters, whereas higher priority is given to remote interpreting technologies. With a few exceptions, the responses also pointed to a lack of knowledge on the topic among trainers, which emerged from the confusion around the term “computer-assisted interpreting”, which is often interpreted to include other technologies applied to interpreting, such as RSI or computer-assisted interpreter training (CAIT). From the survey emerges, however, a certain openness towards conducting more research into the subject. 15Conférence Internationale Permanente d’Instituts Universitaires de Traducteurs et Interprètes 16As mentioned by some respondents who may, however, be actually making reference to remote interpreting systems. 2.4 Research on CAI tools 2.4 Research on CAI tools 2.4 Research on CAI tools and best practices rooted in socio-constructivism. As the authors observe, “the introduction of CAI tools in interpreting courses should serve the purpose of exposing students to these solutions, and of providing them with the means to make an informed use of such tools. Thus, an introduction to CAI tools can be beneficial, even if trainees will not reach complete mastery of the tools.” (Fantin- uoli & Prandi 2018: 175). The authors thus put forward ideas on how to structure a training programme focused on CAI tool usage in the booth. The proposal in- volved a theoretical module to introduce students to the rationale behind CAI tool creation and integration into the interpreter’s workflow, and a set of practi- cal exercises of increasing complexity targeting the different sub-skills involved in SI with CAI. This approach aimed to foster the students’ awareness of poten- tial advantages and pitfalls in CAI usage, such as the risk of relying too much on the tool when faced with highly specialised terminology to be rendered into the target language (Prandi 2015b: 53). Another strain of research, currently among the most prolific ones in the field of CAI research, focuses on exploring the impact of CAI tools usage on the quality of SI. The studies conducted so far are mostly pilot studies of exploratory nature, and do not attempt a holistic evaluation of the quality of the interpreting perfor- mance in broader terms, but rather narrow it down to terminological accuracy and omissions. The study by Prandi (2015a,b) included an evaluation of this type. She considered the percentage of terms included in the glossary that had been looked up and correctly identified. She found that the participants did not seem to have difficulties in using the tool to look for terms, as shown by the high percent- age of terms looked up and found. The percentage of terms found and correctly interpreted as per glossary is not as high as the percentage of terms found, which could point to a difficulty in coordinating the look-up effort with the other sub- processes involved in SI. Drawing on Prandi’s (2015a, 2015b) experimental design, Biagini (2015) was the first to empirically test the use of CAI tool glossaries in comparison with paper glossaries during SI. His study also involved a relatively small sample of advanced interpreting students. 2.4 Research on CAI tools The reasons why CAI has not yet been included in the curriculum are often of financial nature16 or organisational, as this would require a restructur- ing of the training curriculum. Very often, the lack of knowledge on the subject among trainers makes it difficult to expand the curriculum to include this topic. There are, however, a few exceptions: notably the Universities of Bologna/Forlì, Mainz/Germersheim, Innsbruck and Heidelberg. These institutions either offer dedicated courses on CAI tools or organise workshops on the subject. The train- ing usually involves a mix of theory and practice, and presents students with several tools. Suggestions for the inclusion of CAI tools and other interpreting technologies (RSI and MI) in interpreter training are offered by Fantinuoli & Prandi (2018), who proposed a training programme providing recommendations 54 2 Information technologies and interpreting 2 Information technologies and interpreting More recently, scholars have started investigating the potential of integrating ASR into CAI tools, a development which would signify a step closer to a third generation of AI-enhanced CAI tools. The goal of integrating an ASR system into a CAI tool is that of leveraging speech recognition to provide interpreters with live support during SI, with a view to reducing interpreters’ active interaction with the machine. As Fantinuoli (2017b: 25) observes, “the main drawback [of traditional CAI tools] is that the database is queried manually, adding more cog- nitive effort to the interpreting process. This disadvantage could be addressed by automating the query through the use of Automatic Speech Recognition (ASR), as recent advances in Artificial Intelligence have considerably increased the qual- ity of this technology.” Fantinuoli first postulated the possibility of integrating ASR into a CAI tool in 2017, although this idea had already been put forward be- fore (see Hansen-Schirra 2012). In his paper, he presents a model and a prototype of an ASR-CAI integration and discusses the requirements of both the CAI tool and the ASR system. In order for ASR integration to be possible, both systems need to fulfil a set of criteria. First of all, the ASR system must be able to deal with the typical disfluencies of spoken language and cope with speaker variability, as well as with foreign ac- cents and mispronunciations, especially in the context of English as lingua franca. Other issues inherent to spoken language are ambiguities, such as homophones, and poor articulation which can occur in the case of fast speech. Finally, speech is not segmented, but continuous, and pauses appear not at word boundaries, but are rather syntactical. The recognition of word boundaries is an issue that the ASR system must be able to deal with, or else the quality of the database querying mechanism necessary for producing the text may be compromised. Es- pecially in consecutive interpreting, the ASR system must also be able to deal with background noise (Fantinuoli 2017b: 28). As Fantinuoli (2017b: 29) observes, an ASR system must also be able to “support large-vocabulary recognition” and “support vocabulary customisation” which is necessary to recognise specialised terminology. It must also have a low word error rate (WER), i.e. be highly accu- rate, and a low real-time factor (RTF), i.e. be fast. 2.4 Research on CAI tools The participants were trained in using InterpretBank while interpreting terminology-dense texts. In the final experiment, they were asked to interpret two similar speeches while looking up terms in a printed paper glossary and using InterpretBank. The test-subjects’ renditions were transcribed and rated on terminological accuracy and omissions. The difference between the paper glossary and the CAI tool proved statistically significant for the criteria “percentage of terms interpreted as per glossary” and “number of omissions”. Based on these initial results, CAI tools seem to provide an advantage in terms of terminological accuracy and completeness of informa- tion (fewer cases of omissions were observed even after performing a search). 55 2.4 Research on CAI tools 2.4 Research on CAI tools 2.4 Research on CAI tools of 5.04% and a F1 score of 0.9717, such a system could be used at least in stan- dard settings. A recent study by Brüsewitz (2019) suggests that the commercial ASR solutions currently available already perform rather satisfactorily, at least for the recognition of numerals and specialised terminology, while for named en- tities there is still a certain margin for improvement. In the study, which tested solutions by Google, Watson (IBM), Sonix and Speechmatics on six parameters (numbers, proper names, terminology, homophones, nonsensical utterances, and speech rate), the Google API was the best-performing system. p ) g p g y Following up on the initial prototype proposition by Fantinuoli (2017b), the first investigations were conducted to further test the potential of an ASR-CAI system in experimental settings. So far, the focus has been on the issue of nu- merals, typically considered one of the most common problem triggers for in- terpreters (Gile 2009). Desmet et al. (2018) conducted a pilot study involving a small sample of advanced conference interpreting trainees. The objective was twofold: “to determine if limited technological support can improve the accuracy of interpreted numbers, and how this improvement breaks down over different number and error types” (p. 18). In the experiment, no commercially available CAI tool with integrated ASR was used. The authors created a prototype using PowerPoint presentations based on the speech transcripts which contained the numerals present in the speech, presented simultaneously to the occurrence of the oral stimulus. Thus, the results describe what an ideal system would be able to achieve, i.e., following the criteria outlined above, an ASR-CAI hybrid with per- fect recognition and very low latency. The results support what has been outlined by Fantinuoli: in the case of numbers, a system with such characteristics is capa- ble to improve accuracy (from 56.5 to 86.5%) with statistical significance, and to drastically reduce the occurrence of approximations (by 90% for the experiment), the second most frequent type of error after non-strategic omissions. Overall, in- telligent CAI tools with integrated ASR seem to offer a promising upgrade to the toolkit currently available to interpreters. 17The best possible value being 1, the worst 0. 2 Information technologies and interpreting CAI tools must also present a certain profile: they must have high precision and recall, with priority given to precision if necessary, be able to deal with morphological variations, and have a sleek and user-friendly interface. y The prototype described recognises numerals and terms that have been added to the event database. The speech is first transcribed and pre-processed. The sys- tem then “queries the terminological database and identifies the entities from the text flow” (Fantinuoli 2017b: 30), which are extracted and visualised on the inter- preter’s screen. This initial prototype showed encouraging results: with a WER 56 2.4 Research on CAI tools As the authors observe, however, there is still a lot to explore in this respect, and “further studies should be carried out on how interpreters deal with discrepancies between auditory input from a speaker and visual input from an automatic recognition system, increased delay or different modes of presentation” (Desmet et al. 2018: 26). A follow-up experiment (Defrancq & Fantinuoli 2021) studied ASR support for numerals in a more naturalistic setting, using real-time transcription from the InterpretBank ASR tool. Data was collected through audio and video recordings and a follow-up questionnaire. Additionally, the study took a first look at the im- pact of a sudden loss of ASR support, which occurred in several cases during the 57 2 Information technologies and interpreting experiment. The results confirm that the tool presents high precision (96%) and low enough latency to fit the interpreter’s ear-voice-span (EVS). Overall, an in- crease in complete renditions (from 67.7% to 90.2%) and a drop in omissions (from 15.8% to 3.5%) due to ASR support was also observed. This aligns with previous findings, although the accuracy gain is less significant. This may be due, among other things, to the fact that numerical information was not presented in isola- tion like in the previous experiment, but rather highlighted within the context of the complete ASR transcription, requiring deeper processing by participants. Training would have probably led to even more significant improvements in the subjects’ rendition. It should be noted that significant accuracy gains were ob- served intra-subject only for two participants out of six. Interesting findings emerged from the questionnaire. The tool was judged pos- itively in terms of ergonomics, although some participants would have preferred a more minimal presentation of the numerical information. A certain tendency to over-rely on the tool for support was also observed, which aligns with first observations by Prandi (2015a,b). Interestingly, the authors speculate that the provision of ASR support might have had a positive psychological effect: know- ing numerals would be shown on the screen might reduce stress and/or boost confidence (Defrancq & Fantinuoli 2021: 93). Van Cauwenberghe (2020) reached similar conclusions in his experiment on ASR support for terminology. However, like Defrancq & Fantinuoli (2021), he also contended that interacting with the tool may not be a trivial feat from a cog- nitive standpoint, and observed cases of imported errors from the ASR tool into the interpreters’ renditions. 2.4 Research on CAI tools 2.4 Research on CAI tools 2.4 Research on CAI tools boothmate for terms and numbers. This second feature presents a similar archi- tecture to the ASR function offered by InterpretBank (see §2.2.3), comprising a cascade of ASR, identification of units of interest through a language model (LM), and automatic display of suggestions for terminology, numbers and named enti- ties on the interpreter’s screen. The terminological suggestions are based on the glossary curated by the interpreter18, while numbers and proper names are ex- tracted directly by the LM. The benchmark tests conducted on a general language and a specialised corpus show promising results in terms of precision, recall and F1 value, both for medium-sized and large glossaries (200 vs. 10,000 terms), with an average F1 value of around 98%. The tool performed better for named entities than for specialised terminology. Although the authors reported quite encourag- ing lowest F1 values of 84% and 81% respectively (Fantinuoli et al. 2022: 7), they also noted that individual terms and the glossary used can impact results con- siderably, as shown by a rather poor performance achieved for a speech about social issues (F1 = 76.19% for the medium and 68.90% for the large glossary). The average latency was of 1.6 seconds (minimum = 1.1s, maximum = 2.3s), i.e. low enough to fit into an interpreter’s average EVS. The CAI tool to be integrated into SmarTerp was tested on two components: its ASR system and its semantic interpretation module, which detects relevant entities of interest for the interpreter. For each interpreted session, the LM is adapted on the bases of the interpreter’s event glossary, from which seed words are extracted to be added to the glossary and to select texts from the training corpora for the LM adaptation. The semantic interpretation module deploys three underlying resources: a multilingual general purpose, a domain-specific and a user-specific knowledge graph (Rodríguez et al. 2021: 105). The system evaluation showed rather positive results for the three languages analysed (English, Italian, Spanish), with F scores ranging from 0.82 (English) to 0.90 (Italian). The tests also showed a positive impact of the adaptation systems for all three languages and for all metrics (Rodríguez et al. 2021: 108). In light of these initial results, the integration of CAI tools, particularly of those based on ASR, appears feasible. 18In the glossary creation tool, the target-language equivalents for specialised terminology are first generated through MT and then validated by the interpreter. 2.4 Research on CAI tools In Defrancq & Fantinuoli (2021) and Van Cauwen- berghe (2020), the system latency was deemed sufficiently short. On the topic of latency, Montecchio (2021) and Fantinuoli & Montecchio (2022) conducted a dedicated study precisely aimed at defining the maximum acceptable latency in ASR-enhanced CAI tools. Using an ASR mockup with increasing latencies vary- ing from 1 to 5 seconds, Montecchio (2021) explored the impact of latency on the rendition accuracy for numerals and referents as well as the effect on the per- ceived delivery flow. She found that both accuracy and delivery flow declined with increasing latency, interpreting this loss in quality as evidence of increased cognitive load to cope with the longer ear-voice span (EVS). A recent development in this area is the integration of AI-enhanced CAI tools into RSI platforms, as exemplified KUDO Interpreter Assist (Fantinuoli et al. 2022) and by SmarTerp (Rodríguez et al. 2021). KUDO Interpreter Assist includes two features designed to support inter- preters working remotely: an automatic glossary creation tool and a virtual 58 2.4 Research on CAI tools Nonetheless, a salient issue to be consid- ered is the trade-off between the amount of information offered by a system of this kind and the additional cognitive load needed to operate a CAI tool during interpretation. This applies both to “traditional” in-booth CAI tools and CAI- ASR integration, and not only to numerals, but also to other problem triggers, such as specialised terminology. Stewart et al. (2018: 109) observe that “while 59 2 Information technologies and interpreting displaying all terminology in a glossary achieves high recall of terms, it suffers from low precision. This could potentially have the unwanted effect of cogni- tively overwhelming the interpreter with too many term suggestions”. For this reason, the authors explore the possibility of integrating an NLP tool capable of predicting the elements likely to be missed by interpreters, with a view to reducing errors and improving performance. Such a system would need to be trained on a corpus of data that have been processed to identify problem trig- gers using the combined criteria of “termhood” (is the term difficult to recall and non-ambiguous?), relevance (should the term necessarily be translated?) and in- terpreter coverage (has it been left untranslated or mistranslated?). In addition, such a system must also consider task-relevant criteria, such as fatigue at the end of the interpreting turn (“elapsed time”), speech rate (“word timing”), terms left untranslated because they are rare and thus more difficult to recall from memory (“word frequency”), or because they are long and thus likely to represent tech- nical terminology (“word characteristics and syntactic features”). With a system of this kind, capable of processing the source speech and identifying potentially challenging terms based on these criteria, users could “theoretically adjust the precision-recall threshold” (Stewart et al. 2018: 115), even from speaker to speaker and for each assignment or presentation, in order to achieve a positive trade-off between useful term suggestions and additional cognitive effort required to deal with an extra source of information.i The findings of the empirical studies conducted thus far on the product of CASI appear encouraging. However, it should be noted that current investiga- tions have been highly focused on individual units of information. Little is known as to how CAI tools’ impact quality on a broader level, although first studies are starting to discuss the effects on the product more holistically and through qual- itative analysis (Frittella 2022). 2.4 Research on CAI tools Regrettably, while research on the product of CASI is increasing, the hypothe- ses on CAI tools’ impact on the cognitive subprocesses involved in CASI are yet to be tested empirically in controlled settings. Yet, gaining a more profound understanding of the CASI process and of interpreter-computer interaction dur- ing the in-process use of CAI tools appears to be essential in light of the postu- lated challenges in integrating CAI tools into the SI process. The insight gained through cognitive studies on CAI tools may prove helpful on different fronts. First, it may promote a better understanding of what discourages interpreters from using tools during SI. Second, it may help develop CAI tools truly targeting interpreters’ needs and addressing the cognitive constraints inherent to SI. In turn, this may result in a wider acceptance of tools which have the potential to improve interpreters’ workflow and the quality of interpretation. 60 2.4 Research on CAI tools 2.4 Research on CAI tools 2.4 Research on CAI tools The assumptions on the impact of CAI on mental processes in SI presented in §2.3.4 point to the idea of interference, of a limited amount of cognitive re- sources and time, and to the concepts of attention sharing and task coordination in view of the additional effort posed by terminology look-up. In order to formu- late hypotheses on the impact of CAI tools on the cognitive processes involved in SI, which the present study aims to test, some attention should first be devoted to the exploration of such constructs both from a cognitive psychology perspec- tive and within the context of simultaneous interpreting. This is the topic of the following chapter. 61 61 3.1 Working memory and information processing The concept of WM was first introduced in the seminal work by Baddeley & Hitch (1974). Until the authors’ investigations into the construct, little empirical evidence had been collected to support the idea of WM. The development of a construct of WM represents a step forward in the understanding of the role of memory in human information processing, specifically in how information is processed over a short period of time, for instance during reasoning, language comprehension, and learning, the three cases on which Baddeley and Hitch con- centrated. The construct of short-term memory (STM, Atkinson & Shiffrin 1968) had been used to represent the temporary retention of a small part of informa- tion over a limited amount of time. As for processing in STM tasks, the role of the short-term store (STS) as WM, as proposed by Atkinson & Shiffrin (1968, 1971), had found little empirical evidence in non-memory related tasks. Baddeley and Hitch provided first evidence supporting the idea of a common WM system re- sponsible for both short-term information retention and processing. Thus, they depart from Aktinson and Shiffrin’s assumption that the STS is a single unit and that it is modality-independent. They describe WM as “a control system with limits on both its storage and processing capabilities” (Baddeley & Hitch 1974: 86): we can only store and process a limited amount of information over a brief period of time. The role of WM is essential to interpreting as it is a highly complex cognitive activity which requires both retention of transient information and processing of said information during a short time-span. In addressing the issue of CAI tools in- tegration into SI, it is often pointed out that interpreters may not always have the necessary resources (see §3.3) or sufficient time (see §3.4.3) to attend to the addi- tional sub-processes required by glossary querying or by the interaction with an ASR system. As a consequence, an interpreter’s WM may not be able to retain and process all stimuli that must be attended to. This hypothesis reflects one of the key assumptions about WM, i.e. its limited capacity. This fundamental tenet of WM plays an essential role in all models of WM, despite the ongoing debate on its nature (Seel 2012: 3474), which is, to some extent, still unresolved. 3 Simultaneous interpreting as a complex cognitive activity As discussed in the previous chapter, the main points of criticism concerning the use of CAI tools for terminology look-up during interpreting, and in partic- ular during SI, refer to the addition of yet another task to the concurrent mental processes involved in interpreting, i.e. that of interacting with a digital support tool to deal with specialised terminology perceived as an element of difficulty during interpreting (see §2.3.4). The hypotheses formulated on the cognitive im- plications of a CASI task, both when manual look-up is involved and when it is replaced by an ASR module are, however, mostly based on personal assumptions or unstructured observations and have not yet been grounded in empirical analy- sis aimed at investigating the effects of technological support on the interpreter’s cognition. Such hypotheses refer to constructs that have been formulated and ex- plored in the context of cognitive psychology, learning psychology and human factors research and which have been widely applied in TPR: the concept of in- terference between tasks, the assumption that humans have access to a limited amount of cognitive resources, the issue that attention must be allocated to and shared among different co-occurring tasks during a limited amount of time, and that this requires coordination and monitoring, and, finally, the construct of cog- nitive load (CL), often mentioned in reference to the presumed additional effort posed by CAI tool usage. The present section will review and discuss these concepts, their application and empirical validation with the aim to provide a conceptual framework for the formulation of hypotheses on SI with digital terminological support and for the definition of the research questions guiding the present study. I will start by discussing some basic assumptions about working memory (WM) and attention allocation (§3.1, §3.2), and discuss their relevance for the investigation of CL (§3.4) on the basis of related empirical research within the field of cognitive psychology and TPR. After defining and exploring these key concepts, I will consider how they have been integrated into cognitivist models of SI which view this process primarily as an issue of multitasking (§3.5), as this is the framework within which I conduct my analysis, define my research ques- tions and formulate hypotheses on CL and task interference in SI with CAI tools (§3.6). 3 Simultaneous interpreting as a complex cognitive activity 3.1 Working memory and information processing 3.1 Working memory and information processing functions, the Central Executive (CE). According to the authors, verbal informa- tion is “phonemically coded” (ibid.), and is rehearsed (either overtly or covertly) in a limited-capacity phonemic buffer. Following Atkinson and Shiffrin’s model, Baddeley and Hitch postulate that rehearsal is necessary for information reten- tion. They propose that WM is modality-dependent: while verbal information is thought to be temporarily stored and encoded in the PL, visual-spatial informa- tion is encoded in the VSSP. Both substructures later received further specifica- tion. The PL was thus subdivided into a storage component, i.e. the phonological store – where phonological traces are stored up to 2s (Baddeley 1990) due to the word-length effect (Baddeley 1975) – and a rehearsal subsystem responsible for refreshing memory traces in order to prevent their decay (the articulatory loop). Similarly, the VSSP was divided by Logie (1995) into the visual cache (retention) and the inner scribe (processing). Later on, Baddeley (2000) introduced the con- struct of an Episodic Buffer (EB) which stores and integrates information not only from the PL and the VSSP, but also from long-term memory (LTM) to form a sin- gle percept. Baddeley and Hitch’s model of WM posits important consequences for our understanding of WM. According to the model, retention and processing are managed by different memory structures. This in turn supports the idea of multitasking, which is only possible if resources can be shared between reten- tion and processing, and, conversely, if retention and processing correspond to separate structures in WM. The assumptions put forward by Baddeley and Hitch have received ample empirical support. Of particular relevance to the task of SI are the PL, the CE and the EB. The double function of the PL, i.e. perception and rehearsal, has been evidenced by two related effects found in empirical testings of the model: the phonological similarity effect (Baddeley et al. 2018) and the word-length ef- fect (Baddeley 1975, Jacquemot et al. 2011). The first shows a reduced recall for phonologically similar words, the second a decay in memory traces and thus an impaired recall for visually presented words when articulatory rehearsal is sup- pressed. Neurological evidence also supports this distinction, as different brain areas have been found to activate for storage and rehearsal tasks (Papagno et al. 2017). Gieshoff (2012) found supporting evidence from interpreting as, in her experiment, longer numbers were more difficult to recall. 3.1 Working memory and information processing One ele- ment of dispute lies in the architecture of WM: some scholars, as exemplified by the very successful modular model by Baddeley and Hitch, postulate a WM com- prising a set of related components covering stimulus processing and attention allocation, while others, most notably Cowan (1988), view memory as a unitary store. In Baddeley and Hitch’s model, the WM system involves a phonological loop (PL), a visual-spatial sketch pad (VSSP) and a structure responsible for executive 64 3.1 Working memory and information processing This aligns with the word-length effect, as in interpreting phonological rehearsal is suppressed due to the concurrent production of the interpretation. The concept of a CE responsi- ble for focusing, dividing and switching attention and interacting with the LTM store is supported by evidence from studies investigating executive functions (e.g. Godefroy et al. 2010), although the notion that it is a unitary system may be oversimplified (Stuss & Alexander 2007, Logie 2016). Finally, the introduction of 65 3 Simultaneous interpreting as a complex cognitive activity an EB was justified by findings that recall for longer sets of words was possible than what the PL would allow (Baddeley et al. 1987). More specifically, Baddeley (2012) postulates a capacity of four chunks of information. Further evidence of an integration and a facilitation effect between visual-spatially and verbal-vocally encoded information is provided by studies on bootstrapping data (Darling et al. 2017, Darling & Havelka 2010). Concurring accounts of WM, exemplified by Cowan’s (1988) unitary store model, propose an alternative to Atkinson and Shiffrin’s multi-storage model, which, as discussed above, was highly influential in Baddeley and Hitch’s devel- opment of their own model of WM. The development of Cowan’s (2009) model was based on evidence that there might be interference between the auditory and the visual buffers, which does not support the notion of separate stores. In Cowan’s model the existence of a single, long-term store (LTS) is postulated which contains units of information. STM is described as the activated portion of long term memory (LTM), while the items currently being processed are at- tended to in the so-called “focus of attention”. Thus, WM, i.e. the processing of STM items, is located in the activated portion of LTM that is receiving attention at a certain moment in time. The main point of criticism for Cowan’s model is its oversimplification inherent to the assumption that STM is only the activated portion of LTM. Additionally, studies on amnesic patients have identified im- pairment in STM, but not in LTM, or viceversa. This model of WM is valuable in that it poses the focus on the relationship between information processing and attention, an essential component in the interpreting process, as I will discuss in the following section (§3.2). 3.1 Working memory and information processing term store emerges clearly in Cowan’s model, which promotes a unitary view of human cognitive architecture, as discussed above. The role of a CE is essential to free up cognitive capacity and avoid cognitive overload when capacity is exceeded. As postulated by Miller (1956), the (limited) capacity of our STM depends not on the amount of information, but on how it is compounded: chunks of information are more easily remembered than single elements of information. The CE is thought to intervene by re-coding informa- tion in order to free up capacity, re-combining single elements in larger chunks and thus reducing WM load. An example of chunking is present in instructional design theory (see Van Merrienboer & Sweller 2005) in the notion of schemata for information and knowledge organisation. The development of higher-order schemata allows learners to process information more easily, thus acting as a CE. This also relates to the translation process, in that the development of strategies to address a text to be translated or common sources of difficulty in speeches to be interpreted can alleviate WM by freeing up cognitive capacity for intentional processing. Studies on translation and interpreting competence have highlighted how one essential difference between novice and experts lies in the fact that the latter have developed more solid schemata (or strategies) which allow them to effectively process complex information and avoid cognitive overload, with im- portant consequences for training (see for instance Riccardi 2005). Both approaches to the definition of the architecture of WM present some shortcomings. As discussed above, the very influential multi-component model by Baddeley and Hitch has received repeated empirical support over the years from studies conducted in the field of cognitive psychology. In addition, it has been widely influential in TPR, where it has been adopted as a framework in a number of empirical studies (Darò & Fabbro 1994, Dragsted 2004, Mizuno 2005, Padilla et al. 2005, Kosma 2007, Hvelplund 2011, Köpke & Signorelli 2012, Tima- rová et al. 2015). In this framework, however, it is not clear how other types of sensory input may be processed in addition to auditory and visual information. This shortcoming emerges also on the basis of recent evidence which found that different areas of the brain are activated when different types of auditory and vi- sual stimuli are processed. 3.1 Working memory and information processing Even if WM is structurally considered part of long term memory, its function is nonetheless distinct from LTM, as is the case in the multi-component model which further postulates a structural separation of the two stores. Another key assumption about the human cognitive architecture is that con- trol processes aimed at allocating attention to the stimuli to be processed must be governed by a dedicated cognitive structure. A CE responsible for voluntary processing and directing attention to relevant items is found both in modular and in unitary models of attention. In Baddeley and Hitch’s model, for instance, the CE supervises the PL, the VSSP and the EB, i.e. the structures responsible for information retention and processing. Similarly, although based on largely different assumptions, Cowan also postulates the existence of a CE responsible for attention and voluntary processing, such as the activation and focusing of items. The CE also has the role to activate items stored in long term memory which are necessary for stimulus processing. The link between WM and a long 66 3.1 Working memory and information processing 3 Simultaneous interpreting as a complex cognitive activity On the other hand, an account of WM which views it as a unitary store also con- flicts with such evidence. Additionally, it appears limited for its assumption that information is merely “activated” through focused attention, but does not offer a framework for the integration of new information in a long term store, present- ing a unidirectional view of human memory architecture which brings informa- tion to the surface, but does not explain how information has entered the store in the first place. The value of Cowan’s account of WM, taken to exemplify unitary models of WM, lies in its holistic view of information processing compared to the more rigid account by the more successful model by Baddeley and Hitch. A unify- ing view of WM may come from Postle (2006), who approaches WM from the per- spective of cognitive neuroscience rather than cognitive psychology. He defines WM as a function that “arise[s] through the coordinated recruitment, via atten- tion, of brain systems that have evolved to accomplish sensory-, representation-, and action-related functions” (Postle 2006: 23). It is this functional view which, in my opinion, best serves the discussion around WM for the translation and inter- preting process. Thus, the separation of processing buffers for audio and visual stimuli is applied here functionally, rather than structurally. When referring to separate buffers for audio and visual stimuli, which will be pivotal in my discus- sion of task interference, I start from the assumption, backed by recent evidence, that there are indeed separate channels for the processing of this information. As I embed my methodological approach in the tradition of TPR, which has widely adopted and operationalised the model of WM by Baddeley and Hitch, I will be referring to two single channels dedicated to processing visual stimuli on the one hand and auditory on the other, while implicitly acknowledging that the cognitive structure underlying such systems may very well be more complex.f To sum up, though the approaches discussed present some differences in how they view the structure of WM, they share common assumptions in terms of its function. First of all, in both cases WM is considered to have both a retention capacity and a processing capacity. If the elements temporarily stored in WM are not attended to, not refreshed nor activated, their traces decay from mem- ory. 3.1 Working memory and information processing As discussed by Postle (2006), the consequence of such evidence for the “standard” model of WM is a potentially unlimited proliferation of subsystems, which would undermine the very role of the model: Followed to its logical extreme, the cognitive architecture of the standard model would eventually depict a WM system organized into hundreds (if not thousands or more) of domain-specific buffers, each responsible for the WM processing of a different kind of information. (Postle 2006: 25) 67 3 Simultaneous interpreting as a complex cognitive activity 3.2 The role of attention in processing visual stimuli Closely linked to WM is attention, which represents a key constituent in nu- merous models of human memory architecture, most notably in Baddeley and Hitch’s and especially in Cowan’s model, and in models of resource allocation among concurrent tasks, as I will discuss in §3.3. Attention is a multifaceted con- struct, which can be studied from different perspectives. Of particular interest for the present research object are the concepts of focused or selective attention and of divided attention. In particular, divided attention and its effects on infor- mation processing and performance has been in focus in Cognitive Load Theory (CLT), as will be discussed in §3.4. Interpreting may be viewed as an activity requiring the allocation of atten- tion to multiple streams of information (Seeber 2017). This is especially true to- day, as communication at multilingual events is rarely achieved through a single medium, but often foresees the combination of oral presentations and, for in- stance, slide presentations. This requires interpreters to process multiple types of inputs and several information flows in parallel derived from the external world (e.g. the speech, the presentation, support materials such as speech transcripts or glossaries), in addition to monitoring their own rendition. When a CAI tool or a conventional glossary is used as a support to deal with terminology, additional visual-verbal information enters the perceptual space of the interpreter. Some of these stimuli capture the interpreter’s attention in a bottom-up way. However, it is required of the interpreter to actively divide their attention between the rele- vant streams of information and, for instance when the rendition of a specialised term requires additional effort, to actively focus their attention on a single in- put while at the same time keeping the other stimuli in the focus of attention. Thus, interpreting may be seen as an activity requiring both focused and divided attention, even more so when the processing of multimodal stimuli is involved – as is the case with a digital glossary or an ASR system – and involving both bottom-up and top-down attentional processes. As a consequence, SI with digital terminological support may be seen as an activity imposing a high perceptual load as well as a high CL (see §3.4.1) on the interpreter’s WM. 3 Simultaneous interpreting as a complex cognitive activity This, in turn, underlines the role of attention allocation, which must take place effectively in order to preserve such memory traces in WM. However, at- tentional capacity is considered to be limited. Thus, attention must necessarily be distributed, shared or switched to the different elements that require process- ing. The task of attention distribution, sharing and switching must be regulated by control functions, which in the models discussed is identified as a CE. How attention is allocated to the stimuli competing for attention is the object of the- ories of attention distribution and resource sharing, which will be the object of the following sections. 68 3.2 The role of attention in processing visual stimuli 3.2 The role of attention in processing visual stimuli The concept of perceptual load is pivotal in Lavie’s (1995, 2000, 2005, 2010) perceptual load theory, which provides an explanation of how atten- tion is selectively allocated to stimuli concurring for human attention. Her pro- posal outcomes the dichotomy of early selection models (Broadbent 1958, Treis- man 1964) and late selection models (Deutsch & Deutsch 1963, Norman 1968). The first postulate, respectively, either an all-or-nothing filter which lets only salient items pass through and enter our focus of attention, or an attenuator which lets 69 3 Simultaneous interpreting as a complex cognitive activity through only relevant enough elements. The latter, on the contrary, propose that selection occurs only at a later stage: everything is first perceived and processed, and then priority is assigned to the most relevant items. In light of supporting empirical evidence for both models (e.g. Simons & Cha- bris 1999 for early selection or Stroop 1935 and Eriksen & Eriksen 1974 for late selection), Lavie put forward a model which overcomes the limitations of the pre- ceding models and integrates their strengths. Her model thus represents a “reso- lution to the long-standing early and late selection debate on the extent to which irrelevant distractors can be ignored” and may be viewed as a “hybrid model of selection” (Lavie et al. 2004: 353). Lavie’s theory is particularly relevant to the present research object as it has been tested within the framework of visual at- tention. According to Lavie, attention allocation is dependent on the perceptual load posed by the task. Low-load conditions allow for late selection (more stimuli can be processed and are only later selected), while high-load conditions require early selection of relevant information. The reduction in distractor processing under high perceptual load has been shown by several studies (for a review, see Lavie 2005). At the same time, in situations of high perceptual load, limited atten- tional capacity is left for processing. Therefore, we can hypothesise that during SI with terminological support, if the suggested terms are presented in a way that imposes high perceptual load (e.g. within the context of a glossary rather than in isolation), fewer resources will be left for the processing of other stimuli, such as the speaker’s words or one’s own rendition. 3.3 Attention sharing and task interference cusing on two models which are reflected in models of the interpreting process seen as a matter of task coordination (see §3.5). Similarly to Lavie, Kahneman (1973) postulates the presence of a single pool of attentional (or cognitive) resources that are shifted from one task to another. Kahneman’s single resource theory can therefore be seen as a “general resource model of task interference” (Wickens 2002: 162), which explains a decrease in performance levels simply with the concurrent performance of the tasks at hand. According to Kahneman, variance in dual-task performance is due either to the level of difficulty of the individual tasks or to the preference given to one of the tasks (if a task is not favoured, its performance suffers, while the favoured task is carried out successfully). He distinguishes between the idea of load, i.e. the amount of resources demanded by a task, and that of effort, i.e. the amount of cognitive resources and energy the subject exerts in order to perform a task. When task demands are higher than the available capacities, performance suf- fers. Kahneman also addresses the role of arousal for the availability of cognitive resources: the higher the level of arousal, i.e. of conscious attention, the higher the amount of attentional capacities available. Kahneman’s assumptions have, however, been challenged empirically (for a discussion in light of neuroscien- tific evidence, see Bruya & Tang 2018) and criticised for their vagueness. One major drawback of his theory is that it does not allow for a modulation of hy- potheses on task interference based on the nature of the tasks. This stems from his equation of attention with effort, which does not allow for the integration of bottom-up processes of attention capturing. It does, however, focus the spotlight on one important quality of attention, which aligns with accounts of WM and is largely supported by empirical evidence: that attention is limited. Addition- ally, while limiting on the one hand, the model stresses the role of intentionality on the allocation of attention, which is essential in a complex activity such as interpreting. 1. every non-automated task produces load; 3.3 Attention sharing and task interference Lavie’s theory is helpful in understanding how visual stimuli to be processed are selected during high-load tasks such as SI. However, it is insufficient to formu- late hypotheses on how multiple tasks interfere with each other. Additionally, in her model attention is viewed as a general resource, with no further specifica- tion. As a general theory of attention, her model is not sufficient to explain why certain concurrently performed tasks interfere with each other more than others, nor at which level this interference occurs, and which attentional resources are shared. The issue of task interference is highly relevant to a complex cognitive task such as interpreting, and even more so when the additional task of looking up terminology or processing a term presented automatically on a screen must be integrated with an already multi-layered task such as SI. Since in the present work SI is approached from the perspective of multitasking, I will now discuss attention allocation as a matter of resource sharing among concurrent tasks, fo- 70 3.3 Attention sharing and task interference 3.3 Attention sharing and task interference In order to overcome the limitations of Kahneman’s model, and to account for the fact that “differences in time-sharing efficiency” may be due to the quality of the resources concurrently recruited by multiple tasks (ibid.), as postulated by Kantowitz & Knight (1976) and Wickens (1976) himself, Wickens (1984, 2002) put forward an alternative model of resource sharing which posits that “time-sharing between two tasks [is] more efficient if the two [utilise] separate structures than if they [utilise] common structures” (ibid.). At the core of Wickens’s model are three underlying assumptions: 71 3 Simultaneous interpreting as a complex cognitive activity 2. two interacting tasks demand a higher amount of resources than the per- formance of any single task; 2. two interacting tasks demand a higher amount of resources than the per- formance of any single task; 2. two interacting tasks demand a higher amount of resources than the per- formance of any single task; 3. tasks recruiting the same type of resources exhibit a higher level of inter- ference than tasks recruiting resources of discrete structures. 3. tasks recruiting the same type of resources exhibit a higher level of inter- ference than tasks recruiting resources of discrete structures. Unlike Kahneman, Wickens does not assume the existence of a single pool of resources, but rather of discrete attentional structures. As a consequence, what is shared among tasks are not the structures themselves, but rather their underlying resources. It follows that two or more tasks recruiting the same resources may be very difficult (or impossible) to perform simultaneously than tasks demanding separate resources. This would explain why we can simultaneously look at a painting and talk with a friend, while listening to the radio and to a friend at the same time will inevitably lead to a loss of information from one of the two sources. In the first case (simultaneous performance of a visual and an auditory task), resource sharing will be easier, and thus more efficient, than in the latter (two auditory tasks), which share the same underlying resources. VISUAL PROCESSING Perception Cognition Responding Manual Spatial Vocal Verbal Visual Auditory Spatial Verbal Focal Ambient MODALITIES RESPONSES STAGES CODES Spatial Verbal Figure 3.1: Wickens’s multiple resource model (Wickens 2002: 163) Spatial Figure 3.1: Wickens’s multiple resource model (Wickens 2002: 163) Wickens’s model distinguishes between four dimensions, each further subdi- vided into two levels, as summarised in Table 3.1. 3.3 Attention sharing and task interference The fourth dimension (visual processing), with its respective levels, is a later addition (Wickens 2002) to the original model. 72 3.3 Attention sharing and task interference 3.3 Attention sharing and task interference A higher 73 3 Simultaneous interpreting as a complex cognitive activity total interference score points to less efficient time-sharing between simultane- ously performed tasks. Several findings support Wickens’s model. For instance, experiments conducted by Treisman & Davies (2012) and McLeod (1977) support the notion that concurrent tasks in the same modality interfere with each other, while the research reviewed by Lu et al. (2013) provides support for reduced in- terference of tasks performed in different modalities. One weakness of Wickens’s model is the subdivision of tasks in a processing and response stage, which may represent too limiting an account of complex ac- tivities such as translation and interpreting, where these stages often overlap. It is, however, more in line with neurocognitive evidence suggesting that different areas of the brain are responsible for processing different types of stimuli, simi- larly to what has been discussed for Baddeley and Hitch’s model (see §3.1), albeit with the same limitations. An additional benefit provided by the model is that it allows to operationalise and empirically test hypotheses on the interference of similar but different tasks, such as SI with different types of digital terminologi- cal support, as I will discuss further in §3.6. In turn, being able to test hypotheses at this level of granularity may prove particularly useful in practice in order to fine-tune the type of support provided to interpreters, where small differences may have a significant impact on the level of interference between tasks experi- enced and on the alleviation of WM demands. Therefore, Wickens’s model will be especially relevant in the formulation of hypotheses for the present analysis (see §5.2) and in the present empirical investigation of CASI. Additionally, as I will discuss in §3.5, the model has already been operationalised in empirical interpreting research concerning itself with issues of attention allocation. 3.3 Attention sharing and task interference Table 3.1: Dimensions and respective levels in Wickens’s multiple re- source model Dimensions Levels processing stages perception & cognition response perceptual modalities visual auditory processing codes spatial verbal visual processing ambient focal Table 3.1: Dimensions and respective levels in Wickens’s multiple re- source model According to the model (see Figure 3.1 for a graphic representation), processing occurs at two stages (perception and cognition are subsumed in a single stage, since one cannot occur without the other, and thus share the same pool of re- sources). Perception can be either visual or auditory. This points to the same dis- tinction posited by Baddeley and Hitch and operationalised through the PL and the visuospatial sketchpad. The perceived information can be processed either as spatial or as verbal information. In addition to the four dimensions, Wickens also “concedes the existence of a residual pool of general resources which, albeit not reflected in his model, is available to and demanded by all tasks, modalities, codes, and stages as required” (Seeber 2007: 1382), and which may be, to some extent, equated to the CE. This model of resource-sharing can be used to operationalise hypotheses on the level of task interference and thus to compute an “interference score”, which can be adopted to predict the efficiency with which two tasks can be performed at the same time. Wickens develops a “conflict matrix” (CM) to show the level of resource recruitment by the single sub-tasks (expressed by “demand vectors”) and the degree to which co-occurring sub-tasks are expected to interfere with each other (represented by the “conflict coefficients”), which is higher if the two tasks both demand the same level of a given dimension (Wickens 2002). Demand vectors can range from a value of 0 (no dependence on a certain resource) to 2 (extreme dependence on a resource) and are assigned for each dimension of a certain task. Conflict coefficients can be assigned a value between 0.2 and 1, where 1 represents two tasks that cannot be performed simultaneously, as they are both entirely dependent on the same resource, and 0.2 is the necessary inter- ference cost of the simultaneous performance of two tasks. A conflict coefficient of 1 corresponds to a quantification of the central bottleneck assumed by several models of WM (Wickens 2002). The total interference score between tasks is thus calculated as the sum of demand vectors and conflict coefficients. 1The interest in the cognitive ergonomics of CAI tools is picking up. A joint project currently underway at the University of Ghent and at the University of Mainz is concerned with the question of defining the best interface for the “artificial booth mate”. See the webpage of the Ergonomics for the Artifical Booth Mate project (EABM 2021) for preliminary results: https: //www.eabm.ugent.be/survey/ (Accessed: 01.11.2021). 3.4 The construct of cognitive load 3.4 The construct of cognitive load been very influential in both areas of study, providing “the basic premise of the behavior– mind correlation” (Jakobsen 2017: 23). As the present study aims to approach the use of digital terminology support tools by interpreters from an ex- plicitly cognitive perspective, investigating the “how” behind the “what” of the interpreting performance, the adoption of models and theories formulated within the field of cognitive psychology is expected to constitute a useful reference for the formulation of hypotheses about this research object. Additionally, although the present investigation does not directly aim to explore the ergonomics of the tools employed as terminological support1, the interpreter’s experience in inter- acting with the tools can reasonably be expected to influence cognition during SI, particularly in terms of the load experienced, similarly to how the interaction between translator/post-editors and CAT tools has been shown to play a crucial role for the disposition towards support technologies and for the load experi- enced by translators (see O’Brien 2012, O’Brien et al. 2017, Moorkens & O’Brien 2016). In this section, I discuss the concept of CL from the perspective of cognitive psychology, focusing on the foundational model developed within the frame- work of CLT (Chandler & Sweller 1991), which has been widely adopted to model the translation and interpreting process and has received ample empirical sup- port in TPR. I thus aim to provide first of all a terminological clarification of the concept, and to delineate a framework for the discussion of how the construct has been adopted by Interpreting Studies to discuss the cognitive implications of interpreting (§3.5). Particular attention will be devoted to the predictions of CLT for the load imposed on the interpreter by SI with digital terminological support and potential effects on attention and WM, which will serve as the theoretical basis for the discussion of results in §6.2. 3.4 The construct of cognitive load The notions of WM and attention allocation are particularly relevant to a com- plex cognitive activity such as SI. As such, they represent fundamental tenets for the construct of CL, ubiquitous in cognitive models of the interpreting pro- cess postulated within the framework of Interpreting Studies. In exploring the cognitive underpinnings of (simultaneous) interpreting, scholars have adopted a rather diverse terminology, not only referring to the notion of CL, but also to the concept of processing load or mental (work)load (see for instance Gieshoff 2018). Additionally, the notion of effort, which has been highly influential in Interpret- ing Studies (see §3.5.1), represents an important related construct.il The adoption of such diversified terminology reflects the interfacing of trans- lation and interpreting studies with other disciplines, primarily cognitive psy- chology and human factors, in particular cognitive ergonomics. The former has 74 3 Simultaneous interpreting as a complex cognitive activity Intrinsic cognitive load (ICL) can be defined as the CL that is experienced by the subject while learning. It “depends on the number of elements that must be processed simultaneously in WM, and [this] in turn, depends on the extent of element interactivity of the materials or tasks that must be learned.” (Van Mer- rienboer & Sweller 2005: 150). ICL thus increases when element interactivity in- creases. The experienced load depends on the task itself and its level of interac- tivity: the higher the interactivity of the task elements, the higher the ICL. The level of interactivity, in turn, depends on expertise: through the development of schemata, several interacting elements can be processed as a single input, thus reducing the experienced load. Interpreting may be considered as a process with a high level of interactivity. When multi-modal input is involved, such as in SI with digital terminological support, the interactivity is even higher. According to CLT, the intrinsic load of the task may be expected to increase (see also §3.4.2.1). Similarly to the field of learning, expertise in SI also leads to the development of schemata, which can prove useful to automatise several subprocesses involved in interpreting. As a consequence, the intrinsic load of the task may be lower for more experienced interpreters, especially if they have received training or enjoyed extensive practice in CASI. While the present study is not conducted within the expert-novice paradigm, the predictions of CLT for the intrinsic load produced by the task in question as a function of interpreters’ expertise will be relevant for the discussion of results, especially in terms of their generalisability. Although the aspect of expertise is not included in Wickens’s model (see §3.2), the total interference score may be seen as a theoretical quantification of intrinsic load as a function of element interactivity. Extraneous cognitive load is a concept that refers to task design and presenta- tion. Unlike intrinsic CL, extraneous CL is not dependent on the nature of the task to be performed, but rather on the way the task is presented. Given that intrin- sic and extraneous CL are additive (Van Merrienboer & Sweller 2005: 150), in the presence of a high ICL, extraneous CL should be kept as low as possible, whereas in a lower load task, performance may not suffer from high extraneous load. 3.4.1 Cognitive load in interpreter-computer interaction The construct of CL was first delineated within the framework of Chandler & Sweller’s (1991) CLT. In developing their theory, they were aiming to answer salient questions in instructional design, i.e. how to facilitate knowledge acquisi- tion by effectively designing learning tasks that support students in their learn- ing process. The authors distinguish between three types of CL: intrinsic, extra- neous, and germane load. 75 3 Simultaneous interpreting as a complex cognitive activity 3.4 The construct of cognitive load 3.4 The construct of cognitive load been shown to negatively affect CL in workplace studies on technology-assisted translation (e.g. Bundgaard et al. 2016, Teixeira & O’Brien 2019). A similar effect may reasonably be expected to occur also in technology-supported interpreting. Finally, germane load is the mental effort that the learner or the subject ex- erts in order to perform a certain task. Germane load is subject-dependent and, like intrinsic load, can be expected to decrease with increased expertise. The concept of germane load is valuable as it highlights the role of individual prefer- ences in interacting with the tools and points to the potentially beneficial effect of higher personalisation of support tools for interpreters, similarly to what has been suggested for translators (Taravella & Vielleneuve 2013). Recent approaches exemplified by Vogler et al.’s (2019) proposal of automatically predicting useful terminology to be displayed by ASR systems for interpreters may offer bene- ficial insights and practical solutions to positively intervene on germane load. Additionally, the notion of germane load as a function of competence emerges in the redundancy effect predicted by CLT, as will now be discussed in §3.4.2.1. been shown to negatively affect CL in workplace studies on technology-assisted translation (e.g. Bundgaard et al. 2016, Teixeira & O’Brien 2019). A similar effect may reasonably be expected to occur also in technology-supported interpreting. y y p gy pp p g Finally, germane load is the mental effort that the learner or the subject ex- erts in order to perform a certain task. Germane load is subject-dependent and, like intrinsic load, can be expected to decrease with increased expertise. The concept of germane load is valuable as it highlights the role of individual prefer- ences in interacting with the tools and points to the potentially beneficial effect of higher personalisation of support tools for interpreters, similarly to what has been suggested for translators (Taravella & Vielleneuve 2013). Recent approaches exemplified by Vogler et al.’s (2019) proposal of automatically predicting useful terminology to be displayed by ASR systems for interpreters may offer bene- ficial insights and practical solutions to positively intervene on germane load. Additionally, the notion of germane load as a function of competence emerges in the redundancy effect predicted by CLT, as will now be discussed in §3.4.2.1. 3.4.2 Attention splitting and audio-visual integration As discussed above, CLT postulates that extraneous load is influenced by the way in which information is presented. The task of interpreting a speech simultane- ously while being able to draw support from a digital terminology tool involves the processing of several streams of input (aural and visual): it is therefore useful to consider several additional principles which have been formulated within the framework of CLT and have received strong empirical support also within the field of TPR: the split-attention effect, the redundancy effect and the modality effect. 3 Simultaneous interpreting as a complex cognitive activity If the intrinsic load of SI with terminology tools cannot be reduced, it may be possible to affect the total load imposed by the task by operating on the extraneous load determined by the tool interface or by the way information is visually organised in the digital glossary. In turn, differences in the load experienced when working with different tools may be traced back to differences in their interface, which may for instance be perceived as non conducive by experienced interpreters who have been using different tools for several years and may be one of the sources of resistance towards new tools among seasoned professionals. This “cognitive fric- tion” (Cooper 2004: 19) may be the source of frustration and irritation, which has 76 3.4.2.1 Split attention The split-attention effect was identified by Tarmizi & Sweller (1988), who found that worked examples of geometry problems were more effective if the various sources of information (e.g. the diagram and the diagram description) were pre- sented in an integrated fashion rather than spatially separated. This is because in the conventional mode of presentation, the various sources of information must be integrated for learning to occur, which strains WM. There is a posi- tive split-attention effect if information processing is more effective when the sources of information are integrated in a single percept. Not only is spatial con- tiguity important to promote effective information processing, but so is temporal contiguity, as demonstrated by Sweller (2005) in a series of experiments on mul- timedia learning. As the time component is of the essence in SI, the temporal 77 3 Simultaneous interpreting as a complex cognitive activity contiguity principle holds the potential to predict important effects on the tim- ing of the terminological information presented on screen. In order to avoid or at least to reduce split-attention between the speaker’s words and the terminologi- cal pair presented on the screen, it is essential that the term be found as quickly as possible by the interpreter when manual look-up is involved, and that the la- tency of the ASR system be kept as low as possible (see Montecchio 2021). At the same time, the information presented by the tool must remain visible long enough to avoid having to retain the visually-presented information in WM (see §3.1), which would cause attention splitting. The robustness of the split-attention effect has been corroborated by numerous empirical investigations (for a meta- analysis, see Ginns 2005) and has been applied and found support in studies on subtitle perception, which bears some similarities to technology-supported SI due to its multi-modality (on the positive effects of integrated subtitles, see for instance Fox 2018). 3.4 The construct of cognitive load 3.4 The construct of cognitive load i.e. expanding WM capacity. In alignment with Baddeley’s (1990) model of WM, this may be achieved by dividing information processing between the subsystem devoted to processing auditory stimuli and the buffer devoted to visual stimuli. As for the split-attention effect, a positive effect of modality only occurs if the information presented across different modes is not redundant. A positive modal- ity effect may occur in SI supported by an ASR system if the interpreter has not heard a term pronounced by the speaker or has not been able to decode it and the term is presented on the screen: in other words, if the system processes the term for the interpreter. Additionally, the visual presentation of terms on the screen may be beneficial due to the transient information effect, which occurs “when learning is reduced as a result of current information such as speech or animations being replaced by new” (Low & Sweller 2014: 242).fi g p y The modality effect has been formulated within studies in the field of learning psychology and instructional design, which have mainly considered the combi- nation of images or graphs and of instructional cues in written or auditory for- mat. As, according to the WM model by Baddeley and Hitch, verbal information is coded by the PL (see §3.1), the presentation of completely redundant verbal information in auditory and visual form may not alleviate WM by distributing information processing across two separate buffers. However, as observed by Seeber (2017), SI is a “noisy environment” (p. 464): a term presented in auditory and visual information may therefore not be perceived as entirely redundant. Rather, the multi-modal presentation may facilitate processing especially dur- ing comprehension through multi-modal integration (for a thorough discussion of multi-modal processing in SI, see Seeber 2017). However, as both the visual and the auditory stimuli are presented in the verbal code, there might be a nega- tive trade-off due to interference, although multitasking may still be possible as the information is presented in different perceptual modalities, as predicted by Wickens (see §3.3). 3.4.2.2 Redundancy and modality An important addition to the split-attention principle came from a study con- ducted by Chandler & Sweller (1991). The study revealed that an important pre- requisite for the split-attention effect to occur is that the information presented in an integrated format must not be redundant. On this basis, the redundancy principle postulates that having to integrate redundant information poses an un- necessary load on WM hindering information processing, as part of the process- ing capacity is devoted to the mental integration of the redundant sources of information. This understanding is also valuable for CASI. As discussed above (§3.4.1), experience affects the germane load of a task. In dealing with specialised terminology, provided that the aural information has been clearly perceived and correctly decoded by the interpreter, a term visually presented on screen may be perceived as redundant if effective terminological preparation has occurred and/or the term is available in the interpreter’s LTM, or if strategies are applied to bypass the difficulty posed by the speaker’s use of a specialised term. This may be expected to occur especially for terms automatically suggested by an ASR sys- tem, which currently does not discriminate between the terms presented (but see Vogler et al. 2019 for an alternative approach), as the interpreter has no control over what is shown on the screen. In other words, unnecessary multi-modal pre- sentation of a stimulus term may impose additional extraneous load which may impair processing, rather than facilitate it by alleviating WM.f The split-attention effect operates on the extraneous load by reducing the strain imposed on WM. The modality effect arises from a different approach, 78 3 Simultaneous interpreting as a complex cognitive activity 3 Simultaneous interpreting as a complex cognitive activity Barrouillet et al. (2004) look at CL in terms of the time allocated to the items that must receive attention in order to be processed. Their model reconciles pre- viously held views that the inherent limitation of WM was dependent on the amount of cognitive resources available (a matter of resource-sharing) and other models which define WM spans as dependent on the duration of the processing component (and thus view it as a question of time-sharing, see Towse & Hitch 1995). Barrouillet and his colleagues posit that: 1. attention is required both by storage and by processing, and, being limited, it must be shared between the two; 2. when attention is switched away from retention, memory traces decay; 3. information retrieval from memory is constrained by a central bottleneck (Pashler 1998, Rohrer et al. 1998), which requires attention (and so mainte- nance suffers); 4. if processing involves retrieval tasks, attention sharing is time-based due to the aforementioned bottleneck, and can thus be viewed as rapid switching between processing and maintenance. Barrouillet and colleagues define CL as the number of retrievals:time ratio for tasks in which retrievals are all of the same type and difficulty. Thus, CL can be measured as the time during which attention is captured. This understand- ing of CL may also be applied to translation and interpreting, which share simi- lar processes of information retrieval from LTM for target text production. The availability of a system which takes care of information retrieval (in the form of a translation memory or of an ASR system for terminology) on the part of the translator and the interpreter may have beneficial effects on processing time and alleviate CL. If, however, processing of the visually presented information re- quires a considerable amount of attentional resources, attention may be diverted from retention of the auditory input, with potentially negative effects on task performance and, as predicted by the model, also in terms of processing speed. 3.4.3 Cognitive load as time-based resource sharing Despite the structural differences in the models of (working) memory proposed by Baddeley and Hitch and Cowan (see §3.1), a notion shared by both theories is that of memory decay. Because of the inherent limitation of WM, due to the processing-retention trade-off, items temporarily stored in memory decay if they are not refreshed or attended to, which leads to performance deterioration. This is particularly relevant in multitasking, where information retention and process- ing compete for cognitive resources. 79 3.5 Cognitive implications in interpreting 3.5 Cognitive implications in interpreting As discussed in the previous sections, the main criticisms towards the possibil- ity of integrating digital support tools into the in-process phase of interpreting may be traced back to a view of interpreting primarily in terms of cognitive pro- cessing and specifically as a matter of resource sharing and attention allocation to several co-occurring tasks (see §2.3.4). Early models of the cognitive processes involved in SI approaching this activity from the perspective of multitasking have favoured a bird’s eye-view of the issue. They have taken the valuable step of iden- tifying relevant sub-processes involved in SI (Lederer 1981) and of underscoring the capacity constraints which affect multiple task performance (Kirchhoff 1976). In the present study, I adopt a “micro-cognitive approach” (Muñoz & Martín 2020: 57) to the analysis of specific conditions under which SI occurs, investigat- ing the effects of consulting digital terminological resources as a way to cope with specialised terms, common problem triggers in SI. In doing so, I will focus on the effects on the CL of the co-occurring tasks of interpreting and interacting with the computer. Therefore, in the following sections, I discuss and compare two cognitive pro- cessing models of (simultaneous) interpreting which have approached the phe- nomenon primarily as a question of multitasking: Gile’s Effort Model of SI and Seeber’s Cognitive Load Model (CLM) of SI. These models interpret issues surfacing in the interpretation as problems in the allocation of limited cognitive resources between competing tasks. There are several reasons for choosing to focus on this subset of models over other cognitive, psycholinguistic and neurolinguistic/neurophysiological mod- els of SI (see p. 3–5 in the Introduction). First, both Gile’s and Seeber’s models are cognitive models from the area of Cognitive Interpreting Studies which the present work is situated into and are grounded in the notions of cognitive psy- chology discussed in the previous sections of this chapter. Second, they have already been adopted as a theoretical reference in CAI research (see §2.4). Third, they allow modelling the interaction with external written traces (e.g. glossaries and transcripts of the ST) in addition to the internal interaction and competition between cognitive subprocesses. As such, they allow modelling several sources of cognitive load, both intrinsic and extraneous (see §3.4.1). Fourth, they both appear directly relevant to the present research object. 3.5 Cognitive implications in interpreting The models and theories discussed so far were developed within the field of cog- nitive and learning psychology. Due to the highly influential role played by these disciplines in the early development of research paradigms for Interpreting Stud- ies (see Pöchhacker 2004: 61, Ferreira et al. 2015), the notions of WM, CL, atten- tion allocation and resource sharing have found wide adoption in the theoretical and empirical inquiry into the black box of the interpreter’s mind. 80 3 Simultaneous interpreting as a complex cognitive activity 3 Simultaneous interpreting as a complex cognitive activity After illustrating and discussing the models, I motivate the theoretical frame- work chosen for the present empirical investigation and discuss its application to the present object of inquiry. After illustrating and discussing the models, I motivate the theoretical frame- work chosen for the present empirical investigation and discuss its application to the present object of inquiry. 3.5 Cognitive implications in interpreting Gile’s Effort Model of SI incorporates specifically specialised terminology as a source of increased ef- fort, whereas Seeber’s CLM has already been applied and validated in studies on simultaneous interpreting with text, which shares many features with CASI (as discussed in §3.6), in particular in terms of the interpretation of specialised termi- nology. Finally, they lend themselves to being tested in a laboratory environment while allowing to study SI under relatively naturalistic conditions. 81 3.5 Cognitive implications in interpreting 3.5 Cognitive implications in interpreting SI is thus defined as the sum of the three efforts: SI = L (or R) + M + P + C Gile assumes that if the sum of the resources required by the different efforts does not exceed the sum of the attentional capacities available, then SI is feasible: SI = R (the sum of attentional resources for each effort) ≤A. The same is true for each sub-process, which must not recruit “more than the specific capacity available to it” (Setton 2003). Otherwise, problems arise. It is important to clarify that the additivity of the efforts is not intended in the arithmetic sense, as the efforts also overlap and compete for resources (Gile 2009: 184, Gile 1999: 4). The equation should not be seen as static either, but can rather vary during the interpreting process according to variations in task difficulty. In this respect, Gile (2008) introduces the notion of “local cognitive load” to indicate that overload may happen around occasional sources of difficulty. Gile’s EM of SI is integrated by the Tightrope Hypothesis (Gile 1999), i.e. the assumption that interpreters work close to saturation levels most of the time (Gile 2009: 198). Like tightrope walkers, interpreters must constantly strive to keep their balance between the individual sub-tasks which require careful coor- dination. According to Gile, EOIs therefore arise when the system is saturated as the interpreter experiences cognitive overload due to the inability to effec- tively deal with what Gile (1999: 157) defines as “problem triggers”, e.g. proper names, specialised terms, numbers and enumerations, as they require increased cognitive resources, or because of suboptimal allocation of resources to the in- dividual efforts. This is particularly true for novices, while expert interpreters may be better capable to effectively deal with problem triggers and may have better resource management, but are nonetheless constantly exposed to the risk of cognitive saturation. A second addition to the effort models is the Gravitational Model of Language Availability (see also §1.2.3.1), which is useful to intuitively explain why spe- cialised terminology may trigger overload or processing issues. Gile added this component based on the observation that the effortfulness of speech compre- hension and production was influenced by the availability of “Units of Linguistic Knowledge”. When words or expressions are used often, they are more readily available (they gravitate to the centre of the model). 3.5.1 Gile’s Effort Model of simultaneous interpreting Gile (1988, 1997, 1999) developed his Effort Models (EM) to explain why errors, omissions and “infelicities” (EOIs, see Gile 2011, 2015), i.e. suboptimal rendition of the source speech, occur in interpreting. As specified by the author, the EMs should not be considered as “operational testing or calibrating” (Gile 1991b: 18) tools but were rather designed for the classroom to intuitively explain certain phenomena that can be observed in SI (Gile 2009: 188). For this reason, they have been regarded as a useful pedagogical tool. Nonetheless, the EMs have encoun- tered the favour of many researchers and have been adopted as a conceptual framework “with explanatory and predictive potential on the level of actual in- terpreting performance” (Gile 1999: 2). In developing his models, Gile refers to a key concept widely accepted in cogni- tive psychology and put forward by Shannon & Weaver (1949), i.e. that controlled processes are managed by an inherently limited system, developing further the considerations expressed by Kirchhoff (1976). Such is the case in interpreting, which involves several sub-processes. These operations are not automatic, but rather require the active allocation of limited cognitive resources. Through the development of interpreting expertise, however, some components of these in- herently controlled operations can become automatic, freeing up more process- ing capacity and reducing the chances of EOIs. Gile defines the sub-processes involved in interpreting as “efforts” and identi- fies: • a listening and analysis effort (L), or more generally, a reception effort (R), as it was later renamed to account for visual perception during sign language interpreting and of additional visual inputs (e.g. PowerPoint pre- sentations) • a memorisation effort, i.e. the storage of the information to be processed in memory (M) • a production effort, i.e. the delivery of the message in the target language (P), including self-monitoring • a coordination effort (C) which is responsible for the allocation of the atten- tional resources and the successful concurrent performance of the above- mentioned sub-tasks 82 3.5 Cognitive implications in interpreting Otherwise, they tend to drift outwards, and their understanding and/or production is more effortful. Accord- ing to the Gravitational Model, technical terms which cannot easily be retrieved 83 3 Simultaneous interpreting as a complex cognitive activity from LTM may impose heightened attentional requirements on the listening ef- fort (as the term may not be easily recognised and understood) as well as on the production effort. Interpreters may cope with such problem triggers by consult- ing electronic dictionaries or digital glossaries, although this may require time and attention (Gile 2009: 219). Although reference is quite often made to “the” Effort Model to designate the EM of SI, Gile developed several effort models to provide an intuitive rep- resentation of the efforts involved in different modes of interpreting, e.g. SI with text, consecutive interpreting, interpreting from sign languages, and even remote interpreting. Where additional components seem to require considerable addi- tional efforts on the part of the interpreter at the detriment of other efforts, Gile adds effort components to the equation, such as the human-machine interaction (HMI) effort for remote interpreting, a notion which may also be applied to SI with digital terminological support. Despite the conceptual nature of the Effort Models, they have been adopted as a research framework in a number of studies. Most of these studies have provided support for the notion that the efforts are not automatic (Gumul 2018: 19) and that the human processing capacity is limited, a generally-accepted principle which is also at the core of Baddeley and Hitch’s model of WM (Baddeley & Hitch 1974, see §3.1). Additionally, the findings by several studies (e.g. Gile 1999, Gumul 2018, Gile 2011, Matysiak 2001), including a recent ERP study by Koshkin et al. (2018) have been interpreted as support for the tightrope hypothesis. Gile’s Effort Models have been subjected to criticism by other interpreting scholars. For instance, Pym (2008) conducted an additional analysis of omissions in Gile’s (1999) experiment stressing their functional role, which is not in contrast with the key principles of the model, but not explicitly included either. Other researchers have criticised the lack of temporal resolution of the model (Pöch- hacker 2016). The harshest critic of the model is Seeber (2011), who focuses on the tightrope hypothesis. 3.5 Cognitive implications in interpreting 3.5 Cognitive implications in interpreting or to equate the memory effort with WM and the coordination effort with Bad- deley & Hitch’s (1974) CE. Since the effort models present some parallels with key concepts of cognitive psychology2, it is easy to understand where the mis- understanding stems from, as discussed by Gile (2009) himself. After all, similar interpretations of Gile’s Effort Models can also be found among cognitive psy- chologists (Gile 2009). Gile responded to these criticisms in subsequent and regular updates of the models (e.g. Gile 2016, 2017b, 2020). He observed that most criticism stems from the use of the models in empirical cognitive explorations of the interpreting pro- cess despite them having been conceived to provide a “holistic and intuitive” (Gile 2017b: 10) explanation of EOIs. The intended use of the models is to provide prospective interpreters with a functional explanation of why problems arise in interpreting, and to help them develop and reflect upon tactics and strategies to be used in interpreting (Gile 2020). p g Nonetheless, Gile (2009: 205) claims that “the competition-between-Efforts principle is consistent with the theory of one central pool of processing capac- ity, not with the theory that there may be several pools that the Efforts can draw upon without there being interference with them”. However, a multiple resource model does not negate interference. Rather, it offers a framework to differentiate between tasks which can be performed simultaneously without in- terference and tasks which interfere, and to describe the extent to which they do. If analysed from the perspective of theories formulated in cognitive psychology, it seems to me that Gile’s (2009) “competition hypothesis” may also be explained by the interference of sub-tasks which draw from separate dimensions but share resources. This would be compatible with Wickens’s model (§3.3). Indeed, empir- ical evidence taken as support of the tightrope hypothesis (Koshkin et al. 2018, Gumul 2018) mainly points to interference between efforts, but does not explore the structural reason for this interference. After all, the EM does not aim to pro- vide an architectural explanation underlying information processing.f The notion that the co-existence of multiple “efforts” may increase capacity requirements (Gile 1999: 156) would also hold true within the framework of an architectural model assuming interference between tasks which share resources in some dimension (e.g. Wickens’s multiple resource model, see §3.3). 2For instance, the non-automaticity of the sub-processes, the inherent limitation of the process- ing capacity, the idea of attention allocation and of task interference. 3.5 Cognitive implications in interpreting Seeber agrees that the tightrope hypothesis aptly de- scribes cases in which the source speech presents specific sources of difficulty, such as high delivery rate, non-native accent or high information density. How- ever, he observes that multitasking may very well be possible in interpreting, as commonly observed among professional interpreters, who, unlike budding interpreters, in some cases may have enough cognitive resources left to spare. The misunderstanding about the EM (e.g. Gile 2017b) probably originates from Seeber’s view of the interpreting process from the perspective of cognitive psy- chology, which leads him to interpret the equation provided by Gile in the Effort Models as a reflection of Kahneman’s (1973) single resource theory (see §3.3), 84 3.5 Cognitive implications in interpreting This is the case for SI, during which auditory-verbal resources are required both in the recep- tion phase (Listening + Analysis Effort) and in the response phase (Production Effort). Additionally, Gile suggests that it might be useful to “develop ‘tuning’ or 85 3 Simultaneous interpreting as a complex cognitive activity ‘scaling’ rules for the quantification of processing capacity or time requirements for interpretation tasks” (Gile 1991b: 18). This may be achieved through the “de- mand vectors” proposed by Wickens (1976, 1984, 2002). Finally, the assumption that interpreters’ resources are at risk of saturation most of the time has not yet received empirical support, as suggested by cases of perfect time-sharing (Schu- macher et al. 2001, see also Seeber 2011). Gile concedes that “though the evidence supports the hypothesis of cognitive saturation, it does not necessarily show that such saturation occurs at global level” (Gile 2008: 61). Seeber & Kerzel’s (2012) eyetracking study also found support for a local increase of CL, rather than a general increase of pupil dilation compared to the baseline. Note, however, that the tightrope hypothesis does not state how close interpreters supposedly come to saturation (Gile 2017a). Gile further observes that pupillometry can only cap- ture instances of heightened load if the interpreter does not intervene to prevent it, and thus the risk of overload may not show in the data. It would be interest- ing to conduct further studies using other physiological measures (see §4.5 for an overview), for instance sensitive to stress, which is likely to be experienced when having to deal with problem triggers, and to verify whether this correlates with pupillary dilation. To sum up, Gile’s EMs interpret EOIs in the TT as evidence of the competition between efforts recruiting resources from a single pool of mental capacity. The models have had the merit of conceptualising the idea of the distribution of re- sources between concurring tasks through a series of components (the tightrope hypothesis, the gravitational model, and the competition-between-efforts princi- ple) which have proved quite productive in training. The main limitation of the models for the present exploration of CASI lies in the lack of components modelling the structural nature of the interference between tasks and explaining why, for instance, the human-machine interaction and the listening or the production effort may compete for resources. 3.5.2 Seeber’s Cognitive Load Model of Simultaneous Interpreting Unlike Gile, Seeber explicitly derives his model from cognitive psychology. He adapts Wickens’s model of task interference to SI, while introducing some vari- ations (Seeber 2007). First of all, he turns the three-dimensional model into a bi-dimensional model. This offers a double advantage. First, it enables to graphi- cally represent all components of the model at once (some were “hidden” in the non-visible faces of the cube). Second, it reintroduces the general capacity, the pool of general resources postulated by Wickens, into the model, which did not previously fit into the cube. Seeber further expands the model by adding a Cog- nitive Resource Footprint (CRF), a visual representation of which resources are shared between the co-occurring tasks. Finally, he keeps Wickens’s conflict ma- trix, with some modifications. Seeber uses this adaptation of Wickens’s model, his own CLM, to represent and explore resource sharing during four language processing tasks: shadowing, sight translation, SI (Seeber 2007: 1383) and SI with text (Seeber 2017). 3.5 Cognitive implications in interpreting This level of detail appears, however, essential for interventions on the design of support tools aiming to reduce such interference. Furthermore, the lack of a clear stance as to the models’ grounding in cogni- tive psychology, coupled with unresolved issues preventing an unequivocal vali- dation of the tightrope hypothesis, represent limitations for the models’ applica- tion to experimental research aiming for an operational testing of its theoretical framework, which the present study aspires to do. For these reasons, I examine Seeber’s CLM of SI in the following section and I illustrate the motivation for choosing the CLM as theoretical framework for the present study in §3.5.3. 86 3.5 Cognitive implications in interpreting 3In the original model (Seeber 2007), the TIS was equal to 9. The higher score in the revised model (Seeber 2017) is determined by the addition of demands on visual-spatial resources due to the inclusion of paraverbal information provided by the speaker. 3.5.2.1 Cognitive Load Model of Simultaneous Interpreting Seeber describes SI as a combination of two main tasks: 1) a listening and compre- hension task and 2) a production and monitoring task. Listening and comprehen- sion recruits auditory-verbal and cognitive-verbal resources at the perceptual- cognitive stage: conference interpreters receive the auditory stimulus, i.e. the source speech as pronounced by the speaker, and must analyse the verbal mes- sage for comprehension. In a more recent adaptation of the model (Seeber 2017), a visual-spatial component was added to account for the paraverbal informa- tion the interpreter perceives while interpreting (for visual information in in- terpreting, see also Seubert 2019). Interpreters then “respond” to this stimulus by delivering the message in the target language. At the same time, they must monitor their own rendition. The production and monitoring task thus demands auditory-verbal and cognitive-verbal resources at the perceptual-cognitive stage and additional vocal-verbal resources at the response stage. Figure 3.2 represents the CRF of SI with visual input. Figure 3.2 represents the CRF of SI with visual input. As described in §3.3, Wickens’s model represents the level of dependence of a task on a resource by assigning demand vectors comprised between 0 (no de- pendence) and 2 (extreme dependence). Seeber postulates a demand vector of 1 for each of the concurring sub-tasks (resources), which in SI occur at both stages of processing (perception + cognition, responding), in both perceptual modali- ties (auditory and visual) and in the two processing codes (spatial and verbal). 87 3 Simultaneous interpreting as a complex cognitive activity recruited resource shared resource Perception Cognition Responding Responding Cognition Perception Ambient Visual Processing Focal Visual Processing Auditory Spatial Visual Verbal Visual Spatial Auditory Verbal Manual Spatial Vocal Verbal Manual Spatial Vocal Verbal General Capacity Figure 3.2: Cognitive resource footprint for SI with visual input (Seeber 2017: 468) General Capacity Visual Spatial A Manual Spatial Figure 3.2: Cognitive resource footprint for SI with visual input (Seeber 2017: 468) The revised model foresees three possible levels (0, 0.5, 1), where the use of half vectors serves to model complementarity between information sources (specifi- cally between the auditory verbal and the visual spatial modalities/codes), while redundant information (see §3.4.2.2) within a stage is represented with a full (1) demand vector (Seeber 2017: 481). 3.5.2.1 Cognitive Load Model of Simultaneous Interpreting The revised model foresees three possible levels (0, 0.5, 1), where the use of half vectors serves to model complementarity between information sources (specifi- cally between the auditory verbal and the visual spatial modalities/codes), while redundant information (see §3.4.2.2) within a stage is represented with a full (1) demand vector (Seeber 2017: 481). The sum of the demand vectors and conflict coefficients thus assigned to SI is equal to a total interference score (TIS) of 11.6 (Figure 3.3).3 The application of the model to the three tasks of shadowing, sight translation and SI operationalises the hypothesis that SI requires the higher amount of indi- vidual resources as compared to the other tasks, as it involves more concurrent sub-processes, which results in the highest level of task interference. The CLM of SI in its original version (without the inclusion of visual informa- tion) was tested by Seeber & Kerzel (2012) in an eyetracking experiment on the 88 3.5 Cognitive implications in interpreting vector 0.5 ∅ ∅ 0.5 0.5 0.5 ∅ ∅ demand visual spatial visual verbal auditory spatial auditory verbal cognitive spatial cognitive verbal response spatial response verbal ∅ visual spatial 0.8 0.6 0.6 0.4 0.7 0.5 0.4 0.2 ∅ visual verbal 0.6 0.8 0.4 0.6 0.5 0.7 0.2 0.4 ∅ auditory spatial 0.6 0.4 0.8 0.4 0.7 0.5 0.4 0.2 1 auditory verbal 0.4 0.6 0.4 0.8 0.5 0.7 0.2 0.4 ∅ cognitive spatial 0.7 0.5 0.7 0.5 0.8 0.6 0.6 0.4 1 cognitive verbal 0.5 0.7 0.5 0.7 0.6 0.8 0.4 0.6 ∅ response spatial 0.4 0.2 0.4 0.2 0.6 0.4 0.8 0.6 1 response verbal 0.2 0.4 0.2 0.4 0.4 0.6 0.6 1.0 listening comprehension perceptual cognitive response perceptual cognitive response production & monitoring Total interference score = demand vectors + conflict coefficients = (1 + 1 + 1 + 0.5 + 0.5 + 0.5 + 0.5) + (0.4 + 0.5 + 0.2 + 0.8 + 0.7 + 0.4 + 0.5 + 0.6 + 0.4 + 0.7 + 0.8 + 0.6) = 11.6 Figure 3.3: Conflict matrix for SI with visual input (Seeber 2017: 469) Figure 3.3: Conflict matrix for SI with visual input (Seeber 2017: 469) 89 89 3 Simultaneous interpreting as a complex cognitive activity effect of asymmetric syntactical structures between English and German on the interpreter’s CL. 3.5.2.1 Cognitive Load Model of Simultaneous Interpreting The authors applied the CLM to represent local variations in CL for verb-final sentences and symmetrical structures, predicting higher CL for asymmetrical structures. The CLMs effectively predict that “the interpretation of syntactically asymmetrical structures causes more CL than the interpretation of syntactically symmetrical structures towards the end of the sentence” (ibid., p. 238). The CLM may thus be used not only to model interference between tasks on a global level (i.e. listening and comprehension vs monitoring and production), but also to formulate more in-depth hypotheses on local CL.f Gieshoff (2018, 2021) conducted an experiment aimed at verifying whether see- ing the speaker’s lip movements during SI reduces CL. Inter alia, she used silent pause duration as an indicator of CL and found that silent pauses were shorter when the speaker’s lips were visible. Her findings lend support to the hypothesis that auditory-verbal and visual-spatial information is integrated, as predicted by the CLM in alignment with abundant empirical evidence on multimodal integra- tion (see §3.4.2.2). It should be noted that pupillometric data did not support the hypothesis, though Gieshoff interpreted larger pupils for the multimodal con- dition as an indicator of higher arousal rather than increased cognitive effort (Gieshoff 2018: 242). Seubert (2019) also investigated the processing of visual information during SI with an exploratory eyetracking study conducted on a sample of 13 professional interpreters. The naturalistic research design, which favours the proximity to a real-life interpreting situation rather than a strict control of empirical variables, does not allow for an in-depth analysis of the cognitive interactions between different types of visual support with the interpreting process. Nonetheless, Seu- bert’s research provides valuable observations, some of which offer support to the predictions of the CLM. In particular, the reported strategic behaviour of ex- perienced professionals in dealing with different sources of visual input supports Seeber’s hypothesis that strategies modulate the allocation of cognitive resources in order to avoid cognitive overload, and that the supportive or distracting po- tential of visual input depends, inter alia, also on the type of input processed. For instance, the observation that Seubert’s test subjects devoted a high proportion of their visual attention to the speaker supports the hypothesis that paraverbal information may provide a valuable integration of the auditory channel. 3.5 Cognitive implications in interpreting 3.5 Cognitive implications in interpreting in CL during the interpreting session is also valuable. Before and after their in- terpreting turn, the interpreters’ visual perceptual field appears to be larger than during the more cognitively demanding phases, which Seubert interprets as a higher amount of cognitive resources available for visual processing due to a lower CL (supporting Lavie’s theory, as discussed in §3.2). This observation also supports the idea that the assumptions of the tightrope hypothesis (Gile 1999) may be valid more on a local level rather than on a global level, as pointed out in Seeber & Kerzel (2012). 3.5.2.1 Cognitive Load Model of Simultaneous Interpreting This is in line with Gieshoff’s findings for lip movements, although in Seubert’s experi- ment the speaker area of interest was much larger due to the situated nature of her experiment. It also aligns with the predictions of the CLM, but it should be noted that Seubert did not explicitly test such predictions. Her finding that infor- mants’ visual processing behaviour varied along with the (postulated) variations 90 3.5.2.2 Cognitive Load Model of SI with text The addition of the visual-spatial component to the CLM does not include written information, which according to standard models of WM is processed in the PL, while gestures, lip movements and expressions are processed in the VSSP (see §3.1). To account for the processing of visual-verbal (written) information during SI, Seeber (2017) applied his model to a specific instantiation of SI with written information, i.e. SI with the full transcript of the speech (SIMTXT). For SIMTXT, the inclusion of the written text requires the addition of visual-verbal resources to the cognitive resource footprint (Figure 3.4).l In the conflict matrix, visual-verbal processing is added to the listening and comprehension phase and receives a full demand vector. The attribution of a demand vector of 1 reflects the duplication of aural information in written form. Hence, the total interference score is higher than for SI without written input (14.3), as can be seen from the expanded conflict matrix (Figure 3.5). The application of Seeber’s CLM of SIMTXT was adopted as a theoretical framework in a recent experiment by Seeber et al. (2020) and by Chmiel, Ja- nikowski & Cieślewicz (2020). Seeber and colleagues report on an eyetracking experiment designed to test how attention is allocated to redundant written in- formation during SI. The authors contrasted SIMTXT with reading while listen- ing (RWL) as a control task. They found that interpreters attend to the visual- verbal support during their production, probably in order to “offload short term memory” (Seeber et al. 2020: 13), rather than exploiting redundancy effects to improve their comprehension. This is in line with Seeber’s CLM which postu- lates the attribution of shared vectors to visual-spatial and auditory-verbal re- source demands (i.e. for complementary information), and not for visual-verbal and auditory-verbal stimuli. In essence, their redundancy may require excessive effort for their integration to be possible in the listening and comprehension phase, but may prove beneficial during the production phase to assist with con- trol processing. 3.5.2.2 Cognitive Load Model of SI with text 91 3 Simultaneous interpreting as a complex cognitive activity recruited resource shared resource Perception Cognition Responding Cognition Responding Perception Ambient Visual Processing Focal Visual Processing Auditory Spatial Visual Verbal Visual Spatial Auditory Verbal Manual Spatial Vocal Verbal Manual Spatial Vocal Verbal General Capacity Figure 3.4: Cognitive resource footprint for SI with text (Seeber 2017: 471) General Capacity Visual Spatial A Manual Spatial Figure 3.4: Cognitive resource footprint for SI with text (Seeber 2017: 471) Chmiel, Janikowski & Cieślewicz (2020) investigated source language inter- ference in an English to Polish SI and sight translation (ST) task. They measured the number of cognates, homographs and passive structures, in addition to time lag and total translation time as indicators of CL, adapting Seeber’s CLM to op- erationalise their hypotheses. They observe that in ST the visual-verbal input is the only source of information, and thus its processing differs from SIMTXT and would deserve different values for the demand vectors and the conflict coef- ficient, though they do not explain what demand vector may be assigned for ST. Additionally, they propose a higher conflict coefficient for cognitive verbal vs re- sponse verbal demands in ST based on the inherent higher complexity of textual processing, but this contrasts with Wickens’s observation that “the adjustment of conflict values should not be based on differences in single task demands, since these [are] captured by the single task analysis shell” (Wickens 2002: 170). Figure 3.5: Conflict matrix for SI with visual-verbal input (Seeber 2017: 472) 3.5.2.2 Cognitive Load Model of SI with text Fur- thermore, in their adaptation, they do not mention removing the visual-spatial component, which Seeber (2017) had added to account for paraverbal information 92 3.5 Cognitive implications in interpreting vector 0.5 1 ∅ 0.5 0.5 1 ∅ ∅ demand visual spatial visual verbal auditory spatial auditory verbal cognitive spatial cognitive verbal response spatial response verbal ∅ visual spatial 0.8 0.6 0.6 0.4 0.7 0.5 0.4 0.2 ∅ visual verbal 0.6 0.8 0.4 0.6 0.5 0.7 0.2 0.4 ∅ auditory spatial 0.6 0.4 0.8 0.4 0.7 0.5 0.4 0.2 1 auditory verbal 0.4 0.6 0.4 0.8 0.5 0.7 0.2 0.4 ∅ cognitive spatial 0.7 0.5 0.7 0.5 0.8 0.6 0.6 0.4 1 cognitive verbal 0.5 0.7 0.5 0.7 0.6 0.8 0.4 0.6 ∅ response spatial 0.4 0.2 0.4 0.2 0.6 0.4 0.8 0.6 1 response verbal 0.2 0.4 0.2 0.4 0.4 0.6 0.6 1.0 listening & reading comprehension perceptual cognitive response perceptual cognitive response production & monitoring Total interference score = demand vectors + conflict coefficients = (1 + 1 + 1 + 0.5 + 1 + 0.5 + 0.5 + 1) + (0.4 + 0.5 + 0.2 + 0.6 + 0.7 + 0.4 + 0.8 + 0.7 + 0.4 + 0.5 + 0.6 + 0.4 + 0.7 + 0.8 + 0.6) = 14.8 Figure 3.5: Conflict matrix for SI with visual-verbal input (Seeber 2017: 472) Figure 3.5: Conflict matrix for SI with visual-verbal input (Seeber 2017: 472) 93 93 3 Simultaneous interpreting as a complex cognitive activity from the speaker. In ST, this component is not relevant. This would have proba- bly resulted in an even lower TIS for ST than for SI. Their findings on time lag (longer for ST than for SI) did not elicit a unequivocal interpretation, as the exter- nal pacing in SI inherently affects time lag in SI. Additionally, total translation time was found to be longer in SI, which, contrary to their predictions, might point to higher CL in SI than in ST. Despite some limitations in the authors’ adaptation of the model, the CLM proved a valuable tool for the formulation of hypotheses on SI with text. In her 2019 experiment, Seubert also included written support material in the wide array of visual input presented to her test subjects. She did so by observing the interpreters’ behaviour during the interpretation of citations, which they had not prepared beforehand (Seubert 2019: 198). 4Although they are not the only models to do so. Other authors from different research paradigms have also included visual information in their models (for a review, see Seubert 2019). 3.5 Cognitive implications in interpreting 3.5 Cognitive implications in interpreting natural environment, and we seem to have evolved to expect and rely on them, the way in which we process them may well depend on the composition of the signal” (Seeber 2017: 463). Conducting targeted studies on visual input during interpreting therefore seems motivated. The analysis of Seeber’s CLM and of relevant research in the area of simulta- neous interpreting with visual support has highlighted how Seeber’s conceptual- isation of SI and SIMTXT may represent a more suited theoretical starting point for the present study than Gile’s EMs (see §3.5.1). The following section reviews the main differences between the two models and illustrates the rationale behind the choice of Seeber’s CLM for the present inquiry. 3.5.2.2 Cognitive Load Model of SI with text Seubert reports that, when faced with longer text passages, 8 out of 11 interpreters averted their gaze from the speaker and the slide containing the quote. She interprets this as an indication of higher cognitive effort, which is corroborated by numerous hesitations and deficient content in the renditions (ibid., p. 209). This observation may lend sup- port to the hypothesis that redundant information replicated across two different channels may lead to an increase in CL, which the CLM predicts. It is interest- ing to observe that for more isolated textual information, which only partially replicates the spoken input (e.g. the presentation table of contents, p. 199), the interpreters’ gaze continuously switched from the speaker to the slides, which Seubert interprets as an indication that this type of visual support may be use- ful both for the planning and for the production and monitoring phase (p. 205). These findings partially contrasts with Seeber et al.’s (2020) findings that redun- dant textual support is used mainly in the production phase. It should be noted that Seeber’s experiment was conducted in a laboratory setting under stringent variable control, which reduced the visual input to the speech transcript. In Seu- bert’s experiment, the sheer amount of sources of visual information may have contributed to the perceptual overload which led the interpreters to avert their gaze and not to consult the slides containing the quotes. Her second observation that isolated information is consulted during both phases should be confirmed under more strict experimental conditions. However, I should mention that her findings were for a very specific type of support, i.e. a table of contents, which is different from a speech transcript and may therefore also require fewer plau- sibility checks on the part of the interpreters, as they may expect it to be a re- liable source of information with fewer discrepancy than the written transcript compared to the spoken discourse. Different findings for different types of vi- sual support further corroborate Seeber’s observation that “although signal com- plementarity and signal redundancy appear to be important components of our 94 3 Simultaneous interpreting as a complex cognitive activity Gile’s Effort Model may be less strict in that in principle it allows to include facilitation effects by, for instance, combining auditory and visual processing in a single reception effort (see Gile 2009). If facilitation effects due to redundancy across different modalities were found (though, at present, this is not yet the case), the lower reception effort would leave more processing resources available for memory, coordination and production. However, it may be argued that, since aural and visual stimuli are processed separately, the elaboration of additional stimuli should be modelled as an extra effort, as suggests Gieshoff (2018: 74). This would require distributing attentional capacities across a higher number of sub- tasks, with fewer resources available for each task. As for terms as problem triggers, they are discussed explicitly in Gile (2009) as potential causes of locally increased CL during SI. They may pose a higher load on the Reception Effort (if the term cannot be quickly recognised or identified) as well as on the Production Effort (if the target language equivalent is not available). As Wickens (2002) postulates a three-tier system for the allocation of demand vectors, it may theoretically be possible to postulate higher demands also within the framework of Seeber’s CLM, either for the listening and understanding task or for the production task, or for both. Despite the differences discussed above, evidence congruent with both ac- counts of multitasking during interpreting has been found (among recent studies, see for instance Chmiel, Janikowski & Cieślewicz 2020, Gieshoff 2021, Seeber et al. 2020). Yet, the two models have often been contrasted as antithetic, as stressed by Seeber (2011) himself. Approaching inquiries into CL from the perspective of cognitive psychology, Seeber views Gile’s model as reflective of Kahneman’s sin- gle resource theory, which contrasts with his CLM based on Wickens’s multiple resource theory. As discussed above (§3.5.1), Gile does not support this inter- pretation of his model and further criticises the CLM as “not indicative of what actually happens in the booth” (Gile 2020: 9). In my view, the two models do not necessarily contradict each other. Rather, Gile’s EM offers a broader, more holis- tic view of multitasking in SI. 3.5.3 Discussion and choice of a model for the present work The two models which have been in focus in the previous sections (§3.5.1, §3.5.2) represent SI essentially as a multitasking activity. They both postulate a lim- ited attentional capacity constraining WM (though Gile refers to a more general “Memory Effort”, see Gile 2009) and explicitly include the processing of textual information during SI (Gile 2009, 2020, Seeber 2017).4 The textual component is pertinent to the present inquiry since terminology support is also presented in the visual-verbal modality and in part replicates the auditory input.f One important difference lies in Seeber’s clear stance as to which framework he grounds his model in, i.e. Wickens’s multiple resource theory. As such, his model is clearly rooted in cognitive psychology. As the present work is con- ducted within a cognitive processing paradigm adopting consolidated methods derived from TPR, which have in turn largely been adopted (and adapted) from cognitive psychology, choosing Seeber’s CLM as the theoretical framework to operationalise my hypotheses seems the more coherent approach. Additionally, the model has been used to effectively predict a number of effects of interaction with visual-verbal input in SI (as discussed in §3.5.2.2) and to my knowledge, no empirical evidence has been found yet to contrast its assumptions. Though the initial model did not provide for facilitation effects due to crossmodal integra- tion (Seeber 2007), Seeber’s adaptation of Wickens’s model through half-scores for the individual demand vectors allows to include such effects and to model hy- potheses in this respect. In this sense, the CLM always posits higher or equal CL with the addition of further sub-tasks (e.g. interacting with a speech transcript or with a computer for terminology look-up), never lower. 95 3 Simultaneous interpreting as a complex cognitive activity Seeber’s CLM may, on the other hand, be seen as a micro-cognitive model of the interpreting process, rooted in the theories and findings of cognitive psychology and allowing for the formulation of fine-grained hypotheses on how the sub-components of interpreting may interfere with each other, and of why certain sub-components do so more than others. As the present study contrasts the provision of terminological support through three different kinds of digital tools, this level of granularity may prove particu- larly suited to identifying specific sources of extraneous load and modelling their impact on the CL of CASI. Adopting a model of the interpreting process origi- nating from a psychological model should further facilitate the interpretation of 96 3.6 A model of SI with digital terminology support results in the present study in view of accepted theories stemming from cognitive psychology. gy dditionally, Seeber’s CLM presents several specificities: Additionally, Seeber’s CLM presents several specificities: 1. It defines local CL as “a function of input and output features” (Seeber 2011: 189) in relation to the amount of parallel processing and the amount of time for which elements must be stored. 2. It shows how, while interpreters might indeed reach maximum CL locally, most of the time they work below this “red line”. 3. It accounts for local variations in CL, which can be explored at a micro- scopic level. The output in SI is seen as the result of strategies aimed at managing the limits inherent to the task at hand and at saving elaboration capacity, as exemplified by his application of the model in the framework of an experiment investigating the impact of syntactical structure for the language pair German-English on CL in SI (Seeber 2011). 4. It is able to “account for the conflict potential posed by an overlap [of tasks] and the interference they cause” (Seeber 2011: 189). 5. It illustrates how the overall cognitive demands are affected by the differ- ent combinations of sub-tasks. 6. It offers a “first attempt at quantifying CL, relying principally on Wickens’s demand vectors and conflict coefficients” (ibid.). It is this level of granularity and flexibility of the model, already adapted to SI with text, which may represent the most valuable aspects of the CLM. 3 Simultaneous interpreting as a complex cognitive activity Addi- tionally, by postulating different demand vectors for a certain sub-task, it would be possible to further differentiate between diverse interpreting scenarios, pre- dicting differences in the CL inherent to SI performed under different conditions and tracing them back precisely to the interference of defined sub-processes. For the above-mentioned reasons, Seeber’s framework is best suited to oper- ationalise hypotheses on digital terminological support during SI through stan- dard digital glossaries, CAI tools, and ASR-CAI hybrids. 3.6 A model of SI with digital terminology support SI can be carried out in forms which add elements to its basic constellation of sub-tasks and resources recruited, as is the case for SI with text, or for assign- 97 3 Simultaneous interpreting as a complex cognitive activity ments during which presentations are shown during speech delivery, an addi- tional input that is also processed by the interpreter. What happens, then, when the interpreter searches for terminology in a glossary while interpreting simul- taneously? Or when an interpreter is automatically prompted with terminology by an ASR system? Which cognitive resources are recruited, and in which of the four dimensions identified by Wickens? How much do said resources inter- fere with each other and why? As discussed in the previous sections (3.5.2), the CLM has been used to formulate hypotheses on “standard” SI (with visual input) and on SI with text. In order to formulate hypotheses around these questions, I propose to apply Seeber’s CLM to SI with digital terminology support. In previous publications (Prandi 2017, 2018), I had suggested an adaptation of Seeber’s CLM to SI with digital glossaries, CAI tools and ASR based on the ini- tial version of the model (Seeber 2007, 2011, Seeber & Kerzel 2012), as at the time of submission I was not yet aware of his more recent publication proposing an application of the model to SI with text (Seeber 2017). In my initial application of the CLM to SI with digital terminology support, I had not discussed the beneficial audio-visual integration effects due to the processing of paraverbal information provided by the speaker. The visual-spatial component had been included to ac- count for visual search operations required in the case of morphological neigh- bours for target terms (Prandi 2018) and had therefore received a full demand vector. I will therefore revise and update my proposed application of the CLM to SI with digital glossaries, CAI tools and ASR based on the more recent expansion of the model (Seeber 2017). 3.6 A model of SI with digital terminology support 3.6 A model of SI with digital terminology support however, has made it relatively quick and easy to perform online queries in ter- minological resources, either peri-process or in-process. Glossaries are only one example of such resources, as mentioned in §1.2.3. For the scope of the present inquiry, I will focus on the consultation of terminological resources as a coping tactic (Gile 2009: 203) to deal with specialised terminology. I will do so by ex- amining three potential types of digital support interpreters may utilise in order to cope with specialised terminology: digital glossaries, CAI tools, and ASR-CAI hybrids. 3.6.1 Application of Seeber’s CLM to SI with terminological support Specialised terminology may not constitute a source of difficulty if the target language equivalent is readily available during the interpretation. In that case, the interpreter recruits cognitive resources to retrieve the equivalent from LTM. This may also result in higher CL, without, however, disrupting the rendition. Thorough preparation and terminological activation ahead of the assignment may help prevent impasses due to the use of specialised terminology. Indeed, assignment preparation, including terminological preparation, is considered an essential constituent of a successful interpreting performance, and is therefore of- ten presented as a valuable strategy that interpreter trainees must learn to apply (e.g. Gile 2009). When the term cannot be retrieved from LTM, a series of coping tactics may be applied, such as paraphrasing, transcoding, generalising, and so forth (for an overview and discussion of different coping tactics, see Gile 2009). The progressive penetration of the digital medium in the interpreting profession, 98 3.6 A model of SI with digital terminology support 3.6.2 Manual terminology look-up When terminology is manually looked up during SI, additional attentional re- sources are recruited. In order to search for a term in a digital glossary or a CAI tool database, the interpreter must perform several operations: (a) type a term or part thereof to query the database (and in most traditional digital glossaries, also click the enter key), (b) locate the relevant term on the page/in the list of terminological pairs, and (c) read the term and integrate it into the rendition. It should be noted that locating the term may not always require visual search in a digital glossary prepared, for instance, with Word or Excel or saved as a PDF. In all of these cases, all occurrences for the string searched may be highlighted, so the interpreter would need to visually identify the relevant term. However, if no orthographic neighbours are present in the glossary, only one term will be highlighted. The visual cue should attract the interpreter’s visual attention through a bottom-up mechanism (see for instance Seubert 2019), rendering the identification of the target term more agile. Of course, much depends on the search settings chosen by the interpreters and on the strategies they adopt. For instance, one may select “search only whole words”, to further reduce the num- ber of results. For bigrams and trigrams, i.e. terms composed of more than one component, the number of results may depend on which element is searched. For instance, in a speech on nuclear energy, several types of reactors may be mentioned: if the glossary contains “boiling water reactor”, “pressurised steam reactor” and “nuclear reactor” and the interpreter looks for “reactor”, all three results will be highlighted and the interpreter may need to skip to the next oc- currence or scroll down in the case of a multi-page glossary, whereas searching for “boiling” will only highlight one term and no further operations will be re- quired. In a CAI tool such as InterpretBank, results which do not correspond to the search string are not displayed thanks to a dedicated search algorithm. How- ever, visual search may still occur, as for digital glossaries, if several orthographic neighbours are shown on the screen. 99 3 Simultaneous interpreting as a complex cognitive activity However, InterpretBank poses three main advantages. First, it reduces the amount of visual input to be processed. Second, it does not require the user to position the cursor in the search field. This happens automatically in the Con- ference Mode: after a query, the programme is ready for the next search. If this operation has not been automatised by the interpreter, a digital glossary may cause the querying process to slow down, as the user has to prepare for the fol- lowing query. Third, the search bar is cleared automatically after each query in InterpretBank, which is not the case for a standard digital glossary. Forgetting to clear the search bar at the end of a query may interfere with other sub-processes during the following query. Using the CRF from the CLM as theoretical framework, there is first of all a “response” to the auditory stimulus during manual look-up. This recruits manual- spatial resources in addition to the vocal-verbal response, i.e. the interpreter’s rendition. It seems correct to assign the terminology query to the production and monitoring stage as the manual query is conducted to support target lan- guage rendition. If the term has not been recognised, no manual query can be performed. Furthermore, visual-spatial resources are recruited at the perceptive- cognitive stage, to identify the term, while visual-verbal resources are required to read the term (Figure 3.6). SI with a glossary therefore shares something with sight translation and with SI with text, which also require visual-verbal resources during perception/cognition. As observed by Gieshoff (2018: 74–75), we may ex- pect the glossary query to “interfere with the speech, the auditory stream, as both elements are verbal and cognitive-perceptual inputs”. The cognitive resource footprint of SI with manual look-up (digital glossary/ InterpretBank) illustrates the recruitment of additional resources (Figure 3.7). It should be noted that the CRF for SI with terminology look-up applies only to the instances in which a query occurs, and not to the entire interpreting task. When no query is performed, one can apply Seeber’s CLM for SI. 3 Simultaneous interpreting as a complex cognitive activity In sum, the CRF posits a recruitment of: • the general capacity available for all tasks • visual-verbal, visual-spatial and auditory-verbal resources at the percep- tion and cognition stage • visual-verbal, visual-spatial and auditory-verbal resources at the percep- tion and cognition stage • manual-spatial, vocal-verbal and auditory-verbal resources at the produc- tion and monitoring stage Thus, auditory-verbal and cognitive-verbal resources are recruited during both sub-tasks. Thus, auditory-verbal and cognitive-verbal resources are recruited during both sub-tasks. 100 3.6 A model of SI with digital terminology support The different levels of recruitment of cognitive resources may thus be visually represented by the conflict matrices 101 3 Simultaneous interpreting as a complex cognitive activity recruited resource shared resource Perception Cognition Responding Cognition Perception Ambient Visual Processing Focal Visual Processing Auditory Spatial Visual Verbal Visual Spatial Auditory Verbal Manual Spatial Vocal Verbal Manual Spatial Vocal Verbal General Capacity Figure 3.7: Cognitive Resource Footprint for SI with manual terminol- ogy lookup in a digital glossary or a CAI tool (Prandi 2018: 37) General Capacity A Manual Spatial Visual Verbal Figure 3.7: Cognitive Resource Footprint for SI with manual terminol- ogy lookup in a digital glossary or a CAI tool (Prandi 2018: 37) of SI performed with the support of different tools, and could in turn result in higher total interference scores. At this stage, however, these remain theoretical speculations which need to be verified empirically. 3.6 A model of SI with digital terminology support 3.6 A model of SI with digital terminology support A comparison with the conflict matrix for “standard” SI (with visual input, Figure 3.3) highlights how manual terminology look-up during SI should produce a higher CL, not only because more resources are mobilised, but also because they are shared across a higher number of sub-tasks. recruited resouce shared resource Look up Identify term Read term Perception Cognition Responding Cognition Perception Ambient Visual Processing Focal Visual Processing Auditory Spatial Visual Verbal Visual Spatial Auditory Verbal Manual Spatial Vocal Verbal Manual Spatial Vocal Verbal General Capacity Figure 3.6: Additional cognitive resources recruited during SI with manual terminology lookup in a digital glossary or a CAI tool (Prandi 2018: 36) Look up General Capacity Identify term Figure 3.6: Additional cognitive resources recruited during SI with manual terminology lookup in a digital glossary or a CAI tool (Prandi 2018: 36) As discussed above, however, CAI tools may offer a series of advantages which have the potential to partly reduce the additional load resulting from the interac- tion with the computer. This does not emerge from the CRF. In his model, Wick- ens (2002: 172) contemplates the use of three levels in the assignment of demand vectors to represent cases in which a sub-task recruits a certain resource to a high degree. Seeber applies this principle in the CLM of SI with visual input and of SIMTXT, where a demand vector of 0.5 is assigned in the case of multimodal integration due to beneficial modality effects (see §3.5.2.1 and §3.5.2.2). In theory, it could be possible to assign a higher demand vector, for instance 1.5, to further differentiate between the tools used for manual look-up, which may differ in the degree of recruitment both of visual-spatial resources and of manual-spatial re- sources at the production/monitoring stage. 3.6.3 Automatic terminology look-up A CAI tool with ASR integration does not require active operation of the techno- logical support by the interpreter. Specifically, the tool does not recruit manual- spatial resources. Nonetheless, visual-spatial and visual-verbal resources are re- cruited to locate and process the term visualised on the screen.f A further differentiation from tools with manual look-up lies in the poten- tially double advantage offered by speech recognition: the tool may support both comprehension and production processes. As observed by Pym (2011), technolo- gies applied to translation and interpreting essentially externalises cognitive pro- cesses, acting as an external memory, or as a second brain. On the one hand, an ASR tool performs operations similar to those performed by the PL: it decodes 102 3.6 A model of SI with digital terminology support 3.6 A model of SI with digital terminology support the acoustic information and maps it against its internal lexicon, which results in “understanding”, i.e. in the presentation of the term pronounced by the speaker on the screen. On the other hand, when coupled with an extraction tool, as is the case for the ASR integration in InterpretBank, it interacts with its “LTM” (the glossary) to retrieve the target-language equivalent. This is similar to the purpose of a translation memory, which retrieves pre-processed text segments facilitating the translator’s work. Using an ASR tool may thus be valuable to fa- cilitate terminology recognition during listening and, as for tools with manual look-up, to optimise the rendition of specialised terms during production. recruited resource shared resource Perception Cognition Responding Responding Cognition Perception Ambient Visual Processing Focal Visual Processing Auditory Spatial Visual Verbal Visual Spatial Auditory Verbal Manual Spatial Vocal Verbal Manual Spatial Vocal Verbal General Capacity Figure 3.8: Cognitive Resource Footprint for SI with ASR support General Capacity Figure 3.8: Cognitive Resource Footprint for SI with ASR support The cognitive resource footprint for automatic terminology look-up during SI is represented in Figure 3.8. Of course, the advantage described above presup- poses that the term recognition is successful. This is not the case in 100% of the cases with ASR: if the wrong term is recognised (for instance due to homophones or to a non-native accent), an imprecise or wrong suggestion, or a lack thereof, might go against the interpreter’s expectations, resulting in irritation and thus potentially causing additional CL. Additionally, the system’s latency also plays 103 3 Simultaneous interpreting as a complex cognitive activity a role: long latencies lead to a drop in accuracy and fluency (for a dedicated analysis, see e.g. Montecchio 2021). Although current systems seem to perform sufficiently fast (Brüsewitz 2019), in real-life laptops do not always perform as expected and may become slow due to a number of reasons. If the term is not pre- sented quickly enough on the screen, the potentially beneficial effect resulting from cross-channel redundancy may be hindered by a lack of temporal conti- guity (see Mayer & Fiorella 2014): the tool’s suggestions may be perceived as a distraction rather than as a facilitation (Van Cauwenberghe 2020). 3.6 A model of SI with digital terminology support A recent study by Chmiel, Janikowski & Lijewska (2020) has focused precisely on the issue of incongruence between auditory and written input, although in the framework of SI with text. The authors’ findings indicate that when both auditory and written stimuli are presented, interpreters tend to focus on the vi- sual modality. When the stimuli are incongruent, the visual modality interferes with the auditory input to the point that interpreters include incorrect units of information in their rendition – in this study, this happens especially for terms. Similar findings of imported errors from incorrect visual support for terms were also found in an experiment by Van Cauwenberghe (2020) on the provision of ASR support for specialised terminology. It is interesting that such effects were found not only for students (Van Cauwenberghe 2020), but also for professional interpreters. Chmiel, Janikowski & Lijewska (2020) also found that incongruent items tended to be fixated longer than congruent ones. It may be argued that similar sources of irritation or distraction may also arise with manual terminology look-up: provided that the interpreters have correctly identified the term heard, they may not find it in the glossary, or may not find it quickly enough: in the case of very large glossaries, or if the interpreters have not had sufficient time to thoroughly prepare, this may not be a seldom occurrence. Even though they exceed the scope of the present study, the above-described sce- narios are more than probable in real-life and would deserve further dedicated explorations, both in terms of their effect on cognition and on target-text pro- duction. 104 4 Measuring cognitive load As pointed out in the literature, CL “cannot be observed and measured directly” (Chen 2017: 648). The reason for this is that CL is a multidimensional theoretical construct, the definition and nature of which are still debated both in the field of psychology, where it originated, and in Translation and Interpreting Studies (see §3.4). When aiming to estimate CL or to assess it empirically, researchers have therefore traditionally relied upon the observation and measurement of phenom- ena that arise as a consequence of high cognitive effort and reflect its variations (Paas et al. 2003, Ehrensberger-Dow et al. 2020). Several methods and measures have been adopted to this aim, good overviews of which are offered by Seeber (2013), based on Paas et al. (2003); Chen (2017), who also presents measures of workload, a construct derived from human workload studies; and Ehrensberger- Dow et al. (2020), who also report on innovative measures and thoroughly dis- cuss their limitations. In this chapter, I review and discuss the methods and mea- sures adopted in TPR for assessing CL as a key factor in the translation and interpreting process, in order to motivate and frame my methodological choices (§5.5.6). Five main approaches may be identified in the measurement of CL or of the cognitive effort required by a particular task. The methods used to measure CL may be placed on a double continuum. On the one hand, methods may be divided according to the amount of intervention by the researcher. On the other hand, they may be classified by whether they require the active involvement of the experiment participant in providing cognitive-load related data, or whether this onus entirely falls on the researcher. This in turn reflects a more subjective ver- sus a more objective type of measure. Due to inherent limitations of each type of measure, cognitive research often adopts a mixed-method paradigm, where several measures are combined “in a complementary fashion to elucidate the phenomenon of cognitive effort from different angles” (Ehrensberger-Dow et al. 2020: 222). For instance, performance measures may not provide reliable esti- mates of effort because a good or stable performance may mask the high mental effort invested to achieve this result. Combining this type of data with more di- rect measures of cognitive effort may unveil effects of a specific intervention on 4 Measuring cognitive load the load imposed by the task. 4 Measuring cognitive load As all methods present both advantages and po- tential drawbacks, as will be discussed in the following sections, a combination of methods represents a good way to maximise the potential of the individual methods while addressing their respective limitations (see §5.3). 4.1 Theoretical approaches The first category of approaches for CL estimation comprises methods that do not require empirical experimentation and focus on the task, rather than on the task performer. They have been defined as analytical methods (Paas et al. 2003: 66) as they rely on a theoretical analysis of the task at hand conducted on the basis of models. Examples of this approach in interpreting research are Gile’s Effort Models (1995, 2009) and Seeber’s Cognitive Load Model of SI (2007, 2011) and of SI with text (Seeber 2017). Analytical methods rely on the analysis of task characteristics and are thus detached from individual performance. They repre- sent an a-priori estimation of the CL generated by a particular task (Chen 2017: 647). They may be useful in producing hypotheses to be tested empirically, espe- cially in order to contrast the relative load imposed by similar tasks (see Seeber 2011, Seeber & Kerzel 2012). However, they may only provide an a-priori estima- tion of CL, not a direct measurement of cognitive or mental effort. Without ex- perimental validation, they remain theoretical accounts of the postulated sources of CL for a task, and as a theory they are “unable to take into account individual differences” (Seeber 2013: 22). 4.2 Subjective approaches Further on the continuum, subjective measures represent approaches which place the spotlight on the individual experience of the subject involved in the experiment. The adoption of subjective methods is based on the idea that “in an attempt of in-depth understanding and in order to find causal relationships, it is necessary to ask the subject” (Carl & Hansen 2011: 3). While this offers the ad- vantage of highlighting individual differences and identifying phenomena not ac- counted for in theoretical models, it is open to “a possible contamination of data by memory and consciousness effects” (Seeber 2013: 19). Subjective measures rely on the active involvement of the participant in providing a personal estimate of the effort experienced during task performance, mainly through introspection, post-hoc questionnaires, retrospective interviews or think-aloud protocols (TAP) (ibid.). 106 4.2 Subjective approaches 4.2 Subjective approaches Subjective data are often reflected in unidimensional or multidimensional psy- chometric rating scales (Chen 2017: 648). The latter can prove particularly use- ful in identifying the individual factors that may contribute to variations in CL. An example is represented by the NASA Task Load Index (NASA-TLX, Hart & Staveland 1988), widely used in mental workload research. It requires rating on six subscales measuring mental, physical and temporal demands, performance, effort and frustration, which can be altered or integrated depending on the re- search design. Chen (2017) identifies the NASA-TLX as a potentially useful asset in research on simultaneous and consecutive interpreting. An adaptation of this rating scale was recently applied by Gieshoff (2018) in combination with pupil- lometry measures and by Sun & Shreve (2014) in a study on translation difficulty. An important caveat of this approach is that the rating should be conducted im- mediately after task completion and “possibly be supplemented with cues (e.g., processed texts or recordings of the process) to facilitate recall” (Ehrensberger- Dow et al. 2020: 224). As suggested by Gile & Lei (2020: 275), the inclusion of subjective methods in the research design, especially retrospection, may provide valuable clues for the interpretation of objective measures. A recent example is the adoption of ret- rospective questionnaires by Chmiel, Janikowski & Cieślewicz (2020) in a study on source language interference in sight translation and SI, which supported the research team in the interpretation of combined temporal measures and product analysis. Subjective measures, however, present several disadvantages (see Ehrensber- ger-Dow et al. 2020). 4.2 Subjective approaches First, careful consideration is required in the choice of re- search devices to collect subjective data. In using questionnaires, one must con- sider that forced-choice questionnaires limit the subjects’ reporting and might prime their answers, while open questions may be more difficult to process sta- tistically. The issue of priming may also affect the outcomes of interviews due to the presence of the researcher, which may also reduce the participant’s willing- ness to report experiences perceived as embarrassing, not noteworthy, or poten- tially undermining their perceived professionalism. This reluctance is an issue often encountered in studies involving professional interpreters as subjects. Ret- rospection must occur close to the experimental intervention to be effective, but this may significantly extend the experiment’s duration and potentially result in fatigue effects. Finally, TAPs have the disadvantage of potentially slowing down the process due to the concurrent verbalisation and task performance (Jakobsen 2003). Additionally, only what is verbalised will become known to the researcher, while other important aspects may not emerge from studies only adopting sub- jective measures. 107 4 Measuring cognitive load 4.3 Performance measures Performance measures combine the advantage of analysing and comparing indi- vidual performances with the measurement of objective parameters. In cognitive psychology, they have traditionally been applied using primary task measures, i.e. by measuring task performance, or by requiring the concurrent performance of a second task and measuring how its performance is affected by variations in the CL generated by the first (secondary task measures). In order to provide reliable indications of CL, experiments conducted using this approach require a stringent empirical design in order to avoid effects due to variables not con- trolled by the researcher (Seeber 2013: 20). This may require sacrificing some ecological validity to avoid uncontrolled effects. When high ecological validity is to be maintained, secondary task measures may be unsuitable, as they interfere with the main task (Ehrensberger-Dow et al. 2020: 227). A downside of using per- formance measures is that they often involve human raters, which may affect the degree of objectivity of these measures and requires strictly defined evaluation criteria. In studies on translation and interpreting, the two most common aspects of performance analysed are accuracy and disfluencies, especially pauses. This is based on the assumption that performance suffers from heightened or excessive cognitive effort and that issues of cognitive resource management may therefore emerge on the linguistic surface. 1The reader may refer to §1.1.2 for an overview of said evaluation frameworks. 4.3 Performance measures 4.3 Performance measures be adopted to evaluate the product of human translations, particularly through a contrastive approach with the MT output. An example is the study by Carl & Buch-Kromann (2010) on the differences between student and professional translators. The study used BLEU scores and human evaluations to assess the accuracy and fluency of the target texts, combining performance measures with behavioural measures (eye movements and keystrokes, see §4.4). Taxonomies of error types aimed at assessing the accuracy of the linguistic output in interpreting were proposed e.g. by Barik (1971), Altman (1994), Waden- sjö (1993: 2002) and Napier (2004, 2016). The dimensions of accuracy range from the word to the text level and the error and omission categories may reach a high level of granularity, but may also be very simple depending on the scope of the study. For instance, in a previous contribution (Prandi 2015a,b), I evaluated the accuracy of the terms interpreted with a simple dychotomic classification of “terms translated as per glossary” (yes/no). An issue in the assessment of accuracy in interpreting, mostly operationalised in terms of errors and omissions, lies in the fact that the taxonomy tends to be modified or devised anew by each researcher. Therefore, “there are nearly as many error classification systems as there are empirical studies requiring an overall assessment of source–target correspondence” (Pöchhacker 2004: 143). A limitation of the application of widely used frameworks for the assessment of translation as discussed above is that they have not been explicitly formulated for the oral modality. Consequently, they need to be adapted for the evaluation of the interpreting performance. An example can be found in Xu (2015). In her doctoral thesis on corpus-driven preparation for interpreters, she incorporated, merged and redefined several error categories from BlackJack, MeLLANGE and SAE J2450. Her error categories included incorrect terms, omissions, inappropri- ate collocations, grammatical errors, pronunciation errors and semantic errors. The scoring rubric was used to assess the accuracy of interpretations for which students had prepared either following a traditional approach or using corpus- driven terminology extraction. In TPR, a popular evaluation framework used to measure the accuracy of the translated product is MQM (see §1.1.2). As will be discussed in §6.2.1, MQM and other standardised frameworks for quality evalua- tion in translation may also be adapted to interpreting (as in Xu 2015) and used as methods for accuracy measurement in the area of CAI. 4.3.1 Accuracy As discussed in §1.1.2, accuracy has been identified as an important indicator of performance, both for translation and for interpreting. Accuracy may be as- sessed either holistically, i.e. on the global level of the text, or componentially (Tiselius 2015: 3), on an analytical level, i.e. by evaluating errors or omissions in individual components of the product (paragraphs, sentences, or words). A po- tential drawback of assessing the accuracy of the target text, be it written or oral, is that it is necessary to clearly define and describe the scale used for the eval- uation. In some cases, the involvement of the researcher in the evaluation may contribute to reduce discrepancies between evaluators (Hansen 2009: 395–397). Studies conducted on the product of machine translation engines have addressed the limitations of human evaluation by adopting established taxonomies for word or phrase-based assessment of MT, for instance in the area of post-editing stud- ies (e.g. Vardaro et al. 2019, Marzouk 2021).1 Such taxonomies may, however, also 108 4.3.2 Dysfluencies and pauses In psycholinguistics, pauses in speech production have been investigated as cor- relates of cognitive effort (e.g. Goldman-Eisler 1972) and this association has been 109 4 Measuring cognitive load confirmed in a number of studies (e.g. Dragsted 2004, Lacruz et al. 2012, O’Brien et al. 2014, Kumpulainen 2015). In spoken discourse, interruptions in the speech flow may, however, also be adopted intentionally as a rhetorical device for em- phasis or simply to draw breath. In addition, pause patterns (duration and fre- quency) tend to vary from each individual to the next (O’Brien 2006). Hence, the interpretation of pauses as indicative of cognitive processing or speech planning is not unequivocal, and is therefore often complemented by other measures in research on translation and interpreting. In SI, since the source speech unfolds continuously during the task, pauses may also suggest that the interpreter is wait- ing for more linguistic material from the speaker. Furthermore, the position of pauses in the rendition may also reflect an alignment of the interpreter’s output to the speaker’s rhetorical style. Methodologically, measuring pauses also poses a number of difficulties, for instance in establishing what should be considered as a filled or silent pause and in determining the minimum duration of pauses. Current audio processing tools represent useful research aids as they allow to automatically identify silent pauses according to a set of criteria selected by the researcher. Other types of dysfluencies usually require manual tagging and counting (Gieshoff 2021: 185). This can make the analysis of dysfluencies other than silent pauses rather labori- ous. For this reason, depending on the amount of data, researchers may choose to investigate macro or micro-locations of pauses (Dragsted & Hansen 2008).f In interpreting, different levels of intentionality behind pause patterns emerge from studies comparing students and professional interpreters. Students’ pauses tend to reflect hesitations and more effortful processing (e.g. Tissi 2001, Mead 2002, 2005). Conversely, experienced interpreters show more deliberateness in their pause patterns (see Cecot 2001, Ahrens 2005, 2007). Therefore, studies in- volving students as test subjects may find the analysis of pauses as indicators of cognitive effort useful. In a study comparing written translation and sight trans- lation, Dragsted & Hansen (2009) calculated the number and duration of pauses to contrast text production patterns for interpreting and translation, suggesting that the inclusion of the oral modality in translator training may improve the output quality. 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches overview, see Lacruz et al. 2016). For instance, a study on the coordination of comprehension and production processes in translation by Dragsted & Hansen (2008) found that pauses indicate a coordination effort due to the transition be- tween the two phases. In her study on post-editing, O’Brien (2006) found that pauses helped in the identification of a correlation between source text quality and post-editing effort (p. 17). In a study contrasting spoken translation (i.e. writ- ten translation obtained through target text dictation) and post-editing of MT, Carl, Lacruz, et al. (2016a) analysed the pause structure of target text production, finding a more coherent generation of translations when dictation was used. In research on the post-editing process, the two most widely used metrics are the average pause ratio (APR) and the pause-to-word ratio (PWR). Both were devel- oped after O’Brien’s (2006) influential work on post-editing, where she suggested a simple pause ratio measured as the total pause duration divided by the total duration of the post-editing task for a certain segment. The APR (O’Brien et al. 2014) is calculated as the average pause time divided by the average post-editing time per word for each segment. The APR was found to be reflective of cogni- tive effort, as the prediction that more numerous, shorter pauses may indicate more effortful post-editing was confirmed in a study on post-editing (Spanish to English) and later found not to be influenced by the pause threshold chosen (Lacruz et al. 2014). The PWR (ibid.) indicates the number of pauses divided by the number of source words per post-edited segment. This measure correlates strongly with APR as well as with other ratings of source quality (e.g. HTER, see Snover et al. 2006). This measure was further refined in Lacruz et al. (2018), where ranges of pauses of different lengths were considered in the analysis. Clusters of short monitoring pauses were found to correlate with cognitive effort during post-editing, suggesting that they might be indicative of monitoring processes. 4.3.2 Dysfluencies and pauses A recent example of studies using pause patterns as correlates of CL is provided by Gieshoff (2018, 2021), who involved 14 interpreting students as her test subjects and used silent pauses to contrast SI with and without visi- ble lip movements. Filled pauses (uhm) were analysed as indicators of cognitive effort in a corpus study by Plevoets & Defrancq (2016, 2018) and were found to correlate with the informational load of the input speech. Pause-related metrics have also been successfully developed and implemented as indicators of cognitive effort in translation and post-editing (for a clear 110 4.4 Behavioural approaches The collection of behavioural data, together with neurophysiological, hematic and cardiac correlates (see §4.5), occupies the other extreme of the objectivity continuum. Behavioural measures do not rely on the subjective evaluation by par- ticipants nor on the product-oriented evaluation of the performance, but rather on the measurement of phenomena arising in the translator’s or interpreter’s brain and body that have been empirically linked with cognitive activity (Jakob- sen 2017: 22). The higher degree of objectivity of these approaches represents a clear ad- vantage compared to other measures: translation events may be measured and 111 4 Measuring cognitive load analysed to gain insight into the translation processes which guide the partici- pant’s behaviour. This fundamental assumption, underlying cognitive psychol- ogy and neuroscience, foregrounds the usefulness of empirical methods derived from these disciplines for the investigation of the translation and interpreting process. However, TPR has also developed its own additional measures and anal- ysis tools to explore the translation process, hence the widespread adoption of mixed-method approaches in this research area. In the following subsections, I discuss the main behavioural measures which have been adopted to evaluate the cognitive load and effort involved in various types of translation activity. Behavioural measures may be divided in the two overarching categories of measures of time and measures of eye movement. 4.4.1.1 Speed and task time The first subgroup of measures concerns the amount of time required to produce a unit of the target text or the entire target text. These measures are based on the assumption that difficult tasks tend to take longer than easier tasks (O’Brien 2008: 87). An example may be word production time (WPT), i.e. the total time required to produce a word in the target text in translation, including revisions (Carl, Lacruz, et al. 2016b: 3). Another example may be found in O’Brien (2008), where processing speed was measured to establish the cognitive effort involved in the translation of segments with the support of different levels of fuzzy matches in the TM of a CAT tool environment. On a broader level, some study designs may obtain useful information by mea- suring the time required to perform the entire task. For instance, Jakobsen & Hvelplund (2008) compared “task time” in reading for comprehension, for trans- lation, while interpreting and while translating. Chmiel, Janikowski & Cieślewicz (2020) measured “translation time” to compare SI and sight translation. This ap- proach presents some limitations for interpreting, since especially SI is inher- ently constrained by the speaker’s speech rate and because individual differences in the participants’ delivery rate may be expected to play a role. Therefore, these factors should be taken into account when adopting overall measures of speed as indicators of effort. 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4.1 Time-related measures Both for translation and for interpreting, it is generally assumed that there ex- ists a “rough correlation between time spent on translating a word or passage and the cognitive effort invested in solving a problem or in making a decision between competing solutions or strategies” (Jakobsen 2017: 30). Metrics based on the overall time required by the language processing task in question or by a sub-process thereof, or related to the transposition of individual elements of the source text into the target language may therefore provide useful and objective insights into the cognitive effort exerted by the participant. In research on writ- ten translation, post-editing and audio-visual translation, the use of time-based measures as indicators of the cognitive effort involved in producing a transla- tion is facilitated by the use of keystroke logging, or key-logging, which enables a direct exploration of how the translation process unfolds over time. The advan- tage of the method is evident for the exploration of various types of translation activities, as testified by the development of dedicated tools, such as Scriptlog (Andersson et al. 2006), Translog and Translog-II (Jakobsen 2006), and Inputlog (Leijten & Van Waes 2013). However, key-logging may also hold the potential to gain further insight into time-related aspects of human-machine interaction dur- ing computer-supported or computer-mediated interpreting. An example may be the automatic recording of the user activity in CAI tools to produce log files which can be used for empirical analysis, as has been done in studies using In- terpretBank (see Biagini 2015, Prandi 2015a,b, 2018). With or without the support of key-logging, time measures have largely been employed in studies on interpreting as well as on translation. Two main types of time-based measures may be identified: on the one hand, speed-based measures, on the other, time-lag measures. 112 4.4.1.2 Time lag: EVS and EKS The second type of time-related measures focus on the time-span elapsing be- tween the time of production of a specific element in the target text (written or oral) and its appearance or perception in the source text. Often described with the general term of “décalage” (Timarová 2015: 418), this time-span may represent different aspects of the process according to the object of investigation. Perhaps the two most widely adopted measures of this kind are the ear-voice span (EVS) in interpreting, especially simultaneous, and the eye-key span (EKS) in transla- tion, which was derived from the first (Dragsted & Hansen 2008). The EVS has been used not only to explore temporal aspects of the interpreting process (Pöchhacker 2004: 117), such as processing speed, but it has also been shown to provide a reliable indication of cognitive effort in SI (e.g. in Treisman 1965, Barik 1971, Shlesinger 1998, Timarová et al. 2011). In interpreting, the duration of the EVS is taken as an indicator of cognitive effort: a shorter time required by the interpreter to produce the target-text equiv- alent of the source text, or of a specific source-text element, is generally under- stood as suggesting faster and, hence, less effortful processing. Timarová et al. 113 4 Measuring cognitive load (2011) reviewed the various methods used to measure time lag in interpreting and compared EVS with EKS. The different methods used for the measurement of time-lag depend on the specific object of investigation, and can range from a broader and more surface-level analysis of EVS to establish reference values for SI, to a comparison of mean or median values across tasks or even to explore whether local variations in EVS reflect variations in cognitive processing. The most common approach is to measure time lag in number of seconds, but num- ber of words has also been used. In literature, the average EVS for professional conference interpreters has been identified as comprised between 2s and 4s (see for instance Barik 1973, Lederer 1978, Oléron & Nanpon 2002, Christoffels & De Groot 2004, Defrancq 2015, Ti- marová et al. 2011), though shorter and longer time lags have also been observed. It may even be negative “where true anticipation occurs” (Timarová 2015: 419). Overall, however, research has linked an EVS longer than 4s with a loss in ac- curacy (Lee 2002, 2003, Timarová et al. 2014). 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches imposed by the subprocesses involved and on the strategies guiding the user’s choices. Studies using EKS have produced telling results on the underlying pro- cesses of translation. For instance, Dragsted (2010) identified two different trans- lating styles typical of professional and budding translators. The first favour an integrated style, where comprehension, production and revision processes are strictly interlinked, as shown by the shorter EKS, comparable to that of simul- taneous interpreters (an average 2.8s in her group of participants). The second tend to work sequentially, as shown by the longer EKS (7.2s on average), which indicates longer time elapsing between the comprehension and the translation of a specific text unit. 4.4.1.2 Time lag: EVS and EKS It is generally assumed that val- ues longer than these point to processing issues on the part of the interpreter and may negatively affect the perception of quality, as they often result in long or frequent filled or silent pauses. However, the EVS has been found to vary both between and within tasks and to be influenced by external factors (e.g. the speaker’s output rate, language combination) as well as by individual factors (e.g. the interpreter’s own delivery rate, individual cognitive makeup, experience). For instance, professional interpreters have been found to have both longer (Moser- Mercer et al. 2000) and shorter EVS (Timarová et al. 2014) than trainees: this discrepancy might be interpreted as an indication that they are better able to adjust their décalage than interpreting students thanks to their expertise. The interpretation of the EVS as indicative of cognitive effort is, therefore, not straightforward. This is particularly relevant when the EVS is not measured as a broad indicator of processing speed, but to contrast the speed of rendition of individual elements of the source text. Nonetheless, the EVS may be considered as a standard measure of cognitive effort in interpreting, and has been widely adopted in interpreting research, often in combination with further measures to address the already mentioned limitations. The use of the EKS to investigate mental effort in translation has become pos- sible thanks to the interfacing of key-logging and eyetracking (see §4.4.2), par- ticularly with the development of dedicated data collection environments such as Translog-II (Jakobsen 2006). Thanks to the integration of a gaze-to-word map- ping tool, Translog-II allows to effectively triangulate gaze data and participants’ typing behaviour, which provides detailed insight into the unfolding of the trans- lation, post-editing or subtitling process and offers valuable clues on the load 114 4.4.2 Eye movement measures and eyetracking The measurement of movements of the eye and of other phenomena linked to the eye physiology is possible thanks to a technique called eyetracking. Put simply, eyetracking is “a technology for recording eye movements” (Jakobsen 2019: 398). eye physiology is possible thanks to a technique called eyetracking. Put simply, eyetracking is “a technology for recording eye movements” (Jakobsen 2019: 398). The adoption of the eyetracking methodology for cognitive inquiries into dif- ferent translation activities rests on two fundamental theoretical tenets, sub- sumed in Just & Carpenter’s (1980: 331) eye-mind hypothesis and immediacy as- sumption. The eye-mind hypothesis postulates that observable eye movements, i.e. physical manifestations of overt attention, are strictly linked with covert at- tention and cognitive processes. Therefore, the object of visual attention is as- sumed to be the object of cognitive attention (Hvelplund 2017: 250) and the mea- surement of eye movements can be indicative of concurring mental processes. According to the immediacy assumption, there is no “lag between what is be- ing fixated [i.e. looked at for a period of time long enough to process it] and the hidden cognitive processes that take place inside the mind” (ibid.). The eye- mind assumption has however been challenged by the notion of mind drifting (see Posner 1980, Smallwood & Schooler 2006): it is possible for the mind to start wandering while looking at an object, a phenomenon commonly experienced in everyday life. This poses an important limitation on the eye-mind hypothesis, namely that this mind drifting cannot be observed nor measured, as it is not reflected in eye movements. Additionally, it has been empirically demonstrated that there is a certain lag between the focus of visual attention and what is being cognitively processed at a given moment (Jakobsen 2019: 409). Essentially, the eyes “seem to behave some- what like a dog on a leash held by the mind rather than there being a perfectly straightforward relationship” (Jakobsen 2017: 34). Nonetheless, as observed by Hvelplund (2014, 2017), while the researcher should be aware of these potential 115 4 Measuring cognitive load drawbacks, mind drifting during a cognitively taxing task such as translation is probably a rare occurrence. In interpreting, the probability of a cognitive shift of this kind can reasonably assumed to be even more remote, particularly during SI due to the immediacy of the task. 4.4.2 Eye movement measures and eyetracking Owing to this, eye movements can reliably be interpreted as “correlates of cognitive processing in translation” and interpreting, as validated in neighbouring disciplines (Hvelplund 2014: 211). Eyetracking has been used in several research areas, chiefly psychology, psy- cholinguistics and cognitive sciences as a way to empirically investigate human behaviour, cognition and attention (Hvelplund 2017: 248), and at a more basic level “to study the physiological mechanics of human eye movements” (Jakob- sen 2019: 398). The potential of this technique, however, makes it suitable to the investigation of behaviour, cognition and attention allocation also in cogni- tively taxing activities such as translation and interpreting, as will be discussed in §4.4.2.3. In the following sections, I describe the equipment used in eyetrack- ing and introduce the main measurements, with a focus on fixation-based and related metrics, as they have found wide application in TPR and in studies on human-computer interaction and may therefore also be applied to cognitive in- quiries into interpreter-computer interactions. 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches They are limited to the monitor they are integrated in or mounted on, but they are relatively unobtrusive as they do not require to be worn by the participant. At the same time, they allow for high-frequency recordings, essential due to the speed of eye movements, especially during translation (Jakobsen 2017: 36). Despite its vast potential for the objective exploration of cognition and atten- tion, the eyetracking methodology presents some limitations. First, it requires strict experimental conditions, as the lack of control of aspects such as light in- tensity or the participants’ distance from the screen may affect data quality and results. Second, while a certain level of control can be achieved through a proper laboratory set-up (Rösener 2016), other aspects, particularly those linked to par- ticipants’ characteristics and behaviour, may pose non-negligible issues to the re- searcher. A common example is the type of glasses or lenses worn by participants, which may create artefacts or impede the correct recording of eye movements. Finally, stress due to the experimental conditions cannot be ignored, although it may be contained by the use of remote eyetrackers which are less invasive and do not require a modification of the task being performed. Despite being contactless, however, they still require calibration, which makes the experimen- tal condition more apparent. Nonetheless, as pointed out in the literature (e.g. Hvelplund 2014: 206, Hansen 2008: 390), stress may be perceived by participants due to the simple fact of being observed during task performance (the so-called white coat effect). While these potential drawbacks and external influencing fac- tors cannot be eliminated entirely and should be considered during experimental design and data analysis, the advantages offered by eyetracking far outweigh its limitations, which explains its popularity in translation process research. 4.4.2.1 Equipment and applications The device used to produce a recording of where the eyes are looking is called eyetracker. What is recorded is not the image of the eye itself, but rather the reflection of infrared light on the cornea of the eye (Duchowski 2017: 54). The reflected light is recorded at a rate comprised between 30 and 2,000 Hz, which corresponds to 30 to 2,000 samples recorded per second (Jakobsen 2019: 399).fi Different types of eyetracker are available, each presenting benefits and po- tential drawbacks. Head-mounted eyetrackers, for instance, often used in com- bination with chin rests, bite bars or forehead rests, allow for high quality data, as the precision of the eyetracker is enhanced by the steady position of the sub- ject’s head. On the other hand, it poses severe restrictions on the participants’ freedom of movement, which may make this type of eyetracker unsuitable for certain types of research. On the other end of the spectrum are tracking eye- glasses, which offer the advantage of being very portable and of allowing re- searchers to take their investigation out of the laboratory, as they make it possi- ble to record eye movements on several planes, i.e. on surfaces different from a computer monitor. These eyetrackers, however, have the limitation of allowing recordings only at slow speed (30 Hz). A compromise between the two is offered by stand-alone or attachable remote eyetrackers (see Ehrensberger-Dow 2014). 116 4.4.2.2 Fixation-based measures Eyetracking allows to conduct several types of measurements. The most popu- lar eyetracking measures are fixation-based measures and saccades, quick eye movements occurring between fixations (Poole & Ball 2005). From fixations and saccades, other metrics may be derived, for instance gaze measurements. Eye- tracking also allows to collect additional types of data related to eye movements and eye physiology, such as blink rate and pupil size/dilation (see §4.5.3). Due to the popularity of the eyetracking technique, a series of metrics have become established as the norm of reference for the investigation of translation in a cognitive framework, often borrowing from neighbouring disciplines. At the same time, TPR has offered further confirmation of the ability of certain indica- tors to account for specific variations in cognitive effort or in the allocation of attentional resources, and to measure the “number and patterns of translation 117 4 Measuring cognitive load process activities, the duration of and switches between activities, or the fluency in production” (Ehrensberger-Dow et al. 2020: 225). In particular, fixation measures and measures derived therefrom have been widely used as dependent variables in TPR to determine the cognitive effort in- volved in the translation process, as they are more sensitive to linguistic factors (Staub & Rayner 2007). For this reason, as they will be relevant for the present inquiry, here I focus on these measures.2 Fixation-based measures provide useful indications concerning the “quality and intensity of cognitive attention in a task” (Jakobsen 2019: 402). A fixation is a period of time during which the eye remains stable (fixated) on a target, which is necessary for the ocular system to bring the object into focus (Duchowski 2017: 46). Its length is usually understood to be comprised between 200 and 300 ms in reading (Rayner 1998: 373, Holmqvist 2011: 381), but it “may be as long as several seconds” (Karsh & Breitenbach 1983, Young & Sheena 1975), and as short as 30– 40ms (Holmqvist 2011: 413). In particular, the duration and number of fixations index the amount of cognitive attention on the fixated object or area of interest (AOI). At the same time, they indicate the intensity of cognitive effort required for the performance of a task, for instance for the processing of individual words or phrases in translation. 2A thorough description of eyetracking measures applicable to translation process research goes beyond the scope of the present contribution. For a comprehensive discussion, a useful reference may be found in Conklin et al. (2018). 4.4 Behavioural approaches 4.4 Behavioural approaches 4.4 Behavioural approaches interpreting. While the following overview has no ambition of completeness, it will contribute to highlight the role of eyetracking as a useful methodology for TPR. 4.4.2.2 Fixation-based measures In literature, longer or more numerous fixations have often been associated with “a deeper and more effortful cognitive processing” (Holmqvist 2011: 413, Hvelplund 2014: 212, Lacruz et al. 2016: 1214). Among early measures, first fixation durations are particularly indicative of the amount of attention generated by an item, especially in reading when the area of interest is a single word (see Conklin et al. 2018: 124). Among late measures, fixation count and total gaze time, also defined as total fixation duration or total fixation time, are also used as indicators of attention and processing effort. If the interest is in how processing unfolds over time, time-related fixation measures may provide useful insights. Two widely used metrics of this kind are time to first fixation, which indicates how much time elapses before the AOI is fixated for the first time, and average fixation duration (Conklin et al. 2018: 129). Studies concerned with the focus of attention of a translator or of an interpreter may investigate gaze or fixation patterns to identify which area of the visual stimulus attracted the participant’s attention (Ehrensberger-Dow et al. 2020). In the following section, I present some examples of how the method and mea- sures discussed above have been applied in TPR, with a focus on inquiries into 118 4 Measuring cognitive load 4 Measuring cognitive load To study interpreting, the eyetracking methodology has been adopted more extensively in research on modes of interpreting which involve a written and/or visual component (e.g. sight translation and consecutive interpreting) or for set- tings which require greater interaction (dialogue interpreting). It has recently seen a revival due to the low intrusiveness, high temporal resolution and rela- tively limited cost of eyetrackers, especially of the remote type, which interfere less with the interpreting process. The first study applying eyetracking to interpreting comes from outside In- terpreting Studies. McDonald & Carpenter (1981) used eyetracking to investigate how ambiguous phrases are interpreted and parsed in sight translation, compar- ing the comprehension and the target text production phase and including the mechanisms underlying the identification of errors during the task. A series of studies adopting the eyetracking method followed, which focused on further testing the applicability of the method (in conjunction with fMRI) to the study of directionality in interpreting (Chang 2009), of sight translation performed by trainee interpreters (Chmiel & Mazur 2013) and professional inter- preters (Hansen & Dragsted 2007), also in terms of “syntactic disruption and visual interference” in sight translation (Shreve et al. 2010). In these studies, fixation-based measures were used, often as a combination of several metrics, chiefly fixation duration and counts. An example is Dragsted & Hansen (2009) who combined these metrics with heatmaps and key-logging to contrast written and sight translation, elucidating differences in how interpreters and translators process the source text. More recent studies using fixation durations are the one by Seubert (2019) on visual input in SI and by Seeber et al. (2020) on multimodal processing during SI with text. Seubert used fixation durations and gaze patterns to explore the effects of various sources of visual and written information on professional interpreters’ attention allocation. Seeber et al. (2020) investigated visual attention measuring the proportion of fixations and mean dwell time on five areas of the written input (see also §3.5.2.2). Another application of eyetracking to research on interpreting lies in the use of pupillary measures, discussed in §4.5.3. Eyetracking has not only been used to explore simultaneous or sight interpret- ing, but may also yield valuable insight into visual attention and cognitive load and effort in dialogue interpreting, characterised by a higher degree of interac- tion. 4.4.2.3 Eyetracking in Translation Process Research The wide application of eyetracking to the study of the translation process is due to its ability to offer objective, real-time measures of the mental processes in- volved in translation, often without requiring excessive manipulation of the task itself. In TPR, eyetracking measures have been widely adopted, often in combi- nation with other behavioural measures as well as subjective and performance measures or with physiological methods such as PET or fMRI (see §4.5). Over the past twenty years, the eyetracking method has been used to study the cogni- tive processes underpinning translation, subtitling, post-editing and interpreting. The eyetracking technique has been applied to the exploration of a variety of as- pects linked to the translation process and profession, such as “translation exper- tise, competence and experience, cognitive effort, reading in translation, human- computer and human-information interaction, metaphor processing, directional- ity, reception of translated material” (Hvelplund 2017: 251). As such, eyetracking research in translation and interpreting represents a relatively recent, albeit very productive innovation in Translation and Interpreting Studies. Particularly relevant to the present contribution is the adoption of eyetracking as a research methodology to study the processing effort required by different translation tasks, specifically to explore how cognitive resources are allocated, distributed, and coordinated during translation or during specific translation sub- tasks (Hvelplund 2017: 254). A large body of research has adopted eyetracking to this aim. For instance, a number of studies have successfully investigated read- ing for and during translation with eyetracking, which has become a standard methodology in this field of research. Jakobsen & Hvelplund (2008) compared reading for comprehension, as preparation for translating, for sight translation and during translation. In addition to time on task, they analysed fixation counts, average fixation duration and gaze times, finding a constant progression in all of these metrics from the first to the last type of translation task. A similar ex- periment contrasting reading for comprehension, for oral summarisation and for sight translation conducted by Alves et al. (2012) found longer fixation durations for sight translation. Schaeffer & Carl (2014) measured translation effort in the production of the target-text equivalent for literal translations by measuring gaze and translation time, while in a comparable study, Dragsted (2012) analysed total reading time and number of fixations as indicators of translation effort. 119 4 Measuring cognitive load A first step in this direction was recently taken by Tiselius & Sneed (2020), who used eyetracking glasses to collect data on the gaze patterns of experienced and inexperienced interpreters during dialogue interpreting. They suggest that gaze aversion during interpreting may index cognitive effort. In addition, the use 120 4.5 Physiological measures 4.5 Physiological measures of eyetracking may yield valuable information on how the note-taking process unfolds, as explored for instance by Kuang (2019). Particularly interesting is the recent turn of eyetracking TPR towards an in- vestigation of the interaction between translators and the tools they use as sup- port for the translation task, from the consultation of digital resources such as websites and online terminological databases, to the use of CAT tools and trans- lation workbenches during post-editing. As observed by Jakobsen (2019: 403), “a fascinating feature of gaze data is that they can be interpreted both as documen- tation of how well an interface design works, how well a translator interacts with a computer program, and as documentation of a translator’s mental effort in carrying out a task”. This characteristic of eyetracking measures may there- fore prove useful to study the interaction between interpreters and CAI tools. Previous research has exploited eyetracking to this aim in studies on transla- tion tasks of different kinds. An early example of this application is offered by O’Brien (2006), who used eyetracking to explore how the use of translation mem- ories affects cognitive effort, finding that exact matches elicit the lowest effort, as indicated by pupil size. More recently, Alves et al. (2016) explored cognitive effort in post-editing tasks conducted in two different environments, i.e. with in- teractive and standard machine translation, using the metrics fixation count and fixation duration (average and median) as correlates of effort. Carl, Aizawa, et al. (2016) contrasted post-editing and translation dictation through speech recogni- tion combining key-logging and time-based measures (e.g. translation duration, translation dictation duration) with gaze durations (see also Carl, Lacruz, et al. 2016a). In a study on the processes of error identification and correction in ma- chine translated texts and post-edited machine translations, Vardaro et al. (2019) combined early measures (first fixation durations and first pass durations) and late measures (total reading time and regression path durations). Finally, in the area of audio-visual translation, Tardel et al. 4 Measuring cognitive load (2021) investigated cognitive effort by combining total fixation count, average visit duration, total reading time, rel- ative attention to video and video replay time (p. 122). 3An excellent introduction to neurocognitive inquiries into translation and interpreting is pro- vided by García (2019). 4.5 Physiological measures Additional methods have also been used to explore cognitive processing in writ- ten and oral translation, such as measures of “cardiac, hematic, electro-dermal, ocular, muscular and cerebral responses” (Seeber 2013: 25). Since they are based on the activation of the autonomic nervous system (Ehrensberger-Dow et al. 2020: 224), these measures are less subjective and are language independent, 121 4 Measuring cognitive load which increases comparability between studies. In addition, like gaze-related metrics, these measures are continuous, thus allowing for a detailed account of local variations of CL. Some methods relying on physiological measures can, however, be rather in- trusive and require expensive equipment as well as a high level of expertise by the researcher for their successful integration into the empirical design. Because of their complexity, cost and intrusiveness, most psycho-physiologi- cal approaches have found rare application in studies on translation and inter- preting. Nonetheless, physiological measures may substantiate claims about CL variations in translation and interpreting tasks because of their objectivity. In studies on CL in interpreting, physiological methods have mainly comprised mea- sures of the brain, of the eye, and of the heart (Chen 2017: 649), which is why I focus on these metrics in the following sections. 4.5.1 Brain measures Brain imaging methods comprise different techniques, either exploiting the elec- trical or magnetic activity of nerve cells (i.e. electroencephalography (EEG) and magnetoencephalography (MEG)) or hemodynamic changes in the brain (i.e. positron emission tomography (PET), functional magnetic resonance imaging (fMRI), and near-infrared spectroscopy (NIS)). Tommola et al. (2000) provide an accessible overview and discussion of the different methods. Research on the neurocognition of translation and interpreting has employed mostly the EEG, the fMRI and the PET methods. For instance, Petsche et al. (1993) used electroencephalography (EEG) to iden- tify which areas of the brain activate in SI and in shadowing. EEG is a non- invasive and affordable method, which provides excellent temporal resolution. When used to investigate the interpreting process, however, EEG presents an important limitation: because mandibular movements can generate artefacts, the tasks must be performed covertly, which may give rise to a series of effects un- accounted for and also prevent its combination with performance measures. Ad- ditionally, its spatial resolution is quite poor, a limitation which MEG does not have, although the issue of artefacts remains also with this method. Due to these limitations, the use of EEG has been more popular in studies on translation. Some examples are Oster’s (2019) work on the translation of cognates or the extensive body of research by García (e.g. 2013, 2015, 2019).3 122 4.5 Physiological measures 4.5 Physiological measures 4.5 Physiological measures Functional magnetic resonance imaging (fMRI) is also a non-invasive tech- nique, which offers the opposite advantages and limitations compared to EEG: it can be useful for the localisation of brain functions due to its high spatial resolu- tion, while its low temporal resolution and high sensitivity to head movements (including overt speech) limit its application to study interpreting, not least be- cause the scanner environment is very noisy (Tommola et al. 2000). Nonetheless, research on the neural basis of interpreting has successfully utilised the method. Some examples are Hervais-Adelman et al. (2011, 2015) or research on direction- ality by Chang (2009) and Kalderon (2017). Kalderon used the method also to explore the effect of interpreting expertise in brain activation. PET is a highly invasive method, as it requires the intravenous administra- tion of radioactive ligands (Seeber 2013: 25). This may severely limit sample size. 4.5.1 Brain measures While its temporal resolution is limited, PET provides good spatial resolution and “seems to be the only one that allows the investigation of the entire multi- effort process of SI” (Tommola et al. 2000: 18). In Price et al. (1999), PET was used to compare translation and interpreting, while Rinne et al. (2000) and Tommola et al. (2000) used PET to compare speech shadowing and SI, including the effect of directionality on activation patterns. 4.5.3 Pupillary measures Among physiological approaches, the collection of pupillometry data represents the least invasive and the more suited method to be applied to investigations of the interpreting process, in particular of SI, which, for its continuous and real- time nature, does not easily allow for the adoption of more invasive techniques, as discussed above. Pupillary measures are also often used in the investigation of cognitive pro- cesses and attention. The pupil size and dilation, measured in millimetres, are taken to index the amount of processing effort exerted by a test subject to per- form a certain task. The rationale behind the use of measures of pupil size and dilation in cognitive studies lies in the assumption that the diameter of the pupil reflects the difficulty of the task at hand or the working load posed by it (see for instance Hess & Polt 1964, Just & Carpenter 1993, Kahneman & Beatty 1966). The wider the pupil, the higher the difficulty and/or the working load. The in- terpretation of pupil measures is, however, far from being this straightforward, as the pupil has been found to contract and dilate in relation to a wide range of stimuli and psychological and affective states. The diameter of the pupil can be influenced not only by workload, but also by emotion and anticipation, fatigue, diabetes, age, pain and drugs (see Holmqvist 2011: 426 for a detailed discussion). However, some of these factors can be controlled experimentally, for instance by excluding participants with relevant pre-existing conditions or of a certain age group from data collection. To ensure good data quality, it is preferable to work with systems that have “a fixed distance between camera and eye” (Holmqvist 2011: 530). Remote eyetrackers may expose the researcher to the risk of introduc- ing artifacts in the data. Most importantly, pupils react particularly to changes in luminance, which requires a stable source of light and a controlled environ- ment during data collection. Ideally, the study design should include stimuli with equal brightness and contrast. A major complication in using pupillometry to in- vestigate cognitive processing lies in the fact that pupillary responses do not occur immediately after stimulus presentation, but with a certain latency. Pupil dilation latency varies according to the stimulus presented, with latencies com- prised between 150 and 400 ms for light stimuli or between 300 and 500 ms for interpreting, as found by Hyönä et al. 4.5.2 Heart rate measures Cardiac system measures such as heart rate, heart rate variability (i.e. variation in time intervals between consecutive heart beats, see Thayer et al. 2012), and blood pressure have been used to measure the impact of a variety of language processing tasks on stress (e.g. Klonowicz 1994, Korpal 2017, Kurz 2003). How- ever, as higher stress experienced may signal increased cognitive effort, such measures may prove useful in explorations on mental effort in the translation and interpreting process. Heart rate measures have been adopted to study the interpreting process, es- pecially after what has been defined as the “psycho-affective turn in Interpreting Studies” (Korpal 2016: 298), i.e. a renewed interest for the role of psycho-affective factors in interpreting, for instance for aptitude testing (see also Chabasse 2009). As reported by Korpal (2016: 304), additional markers of stress used in interpret- ing research are skin conductance level (e.g. Kurz 2002, 2003), cortisol concen- tration (used in Blumenthal et al. 2006 and in AIIC 2002), and IgM levels (Moser- Mercer 2005), often used in combination with and integrated by subjective mea- sures, such as self-reported measures of stress. In TPR, an example of the com- bination of multiple types of measures, including an extensive set of heart rate 123 4 Measuring cognitive load measures, can be found in Herbig et al. (2021), who explored how objective mea- sures relate to the subjectively reported CL in MTPE. 4.5.3 Pupillary measures (1995: 605) (see Hvelplund 2014: 215). This means that dilation data might refer to what the participant was processing be- fore stimulus onset. For the aforementioned reasons, measures of pupil size and 124 4.5 Physiological measures 4.5 Physiological measures dilation are best analysed in combination with other metrics, and caution should be exercised when interpreting pupillary measures. In the 90s, three groundbreaking studies sought to establish the feasibility of the application of pupillometry to the exploration of processing effort in SI (Tom- mola & Niemi 1986, Tommola & Hyönä 1990, Hyönä et al. 1995). As pupil diameter had been found to positively correlate with CL (the larger the pupil, the higher the load) during cognitive processes, including language comprehension, the re- searchers sought to establish whether pupillometry could be adopted as a diag- nostic method to index variations in processing effort during interpreting. The choice of this physiological indicator was motivated by the fact that it was hy- pothesised to comply with the key requirements identified by Kahneman (1973): the ability to reflect load differences between and within tasks and between sub- jects. In other words, capturing differences in processing load during interpreta- tion required an online, real-time and highly sensitive metric capable to co-vary with global and local variations in CL. In those first attempts at integrating this measure into the exploration of the interpreting process, pupillometry was used to compare listening (without comprehension testing), shadowing and SI into the participants’ mother tongue (Tommola & Hyönä 1990) and, later, to investigate the effect of word difficulty and directionality on CL during SI (Hyönä et al. 1995). In the first study, the hypothesis was that SI would yield the largest average pupil diameter, followed by shadowing and listening. In the second experiment, pupil size was expected to be larger for the words classified as difficult to translate and for the repetition of words in the foreign language. The intention behind these initial studies was rather of methodological nature. The results confirmed the adequacy of pupillometry as a method to investigate mental load during interpreting and related language tasks. Specifically, mean pupil size was found to be larger in the interpreting task. An order effect was also found, meaning that pupil dilation is larger at the beginning of the task. In the second experiment (Hyönä et al. 4.5.3 Pupillary measures 1995: 609), pupils were found to dilate on average by 0.40 mm during the translation of difficult words, while maximum di- lation was found to be of 0.57 mm. Another important finding was that pupillary latency, i.e. the delay in pupillary response to a task, was between 300 and 500 ms (Hyönä et al. 1995: 605), which aligns with previously reported values (Beatty 1982, Hoeks & Levelt 1993). These pioneering studies paved the methodological way for the adoption of pupillometry in interpreting process research, showing its feasibility for the investigation of this cognitively-taxing task. Pupillometry was recently adopted by Seeber & Kerzel (2012) and by Gieshoff (2018) to study, respectively, the effect of verb-final and verb-initial syntactical 125 4 Measuring cognitive load structures in German to English SI on local CL and of multimodal input presen- tation on mental workload in SI. In their discussion of results, Seeber and Kerzel highlight an important aspect of the analysis of SI conducted under different conditions or on different source materials (as in the case analysed by the authors). While the differences in mean pupil diameter and dilation identified by Hyönä et al. (1995) were relatively large, which was to be expected due to the comparison of different tasks, when the comparison is made between the same task with more subtle modifications, the differences in pupil-based measures are likely to be less obvious. This was the case in Seeber and Kerzel’s study, which found that pupils dilated by 3.996 and 4.048 mm while interpreting symmetrical and asymmetrical syntax respectively, as measured at the end of the task, i.e. in the temporal break between the stim- uli sentences. This is an important methodological understanding, which should also be taken into account in the present study seeking to compare SI conducted with varying digital support for terminology look-up. g g g Gieshoff (2018) found an effect of audio-visual input (i.e. the presence of the speaker’s video) on pupil size: its decrease was less marked than in the audio- only condition. On the other hand, background noise did not have an effect. She also found an effect of task, as she compared listening to interpreting. She found that pupil sizes decreased more largely in listening than in the SI task. 4.5.3 Pupillary measures Results from the studies presented “illustrate the great potential of pupillome- try as a method and [task-evoked pupillary responses] as a measure of cognitive load in simultaneous interpreting” (Seeber 2013: 27). However, the methodologi- cal challenges highlighted in the aforementioned studies underline how, despite its strengths, pupillometry requires a stringent experimental setup to allow the researcher to isolate stimulus responses. Additionally, pupillometry provides the best results when applied to isolated and short stimuli. When used over longer periods of time, local variations are averaged out and the technique does not offer useful insights into the underlying cognitive process (see for instance Schultheis & Jameson 2004). The present chapter has reviewed the different methods and measures used to estimate CL in translation and interpreting. As discussed, all measures present advantages and disadvantages. A combination of several types of metrics may therefore prove beneficial to address the inherent limitations of each and max- imise the benefits provided by the individual metrics. This is the approach com- monly followed in TPR and it will also be adopted in the present study. The fol- lowing chapter illustrates the methodology used in the experiment and discusses the measures chosen for data collection in §5.5.5. 126 5 Method The present chapter illustrates and discusses the methodology used in the present study, which was conducted to compare the effects of digital terminological sup- port tools on the product and process of SI. The research work comprised a pilot study and a main study. The pilot study was aimed at validating the methodology. Its results, already published in Prandi (2017, 2018), and the adaptation of the design for the main experiment are discussed in §5.4. In §5.5, the experimental materials and setup of the main study are presented. The results of the main study are presented and discussed in Chapter 6. 5.1 Research gap As discussed in the literature review (§2.4), arguments in favour of and against the introduction of CAI tools and technologies such as ASR into the interpreting process are substantiated by empirical data only to a limited extent. The lack of data is accompanied by the lack of an empirically-validated research methodol- ogy for the combined collection and analysis of process and product data on SI with the support of digital tools. The existing empirical contributions have fo- cused primarily on a product-oriented analysis through the collection of perfor- mance data. This currently represents a major barrier to the further development of empirical CAI research. Research has addressed the evaluation of terminologi- cal quality (Biagini 2015, Prandi 2015a,b, Van Cauwenberghe 2020) and, especially in the case of ASR support, the accuracy of number renditions (e.g. Defrancq & Fantinuoli 2021, Pisani & Fantinuoli 2021). Furthermore, most of these analyses have not expanded the focus beyond the unit of information for which support is offered, except for Biagini (2015) (who also considered whether glossary searches produced omissions or serious errors), Montecchio (2021) (who evaluated the im- pact on perceived fluency in addition to the accuracy of numerals rendition), and Frittella (2022), who conducted a qualitative evaluation of the interpreta- tions of number-dense speeches. In some publications, subjective data were col- lected through retrospective questionnaires to gain further insight into the in- teraction with digital terminology support solutions (e.g. Defrancq & Fantinuoli 5 Method 2021, Pisani & Fantinuoli 2021). To the best of my knowledge, no behavioural methods have been employed to date to substantiate these observations with more objective measures (as recommended e.g. by Hansen 2008: 391–392), such as time-lag or fixation-based metrics. In addition to these methodological limitations, an evident research gap emerges. Research on CAI tools has focused mostly on the product, while the cognitive processes leading to specific phenomena in the target text have re- mained largely unexplored. The present contribution represents a first step in this direction by attempting a look into the cognitive inner-workings of CASI. 5.3 Research approach 5.3 Research approach 5.3 Research approach able to identify the terms more quickly, at least when orthographic neigh- bours are shown on screen, resulting in shorter time to first fixation. Addi- tionally, InterpretBank should impose a lower load on task coordination, as it does not require preparation for subsequent queries (see §3.6.2). This should result in a higher number of terms interpreted as per glossary and fewer serious errors and omissions compared to a digital glossary. It should also elicit more queries than a PDF glossary due to its posited higher user- friendliness and higher querying speed thanks to the incremental search. 3. ASR support may facilitate not only production, but also comprehension, due to beneficial redundancy effects (if temporal contiguity is ensured, see §3.4.2.1 and Seeber et al. 2020, Chmiel, Janikowski & Lijewska 2020). This should result in a higher number of terms interpreted as per glossary, and fewer serious errors and omissions compared to CAI tools and digital glos- saries. 4. ASR support should reduce negative split attention effects (see §3.4.2.1) due to increased temporal contiguity and to the absence of manual look-up compared to CAI tools and digital glossaries, resulting in shorter fixation time on the tool area and promoting attentional focus on the speaker. 5.2 Hypotheses on cognitive load in SI with terminology look-up As previously discussed (§3.4), CL should not be seen as a static construct, but rather as a variable that changes constantly during the interpreting task as a function of the cognitive resources recruited by the co-occurring tasks. For the scope of the present inquiry, I chose to focus my analysis on the effect of termi- nology look-up and of automatic terminological prompts on the CL experienced during SI. As highlighted in the previous sections, different digital solutions for the terminological support of the interpreter may be expected to have a differ- ent impact on cognitive effort and, from a product-oriented perspective, also on the terminological accuracy of the interpreter’s rendition. In the context of the present contribution, I focus on the differences resulting from the adoption of a digital glossary, of a CAI tool (InterpretBank), and of ASR as terminological support in the booth. On the basis of what has been discussed so far and as predicted by Seeber’s CLM applied to CASI (see §3.5.2), it may be hypothesised that: 1. Tools which require manual terminology look-up (digital glossaries and standard CAI tools) may impose a higher CL due to the recruitment of ad- ditional manual-spatial resources during production and monitoring. Con- versely, lower CL may be expected for ASR tools, which do not pose de- mands on manual-spatial resources. This should result in shorter EVS, shorter time on task, and shorter average fixation durations. 2. As the CAI tool InterpretBank provides a series of postulated advantages (see §3.6.2) compared to standard digital glossaries, the test subjects may be 128 5.3 Research approach The combination of product- and process-related data has been proposed as a means to improve TPR (Hansen 2007). As advocated for instance by Mellinger (2019), a product- and process-oriented perspective to the exploration of CAI ap- pears necessary, if we are to advance our understanding of the impact of CAI tools on cognition in (simultaneous) interpreting and to develop tools truly tai- lored to the needs of interpreters and capable of addressing the inherent con- straints of SI. In order to contribute to bridging the current research gap, both in terms of research object and of methodology, I therefore developed and tested a convergent mixed-method design for the investigation of the product and pro- cess of CASI with the support of different terminology look-up tools. The study adopts an experimental approach for the generation and collection of data under controlled laboratory conditions. The approach chosen is predominantly quanti- tative. Due to the complexity of the present object of study, “so much less con- venient to study than language fixed in writing” (Pöchhacker 2004: 48), in the present contribution multiple types of quantitative data related to the CL gen- erated by CASI are collected and combined, and are corroborated by qualitative data related to the participants’ experience in the interaction with the tools. 129 5 Method 5 Method 5 Method The study adopts methods derived from the TPR framework and includes per- formance, subjective, and behavioural measures combined in order to address the inherent limitations of each method. The choice of measures was motivated by the review of the methods used to measure cognitive load and their application in TPR (see Chapter 4). Table 5.1 presents an overview of the metrics selected for the study and the type of data collected. The measures chosen are discussed in detail in §5.5.5. Table 5.1: Metrics included in the main study, type (qualitative/ quantitative) and category (subjective, performance, or behavioural) Table 5.1: Metrics included in the main study, type (qualitative/ quantitative) and category (subjective, performance, or behavioural) Measure Type Category Terminological accuracy quantitative performance Errors/omissions quantitative performance Inter-cluster pause duration quantitative behavioural EVS quantitative behavioural Glossary queries quantitative behavioural Time to first fixation quantitative behavioural Average fixation duration quantitative behavioural Fixation time quantitative behavioural Debriefing questionnaire qualitative subjective The hypotheses concerning digital terminological support in the booth (§5.2) were explored using a within-subject design: in this way, as discussed in Lazar (2017: 65), “the impact of individual differences is effectively isolated and the ex- pected difference can be observed with a relatively smaller sample size” (see also §5.5.1.1). A small-N design (Smith & Little 2018) need not necessarily be consid- ered a limitation. In other disciplines, such as psychology, “there is a long history of research [...] employing small-N designs” (ibid., p. 2083). A small sample where multiple observations are performed on a limited number of participants may be even more informative than a large sample, as combining quantitative analysis with qualitative observations (e.g. from questionnaires) and exploring the same dataset through multiple metrics becomes more feasible. As a consequence, a researcher may be able to paint a more refined picture of the mechanisms un- derlying a specific process or observations made on the product. Considering that interpreter-machine interaction is still a largely unexplored topic in CAI, working with a small sample seems valid. As observed for instance by Hansen- Schirra & Nitzke (2020: 422), small-N studies “are valuable to build hypotheses 130 5.4 Pilot study: Method and results for larger studies”. The benefits of a small sample size have been recognised also in the area of human-computer interaction, where skilled participants are usu- ally needed and cannot always be found easily (Lazar 2017: 65). 5 Method At the time of the experiment, CAI tools and ASR in the booth were still a mirage for many inter- preters and only included in training to a limited extent (Prandi 2020). When the use of CAI tools becomes more mainstream (and this may already be the case due to rapidly changing working conditions during the COVID-19 pandemic), it may be more feasible to recruit larger samples. In order to test the hypotheses of the present study, involving well-practiced participants was deemed more valuable than recruiting a large sample. Whenever possible during the study, I strove to preserve ecological validity, i.e. “the naturalness of the investigated process” (Hansen 2008: 386). For this reason, real-life conditions were replicated insofar as they allowed for sufficient experimental control to be able to draw inferences from the collected data (see Spinner et al. 2013 for an empirically-based discussion on ecological validity in reading research, largely applicable to experiments on translation and interpret- ing). However, as addressed in §5.4.3, 5.4.4, 5.5.3, 5.5.6, it was necessary to limit the potential impact of confounding variables on results. Therefore, the stimuli were designed to reduce such impact (§5.4.3 and §5.5.3). The review of the research conducted using eyetracking on translation and interpreting (see §4.4.2.3) shows that the technique may provide valuable insight into the mental processes at the core of translation and interpreting. In particular, the large body of TPR conducted using this technique, which has contributed to establishing a set of reference metrics taken to index cognitive load and effort, may prove especially valuable also for the present object of inquiry, which is why I opted to conduct an eyetracking study to address my research questions.i Since the present study is the first to explore SI with digital support for ter- minology from a preeminently cognitive perspective, it has an exploratory char- acter and also aims at formulating further hypotheses in addition to providing initial findings on this complex research object. 5.4.1 Participants For the pilot study, I was able to recruit six advanced students of the master’s de- gree in conference interpreting of the University of Mainz/Germersheim. They had all received at least three semesters of instruction in simultaneous and con- secutive interpreting and had English as their B (active) or C (passive) language. The sample was made up of three German natives (one male, two females) and three Italian natives (one female, two males). The two language combinations, comprising one Germanic and one Romance language, were chosen to ensure that the stimuli could be considered challenging independently of structural and linguistic similarities in the language pair, e.g. when cognates were included as stimuli. Participation in the study was voluntary and not remunerated, but com- pensated instead by allowing students to learn how to use a CAI tool which is not usually included in the regular interpreting curriculum. An additional benefit of participating in the study consisted in the added amount of practice hours with a laptop in the booth, which is rarely done systematically in class. This was, how- ever, also a limitation, as students were required to schedule time for the practice sessions. This is probably the main factor contributing to the difficulties experi- enced during participant recruitment, a difficulty which was unfortunately also confirmed for the main study, as will be discussed in §5.5.1. 5.4 Pilot study: Method and results In order to test the methodology for the main study, a pilot study was conducted at the University of Mainz/Germersheim between May and July 2017. The exper- imental design combined process and product-oriented data collection methods. Its primary focus was the validation of the stimuli designed for data collection. In this section, I describe the experimental setup illustrating the stimuli used and 131 5 Method analysing the participants’ deliveries and queries performed. I then highlight the limitations identified in the approach chosen and discuss the modifications ap- plied to the experimental design ahead of the main experiment. 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results training. The preparatory meeting focused in particular on the search functions in Word, Excel and InterpretBank. The main difference between Word and Excel, both traditional tools for organising terminology, was that with Word partici- pants could visualise all the results of a query in a column on the left hand-side of the screen, similarly to InterpretBank, but without the explicit optimisation for interpreters offered by InterpretBank. When working with Excel, they had to skip to the next occurrence manually. The five practice sessions that followed were organised to develop the ability to conduct queries while interpreting. All participants attended all practice sessions, which took place once a week for five weeks. During each session, their task consisted in interpreting three short speeches from English into their respective mother tongue (German or Italian) while using one of the tools to look up terminology. The order of the tools was changed at every session. The speeches used for the training were selected and adapted from the training material used in a previous study (Prandi 2015a,b), to which several authentic speeches were added that had been selected in order to ensure a certain progression from a more controlled to a more naturalistic prac- tice environment. The topics chosen were medicine and biology. I prepared the glossary used for terminology look-up and made it available to all students for both language combinations and each tool, so that they all practised with the same material. The training was thus designed to guarantee equal practice time for each tool. It should be noted that I use the term “training” to indicate that practice was ensured before data collection. However, no specific input on the development of strategies for effective interpreter-machine interaction during SI was given. After the last training session, the participants took a short proficiency test to verify whether they had all learned the basics of glossary querying with the three tools. The test consisted in a series of tasks to be performed with the tools and focused on the search function. Students were asked to record their screens while accomplishing the tasks. The screen recordings were later analysed: all students passed the test and could take part in data collection. 5.4.2 Test subjects’ training Before taking part in data collection, it was necessary to prepare participants in order to ensure that they were all equally proficient in the use of the three tools compared in the study, so that I could exclude this variable during data analysis. The individual disposition towards technology continues to play a role despite training, of course, but no subject preparation can change this highly personal factor. Nonetheless, it is certainly easier to exclude the variable of lack of tool expertise if the participants have gone through a dedicated training beforehand. The students were therefore invited to attend a preliminary preparatory meet- ing during which they were instructed on the basics of terminology manage- ment for conference interpreters. The training had an explicit practical focus, as previous studies (Prandi 2015a,b) had highlighted the role of practice as more beneficial to familiarise the participants with the tool than a more theory-prone 132 5.4.3 Materials The speeches drafted for the pilot study were prepared based on a series of known effects and hypotheses related to features of oral speech perception and produc- tion and to linguistic and morphological characteristics of terms as stimuli. These considerations are described in the following subsections, while §5.4.3.6 illus- trates the features of the speeches validated in the pilot study. 133 5 Method 5 Method 5.4.3.1 Speech rate and presentation Speech rate is a prosodic factor which can be essential in determining the feasibil- ity of the SI task. Especially excessively high input rates have been identified as a major stress factor in interpreting and may render the performance of SI impos- sible (e.g. Riccardi 2015). In SI, the ideal speech rate is comprised between around 95 and 120 wpm (Seleskovitch 1978, Gerver 1976, Seleskovitch 1978, Lederer 1981, Pio 2003, Seeber 2005). Therefore, for the experiment it was necessary to select a speech rate which would make looking up terminology during SI challenging, but not impossible. This was necessary to test the trade-off between looking up terminology and delivering an acceptable interpreting performance. An additional aspect to consider was the mode of presentation. Pre-prepared, read-out speeches tend to be associated with a faster speaking rate and a less spontaneous intonation, which can make the input more difficult to process. However, as required by the experimental nature of the present study, it was nec- essary to ensure consistency between subjects and to use comparable speeches. Therefore, while it would be possible to have the speaker deliver a speech live to maintain higher ecological validity and collect data in a more naturalistic set- ting, in an experimental setting, video recordings can be a way of maintaining the necessary degree of control while at the same time approximating the mode of presentation of the source speech to real-life conditions. Additionally, video- recorded speeches are usual training material for interpreting students. There- fore, using a video-recorded read-out speech at a comfortable speaking rate was expected to be perceived by the participants as similar to their asynchronous training sessions in terms of speed and prosodic aspects. 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results on their own. In contrast, using unedited speeches and conducting the analysis at the text level would have likely introduced an excessive amount of potentially confounding variables and made stimulus-response correlations difficult to iden- tify. For these reasons, I aimed for highly controlled stimuli while at the same time trying to make the task as realistic as possible. A useful reference was identified in Seeber & Kerzel’s (2012) methodology. In their study on cognitive load effects in symmetric and asymmetric sentence structures in SI, they presented stimulus sentences embedded in sentence clus- ters, i.e. the critical (target) sentence is enclosed in a pre-critical (introductory) and a post-critical (continuation) sentence. This approach presents several ad- vantages: it makes data analysis more efficient by allowing for a focus on the target sentences (i.e. the sentences including the stimulus-term); it promotes the creation of comparable speeches, as they all present the same structure; and fi- nally, it provides participants with the impression that they are interpreting a continuous speech rather than isolated sentences. In experimental research, the inclusion of filler items is essential to minimise unintended repetition effects due to noticeable patterns in the stimuli (see Con- klin et al. 2018: 44–45, Keating & Jegerski 2015). In other words, if every sentence contained a stimulus, the participants may notice it and adjust their look-up strat- egy when working in the PDF or CAI condition. The terminological density may also be too high to allow for queries during the interpreting task and the partic- ipants may stop looking up terms altogether, which would defeat the purpose of the study. To avoid eliciting unwanted effects, the pre-critical (introductory) sentences and the post-critical (continuation) sentences would have to be super- ficially as similar as possible. Crucially, they would need to be highly comparable in terms of length, number of clauses and syntax (see Keating & Jegerski 2015: 16). While one sentence between each stimulus and the next may have been suffi- cient to give participants some respite between two queries, the introduction of a continuation sentence was deemed useful to ascertain whether a glossary query may cause a trickle-down effect due to imported load (see Gile 2008), leading to severe errors and omissions in the following sentence. 5.4.3.2 Speech structure: Sentence processing When designing the experiment, I was also faced with the challenge of present- ing participants with clearly defined stimuli that could then easily be correlated with responses. At the same time, it was necessary to preserve a certain degree of ecological validity in order not to alienate the test subjects with a too unfamil- iar task – a common challenge in the investigation of the interpreting process in a laboratory setting. While I am interested in analysing CL at a local level (see §3.4), presenting participants with individual terms to interpret without any context would have excluded the element of simultaneity from my study. This would have counteracted the very goal of my investigation, as it is the simultane- ous performance of cognitive tasks that makes challenges and limitations arise which would otherwise not be noticeable when considering the individual tasks 134 5.4.3.3 Compound processing In English and German, all compounds are right-headed (Arcara et al. 2014, Se- menza & Luzzatti 2014), while in Italian they can be either left- or right-headed (e.g. Ghiselli 2015). The morphological structure of the terms was expected to af- fect search behaviour. Since a word-length effect has been observed for acoustic understanding, i.e. shorter words have been found to be more difficult to process (e.g. Barton et al. 2014), short unigrams (monosyllabic or bi-syllabic) may have posed a difficulty for a glossary search. Therefore, if they had not been under- stood, they would offer less linguistic material to perform a query (which can be conducted also for incomplete words). However, as working memory span has been shown to increase for short words, if the stimuli were correctly understood they may have interfered less with the other subprocesses, since they would pose a lighter load on WM. Conversely, I expected longer terms to be easier to pro- cess acoustically. In particular, multi-word terms (bigrams and trigrams) would facilitate the glossary search by offering more linguistic material: for instance, participants may choose not to look for the first element of the term, but rather for one of the other elements (e.g. the modifier instead of the head of the com- pound expression). Additionally, compound expressions have been identified as an element of difficulty in SI (see Ghiselli 2015), and I therefore expected them to be looked up more often than unigrams. 5.4 Pilot study: Method and results In studies on sentence pro- cessing, “the processing of a critical region in a sentence oftentimes continues or spills over onto the words immediately following the critical region” (Keating & Jegerski 2015: 6). This “spillover effect” (e.g. Rayner & Duffy 1986) is often notice- able not in the target region (i.e. the target sentence in this case), but rather in the spillover regions. Continuation sentences would therefore serve as a test-bed to verify whether coping with the stimulus term in the target sentence may lead to effects in the following textual material. 135 5 Method 5 Method 1For the scope of the pilot study, the stimuli were not further controlled for linguistic aspects such as cognate status or level of concreteness, which can affect linguistic processing (see §1.1.1 for a discussion of the features of language for special purposes). These aspects were considered in the main study (see §5.5). 5.4.3.5 Frequency of the stimuli Finally, the frequency of the stimuli used was expected to affect the participants’ search behaviour. It was reasonable to expect that more frequent terms may have been known to the students and not require a query, whereas less frequent terms may have required glossary look-up. CAI tools are generally considered to be more user-friendly and to take up fewer cognitive resources than traditional glos- sary tools (see §2.4), thus allowing for a higher number of queries. In his study contrasting paper glossaries and InterpretBank, Biagini (2015) had found a higher number of terms searched in InterpretBank compared to the paper glossary con- dition. Introducing both frequent and infrequent terms as stimuli was expected to help explore the postulated advantages of CAI tools, which are designed to facilitate look-up. In particular, I hypothesised that participants would look up a higher percentage of no-query terms when working in the CAI condition due to a lower load on working memory (see §3.6.2, 5.2).1 5.4 Pilot study: Method and results of the target sentence, the interpreter would also have to start processing the following sentence: having to store the term in WM in addition to the previous information during the query and to finish interpreting the target sentence may increase the CL to the point of overloading or of resulting in serious issues or omissions in the continuation sentence. However, sentence-final stimuli offer the advantage of being preceded by more context than sentence-medial stim- uli. This may facilitate semantic anticipation (e.g. Gile 2009). Hence, interpreters may be able to avoid a query altogether or apply the strategy of formulating their rendition with a less compromising syntactical structure, facilitating a seamless integration of the term after performing a query. For this reason, sentence-final stimuli may actually elicit lower additional load. If, on the other hand, the test subject does not anticipate that an unknown term requiring a glossary query is coming up, self-corrections may be necessary. This may be expected both for sentence-medial and sentence-final stimuli, resulting respectively in false starts and self-corrections or reformulations. 5.4.3.4 Position of terms in the speeches The position of the terms in the sentence was also expected to play a role. In read- ing research, sentence-final stimuli have been found to elicit longer fixations, which suggests higher cognitive load. This is known as the sentence wrap-up effect (Warren et al. 2009), which has been explained with the integration of sen- tence meaning with preceding and following context: at the end of the sentence, the reader receives all the necessary clues to correctly gauge the meaning of the sentence. It should be noted that this effect has been found not only for sentence- final stimuli, but also for clause-final stimuli. While the phenomenon has been studied mainly with reference to written textual processing, a similar effect has been found in the eyetracking study by Seeber & Kerzel (2012): verb-final sen- tences in German elicited significantly larger pupil dilations than verb-medial sentences in a study on SI from German into English. Similarly, a stimulus term placed at the end of the sentence may pose a higher strain on WM and result in CL possibly due to the need to store in memory previous units of information se- mantically or morphologically depending on the term. Furthermore, at the end 136 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 2https://webgate.ec.europa.eu/sr/ (Accessed: 01.11.2021). 2https://webgate.ec.europa.eu/sr/ (Accessed: 01.11.2021). 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results By focusing, for instance, on woody biomass fuels, we can truly make a difference. By focusing, for instance, on woody biomass fuels, we can truly make a difference. They have the potential to help us respond to the challenges we’re facing. The stimuli introduced in the speeches present a set of features that were cho- sen to control for additional confounding variables. Each text contained 36 terms, one per target sentence. The terms were symmetrically grouped by morpholog- ical structure, by position in the sentence, and by necessity of a query based on their frequency. Specifically, of the 36 terms, 12 were unigrams (e.g. “bioenergy”), 12 were bi- grams, i.e. made up of two elements, usually noun-noun (e.g. “energy poverty”) or adjective-noun compounds (e.g. “tidal barrage”) and 12 were trigrams, in differ- ent combinations of adjectives and nouns (e.g. “pressurised water reactor”). For the above-mentioned reasons, I decided to control for the presence of multi-word and single word expressions in the speeches by equally distributing the stimuli across the three categories. In order to control for potential sentence-final effects and to test my hypotheses on anticipation, half of the terms were placed at the end of the sentence and half in the middle. Finally, I divided the terms equally into terms requiring a query and terms for which a glossary search was not deemed to be necessary. The terms classified as requiring a glossary search were highly technical terms that do not belong to the 10,000 most common English terms as per their frequency in the Leipzig Corpora Collection (2012, 2016, 2021) corpora. 5.4.3.6 Features of the speeches used in the pilot study The design of the speeches was motivated by the considerations illustrated above and the intention to test how specific features of the speeches would affect the 137 5 Method participants’ performance and interaction with the tools. Additionally, the natu- ralistic base material was modified in order to control for potential confounding factors. The speeches used for data collection were based on a corpus of speeches col- lected ad-hoc from the speech repository of the EU Directorate General for Inter- pretation2 on the topics of energy sources. The speeches were on average 12 min- utes long at an average speed of 122.26 words per minute (wpm). The speeches were read out by a male native speaker of British English and video recorded. Additionally, the speeches used as training materials ahead of the test were also presented in the same way. As the speaker was himself an interpreter trainer and conference interpreter, he was familiar with the use of pre-prepared material to deliver speeches for training purposes. Each speech was made up of 36 sentence clusters, containing one stimulus each. Each sentence cluster was made up of: 1. An introductory sentence, which provides context but should not elicit glossary queries; 1. An introductory sentence, which provides context but should not elicit glossary queries; 2. A target sentence containing the stimulus to be looked up; 3. A continuation sentence, with the same features and role of the introduc- tory sentence. Thus, each stimulus was separated from the following one by two filler sentences. Due to the presence of introduction and continuation sentences, each speech thus contained 66.7% noncritical sentences and 33.3% critical (target) sentences, as recommended in literature (e.g. Keating & Jegerski 2015: 17). The structure was repeated throughout each speech. I report an example from Speech 1 to illustrate how the speeches were structured: (1) So we need to change this basic trend and this is why the urgency is there. In our policies, we should definitely address the need to improve vehicle efficiency. But there is still much more I can do, in many other areas, as you are aware. fi But there is still much more I can do, in many other areas, as you are aware. (2) At the EU level, there is another policy option that can help us. 2https://webgate.ec.europa.eu/sr/ (Accessed: 01.11.2021). 138 5.4.4 Procedure At the time of the pilot study, the adoption of ASR as in-booth support had just been theorised for the first time (see Fantinuoli 2017b, Cavallo & Ortiz 2018). For this reason, and based on the results of the available inquiries into conference interpreters’ terminological practices (see §2.3), the in-booth digital support so- lutions compared in the pilot study were Word glossaries, Excel glossaries and the CAI tool InterpretBank. Data collection took place at the Translation and Cognition (TRA&CO) Center of the University of Mainz/Germersheim, the university’s neurolinguistic labo- ratory dedicated to Translation Studies. Before the start of the experiments, the students were briefed about the structure of the study and signed a waiver on the collection and treatment of their data. They were informed that they could decline their consent to the use of their data at any time. They were informed about the interpreting task they were going to complete during the experiment 139 5 Method 5 Method and were told the topic of the speeches (renewables and other energy sources). I am aware that not having participants prepare ahead of the interpreting task does not reflect usual professional practice. Nonetheless, I decided to sacrifice this aspect of ecological validity in favour of a more controlled experimental de- sign. Preparation is highly personal, depending on several variables, not least also on motivation, and attempting to control and standardise the test subjects’ preparation time and thoroughness would have probably proved to be impossi- ble. An additional reason for excluding preparation from the experiment was the need to ensure that participants would be presented with a sufficient number of stimuli to look up in the glossary. As a result, the number of queries is probably much higher in the experimental setting than in a real-life interpreting assign- ment, but having sufficient data points per participant was necessary considering the small sample, and essential to draw initial conclusions on the comparison of the three tools. The students were asked to approach the interpreting task as if they had to interpret at a real conference on renewable energy and had had little time to prepare, but a colleague had made available a glossary at the last minute. They were not explicitly encouraged to look up terms while interpreting, but were asked to consider the glossary simply as an aid that they could resort to when necessary. 3During the pilot study, the eyetracking set-up used during the main study was also tested. Therefore, the gaze replay videos exported from the data analysis tool of the eyetracker (SMI BeGaze) were available and could be used to reconstruct the queries in the Word and Excel glossaries. However, during the pilot study, gaze data was not analysed. The eyetracking setup is therefore described and discussed in greater detail in §5.5.3 and §5.5.6. 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results and Excel condition by reviewing the screen recordings.3 The participants’ inter- pretations were recorded with Audacity (2021) and then transcribed following the HIAT transcription conventions (Rehbein et al. 2004) in Partitur Editor, the transcription tool of the Exmaralda Suite (Schmidt & Wörner 2014). 5.4.4 Procedure I prepared the glossary, which contained 421 terms, i.e. all the specialised ter- minology included in the texts plus additional terms related to the topic in ques- tion. The same terms were available for each tool and language combination. The glossary, prepared in InterpretBank and then exported and converted into .xlsx and .docx files, presented a simple tabular structure which only included the term and its equivalent in the target language. It could be argued that inter- preters’ glossaries may also be more complex, containing additional information such as synonyms, definitions and collocations. Moreover, their structure and content are very personal, as they reflect the individual preparation style, pref- erences and strategies. Accounting for each of these variables would have been extremely complex empirically. Therefore, I chose to focus on the minimum com- mon denominator for the test, i.e. the terminological pair. During the experiment, participants interpreted three speeches from English into their A language, as they had done during the practice sessions. They used a different tool each time. The order of speeches and tools was randomised to avoid effects due to fatigue ensuing during the test (Spinner et al. 2013: 400, Keating & Jegerski 2015: 18) or to the individual speech/tool combination (Conklin et al. 2018: 42–43). The participants’ glossary queries were collected automatically in the log file generated by InterpretBank after each trial and manually for the Word 140 5.4.5 Stimulus validation QN = query necessary, NO QN = no query necessary (Prandi 2018: 47) Figure 5.1: Percentage of terms searched per stimulus category in the pi- lot study. QN = query necessary, NO QN = no query necessary (Prandi 2018: 47) 5.4.6 Preliminary results A thorough description of the results of the pilot study is reported in Prandi (2017, 2018), including a preliminary analysis of the strategies adopted during SI. While verifying hypotheses was not the main aim of the pilot study, I conducted a series of preliminary analyses on the transcribed interpretations to gain first insight into several expected phenomena in preparation for the main experiment. I first present the results of the analyses related to the stimuli classification and then continue with an analysis of product-related aspects. 5.4.5 Stimulus validation The primary aim of the pilot study was verifying the validity of the a-priori classi- fication of the stimuli in relation to glossary queries. In order to obtain sufficient data for a comparison between the Word, the Excel, and the CAI tool glossaries, it was necessary that the stimuli classified as requiring a query actually elicited a sufficient number of searches for all participants. While a certain degree of inter-subject variability should be expected, as each participant differs in terms of SI skills, world knowledge, and interpreting strategies, it was important for my classification to hold true at least in the majority of cases. I therefore transcribed the students’ deliveries and analysed them to verify: • the total number of terms searched; • the number of terms searched classified as requiring a query (QN); • the number of terms searched classified as not requiring a query (NO QN). • the number of terms searched classified as not requiring a query (NO QN As was to be expected, the percentage of terms searched classified as requiring a query varied considerably between participants, while it was quite similar for the terms not expected to elicit a query in the glossary. This percentage was similar for the German natives (PS1-01, PS1-02, PS1-03), although different terms were looked up. As can be seen in the graph below (Figure 5.1), one participant (PS1-06) looked up a considerably lower number of terms. Additionally, it was important to verify which terms which had been classified as requiring a glossary search had not been looked up by any participant and vice versa. Out of 54 terms, five QN terms were not looked up in any case, while one NO QN term was looked up by all subjects. The terms in question would therefore require replacement or reclassification. 141 5 Method PS1-01 PS1-02 PS1-03 PS1-04 PS1-05 PS1-06 0 20 40 60 80 63 63 70 50 76 31 24 24 26 22 22 0 % Terms searched QN (%/54) Terms searched NO QN (%/54) Figure 5.1: Percentage of terms searched per stimulus category in the pi- lot study. QN = query necessary, NO QN = no query necessary (Prandi 2018: 47) Figure 5.1: Percentage of terms searched per stimulus category in the pi- lot study. 5.4.6.1 Preliminary effects on search behaviour As previously stated, the position of the stimuli was expected to affect the partic- ipants’ search behaviour (see §5.4.3.4). Specifically, I expected the terms placed at the end of the sentence to result in more glossary queries than the terms in the middle of the sentence. In the small sample tested, sentence-final terms seemed to elicit more queries. This could be due to several reasons: first of all, partici- pants could anticipate that a specialised term was going to come up based on the context and would prepare themselves to query the glossary. Additionally, it is possible that, anticipating the need for a glossary search, they would pre- fer a sentence structure in their delivery that would favour a search. This would certainly require the development of ad-hoc strategies to facilitate terminology 142 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results look-up during SI and it could be argued that this would have been difficult for trainee interpreters. It may however be a behaviour likely to be observed in expe- rienced interpreters who are highly proficient in the use of technology during SI. On the other hand, stimuli placed within the sentence could be more difficult to handle through a glossary query, as this would likely generate more noticeable disfluencies in the participants’ rendition. It is possible that interpreters would therefore choose to use an alternative strategy, resulting in lower precision but in a more natural delivery and lower CL experienced. In considering the morphological complexity of the terms (see §5.4.3.3), I could identify a prevalence in searches for unigrams belonging to the “NO QN” cate- gory. This could be due to the fact that querying the glossary for a unigram is more straightforward than for bi- or trigrams: either the term is understood and looked up, or it is not. In the case of bigrams and trigrams, there is an additional layer of decision as interpreters may first have to decide which element of the term should be looked up (provided that all components have been understood). This would require additional cognitive resources unless a dedicated strategy has previously been developed and automatised. Therefore, participants may choose to adopt a different approach altogether. For terms requiring a query, no clear trend could be identified. 4Additionally, in MQM omission is included as a sub-dimension of accuracy separate from mis- translation, which does not fit into the present operationalisation of incorrect renditions. 5.4.6.2 Preliminary effects on accuracy I also conducted additional observations on this preliminary data concerning the accuracy achieved when working with the three different tools. In the context of the present experiment, accuracy is adopted as a performance measure (see §4.3 and §4.3.1) to identify instances of cognitive effort in the interpreter-tool interaction, not as a component of quality to evaluate the target text. Accuracy is therefore operationalised as the fraction of terms translated as per glossary or with an equally acceptable term, which would indicate an effective use of the tool and/or limited cognitive effort exerted to produce the term. The percentage of terms searched translated as per glossary was calculated and the results were analysed following a methodology already used in previ- ous studies on CAI tool use in SI (Prandi 2015a,b, Biagini 2015). I classified the term renditions adapting a classification put forward by Wadensjö (1998) for in- terpreting (see §1.1.2). This framework was chosen for two main reasons. First, to promote comparability with previous studies. Second, because it adopts a broad categorisation and stresses the effectiveness of the rendition, which facilitates the operationalisation of accuracy as a correlate of effective interaction with the tool. For instance, as can be seen below, omissions and unacceptable renditions 143 5 Method 5 Method are grouped together as they both index unsuccessful or effortful queries. Other evaluation frameworks, such as MQM (see §1.1.2), also include accuracy and ter- minology as issue types, but they consider them as distinct dimensions.4 These frameworks may prove very useful in studies aiming to assess the target text quality in CAI, especially if it has been specified that the interpreter should use the equivalent contained in the glossary, but this kind of analysis goes beyond the scope of the present experiment. According to the framework chosen, terms could be classified as: Close renditions: no information is lost, the rendition is precise, the term is trans- lated with the glossary equivalent or an adequate synonym; Close renditions: no information is lost, the rendition is precise, the term is trans- lated with the glossary equivalent or an adequate synonym; Acceptable renditions: some information is lost (e.g. through paraphrasing, the loss of an adjective in complex terms, a drop in register), but the general meaning is maintained; Zero/unacceptable renditions: this category groups renditions that completely or largely deviate from the original (the content is different) and terms that were left untranslated. 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results 5.4 Pilot study: Method and results The analysis of accuracy was conducted first at a microscopic level, that of terminology. However, a successful query that results in an accurately trans- lated term may nonetheless require too much time and attention, thus leading to serious errors or even to the complete omission of information. I therefore expanded my analysis to the sentence level. The transcriptions were annotated following Barik’s (1971) classification of omissions, additions and errors. Barik’s classification is very fine-grained. I decided to adopt only three of the categories identified by the author, as lower-level errors may have been due to other factors and may have been exposed to a higher degree of subjectivity in their identifi- cation, especially since I did not involve a second rater at this stage. I annotated the renditions according to the following elements: • substantial phrasing change (category E4 in Barik 1971); • gross phrasing change (category E5 in Barik 1971); • complete sentence omission (labelled M5). • complete sentence omission (labelled M5). Analysing partial omissions would certainly be interesting, but it would pose the methodological issue of defining units of information and of excluding the strate- gic use of omissions, which is difficult to evaluate post-hoc without involving the participants directly with a retrospective interview. Given the length of the ex- periment per participant (around 1.5 hours including the pre- and post-tests), I decided to reserve this analysis for future studies. The renditions that did not present any issue or only minor issues (skipping omissions and mild phrasing changes) were grouped in a single category to streamline analysis. From this initial analysis, no clear trend could be identified. This is probably due to the variable number of terms searched, which made it difficult to compare performance between subjects. 5.4.6.2 Preliminary effects on accuracy I assigned a value of 2 to close renditions, of 1 to acceptable renditions and of 0 to unacceptable and zero renditions. As for the degree of terminological precision achieved with Word and Excel glossaries and with InterpretBank, inter-subject variability is high. Nonetheless, Excel seems to lead to the worst terminological performance, i.e. to more frequent unacceptable renditions or terms left untrans- lated. This is probably due to the fact that when working with Excel it is neces- sary to manually skip to the next result, while with the other tools the results are all visualised together. This might make queries in Excel excessively cumber- some and time-consuming. InterpretBank seems to perform slightly better than Word in this respect. I assigned a value of 2 to close renditions, of 1 to acceptable renditions and of 0 to unacceptable and zero renditions. As for the degree of terminological precision achieved with Word and Excel glossaries and with InterpretBank, inter-subject variability is high. Nonetheless, Excel seems to lead to the worst terminological performance, i.e. to more frequent unacceptable renditions or terms left untrans- lated. This is probably due to the fact that when working with Excel it is neces- sary to manually skip to the next result, while with the other tools the results are all visualised together. This might make queries in Excel excessively cumber- some and time-consuming. InterpretBank seems to perform slightly better than Word in this respect. I also considered the terminological accuracy in relation to the term struc- ture and found that InterpretBank leads to higher accuracy for unigrams in five cases out of six (with the exception of PS1-06, whose search behaviour however differed greatly from that of the other participants). My hypothesis was that In- terpretBank would prove more effective, leading to higher accuracy, especially for complex terms (bigrams and trigrams). I did not find significant differences for bigrams and trigrams: queries with InterpretBank proved more effective than Word and Excel in half of the cases. 144 5.4.7 Discussion of pilot study results The pilot study overall validated the a-priori categorisation of the stimuli chosen for data collection. However, it highlighted a limitation of the initial design cho- sen for the experiment: given that only half of the stimuli introduced in the text were classified as requiring a glossary query due to their low frequency (i.e. 18 per condition, a total of 54), if participants only looked up a very limited number of terms, the data points collected for those participants would be too scarce to allow for a comparison with the rest of the sample. In order to overcome this limitation, I decided to replace the 54 stimuli classified as not requiring a query 145 5 Method with less frequent, more specialised terms. This involved rewriting a consider- able part of the speeches, but was nonetheless deemed necessary to ensure a more comparable search behaviour among participants.f As for the small effects observed for the stimuli position (see §5.4.3.4 for po- tential position effects on cognitive processing), it could be argued that the dif- ference in search behaviour observed might be due to the terms chosen (e.g., cog- nate status, see also §5.5.3) rather than only to their position, even though the terms were equally distributed within the speeches (half were sentence-medial and half sentence-final). This is possible especially if one considers that only half of the terms in each position had been classified as requiring a query. This aspect could be further tested by switching the position of the stimuli used or by choos- ing a different set of terms. However, this would require a complete rewriting of the speeches, as in the present study the terms are not presented in isolation. Re- placing the 54 stimuli classified as not requiring glossary look-up was expected to counterbalance this random effect. Finally, while at the time of the pilot study ASR was still a hypothetical im- provement of CAI tools, in the following months the first prototypes of ASR- CAI integrations started to appear (Fantinuoli 2017a, see §2.2.2). Word and Excel glossaries may be considered both as examples of “traditional” digital ways to organise and consult terminology for interpreters. Furthermore, they may be ac- cidentally modified by participants during data collection, an unforeseen issue which occurred several times during the pilot study. 5.4.7 Discussion of pilot study results For this reason, in the main experiment, I decided to prepare a tabular glossary and present it as a PDF file to overcome these limitations. I also added a mock-up of an ASR-CAI hybrid to include the most recent development in CAI technology, which I hypothesised to be the most cost-effective in terms of additional CL and degree of accuracy achieved in the interpretation (see §5.2). To sum up, the pilot study represented a valuable step in the study design. The methodology developed and tested in the study is a novelty in the area of CAI research, especially in terms of the tools compared in the study, and of the rigour in the design of the materials used for data collection. The pilot study was the first to specifically address solely digital support tools for terminology. Previous studies had either compared paper and digital glos- saries (e.g. Biagini 2015) or focused on a single CAI tool explored through be- tween-group designs (e.g. Prandi 2015a,b, Gacek 2015). The pilot study explored three different terminology support solutions, i.e. Word glossaries, Excel glos- saries and CAI tools. The study also helped shed first light on the hypotheses guiding the present research work and refine the hypotheses, for instance by highlighting potential commonalities between Word and Excel glossaries, which 146 5.5 Main study: Rationale and method informed the choice of the conditions to compare in the main study as discussed in §5.5. The empirical validation of the methodology laid first ground for the collec- tion of process and product data under controlled experimental conditions. The main methodological gain for the field consists in the development and valida- tion of a new method of speech design (based on Seeber & Kerzel 2012). The level of control and the number of variables considered in the development of the speeches is unprecedented, as previous studies had worked with naturalistic materials without controlling for potential variables affecting the outcome. The usefulness of the methodology proposed in Prandi (2017, 2018) has been recog- nised by a number of recent studies, which have drawn on the methods presented in this section. For instance, a recent study by Van Cauwenberghe (2020) on ASR support for terminology explored the role of morphological complexity for the accuracy achieved by his participants. He also derived several aspects of his evaluation framework from the categories used in the pilot study. 5.4.7 Discussion of pilot study results Investigating number renditions in ASR-supported SI, Frittella (2022) based the design of her speeches on the principles presented here. Overall, the pilot study confirmed the validity of the design choices for the scope of the present research work, while at the same time highlighting weak spots in the design and supporting the optimisation of the methodology for the main experiment. 5.5.1 Participants In this section, I describe the recruitment process for the experiment, highlight- ing the challenges encountered and how they motivated my choice of research de- sign. I further provide information on the demographics of the sample on which the experiment was conducted. 5.5 Main study: Rationale and method This section describes the methodology and results of the main experiment con- ducted to compare the process and performance of SI with the support of tradi- tional digital glossaries, CAI tools and CAI tools with integrated ASR. 5.5.1.1 Participants recruitment Like in the pilot study, participants were recruited among second-year students of the Master’s Degree in Interpreting of the University of Mainz/Germersheim 147 5 Method 5 Method with German as their native language and English either as their B or their C lan- guage. This language combination was chosen because it is the most represented in the master’s course and I thus hoped to reach a larger number of students to recruit for the experiment. An e-mail was sent to the teaching staff of the English department, whom I asked to circulate the call among advanced interpreting students. I also asked for the collaboration of the student body representatives in circulating the e-mail and a leaflet containing the basic information about the experiment. Participants were also recruited through the University’s Facebook groups, where the same information was posted in digital format. After a first round of e-mails sent dur- ing the Winter semester 2018/2019, which led to a few expressions of interest, a second round of e-mails was sent at the end of the Summer Semester 2019. Ad- ditionally, a recruitment e-mail was also sent to the University of Heidelberg, where I held a presentation of the study design and encouraged the students to take part in the study. Participants were remunerated (€50) and received a three-month free Inter- pretBank license. In addition, they had the opportunity to attend a free course on a very relevant and topical subject not usually offered during regular instruc- tion. Despite the considerable recruiting efforts, only 15 students expressed their interest in taking part in the study. Of the initial 15 students recruited, it was pos- sible to collect data from nine students. Data by two students had to be discarded due to technical issues which emerged during data collection, one student could not be calibrated due to an eyesight condition, two resigned after having con- firmed participation and one could not be tested due to the university facilities being closed in the wake of the COVID-19 pandemic. The small size of the sample recruited highlights a certain lack of interest for the subject, probably due to a general lack of awareness of the existence of CAI tools and the increasing relevance they are gaining in the interpreting profession (see also Prandi 2020). Another aspect that certainly had a negative impact on participant recruitment was time. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method & Fantinuoli’s (2021) study, numerical suggestions on the screen while interpret- ing. The interaction with a PDF glossary or a CAI tool certainly requires more extensive training, the benefits of which had been identified in previous work on the topic (Prandi 2015a,b). Structured training on all three conditions was there- fore considered essential to ensure that equal practice time be devoted to all three tools, a variable which could affect results. Based on the considerations on small sample sizes outlined in §5.3, the speeches used for data collection contained a large number of stimuli, allowing numerous observations per participant. For each condition, participants were presented with 36 stimuli. The highest number of data points collected is for the ASR condition (36 for each participant, because the terms were shown automat- ically on the screen). For the two conditions which required an active search in the glossary (PDF and InterpretBank), the minimum number of data points per participant was 22. 5.5.1.1 Participants recruitment Due to the nature of my investigation, which presupposes a sufficient level of expertise in tool usage, training was necessary prior to data collection. The training required the availability of around one hour of time per training session, and it is possible that second-year students may not have had enough time in their busy schedule due to exam preparation. A simi- lar experiment conducted recently on the simultaneous interpreting of numbers with ASR support (Defrancq & Fantinuoli 2021) did not foresee practice sessions prior to data collection, but it could be argued that for ASR, the interaction with the tool mainly consists in getting used to seeing terminological or, in Defrancq 148 5.5.1.2 Characteristics of the participants recruited The final sample was made up of nine advanced interpreting students. Working with students was a choice of practical nature, since they had to be trained on the tools before the experiment and the equipment and facilities for data collection were available at the university (see §5.5.6). Additionally, having all been trained at the same institution and having received comparable amounts of training in SI, they may be considered a relatively homogeneous sample. The challenges in recruiting representative samples of professionals are well-known in TIS (e.g. Hansen-Schirra & Nitzke 2020). The main issue in working with students is that they “exhibit different cognitive processes and behavioural patterns than profes- sionals, which might in turn affect the generalizability of the results” (ibid.). For this reason, care should be taken in generalising the results of the present study to the population of professional conference interpreters. As will be described in more detail in §5.5.6, before collecting gaze data, I had my participants go through a series of tests to gain a more detailed picture of my sample. I measured their typing speed and the size of their English vocabulary (§5.5.1.2.4). I also asked them a series of questions to gauge their knowledge of the subjects of the speeches used during the experiment (§5.5.1.2.5). With a view to provide a more nuanced picture of the sample at hand and identify additional confounding variables which may play a role in the interpre- tation of the experimental results, at the end of data collection I asked the par- ticipants to fill out the Translation and Interpreting Competence Questionnaire (TICQ) developed by Schaeffer et al. (2020). The main benefits offered by the tool 149 5 Method 5 Method consist in its easy customisation, its availability in four languages (English, Ger- man, Spanish and Chinese) and the fact that it can be administered online. As a quantitatively valid tool, its use has been advocated to promote greater compara- bility in T&I research (Schaeffer et al. 2020: 102). Only module A (demographics) and module C (interpreting competence) were administered in the online format. 5For instance, the questions in section C2 concerning their professional experience. 6Schaeffer, personal communication (2021). 5.5.1.2 Characteristics of the participants recruited While compiling the questionnaire, some students informed me that they had either misinterpreted some questions or that they were uncertain on how the questions could be applied to their status of interpreting students.5 This resulted in missing data for some questions and in a mismatch between the answers to several questions for the same test subject. Competence was not explicitly oper- ationalised as a variable in the present study, and I was later informed that the interpreting competence score was still under development.6 For these reasons, I do not report the total score here, but rather extrapolate from the question- naire the answers which are related to my research questions and which may provide further insight into the make-up of my sample. These answers were also expected to provide useful additional qualitative data to interpret findings on the participants’ interaction with the tools in the booth. I report relevant results in the sections below and then discuss the results of the pre-tests. 5.5.1.2.1 Demographics The nine students recruited for the study were attending the Master’s Degree in Conference Interpreting or the double Master’s in Translation and Conference Interpreting at the time of the experiment. They had all received at least three semesters of instruction in simultaneous and consecutive interpreting. Seven were females, two were males (age M = 27.33 years, Mdn = 25). They were all native speakers of German; two of them were bilingual (with Arabic and Polish). 5.5.1.2.2 Language proficiency In Module A3, the TICQ asks survey respondents to provide a self-rating of their knowledge of their L2 and L3 languages. In my sample, a small drawback con- sisted in the fact that two participants had indicated English as their L3, rather than as their L2. In the self-assessment question, no distinction is made between active and passive knowledge for L2, but only for L3. Therefore, if English had been selected as the participants’ L3, the average rating was used. The mean self-reported value was 79.17 (on a scale from 0 to 100, SD = 7.71). 150 5.5 Main study: Rationale and method P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Self-reported knowledge of English (%) Figure 5.2: Bar plot of the self-reported knowledge of English (%). 𝑀= 79.17, SD = 7.71 P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Self-reported knowledge of English (%) Figure 5.2: Bar plot of the self-reported knowledge of English (%). 𝑀= 79.17, SD = 7.71 Figure 5.2: Bar plot of the self-reported knowledge of English (%). 𝑀= 79.17, SD = 7.71 5 Method Only one participant indicated that they look up the term in their resources when they cannot recall it from memory, or they do not know its equivalent, which I found quite interesting considering the intensive training all students had attended prior to participating in the experiment and answering the TICQ. This might indicate that most of them prefer other, less demanding coping strategies in terms of attentional resources than performing a query in a digital glossary, which is probably considered an emergency tactic by most students in my sample. At the same time, the strategy of using an ASR-CAI hybrid to receive real-time terminology suggestions was not an available option for this question, which may have otherwise yielded different results. 5.5.1.2.3 Interpreting experience No students identified themselves as conference interpreters, although some of them reported having some professional experience. Their estimated décalage (Section C3.2 of the TICQ) was comprised between 2 and 3s (𝑀= 2.61, SD = 0.48). A question present in the TICQ particularly rel- evant to my research desiderata concerned the strategies adopted to deal with an unknown term while interpreting. The preferred strategies was “Ask my part- ner”, i.e. seeking help from the boothmate, followed by “Stalling” (Figure 5.3). Ask my partner Look up in resource Stalling 0 20 40 60 80 100 56 11 33 % Figure 5.3: Preferred strategy to deal with unknown terms during in- terpretation Figure 5.3: Preferred strategy to deal with unknown terms during in- terpretation 151 5.5.1.2.4 Typing speed and English vocabulary size The participants’ typing speed was measured ahead of the experiment (in words per minute, wpm). In addition, I assessed the size of their English vocabulary using the online vocabulary test WordORnot (Center for Reading Research 2014). The test takes around four minutes and requires the participant to judge whether the words shown on the screen are actual English words or non-words. The final test score is an estimate of the participant’s English vocabulary size. The results are illustrated in the graphs below (Figures 5.4 and 5.5). P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Typing speed (wpm) Figure 5.4: Typing speed of participants measured during the pre-test (wpm) Figure 5.4: Typing speed of participants measured during the pre-test (wpm) The participants’ typing speed and the size of their English vocabulary was tested as these factors may have an impact on their search behaviour, in particu- lar on the amount of terms looked up and on the amount of terms that they were 152 5.5 Main study: Rationale and method P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant English vocabulary (%) Figure 5.5: Estimated size of English vocabulary per participant (%) Figure 5.5: Estimated size of English vocabulary per participant (%) able to find in the glossary. It should be noted that a correlation between these two factors and the search behaviour may not be straightforward, as a number of scenarios may arise: a fast typer may search a low percentage of terms if he or she has a wide English vocabulary, but the opposite may also be true, because he or she may choose to only look up the terms considered strictly necessary and to adopt alternative strategies, such as paraphrasing, which should be easier to adopt if one has high language flexibility. Nonetheless, I preferred to be on the safe side in order to consider also these aspects, if necessary. 5.5.1.2.5 Knowledge of speech topics In addition to these data, participants had to answer a few questions which served a double purpose: to ascertain the participants’ background knowledge of the topic of the speeches to interpret, which were different from the topics chosen for the training, and to provide the participants with at least a general overview of the speech topics. Not knowing the subject of the speeches may have caused additional stress ahead of the interpreting task. I first asked the participants to rate their knowledge of the key topics discussed in the speeches. As can be seen in Figure 5.6, no topic was completely unknown to the students, and moderate differences could be found between the students’ knowledge about the six topics. However, overall, none of the students considered themselves experts in the topic presented. Therefore, I could reasonably expect the students to look up a considerable number of terms during the experiment. Additionally, I verified the participants’ terminological preparation on the topics in question. To this aim, 153 5 Method Renewable energy Fossil fuels Coal mining Solar energy Wind energy Nuclear energy 1 2 3 4 5 Knowledge (self-rating 1–5) Figure 5.6: Self-ratings of knowledge of the main topics included in the speeches. Mean values for the sample on a scale from 1 (no knowledge) to 5 (expert knowledge) 5 Method Figure 5.6: Self-ratings of knowledge of the main topics included in the speeches. Mean values for the sample on a scale from 1 (no knowledge) to 5 (expert knowledge) I asked the students to write down three terms (with their German translation) which they associated with each of the six speech topics. I then checked which terms had been included and whether the students had indicated any term that was also present as a stimulus in the speeches. Of all the terms added by the participants, only four were similar or closely related to the stimuli selected. Of course, I cannot exclude that, just because a certain term was not mentioned during the briefing task, it was unknown to the student. Nonetheless, this task in the pre-test proved useful to verify which terms would first come to mind when thinking of a certain topic. 5.5.3 Materials The speeches that the participants were asked to interpret during data collection were very similar to the ones used during the pilot study, with a few alterations. The speeches are reported in the Appendix. First, as discussed in §5.4.7, the 54 stimuli which had previously been classified as not requiring a query were replaced with less frequent, more specialised terms. This required an almost complete rewriting of the speeches. Table 5.2 shows the distribution of terms in the final speeches. i p It should be noted that the terms selected as stimuli do not represent all do- main-relevant terms which could be of interest to interpreters. A domain-specific but highly frequent word could also elicit a glossary query. Ultimately, the terms were selected with the aim to generate sufficient data points for each participant. As shown in Table 5.2, the ratio for each category (position and morphological complexity) was maintained also in the main experiment. All proper names and figures were removed from the text, as they could constitute further sources of difficulty in addition to specialised terminology (Gile 1995, 2009), which was in focus in the present study. Since the stimuli selected all had a very low frequency, they were expected to require a query in the glossary or a glance at the ASR suggestions in most cases. However, since the corpus of texts selected for the extraction of the stimuli was highly specialised, cognate facilitation effects may have occurred (Costa et al. 2000), for instance due to the prevalence of Latinisms, It should be noted that the terms selected as stimuli do not represent all do- main-relevant terms which could be of interest to interpreters. A domain-specific but highly frequent word could also elicit a glossary query. Ultimately, the terms were selected with the aim to generate sufficient data points for each participant. As shown in Table 5.2, the ratio for each category (position and morphological complexity) was maintained also in the main experiment. All proper names and figures were removed from the text, as they could constitute further sources of difficulty in addition to specialised terminology (Gile 1995, 2009), which was in focus in the present study. Since the stimuli selected all had a very low frequency, they were expected to require a query in the glossary or a glance at the ASR suggestions in most cases. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method more flexible and customisable training schedule would encourage participation in the study. Unfortunately, as described above, this was not the case. In terms of the training content, the only modification introduced was the substitution of the Word and Excel glossaries with a PDF glossary, and the intro- duction of the ASR-CAI mock-up to account for a third generation of CAI tools which participants would likely be able to work with in their foreseeable future careers. The condition that equal practice time be ensured for each tool was also maintained for the main experiment. I made sure that the students had actually completed the five practice sessions by asking them to upload the recordings of their interpretations online. The students were able to proceed to the following training session only after completing the previous one. After the training had been completed, the participants were asked to take a short quiz to demonstrate their proficiency in the use of the tools. The questions were formulated in or- der to verify that the students had acquainted themselves with the user interface and were able to correctly operate the PDF tool and InterpretBank. All students passed the test and could hence take part in the data collection. 5.5.2 Participants’ training As in the pilot study, participants were required to attend a training course prior to data collection. The structure of the training was the same as during the pilot study, with the only difference that it was held online as a self-paced course. To this aim, I created an online course on the University’s Moodle platform to which the study participants were enrolled after confirming their participation in the study. The introductory meeting was offered as a webinar that the students could attend live or watch later as a recording. The online format of the training offered a double advantage: the participants could re-watch the recording as many times as necessary and they could more easily integrate the training in their schedules, with the additional option of re- peating practice sessions if they needed additional training. I had hoped that a 154 5.5.3 Materials However, since the corpus of texts selected for the extraction of the stimuli was highly specialised, cognate facilitation effects may have occurred (Costa et al. 2000), for instance due to the prevalence of Latinisms, 155 5 Method Table 5.2: Distribution of terms for each speech in the main experiment. The morphological complexity and position of terms were equally dis- tributed within the speeches. Total Morph. complexity Position 36 Unigrams 12 End 6 Middle 6 Bigrams 12 End 6 Middle 6 Trigrams 12 End 6 Middle 6 which are frequent in specialised discourse (see §1.1). Since precedence was given to providing continuum by drafting entire speeches, it was not feasible to replace all cognate stimuli with non-cognates. Therefore, to control as much as possible for cognate facilitation effects, when a non-cognate translation was available, it was selected as target language equivalent for the glossary. which are frequent in specialised discourse (see §1.1). Since precedence was given to providing continuum by drafting entire speeches, it was not feasible to replace all cognate stimuli with non-cognates. Therefore, to control as much as possible for cognate facilitation effects, when a non-cognate translation was available, it was selected as target language equivalent for the glossary. In addition, the cognate status of the stimuli across the three speeches used during the experiment was checked. To calculate the cognate status of the se- lected stimuli, I calculated the normalised Levenshtein ratio as: Levenshtein ratio = 1 − Levenshtein distance max(length source, length target) The Levenshtein distance (Levenshtein 1966), i.e. the smallest number of dele- tions, additions and substitutions required to transform the source into the target string, is divided by the maximum length of the strings compared to account for word length. The formula above yields a score comprised between 1 (perfect cog- nate) and 0 (no overlap). A higher score indicates that fewer deletions, additions or transpositions are required to obtain the target term (Schepens et al. 2012). The ANOVA of the normalised Levenshtein scores for the stimuli of each text was found to be non significant (𝐹(2, 105) = 0.243, 𝑝= 0.785). On this basis, I did not expect cognate facilitation effects to significantly affect the participants’ interaction with the support tools. The speeches were recorded once again by the same speaker as in the pilot study. The structure made up of sentence clusters comprising an introductory, a target (T) sentence and a continuation (C) sentence was maintained. However, 156 5.5 Main study: Rationale and method in order to facilitate the analysis and isolate the effects of each stimulus on dé- calage and CL, I introduced a 5s pause between each sentence cluster and the next. In literature, the average EVS for professional conference interpreters has been identified as comprised between 2s and 6s (e.g. Barik 1973, Lederer 1978, Oléron & Nanpon 2002, Christoffels & De Groot 2004, Timarová et al. 2011, Defrancq 2015), so a 5s pause was deemed sufficient. The average length of the speeches used in the main experiment was 14’39”, with an average WPM of 104.3 (164.4 average syllables per minute). As shown in Table 5.3, although the content of the speeches varied, they were highly comparable in terms of length and speed. To further verify whether the speeches were comparable, I analysed them us- ing Coh-Metrix (McNamara et al. 2014). Coh-Metrix automatically computes a series of indices of text cohesion and coherence which provide an estimation of text difficulty and can be used both for written and for oral texts. The Flesch Reading Ease Index (Flesch 1948), a common index of text readability, and the SYNLE index, a reliable index of WM load (McNamara et al. 2014: 70), were con- sidered as indices of difficulty. The texts were highly comparable also according to these two indices. Therefore, the speech difficulty was not expected to affect performance. Levenshtein ratio = 1 − Levenshtein distance max(length source, length target) Table 5.3: Features of the speeches used in the main experiment. For each speech, the table reports the duration, the number of words per minute (wpm) and of syllables for minute (spm), the Flesch Reading Ease score and the SYNLE score (left embeddedness, mean). Table 5.3: Features of the speeches used in the main experiment. For each speech, the table reports the duration, the number of words per minute (wpm) and of syllables for minute (spm), the Flesch Reading Ease score and the SYNLE score (left embeddedness, mean). Speech Duration wpm spm Flesch Reading Ease SYNLE A 14’34” 105 165.8 59.92 3.57 B 14’46” 104 162.5 62.47 3.41 C 14’37” 104 164.9 61.35 3.71 While interpreting, the students could see both the video of the speaker and the tool with the glossary on a split screen. The video was placed on the left hand-side, the glossary on the right-hand side of the monitor. I decided to also include the video of the speaker to reproduce a typical configuration of an SI task, where the speaker is usually visible (see Seeber 2017). Additionally, visualising the speaker video and the tool on screen may be compared to a typical RSI con- figuration (see §2.1.1), a setting in which the use of CAI tools, particularly with ASR integration, may be envisaged in the near future. The speaker video was also included for another purpose related to the hypotheses guiding the experiment: since the video was expected to keep the participants’ attention on the screen, 157 5 Method essentially serving as a fixation cross (Conklin et al. 2018), it would support the collection of gaze data when the participants were not interacting with the tools. This data proved useful in investigating effects on attention allocation during CASI (see §6.2.2.6). 5.5.4 Apparatus used for data collection Gaze data was collected with an SMI red250 remote eyetracker mounted on an ex- ternal desktop monitor. No chin rest or head support was used, although it would have improved data quality (Holmqvist 2011: 83), because it would have con- stricted the participants’ head movements during speech and limited their ability to look at the keyboard while typing. The use of a head-mounted tracker and of a chin rest would have promoted more precision in the collection of gaze data (Conklin et al. 2018: 17), but it would have limited movement. The increased intru- siveness perceived may have caused greater stress. As pointed out by Hvelplund (2014: 206), “the [interpreter]’s eye movements may thus well be related to stress from not being able to look at what key is being pressed in addition to actual problem-solving activities arising from the [interpretation] itself”. The students’ interpretations were recorded in Audacity with a microphone placed on the table, next to the participant. At the beginning of each video, whether with or without the ASR-CAI mock-up, I added a set of instructions that the participants were required to follow during data collection. Participants were instructed to wait until they would hear a loud “beep” coming from the computer before putting on their headphones. This acoustic marker was added to the recording to facilitate the synchronisation of the original speech with the interpretations. It was loud enough to be picked up by the microphone used to record the participants’ deliveries. 5.5.5.1.1 Terminological accuracy Assessing the terminological accuracy of terms consulted in the glossary or au- tomatically prompted by ASR has become a standard measure in CAI research focused on terminology (e.g. Biagini 2015, Prandi 2015a,b), as one of the main claims of such applications is that their use may lead to greater target text (TT) accuracy. To conduct the evaluation, I used the same three-level grading scale adopted in the pilot study. The three categories were defined as in the pilot study, i.e. a value of 2 would be assigned to “close renditions” (terms translated as per glos- sary or through a synonym), a value of 1 to acceptable renditions, and a value of 0 to wrong renditions or to an omission of the term which had not been replaced by a paraphrase. For the evaluation, the terms were presented to the raters in an Excel table. The terms were listed in isolation, not embedded in the respective utterance, in their order of appearance in the source speech and with their respec- tive glossary equivalent. In order to facilitate the evaluation, adequate synonyms were listed (if available) next to the glossary equivalent. Next to the columns with the terms, the glossary equivalent and the synonyms, the participants’ renditions were listed. Next to each participant’s rendition, the raters could enter the accu- racy scores. For this evaluation, the prosodic component was excluded (no audio recordings of the renditions were provided). 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method goes beyond the scope of the present study. For instance, in the experiment only duration-based eyetracking measures were used, while future studies may adopt additional metrics such as the number of fixations. Similarly, the only subjective measure used was a short qualitative questionnaire, but future research may, for instance, also test the use of cued retrospection to gain further insight into the CAI process. The present section provides the rationale for the selected metrics and describes how they were applied to the analysis of the data collected during the main experiment. 5.5.5.1 Performance measures The study adopted the two performance measures included in the pilot study: terminological accuracy and number of errors and omissions. 5.5.5 Measures Chapter 4 reviewed the main methods used to measure cognitive load and effort in translation and interpreting and highlighted how a combination of methods may represent a valuable experiment design choice to address the inherent limi- tations of each metric. As introduced in §5.3 (see Table 5.1), the present study combined performance, behavioural, and subjective measures. It appears trivial to say that the measures selected for the main experiment are not the only measures which may be used to explore CASI from a cognitive processing standpoint. Conducting a comprehen- sive evaluation of how all measures of CL apply to the research object in question 158 5.5.5.1.2 Errors and omissions In analysing the quality of the deliveries, it is paramount to consider not only whether the stimulus term was correctly rendered, but also whether the glossary 159 5 Method 5 Method query occupied cognitive resources which could not be allocated to the processes of listening and processing of the adjacent units of information. It is theoretically possible that a correctly identified term co-occurs with contresens or the com- plete omission of target and/or continuation sentences, as discussed in §5.4.3.2. Therefore, in addition to the analysis of the terminological accuracy, I expanded the investigation to the sentence level. Since the individual interpreting skills also play a role, I chose to conduct the analysis at a more superficial level, only checking the presence of severe errors and complete omissions for the terms queried (in the InterpretBank and PDF glossary) and not venturing into a detailed analysis of other lower-level issues such as partial omissions in the interpreted sentences. This approach would require controlling for the individual interpret- ing competence and disambiguating between strategic and non-strategic omis- sions, additions and generalisations. These categories would be difficult to oper- ationalise without subjective methods such as retrospective interviews, which I chose not to conduct because it would have further extended the duration of the experiment. In analysing errors and omissions, I only considered the target and continua- tion sentences for which a glossary query had occurred. The goal was to establish whether a glossary query may result in an error or omission in the target or con- tinuation sentence, which would indicate an interference between the two tasks. Therefore, I considered how the use of different tools affected this variable in order to answer the question: are there fewer complete omissions and severe er- rors when terms are presented on the screen automatically? Finally, in line with my hypotheses (see §5.2), I explored where the majority of errors and omissions occurred, whether in the target sentence or in the continuation sentence. 5.5.5.2 Behavioural measures Six behavioural measures were included in the main experiment: glossary queries, EVS, inter-cluster pause duration (ICPD), time to first fixation, average fixation duration and fixation time. 5.5.5.2.3 Inter-cluster pause duration (ICPD) A short EVS may also suggest overall faster processing, but it does not measure it directly. Hence, I also analysed the duration of pauses between sentence clus- ters, i.e. between the last word of the continuation sentence pronounced by the speaker and the last word of the delivery pronounced by the participant for each sentence cluster. This may be considered as an additional measure of time lag and is geared towards a general analysis of how the speed of processing is influenced by the explanatory value of the tool used as support. Since the speeches used in the main experiment presented a 5s pause between two sentence clusters (see §5.5.3), the ICPD may be used as an additional indica- tor of cognitive effort. More specifically, a long ICPD (of 5s or longer) may be interpreted as an indicator of fast processing. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method This metric was included as its usefulness had been highlighted both by pre- vious research (Prandi 2015a,b, Biagini 2015) and by the pilot study (see §5.4.7). The categories of terms analysed are reported in §6.1.2.1. 5.5.5.2.2 EVS A widely-used indicator of CL in interpreting is the EVS, as discussed in §4.4.1.2. In the main experiment, I expected significant differences between tools in the EVS (see §5.2). This may point to a greater ease of use of the ASR-CAI mock-up compared to the CAI tool and the PDF glossary, and of the CAI tool compared to the PDF glossary. In the main experiment, I calculated the EVS for the terms searched by the par- ticipants and compared it for the three tools, both for each participant and for the whole sample. The EVS was calculated as the begin time of stimulus translation minus the end time of stimulus utterance by the speaker. The availability of time stamps for individual words offered by the automatic transcription obtained with the Speechmatics tool (see §5.5.7.1) allowed overcoming a methodological chal- lenge identified by Timarová et al. (2011: 155), i.e. the need to manually identify the beginning of the stimulus. Since this the timestamp was available automati- cally, the EVS could be measured with millisecond precision. 5.5.5.2.1 Glossary queries In order to gain a comprehensive picture of the test subjects’ interactions with the tools during SI, I collected data pertaining to the queries performed in the PDF glossary and in InterpretBank. As will be discussed in §6.1.2.1, all participants fixated all stimuli terms on the screen, therefore the analysis focused on the tools with manual look-up. 160 5.5.5.2.6 Fixation time In addition to the average fixation durations on the tool area, I also considered the total time spent fixating the speaker and the tool area. SMI BeGaze provides the metric “Fixation time”, i.e. the “sum of the fixation durations inside the AOI” (SensoMotoric Instruments 2017a: 370). In the present study, the metric repre- sents the amount of time spent processing the speaker video and the side of the screen where the tool was displayed. This metric was expected to provide in- formation on how attention was allocated to the different sources of visual and visual-verbal information available to the participants, specifically on how much the tool distracted the participants from the speaker (see §5.2). 5.5.5.2.5 Average fixation duration on term and tool AOIs Fixation duration has been traditionally used as an indicator of cognitive effort in the processing of the ST (see §4.4.2.2). In the present study, the focus was on the effort exerted by participants while interacting with the tools. The results therefore do not indicate how easy it is to interpret the target terms in differ- ent conditions but should rather be understood as an indication of the effort of human-machine interaction. In each condition and for each participant, I calculated the average fixation duration on the term AOIs. Additionally, I also considered the duration of the fixations on the tool area. 5.5.5.2.4 Time to first fixation The time to first fixation on the term AOI was expected to provide insight into the question of how the tool use affects the participants’ speed in identifying the 161 5 Method target term on the screen. SMI BeGaze provides the metrics “Time to first appear- ance” and “Entry time” for each AOI. The metrics, in ms, represent respectively the moment in which the AOI first becomes visible on the monitor and the start time of the first fixation to enter the AOI. For each participant and under each condition, I therefore calculated the mean time to first fixation for the term AOIs by subtracting the timestamp for “Time to first appearance” from the timestamp for “Entry time”. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method quantitative analysis conducted with the above-mentioned performance and be- havioural measures. In the questionnaire, the participants were first asked to rank the speeches from the easiest to the most difficult. This was done to check for potential ef- fects due to the nature of the speeches interpreted and not to the other variables controlled in the experiment. Additionally, the subjects were asked to rank the tools from the most useful to the least useful and from the most distracting to the least distracting. I also asked participants to include further details of which as- pects of each tool they had found to be most useful and which most problematic. This was expected to help identify the reasons behind their preference. Finally, I explored their preferences as to the tool which they would bring with them into the booth during future assignments. 5.5.5.3 Subjective measure: The debriefing questionnaire Only one subjective measure was included in the main experiment: a short post- hoc qualitative questionnaire. The questionnaire was kept short to avoid extend- ing the duration of the already relatively long experiment. The inclusion of the debriefing questionnaire was expected to provide additional data to help frame the quantitative analysis and to highlight phenomena not emerging from the 162 5.5.5.4 A note on pupil size In previous publications reporting on the envisaged methodology for the present study Prandi (2017, 2018), I had postulated the inclusion of the metric “pupil size” in addition to the measures discussed in the previous section and selected for the main experiment. Pupil size, or pupil diameter, was used in previous studies investigating the interpreting process (see §4.5.3), both as an indicator of global cognitive load generated by the task (e.g. Hyönä et al. 1995) and to explore local variations in cognitive load as a response to specific features of the source speech (Seeber & Kerzel 2012). As discussed in §4.5.3, however, correlating variations in pupil size to varia- tions in cognitive effort is not straightforward. There are a number of factors which may influence the pupillary responses to the stimuli presented. This rep- resents an important limitation for the adoption of pupillometry as a method to explore cognitive load in interpreting. The specific research object of the present investigation presents additional limitations to the ones discussed in §4.5.3. For instance, the way in which the terminological information was presented in the PDF glossary and in the CAI tool may have resulted in higher arousal in the participants than during the use of the ASR-CAI mock-up. In the PDF condition, the participants could see a large number of terms on the screen. In the CAI condition, the screen changed very quickly during the query. It is possible that participants may have been more visually stimulated by the interfaces in the PDF and the CAI condition due to the highly dynamic stimuli. Variations in pupil size 163 5 Method would therefore reflect different levels of arousal rather than higher or lower cognitive load. Additionally, the ASR-CAI mock-up window was relatively empty: it was white most of the time apart from the moments in which the terms appeared. The higher screen luminance in this condition may have been expected to re- duce the pupil size. This reduction in pupil size may therefore simply reflect a physiological reaction of the pupil to the brightness of the screen, not variations in cognitive effort. This is a common issue in human factors studies, for instance when participants interact with web pages (Holmqvist 2011: 530). Finally, it should be noted that, in most studies on interpreting which adopted pupillometry to explore cognitive load (e.g. Hyönä et al. 5.5.5.4 A note on pupil size 1995, Seeber & Kerzel 2012), the informants were not interacting with visual stimuli. The only excep- tion is the study by Gieshoff (2018). However, she was able to control for lumi- nance across conditions and participants as the stimuli were not co-created by her participants, unlike in the present study for the two conditions which involve manual look-up. For the limitations discussed above, I opted for a more cautious approach and chose not to include pupil size as a metric for the present study. 5.5.6 Procedure The experimental design was very similar to the pilot study. I introduced several modifications and additional tests to round up and facilitate my subsequent data analysis. As described in §5.5.2, the main difference consisted in the replacement of the Word and Excel glossaries with a PDF glossary and the introduction of an ASR-CAI mock-up as the third condition. In the PDF glossary, the font size cho- sen was 16, as recommended in literature (e.g. O’Brien 2009: 261, Hvelplund 2014: 20, Conklin et al. 2018: 37). A large font size was also chosen for the InterpretBank glossary and for the ASR-CAI mock-up (see Figure 5.7). The ASR-CAI mock-up was prepared following a previous study on the auto- matic speech recognition of numbers by Desmet et al. (2018), who had prepared a PowerPoint presentation containing all the numerals mentioned during the speech. I prepared a version of the recorded speech positioned on the left-hand side of the screen, as for the other conditions, while on the right-hand side of the screen, on a white background, I added the terms and their glossary equivalents shortly after they had been pronounced by the speaker. This made it possible to simulate a constant system latency (ca. 1s) and ensure synchronisation between the speaker video and the ASR-CAI mock-up. 164 5.5 Main study: Rationale and method (a) PDF (b) CAI (c) simASR Figure 5.7: Experimental conditions (PDF, CAI and simASR). In all con- ditions, the speaker was shown on the left hand-side of the screen, while the terminology was presented on the right. AOIs were placed on the speaker video, the tool area, and on each stimulus term. (c) simASR (c) simASR Figure 5.7: Experimental conditions (PDF, CAI and simASR). In all con- ditions, the speaker was shown on the left hand-side of the screen, while the terminology was presented on the right. AOIs were placed on the speaker video, the tool area, and on each stimulus term. 165 5 Method 5 Method As discussed in §5.5.3, only the terms present in the glossary were shown on the screen to simulate the behaviour of the ASR module, which uses the underly- ing glossary to identify relevant items in the generated transcription. The target language equivalents are those present in the glossary and which had been pre- viously selected and validated during glossary creation. 5.5.6 Procedure This architecture corre- sponds to the current state of the art for CAI tools (see §2.2.3 and §2.4). The font and colour chosen were the same as in the only available prototype of this kind (InterpretBank with ASR). In the mock-up, only one term was shown at a time on the screen and it disappeared after 3000ms. I decided not to use the actual ASR integration in the CAI tool InterpretBank, but rather to simulate it, as I would be comparing two mature and stable systems (PDF glossary and In- terpretBank) with a system still under development and which at the time of the experiment did not yet work optimally for terminology recognition. Mistakes in the use of the PDF glossary and InterpretBank would be due to suboptimal usage, not to system performance, unlike for the ASR-CAI hybrid. Additionally, using the real system could have resulted in different levels of performance, making it difficult to compare participants’ data for this condition. While it would have been possible to simulate system failures experimentally, for instance by testing the real tool on the speaker’s video, following this approach would have intro- duced yet another variable into the experimental set-up and further complicated data analysis. As this exceeds my research questions, I decided not to explore this facet of interpreter-tool interaction in the present experiment. Thus, the results for the ASR condition represent the ceiling performance that participants may achieve when using this kind of support. To denote that the ASR-CAI mock-up represents the best performance obtainable with an ASR-enhanced CAI tool, this condition will henceforth be defined as “simASR”. On the day of the experiment, participants were briefed on the structure of the experiment and signed a consent form for the collection and use of their data. They were informed that they would be able to revoke their consent at any time and that their data would be made anonymous for analysis. Participants were remunerated with €50 after the end of the experiment. Before interpreting, they went through a brief pre-test comprising a measurement of their typing speed and the size of their English vocabulary (see §5.5.1.2.4). Afterwards, they answered a series of questions to help them prepare for the interpreting task (see §5.5.1.2.5). After the preliminary tests were completed, participants were asked to sit in front of the monitor where the stimuli were going to be presented. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method under stable light conditions. Participants were seated at a distance of around 60– 65cm from the monitor. To keep the participants’ position stable relative to the eyetracker and the monitor, a backpack was tied to the chair where they were sit- ting. They were asked to wear the backpack, which was adjusted in order not to restrict the participants’ movements excessively, but to provide feedback if they shifted too close to the monitor. Nonetheless, a certain variation in their prox- imity to the screen was inevitable. Participants were seated on a chair without wheels to avoid them changing position during the experiment. I explained what they were going to see on the monitor before the actual in- terpreting task would start and clarified any remaining doubts they may have on data collection. The microphone was placed on the table next to the participants to record their interpretation. Afterwards, the experiment was started. The experiment was prepared in SMI Experiment Center (SensoMotoric Instru- ments 2017b). Participants were calibrated at the beginning of the experiment. A nine-point calibration was performed at the beginning of the experiment and then repeated at the start of each new trial. During lengthy recording sessions, drift can occur, i.e. a gradual loss of synchronisation between the participant’s gaze and the recorded position on screen (see Hvelplund 2014: 210), which war- rants a re-calibration between trials. After the calibration, a dry screen recording was run during which I prepared the speech and the tool on the participant’s monitor. Then, the second screen recording started, for which gaze data was recorded. I asked the participants to start the video and to follow the instructions on the screen. They would hear a loud “beep” coming from the computer and would then put on their headphones. I then asked participants to place their headphones on the desk in front of them at the end of each trial. To avoid fatigue effects, which can affect results (see Spinner et al. 2013: 400, Keating & Jegerski 2015: 18), I allowed participants to take a short break or drink a sip of water between trials. I then proceeded with the second and the third condition. In total, the eye-tracking part of the experiment lasted around one hour including breaks. Conklin et al. (2018) recommend not to exceed a duration of one hour of time for eyetracking studies. 5.5.6 Procedure In the Tra&Co laboratory, a dedicated room is available for the collection of eyetracking data 166 7https://www.speechmatics.com/ (Accessed 25.08.2020) 5.5.7 Data preparation In this section, I illustrate how the data was prepared and systematised ahead of the analysis, the results of which are reported in §6.1. 5.5 Main study: Rationale and method As in the pilot study, the order of the speeches and of the tools were coun- terbalanced using a randomised Latin square design to minimise order effects (Conklin et al. 2018: 42–43). The impact of learning effects (e.g. Lazar 2017: 65) was minimised by the training attended by participants for all three conditions (see §5.5.2).t After the eyetracking section of the experiment was completed, I asked partic- ipants to answer some debriefing questions and sent them a link to fill out the TICQ questionnaire (see §5.5.1.2). 167 5 Method 5 Method 5 Method 8www.exmaralda.org (Accessed 2021-07-13) 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method terms (source speeches) or their renditions by the test subjects. This facilitated the analysis, as it was possible to export each track separately, time stamps in- cluded. This was necessary to calculate the EVS between the terms pronounced by the speaker and the terms interpreted by the participants (see §6.1.2.2). To calculate the pause length between the sentence clusters (see §6.1.2.3), the audio tracks and their transcriptions were imported into the transcription tool Partitur Editor (Schmidt & Wörner 2014),8 which offers a feature to calculate pause length between two annotations automatically. 5.5.7.1 Recordings First, the audio recordings of the students’ interpretations were cut in order to only contain the relevant material for the analysis. Due to the structure of the experiment prepared in SMI Experiment Center and in order to avoid acciden- tal interference with the software during data collection, the recording had to be started in Audacity prior to the screen recording section of the experiment and therefore contained additional bits of audio. The trimming operation was not necessary for the original speeches, as they had been recorded prior to the experiment. After trimming the recordings, the second step was aligning them with the original speech. In order to facilitate this step of the process, the original record- ing contained the auditory marker that was picked up also by the microphone used to record the students’ renditions. After importing the two audio tracks into Audacity, the audio peak was identified in both audio tracks and used as a reference to align the source speeches and the renditions. In hindsight, a short and sharper sound would have been preferable, as it was sometimes difficult to identify the start of the peak, but the method proved nonetheless effective. Afterwards, the recordings of the students’ interpretations were transcribed au- tomatically using the commercial Speechmatics transcription service.7 The tran- scripts thus generated were later corrected manually in order to remove any transcription error. As in the pilot study, the HIAT transcription conventions were followed (Ehlich & Rehbein 1976). Like other transcription systems, the Speechmatics ASR service provides the advantage of assigning time stamps to each word of the transcript, a feature which proved useful during data analy- sis. Even though the transcription was already available for the source speeches, they were nonetheless aligned using the dedicated feature also provided by the Speechmatics ASR service, in order to obtain time stamps for each word in the original speeches. The third step consisted in importing the audio tracks and the relative tran- scriptions into the ELAN (2020) transcription software. For each audio recording, a dedicated track was added containing the aligned transcription. An additional dependent track was added for the interpretations: it contained only the stimulus 168 5.5.7.2 Areas of interest The other set of data to be prepared consisted in the eyetracking recordings. There were a total of 27 recordings, three for each of the nine participants, who worked in the three different conditions (PDF, CAI, simASR). The choice of areas of interest (AOIs) was motivated by my hypotheses con- cerning attention allocation under the three conditions (see §5.2). As I expected participants to be least visually (and cognitively) engaged in the simASR prompts and most focused on the glossary in the PDF condition, two large rectangle- shaped AOIs were placed on the main windows visualised by the participants: the speaker video and the tool window, to measure the time spent looking at the two areas of the screen. Additional AOIs were placed on the stimulus terms. For all AOIs, I used the hand-drawn method (see Hessels et al. 2016: 1695). At this stage of analysis, I chose a large AOI for the speaker video, although it would be interesting to conduct more fine-grained inquiries exploring how visual attention is shared, for instance, between the speaker’s lip movements (Gieshoff 2018, Seubert 2019) and gestures (Seeber 2012) and the glossary, i.e. between dif- ferent sources of visual input. The stimulus term AOIs extended over the termi- nological pair. This was done in first instance for a practical reason: drawing separate AOIs on the source term and on the target term would have doubled the number of AOIs per stimulus. With nine participants each potentially pro- cessing a total of 108 terms (36 per speech), this would have resulted in a total of 1944 AOIs, of which 1296 would have had to be hand-drawn anew due to each participant’s idiosyncratic search behaviour. Moreover, for moving terms (i.e. in the PDF and CAI condition), the AOIs would have been dynamic, requiring man- ual adjustment for intervening frames. Having double as many AOIs would also have doubled the amount of data to be analysed. To keep data analysis feasible, I opted for a larger AOI comprising both the source and the target term. 169 5 Method 5 Method SMI BeGaze requires AOIs to be drawn post-hoc in the AOI Editor. This was necessary also because each participant worked with a different combination of speech and tool and because the term AOIs were determined by the partici- pants’ search behaviour. 5.5.7.2 Areas of interest Because I had to use the screen recording component in Experiment Center and each participant conducted different queries in the glos- sary, data could not be collected for all participants on the same stimuli. To over- come this limitation, the AOIs for the elements shared by all participants (speaker video and tools) were created as global AOIs. The AOIs placed on the stimulus terms had to be created as local AOIs (i.e., not shared across participants), but each term received a code to facilitate the comparison across participants. The recording started slightly before the moment in which the students started the video of the original speech and the first 30 seconds of the recording contained the instructions for the participants. Therefore, the areas of interest placed on the speaker video and on the tool were activated (made visible) as soon as the speaker appeared on screen by adding a key frame. The same operation was repeated at the end of the speech, when the video ended and a second key frame was added when the AOIs were deactivated. In this way, the data was only analysed for the time window in which the AOIs were visible on the screen. The same principle was followed while placing the AOIs on the terms pre- sented automatically on the screen in the simASR condition: the AOI was ac- tivated as soon as the term appeared on screen and deactivated when it disap- peared. It was not possible to standardise this step completely across all partici- pants as the speech/tool combination was randomised, but I was nonetheless able to apply the AOI definitions for each video and tool for subsequent participants and merely had to adjust the timing. The most challenging step consisted in drawing the AOIs on the stimulus terms looked up by the participants during the PDF and CAI trials. The two conditions presented similar issues, although the CAI condition was particularly complex. In principle, the same procedure was followed as for the simASR con- dition: the AOIs were placed on the terms searched and found in the glossary, activated as soon as they appeared and deactivated when they were no longer visible on screen. 5.5.7.2 Areas of interest Due to the nature of the PDF glossary, which showed a large amount of terms on screen at the same time, when two or more terms were alphabetically close to each other, several terms were already visible on screen while the first query was being performed. I had initially planned to activate the AOIs for all terms as soon as they were present on screen and to deactivate them once they were no longer visible. However, this would have resulted in AOIs visible on the screen for a very long time. As they would not represent the current stimulus, they would confuse 170 6.1.1 Performance measures: Analysis The first type of data analysed comprises the two product-related performance measures selected for the experiment (see §5.5.5.1): terminological accuracy and errors and omissions. 5.5 Main study: Rationale and method 5.5 Main study: Rationale and method the analysis. This would have also been problematic from a practical standpoint. Since the AOIs on the terms had to be placed manually for each individual term and participant, and the screen view could change multiple times due to inter- mediate searches before the stimulus term was mentioned by the speaker, the procedure could have been highly prone to human error, as each video frame would have had to be checked for the presence of stimuli not yet mentioned by the speaker. The issue was even more evident in the CAI tool condition. Due to the pro- gressive search, in InterpretBank the screen view can change numerous times before the user has completed the query. Therefore, it would have been neces- sary to verify which terms were present on the screen at each intermediate view and to check whether they were yet to appear in the speech, requiring an incom- mensurate amount of time. These factors would have made the whole procedure unfeasible and difficult to reproduce in future studies. For this reason, for the PDF and CAI conditions, I decide to activate one AOI at a time, according to the stimulus term currently eliciting a response by the par- ticipant. Despite this necessary simplification, it was nonetheless indispensable to adjust the position of the AOI for each key frame in which the term changed position on the screen, and to ensure that the correct term was being tracked dur- ing the whole typing burst until the query was completed and the term reached its final position on the screen. The use of dynamic AOIs in SMI BeGaze makes it possible to track moving objects on the screen, but nonetheless requires manual adjustments for objects not moving smoothly, which was the case in the present study. It should be noted that the AOIs thus placed on the recording only reflect the cases in which the term searched was found by the tool. In the following chapter, I illustrate the results and discuss them in relation to my hypotheses. 171 6.1 Results The present section illustrates the results of the analysis of the data collected dur- ing the main experiment. The results are presented for each dimension analysed in the main study and explored through the measures introduced in §5.5.5. For an overview of the measures explored in the experiment, the reader may refer to Table 5.1 in §5.3. 6 Results and discussion The present chapter presents and discusses the results of the main experiment conducted following the methodology illustrated in Chapter 5. The results of the pilot study were already presented in §5.4.6. The data analysis is illustrated in §6.1. In §6.2, the results are interpreted and discussed against the hypotheses formulated in §5.2. §6.3 addresses the limitations of the study. 6.1.1.1 Terminological accuracy The first metric analysed was the degree of terminological accuracy achieved for the PDF, the CAI, and the simASR condition. A total of 108 terms (36 for each speech) were evaluated by the author and by an experienced conference interpreter trainer working in the language pair English-German. The evaluation was conducted for all nine participants, for a total of 972 evaluations.i Before conducting the evaluation on the whole data set, I verified the adequacy of the grading scale by calculating Cohen’s 𝜅value on a sample of 22 terms (20% of all stimuli, which is a recommended amount in literature. See Mellinger & Han- son 2017: 326). Cohen’s 𝜅provides an indication of the level of agreement of two raters on the categories used for the evaluation. The maximum value achievable 6 Results and discussion is 1, which indicates perfect agreement, while values below 0.6 are often consid- ered problematic. This measure is more reliable and indicative of agreement than a mere calculation of the percentage of identical evaluations by two raters. The inter-annotator agreement calculated on the sample was substantial at 0.70 (see Landis & Koch 1977). This confirmed the adequacy of the definitions chosen to describe the three categories in the grading scale. The evaluation was then com- pleted for the remaining stimuli and the percentage of terms for each category (0, 1, 2) was calculated for each subject. The percentages thus calculated were averaged between the two raters. The main discrepancies in the coding by the two raters was due to mispronunciations of the target terms, for which the first rater was more benevolent and assigned a value of 2 if the terms used were the ones presented in the glossary, while the second rater was stricter and usually assigned a code of 1 or 0. On the whole sample, the average intercoder agreement for the accuracy evaluation was nonetheless substantial (𝜅= 0.77). 6.1.1.1.1 Accuracy of terms searched In this category, I considered the terminological accuracy achieved when termi- nology equivalents were suggested on screen, either automatically (simASR) or after a glossary query (PDF and CAI). This analysis should provide a first insight into how much the interaction with the tool interferes with the interpreting pro- cess and into how often a glossary query is successful and corresponds with an accurate rendition. If accuracy scores are high, this may provide a first indication that the integration of digital support into the SI process is feasible. This hypothe- sis requires, however, also the examination of errors and omissions co-occurring with searches in the glossary (see §6.1.1.2). In order to determine the effect of the tool on the degree of terminological accuracy achieved, the next category of terms (searched/found) will also be analysed. For each participant and each tool, the bar charts below (Figures 6.1–6.3) illus- trate the distribution of the three levels of terminological accuracy (0, 1, 2) for the terms searched in the glossary. As expected, the best performance was achieved with simASR support: on av- erage, 96.3% of terms received a score of 2, 2.62% a score of 1 and only 1.08% a score of 0. The aggregate mean value for the 2 and 1 scores was 98.92%, which in- dicates on average a very high degree of accuracy in the term renditions. This is to be expected, as for the simASR condition all stimulus terms were also “found” by the simulated tool. 6.1.1.1.1 Accuracy of terms searched After a glossary query, the CAI tool led to a slightly worse performance (86.26% of terms obtained a score of 2 on average), followed by the 174 6.1 Results P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.1: Accuracy scores for terms searched in the PDF condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.1: Accuracy scores for terms searched in the PDF condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.2: Accuracy scores for terms searched in the CAI condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.3: Accuracy scores for terms searched in the simASR condition Figure 6.1: Accuracy scores for terms searched in the PDF condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.2: Accuracy scores for terms searched in the CAI condition Figure 6.2: Accuracy scores for terms searched in the CAI condition Figure 6.2: Accuracy scores for terms searched in the CAI condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Rank 2 Rank 1 Rank 0 Figure 6.3: Accuracy scores for terms searched in the simASR condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Accuracy (%) Figure 6.3: Accuracy scores for terms searched in the simASR condition 175 6 Results and discussion PDF tool (78.00% rated as close renditions, and a notable 16.36% were either un- translated or mistranslated). Overall, the terminological accuracy achieved with digital support was quite high, also in the PDF condition. With reference to within-subject differences, a first inspection of the data showed a more nuanced picture, with cases in which greater accuracy (i.e. a higher percentage of 2 scores for terms) was achieved in the PDF condition than in the CAI condition, and fewer omissions or unacceptable translations were recorded for the PDF glossary than for the CAI tool. 6.1.1.1.1 Accuracy of terms searched I proceeded to verify whether the differences observed on the raw data be- tween the scores achieved with the PDF glossary, the CAI tool, and the ASR-CAI mock-up were significant. Since each participant was tested multiple times on the three conditions, I conducted a Friedman test. Choosing a non-parametric test seems appropriate, as the sample tested was small and the normality of distribution could not be assumed for all conditions. Non-parametric tests were chosen for the analysis also for the other metrics analysed in the study. A Friedman test was conducted on the rank-transformed data for each of the categories of the grading scale. Since each participant worked in three conditions, ranks ranged between 1 (lowest value) and 3 (highest value), with the appropriate half scores in the case of ties.fi For the terms searched, the difference between tools was not significant for the category “acceptable renditions” (score = 1): 𝜒2(2) = 3.182, 𝑝= 0.149. A significant effect of tool was found for the category “close renditions” (score = 2), 𝜒2(2) = 14.89, 𝑝= 0.001, and for the category “zero or unacceptable renditions” (score = 0), 𝜒2(2) = 16.22, 𝑝< 0.001. Post-hoc Wilcoxon signed-ranks tests with a Bonferroni correction were con- ducted for the significant categories, to identify which differences had led to the significant result. More close renditions were observed for the simASR condition (Mdn = 95.83%) than for the CAI condition (Mdn = 86.07%) and for the PDF con- dition (Mdn = 73.91%). This difference was statistically significant between the simASR tool and the PDF glossary (𝑝= 0.005) and between the CAI tool and the simASR tool (𝑝= 0.005). No statistically significant difference was observed between the PDF and the CAI condition (𝑝= 0.096). More 0 scores were assigned to terms searched in the PDF condition (Mdn = 17.19%) than in the CAI condition (Mdn = 7.14%) and in the simASR condition (Mdn = 1.39%). The Wilcoxon signed-ranks tests showed a statistically significant difference between the simASR and the PDF condition (𝑝= 0.004) and between the simASR and the CAI condition (𝑝= 0.004). A statistically significant differ- ence was also observed between the PDF and the CAI condition (𝑝= 0.02). 176 6.1 Results 6.1 Results 6.1.1.1.2 Accuracy of terms searched/found The observations made for the terms looked up were confirmed by the analysis of the accuracy scores for the terms searched/found in the glossary. 6.1.1.1.1 Accuracy of terms searched In this case, the differences between the tools were smaller and the results overall encouraging, as the average percentage of terms rated as close renditions ranged from 91.35% in the case of the PDF glossary to 96.30% for the ASR-CAI mock-up.f As for within-subject differences, the trend observed for the terms queried also emerged at a first inspection of the data for the terms queried and found. Note that for the simASR condition, all terms queried were also “found” (the ASR-CAI mock-up did not contain recognition errors). However, the terms queried in the PDF glossary and with the CAI tool may not be found. Thus, a query may not lead to an accurately translated term. In order to explore this further, I repeated the Friedman test also for the terms searched/found. The test yielded similar results to those obtained for the terms searched. Significant differences between tools were found for the categories “close rendition” (𝜒2(2) = 7.52, 𝑝= 0.023) and “zero/unacceptable rendition” (𝜒2(2) = 14.89, 𝑝= 0.001). Wilcoxon signed-ranks tests with a Bonferroni correction for repeated mea- sures were conducted also for the terms searched/found in the glossary. After a Bonferroni correction (3 comparisons), statistically significant differences were found both for the category of close renditions and for the category of zero/ unacceptable renditions. In the “close renditions” category, a statistically sig- nificant difference was found only between the simASR and the PDF condition (𝑝= 0.026). For the “zero/unacceptable renditions”, a statistically significant dif- ference was found between the CAI and the PDF condition (𝑝= 0.005) and the simASR and the PDF condition (𝑝= 0.005). 6.1.1.2.1 Evaluation framework As for the analysis of the terminological accuracy, the assessment framework for errors and omissions was validated before proceeding with the evaluation. The level of intercoder agreement was calculated on a sample of sentences. The evaluation task included two subtasks: the identification of target and continua- tion sentences that contained a severe error, and the identification of sentences omitted in their entirety. The evaluation was conducted by the same raters as in §6.1.1.1. The sentences to be evaluated were presented in an Excel table. The original sentence and the rendition for each participant were juxtaposed, with an additional column for comments. For the scope of the present study, the partici- pants’ renditions were presented only in written format (but see e.g. Montecchio 2021 for an evaluation including effects on fluency).i The average Cohen’s 𝜅value was calculated for the first sample and was found to show moderate agreement (𝜅= 0.50). The inspection of the ratings assigned by the second evaluator showed that some sentences were marked as erroneous renditions, even though the only mistake consisted in the term used. At first, I had aimed to separate the evaluation of errors in the term renditions from the evaluation of the rest of the sentences. However, I realised that it would have been impossible to judge a sentence as “not erroneous” when the choice of term determined a complete upturning of the sentence meaning. Therefore, the eval- uation task was defined as follows: “Please highlight in yellow the sentences that have been interpreted incor- rectly (= the sentence does not make sense OR its meaning is entirely or largely different when compared to the original)”. This led to a higher number of sentences coded as erroneous, but also to sub- stantial agreement between the raters after the evaluation was repeated on the sample and on the rest of the data set (𝜅= 0.66). The reliability of the yes/no scale for omissions was very high (mean 𝜅in the sample = 0.99), as was to be expected. 6.1.1.2 Errors and omissions The participants’ interpretations were further analysed to explore how the inter- action with the tools interfered with the subjects’ interpreting performance on a global level. In this analysis I considered two types of issues: severe errors and complete omissions. The following sections report the results of a series of tests conducted on the data. I first considered all sentences (without distinguishing between target and continuation sentences) and analysed the number of severe errors and complete omissions occurring after a query. I then explored where the majority of errors and omissions occurred, whether in the target sentence or in the continuation sentence. Finally, I analysed how the tool used influenced the prevalence of errors and omissions after a glossary query. 177 6 Results and discussion 6.1.1.2.1 Evaluation framework The omitted sentences had been left blank in the evaluation table – the task merely consisted in confirming that the content of the sentences had been omitted, and was formulated as follows: “Please high- light in grey the sentences that were omitted completely (this might seem quite straightforward if the field is left blank (“/”), but there might be cases in which you feel that the “T” sentence already provides the information contained in the “C” sentence)”. On the whole data set, the agreement between the two raters was almost per- fect, apart from one case in which the second rater did not code the continuation 178 6.1 Results 6.1 Results sentence as omitted because the sentence of the ST consisted in a reformulation of the preceding target sentence. sentence as omitted because the sentence of the ST consisted in a reformulation of the preceding target sentence. 6.1.1.2.2 Global analysis To start, I calculated the percentage of severe errors and complete omissions coinciding with a stimulus query. The values were calculated for each rater and then averaged. For this first global analysis, I did not consider the effect of the tool on the participants’ performance. Rather, the objective was to explore how much the interaction with the tool resulted in evident issues in the participants’ renditions. In the whole sample, on average, around one sixth of all sentences including a term query led to severe errors (Mdn = 16.91%). Slightly more errors were com- mitted in continuation sentences (Mdn = 16.67%) than in target sentences (Mdn = 13.50%), as shown by Figure 6.4. Figure 6.4: Boxplot of the percentage of errors in target and contin- uation sentences after a query. On average, more severe errors were committed in continuation sentences. Figure 6.4: Boxplot of the percentage of errors in target and contin- uation sentences after a query. On average, more severe errors were committed in continuation sentences. As for the number of omissions, a glossary query resulted in a complete omis- sion of either the target or the continuation sentence (or both) in a limited num- ber of cases (Mdn = 2.66%). More continuation sentences were omitted than tar- get sentences (Mdn = 5.32 % vs. 1.05%, see Figure 6.5). 179 6 Results and discussion Figure 6.5: Boxplot of the percentage of omissions in target and con- tinuation sentences after a query. On average, more continuation sen- tences were omitted completely. Figure 6.5: Boxplot of the percentage of omissions in target and con- tinuation sentences after a query. On average, more continuation sen- tences were omitted completely. Despite the apparent differences observed when considering the sample aver- age, the bar plot of the data (Figure 6.6) showing each participant’s performance did not suggest a common trend as to whether more errors were committed in target or continuation sentences. To confirm the lack of difference, I conducted a Mann-Whitney 𝑈test. As expected, the difference was not statistically signifi- cant (𝑈= 34.5, 𝑧= −0.53, 𝑝= 0.59). The bar plot did however show that all participants omitted more continua- tion sentences than target sentences. I therefore proceeded to verify whether the difference was significant. A Mann-Whitney U test confirmed that the difference was statistically significant (𝑈= 2.00, 𝑧= −3.41, 𝑝= 0.001). 6.1.1.2.3 Effect of tool on errors and omissions: Terms searched To verify how the individual tools affected the occurrence of severe errors and of complete omissions, I first analysed all cases in which a term query had been per- formed. As remarked for the analysis of the terminological accuracy (see §6.1.1.1), for the simASR condition, all terms searched were also “found”. As shown by Figure 6.7, fewer severe errors were committed under the simASR condition (Mdn = 11.11%). The worst performance was obtained with the PDF tool (Mdn = 19.85%), while in the CAI condition a glossary query yielded more severe errors than in the simASR condition, but less than in the PDF condition (Mdn = 15.28%). 180 6.1 Results P1 P2 P3 P4 P5 P6 P7 P8 P9 0 10 20 Participant Errors – targent sentence Errors – continuation sentence Omissions – target sentence Omissions – continuation sentence Figure 6.6: Percentage of errors and omissions in target and continua- tion sentences (Terms searched, mean value per participant) Figure 6.6: Percentage of errors and omissions in target and continua- tion sentences (Terms searched, mean value per participant) Figure 6.7: Mean percentage of errors for terms searched (Terms searched, mean value by tool) Figure 6.7: Mean percentage of errors for terms searched (Terms searched, mean value by tool) 181 6 Results and discussion Figure 6.8: Mean percentage of omissions for terms searched (Terms searched, mean value by tool) Figure 6.8: Mean percentage of omissions for terms searched (Terms searched, mean value by tool) With reference to the number of sentences omitted completely due to the inter- action with the tool (see Figure 6.8), the simASR yielded the best results (Mdn = 0.00% omitted sentences). Queries with the PDF glossary resulted in 4.55% (Mdn) omitted sentences, while the CAI tool led to better results (Mdn = 2.00%). The bar plots in Figures 6.9 and 6.10 show the differences for each participant both for severe errors and completely omitted sentences.fi To test whether the observed differences were significant, I conducted non- parametric Friedman’s tests for the errors and omissions data. A non-parametric test was used since the omissions data were not normally distributed for the simASR condition and due to the limited sample size.iff A significant difference was found for the effect of tools on the percentage of errors in the sentences for which a glossary query had been conducted (𝜒2(2) = 8.00, 𝑝= 0.018). 6.1.1.2.3 Effect of tool on errors and omissions: Terms searched The post-hoc Wilcoxon signed-ranks tests with a Bonferroni correction showed a statistically significant difference only between the simASR and the PDF condition (𝑝= 0.014). A statistically significant difference between tools was also found for the per- centage of omitted sentences after a glossary query (𝜒2(2) = 8.97, 𝑝= 0.011). The post-hoc Wilcoxon signed-ranks tests with a Bonferroni correction showed a significant difference between the simASR and the PDF condition (𝑝= 0.010). 182 6.1 Results P1 P2 P3 P4 P5 P6 P7 P8 P9 0 10 20 30 40 Participant Severe errors (%) PDF CAI simASR Figure 6.9: Mean percentage of errors for terms searched (Terms searched, mean value by tool per participant) P1 P2 P3 P4 P5 P6 P7 P8 P9 0 10 20 30 40 Participant Severe errors (%) PDF CAI simASR PDF CAI simASR Figure 6.9: Mean percentage of errors for terms searched (Terms searched, mean value by tool per participant) P1 P2 P3 P4 P5 P6 P7 P8 P9 0 5 10 Participant Complete omissions (%) PDF CAI simASR Figure 6.10: Mean percentage of omissions for terms searched (Terms searched, mean value by tool per participant) P1 P2 P3 P4 P5 P6 P7 P8 P9 0 5 10 Participant Complete omissions (%) PDF CAI simASR Figure 6.10: Mean percentage of omissions for terms searched (Terms searched, mean value by tool per participant) 183 6 Results and discussion 6.1.1.2.4 Effect of tool on errors and omissions: Terms searched/found 1.1.2.4 Effect of tool on errors and omissions: Terms searched/found To further the analysis, I explored whether there were differences between the tools in the percentage of severe errors and complete omissions after successful queries (terms searched/found). Figure 6.11: Mean percentage of errors for terms searched/found (Terms searched, mean value by tool) Figure 6.11: Mean percentage of errors for terms searched/found (Terms searched, mean value by tool) For terms searched/found, the simASR results are the same as the ones re- ported in §6.1.1.2.3. As shown by Figure 6.11, slightly more severe errors were committed under the CAI condition (Mdn = 13.97%) than under the PDF condi- tion (Mdn = 13.39%). 6.1.2 Behavioural measures: Analysis In addition to the two performance measures discussed in the previous sections, six behavioural measures were analysed. In this section, I report and discuss the results of the tests conducted on the glossary queries, on the two time-lag mea- sures of EVS and ICPD, and on three gaze-related measures: time to first fixation, average fixation duration, and total fixation time. 6.1.1.2.3 Effect of tool on errors and omissions: Terms searched As for the number of sentences omitted completely despite a successful term query (Figure 6.12), the CAI tool yielded better results (Mdn = 1.92% omitted sentences) as compared to the PDF condition (3.70%).f The bar plots in Figures 6.13 and 6.14 show the differences for each participant both for severe errors and completely omitted sentences when the terms were searched/found in the glossaries. As for the previous analyses on the data for the terms searched/found, I con- ducted two Friedman tests (for the errors and the omissions) to test for signifi- cance. 184 6.1 Results Figure 6.12: Mean percentage of omissions for terms searched/found (Terms searched, mean value by tool) Figure 6.12: Mean percentage of omissions for terms searched/found (Terms searched, mean value by tool) Figure 6.12: Mean percentage of omissions for terms searched/found (Terms searched, mean value by tool) P1 P2 P3 P4 P5 P6 P7 P8 P9 0 10 20 30 40 Participant Severe errors (%) PDF CAI simASR Figure 6.13: Mean percentage of errors for terms searched/found (Mean value by tool per participant) P1 P2 P3 P4 P5 P6 P7 P8 P9 0 10 20 30 40 Participant Severe errors (%) PDF CAI simASR Figure 6.13: Mean percentage of errors for terms searched/found (Mean value by tool per participant) 185 6 Results and discussion P1 P2 P3 P4 P5 P6 P7 P8 P9 0 5 10 Participant Complete omissions (%) PDF CAI simASR Figure 6.14: Mean percentage of omissions for terms searched/found (Mean value by tool per participant) Figure 6.14: Mean percentage of omissions for terms searched/found (Mean value by tool per participant) A significant difference was found for the effect of tools on the percentage of errors in the sentences for which a glossary query had been conducted (𝜒2(2) = 6.88, 𝑝= 0.032). The post-hoc Wilcoxon signed-ranks tests with a Bonferroni correction showed a significant difference between the simASR and the PDF con- dition (𝑝= 0.029). A statistically significant difference between tools was also found for the per- centage of omitted sentences after a glossary query (𝜒2(2) = 8.47, 𝑝= 0.014). The post-hoc Wilcoxon signed-ranks tests with a Bonferroni correction showed a significant difference between the simASR and the PDF condition (𝑝= 0.014). 6.1.2.1 Glossary queries In order to gain a picture of the test subjects’ interactions with the tools during SI, I collected data pertaining to the queries performed in the PDF glossary and in InterpretBank. For the CAI-ASR mock-up, no analysis was conducted, as in this condition no action was required from the participants. 186 6.1 Results In addition, the analysis of the glossary queries was used to test how the two criteria followed in the speech design, i.e. stimulus position in the sentence and morphological complexity (see §5.5.3), affected the participants’ search be- haviour. In this case, no distinction was made between the tools, as my goal was to verify whether my assumptions about potential effects of position or morpho- logical complexity (see §5.5.5.2) would be supported by the analysis. Based on a methodology already adopted in previous studies (Prandi 2015a,b, 2017, 2018), I therefore considered the following categories of terms: • Terms searched by tool; • Terms searched/found by tool; • Terms searched by position (sentence-medial and sentence-final); • Terms searched/found by position (sentence-medial and sentence-final); • Terms searched by morphological complexity (unigrams, bigrams, tri- grams); • Terms searched by morphological complexity (unigrams, bigrams, tri- grams); • Terms searched/found by morphological complexity (unigrams, bigrams, trigrams). • Terms searched/found by morphological complexity (unigrams, bigrams, trigrams). 6.1.2.1.1 Terms searched This category represents the percentage of terms searched with each tool. Each speech contained 36 terms. For the CAI-ASR integration, the percentage was al- ways 100%, since the terms were all automatically shown on the screen. For this category, it was therefore interesting to verify whether there was a difference in the percentage of terms looked up with the PDF glossary and with Interpret- Bank, since for both conditions participants were required to choose whether they wanted to conduct a glossary query or not. The simASR condition was there- fore not considered in this analysis. When considering the whole sample, slightly more terms were searched when working with the PDF glossary (Mdn = 88.89%) than with InterpretBank (Mdn = 83.33%), precisely in four cases out of nine. One participant looked up the same number of terms with both tools (83.33%). In another case (P7), all 36 terms were searched also when working with the CAI tool (see Figure 6.15). A Wilcoxon signed-ranks test was conducted to test for significance. As ex- pected, no statistically significant difference was found in the percentage of terms searched with the PDF and with the CAI tool (𝑇= 20.5, 𝑝= 0.67). 187 6 Results and discussion P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Queries (%) PDF CAI simASR Figure 6.15: Percentage of terms searched in the PDF, CAI and simASR condition for each participant. The value is always 100% for the simASR condition. P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Queries (%) PDF CAI simASR Fi 6 15 P t f t h d i th PDF CAI d i ASR Figure 6.15: Percentage of terms searched in the PDF, CAI and simASR condition for each participant. The value is always 100% for the simASR condition. 6.1.2.1.3 Terms searched by position In the speeches prepared for the experiment, the stimuli were distributed equally between terms in the middle of the sentence and terms at the end of the sentence. In the speeches prepared for the experiment, the stimuli were distributed equally between terms in the middle of the sentence and terms at the end of the sentence.i At a first visual inspection of the data per participant, I did not identify a clear trend supporting either of the hypotheses formulated in §5.2, as in some cases the terms at the end of the sentence were looked up more often, while in others the opposite was true. On average, however, the participants looked up more sentence-final terms (Mdn = 52.24%) than sentence-medial terms (Mdn = 47.76%), as shown by Fig- ure 6.17. A Mann-Whitney U test showed a statistically significant difference between the two conditions (𝑈= 5.00, 𝑧= −3.14, 𝑝= 0.002). 6.1.2.1.2 Terms searched/found One of the main indicators of how useful the tools prove as terminology support in the booth is the percentage of terms searched that were actually found by the tool. For this analysis, I did not consider whether the participant found the term in the glossary, but rather whether the term was visualised on the screen after having been looked up. It is possible that a term found by the tool may nonetheless be left untranslated or mistranslated if the participant was unable to localise it on the monitor. Also in this case, the percentage is of course 100% for the ASR-CAI mock-up for all participants, so I only considered the PDF and the CAI tool in the analysis. After a visual inspection of the data (see Figure 6.16), when comparing the percentage of terms searched/found by each tool, the CAI tool seemed to perform better than the PDF glossary for each participant. I therefore decided to conduct further tests to verify whether the observed difference was significant. I conducted a Wilcoxon signed-ranks test to verify whether the observed difference was statistically significant. A non-parametric test was selected due to the small sample size. The results show a significant difference (𝑇= 45.00, 𝑝= 0.008) between the CAI (Mdn = 96.88% successful queries) and the PDF condition (Mdn = 79.41% successful queries). 188 6.1 Results P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Queries (%) PDF CAI simASR Figure 6.16: Percentage of terms searched/found in the PDF, CAI and simASR condition for each participant. The value is always 100% for the simASR condition. P1 P2 P3 P4 P5 P6 P7 P8 P9 0 20 40 60 80 100 Participant Queries (%) PDF CAI simASR Figure 6 16: Percentage of terms searched/found in the PDF CAI and Figure 6.16: Percentage of terms searched/found in the PDF, CAI and simASR condition for each participant. The value is always 100% for the simASR condition. 6.1.2.1.5 Terms searched by morphological complexity I then considered whether the morphological complexity of the terms used as stimuli (see §5.5.3) may have had an impact on the participants’ search behaviour.i The first category of terms analysed were the terms searched. The participants looked up more unigrams (Mdn = 34.33%) than bigrams and trigrams. More tri- grams (Mdn = 33.33%) were looked up than bigrams (Mdn = 32.84%). However, the differences were small, as shown by the plotted data (Figure 6.19). Figure 6.19: Percentage of terms searched according to their morpho- logical complexity. Approximately the same percentage of unigrams, bigrams and trigrams were looked up. Figure 6.19: Percentage of terms searched according to their morpho- logical complexity. Approximately the same percentage of unigrams, bigrams and trigrams were looked up. A Kruskall-Wallis test revealed that the difference was not statistically signif- icant (𝐻(2) = 2.41, 𝑝= 0.30). In other words, the morphological complexity of the terms did not affect the number of terms looked up. 6.1.2.1.4 Terms searched/found by position In order to further explore whether the position of the terms may have affected the success rate of terminology look-up, I calculated the percentage of terms searched/found in each of the two positions (sentence-medial and sentence-final) for each participant. No clear trend could be identified between the PDF and the CAI tool after a visual inspection of the data.i For the participants, it was overall easier to find terms when they were placed at the end of the sentence (Mdn = 53.42%, sentence-medial terms: Mdn = 46.58%), as shown by Figure 6.18. 189 6 Results and discussion Figure 6.17: Percentage of terms searched according to their position in the sentence. More sentence-final terms were looked up than sentence- medial terms. Figure 6.18: Percentage of terms searched/found according to their po- sition in the sentence. More sentence-final terms looked up were found than sentence-medial terms. 190 Figure 6.17: Percentage of terms searched according to their position in the sentence. More sentence-final terms were looked up than sentence- medial terms. Figure 6.17: Percentage of terms searched according to their position in the sentence. More sentence-final terms were looked up than sentence- medial terms. Figure 6.17: Percentage of terms searched according to their position in the sentence. More sentence-final terms were looked up than sentence- medial terms. Figure 6.18: Percentage of terms searched/found according to their po- sition in the sentence. More sentence-final terms looked up were found than sentence-medial terms. Figure 6.18: Percentage of terms searched/found according to their po- sition in the sentence. More sentence-final terms looked up were found than sentence-medial terms. Figure 6.18: Percentage of terms searched/found according to their po- sition in the sentence. More sentence-final terms looked up were found than sentence-medial terms. 190 6.1 Results A Mann-Whitney U test showed a statistically significant difference between the sentence-medial and the sentence-final terms (𝑈= 7.50, 𝑧= −2.915, 𝑝= 0.004). 6.1.2.1.6 Terms searched/found by morphological complexity In order to further explore my hypotheses on how term structure may affect the search behaviour and the success rate of the queries performed, I consider the In order to further explore my hypotheses on how term structure may affect the search behaviour and the success rate of the queries performed, I consider the 191 6 Results and discussion percentage of uni-, bi- and trigrams searched/found with the PDF and the CAI tool, while I excluded the simASR condition for which the distribution was equal between the three categories. In the sample, trigrams were the category of terms looked up that were found more often (Mdn = 36.21%), followed by bigrams (Mdn = 31.91%) and unigrams (Mdn = 31.03%), as shown by the boxplot in Figure 6.20. Figure 6.20: Percentage of terms searched/found according to their morphological complexity. Significantly more trigrams were found than unigrams. Figure 6.20: Percentage of terms searched/found according to their morphological complexity. Significantly more trigrams were found than unigrams. Figure 6.20: Percentage of terms searched/found according to their morphological complexity. Significantly more trigrams were found than unigrams. I tested whether the observed difference was statistically significant using a Kruskal-Wallis non-parametric test as I had done for the terms searched. The difference was statistically significant (𝐻(2) = 11.85, 𝑝= 0.003). Post-hoc Mann- Whitney tests with a Bonferroni correction revealed a statistically significant difference only between the number of unigrams and trigrams found in the glos- sary (𝑝= 0.002). It appears that it was easier to find trigrams than it was to find unigrams. 6.1.2.1.7 Additional observations A series of additional phenomena could be identified while reviewing the Gaze Replays for data preparation. Given the small sample size, it is possible that these phenomena are isolated occurrences, but I report them nonetheless as they may 192 6.1 Results 6.1 Results be useful in formulating additional hypotheses on SI with digital terminology support to be tested in future studies. In four cases out of nine, a handful of terms (nine in total) were shown on the screen when the PDF and/or the CAI tool was used, even though they had not been looked up by the participants. In most cases, they were left untranslated or mistranslated, probably because they had not been seen. In 3 cases, the glossary equivalent was used, which would indicate that they had been seen by the par- ticipants. Even though this was a very rare occurrence during the experiment, it is interesting to notice that this could represent an additional advantage of using digital tools for terminology look-up. A dedicated experiment would be necessary to explore this aspect further. When searching for terms, several participants could not find the term they were looking for as they did not know its spelling. This was shown by the repe- tition of the queries with different spellings of the same term. In some cases, the mistakes were quite naive and with all probability due to the participants having misheard the term. An experienced conference interpreter would probably have recognised that the term pronounced by the speaker was “fission”, and not “fish- ing”, especially in the context of a speech about nuclear energy. For other terms, such as “boule”, which may have a complex spelling for non-native speakers of English, the difficulty is more understandable. This phenomenon further stresses the importance of preparation for effective terminology search during interpret- ing, as well as the role of CAI tools as digital aids and not as a replacement for preparation strategies. Additionally, I observed several instances of terms queried which I had not selected as stimuli, such as “wheat”, “power plant” and “coal”. Nonetheless, they apparently represented additional difficulties for some participants. This might explain errors or omissions and in general the breakdown of the interpreting process even in sentences where no stimulus query had been performed. 6.1.2.2 Ear-voice span For the PDF and CAI conditions, I only considered the terms which had been looked up and translated. The cases in which a search resulted in an omission of the term were not considered, as no relative time stamp was available. For the simASR condition, all terms were taken into consideration, as there were no cases of zero rendition for any participant in this condition. As pointed out by Timarová et al. (2011), the EVS can vary considerably during an interpreting session and the average value is susceptible to minimum and max- imum values. In my sample, I observed cases of very long EVS when compared 193 6 Results and discussion to other values registered for the same participant. Therefore, for this analysis I report not only the average values, but also the median values, as suggested in Timarová et al. (2011). The mean and median EVS varied considerably between subjects, which was to be expected as it can reflect different cognitive make-up, interpreting styles, strategies and reaction times. Nonetheless, all participants showed the same pat- tern, both for the mean and the median values: the EVS was consistently shorter when the simASR support was provided, followed by the CAI tool and the PDF (see Table 6.1). The difference appeared even more evident at the analysis of me- dian instead of mean values. To confirm whether the differences observed at a first inspection of the data were statistically significant, I proceeded with a series of tests. Data were not nor- mally distributed as one participant had longer EVS than all other participants. Therefore, I opted for a non-parametric Friedman test on the rank-transformed data for the terms searched and searched/found in the PDF, CAI and simASR condition. Since the ranking was the same both for mean and median values and for the terms searched and searched/found, as the same pattern was observed for all participants, I conducted only one test. The differences were statistically significant (𝜒2(2) = 18.00, 𝑝< 0.001). Pair- wise comparisons were conducted post-hoc as Wilcoxon signed-rank tests with a Bonferroni correction. They confirmed that the difference in EVS length be- tween each tool and the other two was statistically significant (𝑝= 0.009 for all comparisons). 6.1.2.2 Ear-voice span Considering the median values (see Figures 6.21, 6.22), which in this case seems more logical due to the higher values for P2 which bias the mean upwards, the average sample value seems to indicate that participants could gain 1.7s if they were working with the CAI tool as compared to a PDF glossary, and 1.2s when they used the ASR-CAI mock-up as compared to a standard CAI tool. 6.1.2.3 Inter-cluster pause duration Terms searched Ptcpt PDF CAI simASR Mean Median Mean Median Mean Median 1 5277 4339 3789 3800 2153 2045 2 7919 7210 6845 6560 3794 3089 3 5563 5068 4264 4000 2378 2445 4 5574 5290 3214 3003 1681 1567 5 5432 5495 5075 4168 2516 2065 6 5085 4690 3832 3300 1991 1875 7 4835 4200 3403 2719 2810 2505 8 5877 5070 2742 2605 1810 1490 9 4447 4210 3224 2860 1833 1650 Terms searched/found Ptcpt PDF CAI simASR Mean Median Mean Median Mean Median 1 5336 4351 3662 3782 2153 2045 2 7390 6950 6650 6350 3794 3089 3 5906 5114 4223 3886 2378 2445 4 5532 5270 3108 2982 1681 1567 5 5304 5208 4868 4166 2516 2065 6 5237 4860 3705 3250 1991 1875 7 4835 4200 3403 2719 2810 2505 8 5876 4785 2742 2605 1810 1490 9 4483 4210 3224 2860 1833 1650 195 6 Results and discussion Figure 6.21: Median EVS for terms searched Figure 6.22: Median EVS for terms searched/found 196 6 Results and discussion Figure 6.21: Median EVS for terms searched Figure 6.21: Median EVS for terms searched Figure 6.21: Median EVS for terms searched Figure 6.22: Median EVS for terms searched/found Figure 6.22: Median EVS for terms searched/found Figure 6.22: Median EVS for terms searched/found 196 6.1 Results Figure 6.23: Boxplot of median ICPD (s) for each condition after a glos- sary query. A longer ICPD suggests faster processing of the preceding sentence cluster. Figure 6.23: Boxplot of median ICPD (s) for each condition after a glos- sary query. A longer ICPD suggests faster processing of the preceding sentence cluster. As I did for EVS, I also considered the median values of the ICPD. A Friedman test was conducted also on the median values and yielded similar results (𝜒2(2) = 11.56, 𝑝= 0.003). As I did for EVS, I also considered the median values of the ICPD. A Friedman test was conducted also on the median values and yielded similar results (𝜒2(2) = 11.56, 𝑝= 0.003). By inspecting the data, I noticed that there were cases in which the average pause duration for the CAI tool was shorter than for the PDF, which I found quite surprising. With respect to the mean values, in one case the pause length was longer for the CAI tool than for the ASR mock-up (P4). 6.1.2.3 Inter-cluster pause duration In calculating the mean and median ICPD (see Figure 6.23), I only considered the cases in which a search had been performed in the tool, while I omitted the cases in which the continuation sentence had been completely left out, as this would have shown a very long ICPD actually due to an omission. Since the same group was tested multiple times on the three conditions, I ran a Friedman test to verify whether the differences in the ICPD were statistically significant. The Friedman test confirmed that the difference was statistically sig- nificant (𝜒2(2) = 14.00, 𝑝= 0.001). 194 6.1 Results 6.1 Results Table 6.1: Mean and median EVS values (ms) for each condition, both for terms searched and for terms searched/found. The values are the same as for the simASR condition, as all terms were displayed on the screen. Table 6.1: Mean and median EVS values (ms) for each condition, both for terms searched and for terms searched/found. The values are the same as for the simASR condition, as all terms were displayed on the screen. 1As in the previous cases, the values for the simASR condition are the same. 6.1.2.3 Inter-cluster pause duration I therefore also con- ducted post-hoc pairwise comparisons between the three conditions, both for the mean and the median values. The Wilcoxon signed-ranks tests with a Bon- ferroni correction for the mean values confirmed that the simASR ICPD was significantly longer (Mdn = 4.52s) when compared to the PDF (Mdn = 3.13s) con- dition (𝑝= 0.001) or to the CAI (Mdn = 3.89s) condition (𝑝= 0.014), but the same was not true for the other two conditions (𝑝= 1.00).f If I consider the median values, a slightly different picture emerges. The dif- ferences in median ICPD were statistically significant only for the simASR tool (Mdn = 4.60s) compared to the PDF (Mdn = 3.15s) condition (𝑝= 0.003). There was no statistically significant difference between the PDF and the CAI condition (𝑝= 1.00) and between the CAI and simASR condition (𝑝= 0.055). The median ICPD value for the CAI condition was 3.75s. 197 6.1.2.4 Time to first fixation To explore how fast the participants could identify the terms on the screen, I calculated the mean time to first fixation for the term AOIs for each participant under each condition. As for the previous metrics, I conducted a non-parametric Friedman test to establish whether the differences between the three conditions were statistically significant. The test showed a statistically significant difference (𝜒2(2) = 13.56, 𝑝= 0.001). Post-hoc Wilcoxon signed-rank tests with a Bonferroni correction showed statistically significant differences between the simASR and the PDF condition (𝑝= 0.018) as well as between the ASR and CAI condition (𝑝= 0.018). However, the difference between the CAI and the PDF condition (𝑝= 0.739) was not significant. 6 Results and discussion 6 Results and discussion The same procedure was also followed for the pause durations between sen- tence clusters when terms had been found in the glossary1, considering both the mean and the median values. Non-parametric tests were used also in this case due to the small sample size. A Friedman test returned significant differ- ences for the mean values (𝜒2(2) = 14.89, 𝑝= 0.001) and for the median values (𝜒2(2) = 11.56, 𝑝= 0.003). Post-hoc pairwise comparisons with Wilcoxon signed-ranks test with a Bon- ferroni correction yielded similar results to what had been observed for the searched terms. With concern to the mean values, the ICPD for the simASR condi- tion (Mdn = 4.52s) were significantly longer (𝑝< 0.001) than in the PDF condition (Mdn = 3.21s) and in the CAI condition (𝑝= 0.029; Mdn CAI = 3.91s). As for the median values (Figure 6.24), the simASR ICPD (Mdn = 4.60) was also significantly longer only as compared to the PDF condition (𝑝= 0.003; Mdn PDF = 3.33s). The other two contrasts did not show statistically significant differences. The median ICPD value for the CAI condition was 3.72s. Figure 6.24: Boxplot of median ICPD (s) for each condition after a suc- cessful glossary query. A longer ICPD suggests faster processing of the preceding sentence cluster. Figure 6.24: Boxplot of median ICPD (s) for each condition after a suc- cessful glossary query. A longer ICPD suggests faster processing of the preceding sentence cluster. Figure 6.24: Boxplot of median ICPD (s) for each condition after a suc- cessful glossary query. A longer ICPD suggests faster processing of the preceding sentence cluster. Figure 6.24: Boxplot of median ICPD (s) for each condition after a suc- cessful glossary query. A longer ICPD suggests faster processing of the preceding sentence cluster. 198 6.1 Results 6.1 Results 6.1.2.5 Average fixation duration In each condition and for each participant, I first calculated the average fixation duration for the term AOIs. P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 Participant FixDur (ms) PDF CAI ASR Figure 6.25: Mean fixation duration (ms) on the term AOIs for each participant and each tool FixDur (ms) Figure 6.25: Mean fixation duration (ms) on the term AOIs for each participant and each tool When data was available for both eyes, I averaged the mean values. After plot- ting the data (Figures 6.25 and 6.26), I noticed that the average fixation duration in the simASR condition seemed shorter than in the other two conditions for all participants. In some cases, shorter fixations were observed for the CAI condi- tion and in others for the PDF condition. I therefore decided to verify whether the observed differences were statistically significant. 199 6 Results and discussion 6 Results and discussion I used a Friedman test to test for significance. The test showed a significant difference between the three conditions (𝜒2(2) = 13.56, 𝑝< 0.001). Post-hoc pairwise contrasts with a Bonferroni correction for multiple comparisons indi- cate a significant difference between the simASR condition (Mdn = 250.04) and the PDF (Mdn = 441.51) condition (𝑝< 0.003). A statistically significant difference (𝑝= 0.007) was also observed between the simASR and the CAI condition (Mdn = 383.20). No statistically significant difference was found between the PDF and the CAI condition (𝑝= 1.00). Figure 6.26: Mean fixation duration (ms) on the term AOIs for each tool Figure 6.26: Mean fixation duration (ms) on the term AOIs for each tool In addition to the average fixation durations on the term AOIs, I also consid- ered the duration of the fixations on the tool area (Figures 6.27 and 6.28).i In addition to the average fixation durations on the term AOIs, I also consid- ered the duration of the fixations on the tool area (Figures 6.27 and 6.28).i I computed the average duration of fixations on the tool AOIs for each partic- ipant and conducted a Friedman test to verify the statistical significance of the differences observed on the plotted data. The test showed a statistically signif- icant difference in the average fixation durations on the tool area between the three conditions (𝜒2(2) = 18.00, 𝑝< 0.001). Pairwise comparisons conducted post-hoc showed a statistically significant difference (𝑝< 0.001) for the simASR condition (Mdn = 247.35) as compared to the PDF condition (Mdn = 380.85). No statistically significant difference was found for the other contrasts (Mdn CAI = 314.85). 200 6.1 Results P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 Participant FixDur (ms) PDF CAI simASR Figure 6.27: Mean fixation duration (ms) on the tool AOIs for each par- ticipant and each tool P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 Participant FixDur (ms) PDF CAI simASRi Figure 6.27: Mean fixation duration (ms) on the tool AOIs for each par- ticipant and each tool Figure 6.28: Mean fixation duration (ms) on the tool AOIs for each tool Figure 6.28: Mean fixation duration (ms) on the tool AOIs for each tool 201 6 Results and discussion 6.1.2.6 Fixation time The last gaze-related metric considered was the fixation time on the tool AOI and on the speaker AOI under the PDF, the CAI, and the simASR condition. The last gaze-related metric considered was the fixation time on the tool AOI and on the speaker AOI under the PDF, the CAI, and the simASR condition. The last gaze-related metric considered was the fixation time on the tool AOI and on the speaker AOI under the PDF, the CAI, and the simASR condition. P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) Speaker Tool (a) PDF P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) (b) CAI P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) (c) simASR Figure 6.29: Fixation time (s) on the speaker and tool area for each condition P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) Speaker Tool (a) PDF P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) Participant (b) CAI P1 P2 P3 P4 P5 P6 P7 P8 P9 0 200 400 600 800 Participant Fixation time (s) (c) simASR (c) simASR Figure 6.29: Fixation time (s) on the speaker and tool area for each condition 202 6.1 Results 6.1 Results The aim was to verify whether the participants spent a significantly different amount of time fixating the tool AOI as compared to the tool AOI under each condition. The total fixation time was measured in seconds. I conducted a Mann-Whitney U test for each condition to verify my hypothesis that the effect of the tool would be greater for the simASR condition, i.e. that when working with the ASR-CAI mock-up, the participants spent significantly more time looking at the speaker than at the tool. For the simASR condition, the difference in the time spent fixating the speaker AOI (Mdn = 503.00) and the tool AOI (Mdn = 118.00) was significant (𝑈= 1.00, 𝑧= −3.49, 𝑝< 0.001). A statistically significant difference was also found for the CAI tool (𝑈= 1.00, 𝑧= −3.49, 𝑝< 0.001) between the speaker (Mdn = 431.00) and the tool AOI (Mdn = 143.00). 6.1.2.6 Fixation time For the PDF condition, the median values for the speaker and tool AOIs were respectively 403.00s and 245.00s. The difference was not significant (𝑈= 20.00, 𝑝= 0.070). 6.1.3 Subjective measure: The debriefing questionnaire As observed by one participant, this also provides the additional advantage of being able to identify a term in the glossary even though the spelling is unknown (provided that at least the first letter or sequence of letters has been typed cor- rectly). Similar commentaries were also present in the responses (2 mentions) emphasising the usefulness of the alphabetical order, which facilitated a manual search whenever the spelling was unknown. The program highlighted the first occurrence for the string searched. This was also considered as a useful visual aid in identifying the searched term. An addi- tional helpful feature was identified in the ability of the program to search partial words and not only complete terms, which reduced the amount of time necessary for a search. In one case, no helpful features were mentioned, while in another case a participant laconically commented that “vocab[ulary]” was there. As for the negative aspects of working with a PDF glossary, some comments reflect the other side of the coin of the features mentioned under “most useful”. Very often (6 commentaries), reference is made to the need to navigate through the glossary if the first occurrence does not correspond to the term searched. This is a consequence of the fact that the results are not filtered out of the glossary, but rather highlighted in the body of the document. As stressed by one participant, this may prove distracting. The way the search occurs within the document was also object of stark crit- icism. Here the main issues were identified in the need to clear the search field after each search, which may cause participants to lose precious time, and to click the enter key to start a search. Typing errors would also require a new search, as mentioned by two participants. These difficulties were summed up in the general commentary that “looking up words is cumbersome” in this modality. Specular to this feedback is the description of the helpful and hindering aspects of the booth function of the CAI tool InterpretBank. The dynamic search, which shows first results already while typing and does not require to actively start the query, was mentioned as a positive feature in seven cases. The main positive con- sequence is that it is “not very time consuming”. 6.1.3 Subjective measure: The debriefing questionnaire After data collection, participants were asked to answer a brief questionnaire de- signed to collect additional qualitative data to help frame the quantitative analy- sis. To start, they were asked to rank the speeches from the easiest to the most difficult. Overall, the coal industry speech (speech C) was considered slightly more difficult, while the transmutation speech (B) was considered slightly easier, as can be seen in Figure 6.30. However, in order to exclude speech difficulty effects, the order of the speeches and of the tools had been randomised. Therefore, I can exclude that the content of the speeches may have had a significant effect on the interpreting performance and on the interaction with the tools. Additionally, participants were asked to rank the tools from the most useful to the least useful and from the most distracting to the least distracting. As can be seen in Figure 6.31, the ASR-CAI mock-up was rated as the most useful and least distracting tool. The PDF glossary was considered the least useful and most distracting. As for the tool that participants would bring with them into the booth during a future assignment, most of them selected the ASR tool, two preferred the CAI tool and none indicated the PDF glossary as their preferred type of support. 203 esults and discussion Speech A: Biocrude Speech B: Transmutation Speech C: Coal industry 1 1.5 2 2.5 3 Difficulty (easy > difficult) Figure 6.30: Bar plot of the participants’ ratings of the perceived diffi- culty of the three speeches used in the main experiment (mean sample values) 6 Results and discussion Figure 6.30: Bar plot of the participants’ ratings of the perceived diffi- culty of the three speeches used in the main experiment (mean sample values) PDF CAI simASR 1 1.5 2 2.5 3 (a) Distraction potential PDF CAI simASR 1 1.5 2 2.5 3 (b) Usefulness Figure 6.31: Perceived distraction potential and usefulness of tools (mean sample values) PDF CAI simASR 1 1.5 2 2.5 3 (b) Usefulness Figure 6.31: Perceived distraction potential and usefulness of tools (mean sample values) 204 6.1 Results 6.1 Results In Table 6.2, for each tool I report the characteristics identified as more use- ful and the key issues mentioned by the participants in descending order of fre- quency. The most useful aspect of working with the PDF glossary was indicated as the possibility to see many entries at once. 6.1.3 Subjective measure: The debriefing questionnaire The same was also mentioned in relation to the fact that the search field is automatically cleared “after a few seconds”, which makes it easy to “proceed to [the] next search”. The ability of the tool to show fuzzy matches was also mentioned as a helpful characteristic. 205 6 Results and discussion 6 Results and discussion Table 6.2: Most useful and most problematic aspects mentioned by par- ticipants in relation to the three conditions (PDF glossary, CAI tool, ASR-CAI mock-up). The number of mentions is indicated in brackets. Participants could indicate more than one aspect. Tool Most useful Most problematic PDF Overview of terminology (4) Distraction due to navigation in the file (6) Alphabetical order (2) Search field not cleared automatically - time loss (5) Highlighting of results (2) Traditional type of search (4) Search for parts of words (1) Cumbersome (1) Terminology is available (1) Nothing (1) CAI Progressive search (7) Distraction due to typing while speaking (4) Search field cleared automatically (4) No correction of wrong spelling (2) Fuzzy search (2) Too many/too few terms (1) Multiple results per query shown (2) No assistance if term missed or misheard (1) Speed (1) Nothing (1) Easier to use than PDF (1) ASR No distraction due to typing (6) Distraction / frustration if expected term not shown (7) Low latency (3) Inability to search for terms not shown automatically (3) Easy to use (2) High latency (1) No typing necessary (2) Article not included (1) No time waste for queries (1) Terminology support if term misheard (1) Font/colour (1) In-picture solution (1) 206 6.1 Results Two general comments summarise the overall impression of InterpretBank, de- scribed as fast and easier to use than the PDF glossary. Despite the many points of praise, the CAI tool also presented several shortcomings. In four cases, partic- ipants pointed out that, despite the speed of the search, typing was nonetheless required, which interfered with their listening and proved distracting, although not as much as the PDF glossary. In one aspect the PDF glossary may provide an advantage: as pointed out by one student, when he/she had not correctly un- derstood the term, he/she was not able to find it in the glossary. 6.2.1 Performance measures In the present study, data was collected using two performance measures (see §6.1.1): • the degree of terminological accuracy achieved; • the number of severe errors and complete omissions. I will now discuss the results for each metric analysed. 6.2 Discussion In the present section, I discuss the hypotheses operationalised through Seeber’s CLM (§3.5.2.2) and presented in §5.2 in light of the results presented in §6.1. The discussion concerns the effects of digital glossaries, CAI tools with manual look- up, and a mock-up, ASR-enhanced CAI tool on the process of SI. These effects are explored by considering the performance, behavioural, and subjective measures adopted in the study (§5.5.5). This represents the basis for the validation of Seeber’s CLM applied to CASI in §6.2.4. §6.3 concludes the chapter with a discussion of the limitations of the present study. 6.1.3 Subjective measure: The debriefing questionnaire The CAI tool is thus less useful when a term had been misheard, unlike in the case of the PDF glossary which may allow the interpreter to search for terms on the page or by scrolling down. In one case, no negative characteristics were mentioned in relation to the CAI tool. As for the ASR-CAI tool, most participants agreed that the main advantage for the interpreter consisted in its very low distraction potential. Several respon- dents remarked that typing took up some of their attention and concentration, while being automatically prompted with terminological pairs during the inter- preting task allowed them to stay fully focused on listening and interpreting, also because they did not have to identify the term in a list of potential candidates. Another positive factor was the very low latency of the system in showing terms on the screen. Interestingly, one participant pointed out that, while the tool was fast, the terms could have sometimes “popped up earlier”. Other helpful features were the absence of typing from the human-machine interaction, the lack of time loss due to glossary queries, and the fact that the term was presented directly next to the speaker video. An issue mentioned by several participants in relation to InterpretBank, partly addressed in the PDF glossary by the ability to search in the whole glossary, lies in the difficulty to find terms when their spelling is unknown. In some cases, students had the necessity to look up unknown terms which had not been shown on the screen. This was the main issue pointed out for the ASR-CAI tool. This expectation and feeling of dependency on the tool is aptly, albeit rather naively expressed in one commentary: “in very few instances, I didn’t understand a word from the original speech and the ASR didn’t recognise it as [specialised] terminology and didn’t translate it for me”. Finally, one participant remarked that the terminological equivalent shown on the screen by the ASR tool did not include the indication of the grammatical gender. 207 6 Results and discussion 6.2.1.1 Terminological accuracy The use of digital terminology support solutions was expected to affect the termi- nological accuracy achieved during SI (see §5.2). Higher terminological accuracy would indicate lower cognitive effort required of the participants to operate the tool during SI (for the PDF and CAI condition) and to integrate the suggestions into their rendition. In particular, I expected significant differences between the three conditions. I proposed that the ASR tool would promote greater accuracy (see §5.2). Similarly, based on the postulated optimisation of the query function for CAI tools (see §3.6.2), I expected the use of the CAI tool to result in better terminological accuracy than the PDF glossary.i In the analysis, I first considered the accuracy achieved for all terms searched and I then focused on the terms searched/found. In considering all terms searched, i.e. both those which were found and those which were not found in the glossary, it appears that using an ASR-CAI tool as support may indeed pro- mote more accurate renditions, both as compared to a traditional digital glossary and to a CAI tool with manual look-up. However, it should be noted that most 208 6.2 Discussion 6.2 Discussion close renditions resulted from successful queries. Therefore, discussing the cate- gory of close renditions when considering only the terms searched/found should be more revealing of the cognitive effort exerted by participants to integrate the suggestions into their renditions. For the category of zero renditions, all tool contrasts showed statistically sig- nificant differences. In accordance with Seeber’s CLM applied to technology- supported SI (see §3.6.2), I postulated that the CAI tool may interfere less than the PDF glossary with the production effort, even if a term was not found in the glossary. As a consequence, a CAI tool with manual look-up may allow the inter- preter to use the available cognitive resources to at least produce an acceptable rendition of the term. On the other hand, I expected the PDF glossary to lead to a higher percentage of omitted terms due to a more complex human-machine interaction. The participants’ ability to translate terms more accurately with a tool, independently of whether the query was successful or not, would suggest that lower cognitive effort is required to interact with said tool during SI. This would of course apply only to the two conditions requiring manual look-up, as under the simASR conditions all stimuli were shown on the screen. 6.2.1.1 Terminological accuracy My hypothe- sis was confirmed, as the participants omitted or mistranslated fewer terms than with the PDF glossary not only when working under the simASR condition, but also under the CAI condition. Unsurprisingly, the fewest terms were omitted or mistranslated under the simASR condition. To sum up, the ASR-CAI mock-up seemed to perform better than the CAI tool and the PDF glossary, and the CAI tools seemed to interfere less with the SI process than the PDF glossary. This emerged especially from the relatively high percentage of terms left untranslated or mistranslated on average in the PDF condition (§6.1.1.1). These results suggest that, provided that no system failures occur, an ASR-enhanced CAI tool may pro- mote more accurate renditions of specialised terminology and result in a lower percentage of omitted or misinterpreted terms as compared to a traditional digi- tal glossary (in this case, in PDF format) and to a CAI tool with manual look-up. The postulated differences in the terminological accuracy achieved with the three support systems were less evident for the successful queries (see §6.1.1.1). Data suggest that if a query is successful, the terminological quality achieved is similar for all solutions. What emerged from data analysis is, however, a su- periority of bespoke tools for interpreters (be it standard or ASR-enhanced CAI tools) in preventing omissions or erroneous renditions of the specialised terms compared to a non-dedicated solution such as a digital glossary. The findings of the present study thus suggest that, even if a term is found in the PDF glossary, it may be more difficult to integrate it correctly into the rendition. One reason 209 6 Results and discussion for this might be the fact that the target term is visually more difficult to identify on the screen (see §3.6.2). ( ) Additionally, a first investigation of the transcriptions revealed that some stu- dents either misheard some stimuli (e.g. “peak plants” for “peatlands”) or had dif- ficulties looking up certain terms because they did not know how to spell them. This issue also emerged from the debriefing questionnaires (see §6.1.3 and the dis- cussion in §6.2.3). This resulted in accuracy scores of 0 due to wrong renditions despite a glossary query, as the wrong term was looked up (“peak plants” instead of “peatlands”, in my example) or to zero renditions, because the term was not found in the glossary. 6.2.1.1 Terminological accuracy I could have excluded those stimuli from my analysis, as they represent a sub-category that had not been explicitly accounted for in the speech design. However, I decided to include them nonetheless since the ASR- CAI mock-up solved the problem for the students who interpreted the text with the support of this tool. This may represent an additional advantage when us- ing CAI with integrated ASR, albeit more for certain languages (such as English) than for others, and only if the term is correctly recognised by the ASR mod- ule. This would support the idea that an ASR-CAI tool might reduce negative split-attention effects due to better temporal contiguity (see §3.4.2) and support both the production and monitoring and the listening and comprehension tasks (see §3.6.3). Van Cauwenberghe (2020) reached a similar conclusion, observing that an ASR tool might be useful especially in the interpretation of unknown unigrams (see also §6.1.1.1). Overall, the results from the analysis of the terminological accuracy (both for terms searched and for terms searched/found) suggest that, when using an ASR tool for look-up, a higher degree of terminological accuracy can be achieved, especially as compared to a PDF glossary. Terms seem less likely to remain un- translated or to be mistranslated if they have been suggested by the ASR system. This is rather unsurprising and is in line with the positive results for the ASR tool observed by Van Cauwenberghe (2020: 105–109). The relatively high percentage of terms searched with an accuracy score of 0 in the PDF condition may point, on the other hand, to greater difficulties in finding the terms in the glossary, as also suggested by the data on the glossary queries (see §6.1.2.1). This would indicate greater cognitive effort in operating the tool during SI. If the term is found, however, the differences between the tools may be less pronounced. Using a CAI tool rather than a traditional digital glossary does not seem to offer a significant advantage in promoting more accurate renditions. It may be possible that the act of actively starting a query in the tool is what poses the real difficulty, while the fact that the search function is supposedly more efficient in the CAI 210 6.2 Discussion tools only provides a limited advantage. After all, in order to perform a query, several cognitive operations must be performed ahead of typing, i.e. 6.2.1.1 Terminological accuracy acoustically understanding the term in question, deciding to perform the query and, when the results are offered, identifying the term to be used. Most of these operations are not necessary when using the ASR tool: the term is recognised by the system (although this depends on the system performance); from a strategic point of view, the decision is not whether to look up the term, but rather whether to accept the suggestion and integrate it into the rendition; in the mock-up used, no selection of the term had to take place as previous results were deleted and the participant could only see the last term pronounced by the speaker.if However, the statistically significant difference found between the CAI tool and the PDF glossary for zero/unacceptable renditions (see §6.1.1.1.2) suggests that it was easier to find terms when working with a CAI tool and that once found, they are less likely to lead to terminological issues or omissions in the rendition. The postulated usefulness of bespoke tools in improving terminological accu- racy was suggested, among others, by Rütten (2007) and Will (2007, 2015) (see §1.2.3.2). Their hypothesis appears corroborated by the results of the present study, albeit rather for programs which do not require manually looking up ter- minology in the glossary during interpreting. This supports Fantinuoli’s (2017b) intuition that including ASR in CAI tools may result in a considerable advance- ment in the usefulness of CAI tools in-process. On the other hand, while feasible in most cases, the manual look-up operation does not always yield positive results, which supports the hypothesis that cog- nitive effort might be higher when more manual-spatial and more visual-spatial resources are recruited and interfere with the other subprocesses involved in SI. The suggestion that operating a software program during SI to look up terminol- ogy is incompatible with the other efforts (see §2.3.4) appears, however, rather unjustified. In sum, interpreters may find in CAI tools, especially with integrated ASR, useful support solutions for dealing with specialised terminology, while be- ing advised that, just as their users, the tools remain fallible. This emerged also when considering the errors and omissions occurring in concomitance with a glossary query, as discussed in the following section. 6.2.1.2 Errors and omissions The analysis of severe errors and complete omissions was conducted first by considering general trends in the sample and by also exploring differences re- sulting from the type of support used. As for the distinction between target and 211 6 Results and discussion continuation sentences, my expectation was that while performing the query, the participants may have sufficient cognitive resources to retain in memory the content of the original target sentence but may not be able to dedicate enough resources to processing the continuation sentence while at the same time finish- ing to deliver the content of the previous sentence. Therefore, I expected more errors and omissions to occur in continuation sentences (see §5.2). Statistical analysis only partially supported my hypotheses. The study partic- ipants omitted more continuation sentences than target sentences. This is con- sistent with the hypothesis formulated above that attentional resources may be sufficient to process the content of the target sentence while a term was shown on the screen (simASR condition) or searched in the glossary (PDF and CAI con- dition), but may prove insufficient to also process the contents of the continua- tion sentence. Surprisingly, there was no statistically significant difference in the number of severe errors performed in target and continuation sentences. This is probably due to the fact that my participants had not prepared on the speech topics. However, it suggests that interacting with a terminology tool during SI is a non-trivial task, which may produce issues throughout the interpretation. This may point to a trade-off between the level of terminological accuracy and overall acceptability of the rendition when a terminology support tool is used during SI, and may also suggest that the interaction with the tool may in some cases exces- sively interfere with the other SI subprocesses, resulting in evident issues in the rendition. One interpretation of the spillover effect for omissions observed in the sam- ple may be the notion of exported load put forward by Gile (2008): even though the issue does not occur in the sentence containing the target term, problems may arise later in the rendition. Since in my study the speeches used presented a high terminological density and the study design prompted the participants to look up a high percentage of the stimulus terms, the decisive element might be the number of queries performed. 6.2.1.2 Errors and omissions It is possible that under more naturalistic conditions, with less dense speeches, the number of omitted or erroneously inter- preted sentences may be lower. At the same time, these findings also underline the importance of preparation, which cannot be replaced by the mere availability of target language equivalents for the specialised terminology contained in the speech. Being able to look up a term in a glossary or visualising it on a monitor is no guarantee of a successful rendition of the sentence in which the term is embedded. Similar issues have been also found in MTPE, where it has been sug- gested that post-editors may focus “more on the microlevel of the text than on the macro-level/the overall text.” (Čulo et al. 2017: 197). 212 6.2 Discussion 6.2 Discussion In addition to this global analysis, I also considered the effect that the different tools had on the number of severe errors and complete omissions during CASI. With reference to the number of errors and omissions coinciding with a glossary query, my hypothesis was that a lower percentage of severe errors and entirely omitted sentences would occur under the simASR condition as compared to the CAI and to the PDF condition. This is because I expected the ASR-CAI mock-up to interfere less with the SI process. The same was expected to occur for the CAI condition as compared to the PDF condition (see §5.2).i Already at a first examination of the data considering all stimuli sentences (see §6.1.1.2), the use of simASR as support for terminology seemed to interfere the least with the overall interpreting process, as fewer errors and omissions were observed for this condition in the sample average. With reference only to the sen- tences for which the participants performed a query, additional trends emerged. In all sentences (both target and continuation), a term query or an automatic suggestion on the screen led to severe errors in 16.11% of cases on average. This result may suggest that in most cases it was possible for the participants to look up a term in the glossary (or attend to the automatic suggestion on the screen) and still produce an acceptable sentence in the target language. 6 Results and discussion 6 Results and discussion To my knowledge, the only other study which analysed the number of omis- sions in addition to terminological accuracy for CASI is Biagini (2015). However, different from the present study, Biagini considered both partial and complete omissions, and reported values for the number of terms searched, found and cor- rectly interpreted which did not lead to an omission. The results were not par- ticularly encouraging, as only between 23.5% and 43.3% of terms (M = 33.43%) correctly interpreted did not lead to an omission. In the present study, the fact that only 2.14% (on average, range = 0.00%–4.55%) of correctly identified terms coincided with a complete omission when the CAI tool was used is a rather pos- itive result. Of course, the two values are not directly comparable, as different categories of terms and different types of omissions were considered. Additionally, Biagini’s (2015) experimental material did not present the same level of control as the speeches adopted in the present experiment, and it there- fore cannot be excluded that the features of the speeches used, or the type and distribution of stimuli terms, may have impacted the result. It should also be noted that, in the present study, the values may have been higher if partial omis- sions had been included (see §5.5.5.1).2 The results of the analysis of both terminological accuracy (§6.2.1.1) and errors and omissions warrants a methodological reflection. The inclusion of a broader- level analysis beyond the term level highlighted phenomena which would have otherwise not emerged. It therefore appears highly valuable to include this type of evaluation in studies concerned with the impact of CAI tools on interpreters’ cognitive effort. While terminological accuracy and errors and omissions were examined in the present study as indicators of cognitive effort, accuracy scores (for terms, but also and especially for numbers) have often been approached from the perspective of quality (e.g. in Defrancq & Fantinuoli 2021). Without a more holistic analysis of the rendition, it seems unjustified to equate improved termino- logical accuracy (or improved numeral precision) with an overall improvement in the quality of the rendition. For studies investigating the quality of SI, it might therefore be useful to adopt evaluation frameworks which go beyond the assess- ment of individual elements of the rendition. 2This observation of course applies not only to the CAI tool with manual look-up also used in Biagini (2015), but also to the PDF glossary and to the ASR-CAI mock-up. 6.2.1.2 Errors and omissions The fact that the ASR-CAI mock-up, on average, led to a lower percentage of severe errors after a query also supports the hypothesis of lower interference, although there were notable exceptions and the same trend could not be observed for all participants.f As for the effect of the tool used on the number of erroneous or omitted sen- tences in target and continuation sentences, the only statistically significant dif- ference observed was between the PDF and the simASR condition: fewer errors and omissions were observed in the simASR condition than in the PDF condition. It hence seems that the participants were better able to integrate into their rendi- tion terms suggested automatically than terms which they had looked up in the PDF glossary. Using an ASR-CAI tool may therefore not only help translating the target term correctly (see §6.2.1), but also effectively rendering the utterance containing the specialised term. The CAI tool also seemed to lead to an overall better performance than the PDF glossary, but the differences observed were not statistically significant. No statistically significant differences were observed also between the simASR and the CAI tool. These results are rather surprising, but it is possible that on larger samples, smaller effects may indeed be found. It should also be noted that the analysis of errors and omissions was conducted from a rather broad perspective. This may explain why the only statistically significant differences observed were between the tool expected to yield the best performance (i.e., the ASR-CAI mock- up) and the tool expected to yield the worst results (i.e. the PDF glossary). 213 6 Results and discussion Translation process research on CAT tools and MT or MTPE has used evaluation frameworks involving multiple elements of quality, such as MQM (see §4.3.1). Using frameworks such as these, provided that they are adapted to the unique features of interpreting, may offer a standardised methodology which would in turn also improve the comparability of studies on CAI. In light of the technological turn (Fantinuoli 2018b) in inter- preting, adopting a standardised methodology for quality evaluation may also 214 6.2 Discussion 6.2 Discussion help compare the quality of CAI not only with traditional SI, but also with MI, as has been done in TPR (e.g. Vardaro et al. 2019) using MQM to evaluate the quality of both NMT and human post-editing of MT. help compare the quality of CAI not only with traditional SI, but also with MI, as has been done in TPR (e.g. Vardaro et al. 2019) using MQM to evaluate the quality of both NMT and human post-editing of MT. Additionally, it appears useful to also perform qualitative, rather than exclu- sively quantitative, analyses of the interpreters’ performance with CAI tool sup- port, as suggested by Frittella (2022) for numbers, and similarly to the adoption of a “communicative approach” to analyse MI (Fantinuoli & Prandi 2021). An approach of this type may also promote greater awareness on the part of inter- preters of the role played by the tool and may help prevent effects such as those observed by Van Cauwenberghe (2020: 123) of unnecessary self-corrections af- ter visualising automatic term equivalent suggestions on the monitor during SI. Self-corrections of this kind were observed anecdotally also in the present study, and suggest an excessive reliance on the tool, which was noted also in the first studies on the topic (e.g. Biagini 2015, Prandi 2015a,b). 6.2.2.1 Terminological queries Since the simASR condition did not involve active glossary look-up, the results of the analysis of the terminological queries may only provide insight into the differences between digital glossaries (PDF condition) and CAI tools with manual look-up. Based on the results of the pilot study and the findings in Biagini’s (2015) study comparing paper glossaries and CAI tools, my hypothesis was that the CAI tool would elicit more queries due to its hypothesised ease of use and higher querying speed thanks to the incremental search (see §5.2). When considering the percentage of terms searched (see §6.1.2.1), the results do not support my hypothesis. The percentage of terms searched in the PDF and the CAI tool did not differ significantly. This may suggest that the querying process in a PDF or in a CAI tool is more similar than expected in terms of effort (see §3.6.2 and §5.2), despite the postulated optimisation of the querying process in the CAI tool. It is, however, also possible that the extensive pre-test practice enjoyed by the study participants may have helped them improve their ability to perform manual queries while interpreting, even if the PDF was not optimised for this scope. Additionally, the informants may already have been used to looking up terminology in traditional digital glossaries during their regular interpreting classes. Another aspect to consider is the nature of the materials used for data col- lection. The speeches developed for the experiment were highly terminology- dense, which may explain the similar percentage of looked-up terms under the two conditions. Furthermore, the lack of preparation on the topic in all likeliness contributed to the overall high number of queried terms. The percentage of term queries may, however, yield more interesting results in designs including a prepa- ration phase, as was the case in Biagini (2015) or Prandi (2015a,b). For instance, Prandi had found a relatively high inter-subject variability in the percentage of queries performed with InterpretBank. Low percentages were due to the fact that only the essential terms were looked up and probably also due to the effective domain-specific and terminological preparation of the students involved in the experiment. Note, however, that in that first exploratory study students worked in pairs and sometimes did not look up terms themselves while interpreting.f A more useful indicator of effort in the present experiment may be the per- centage of terms searched/found (see §6.1.2.1). 6.2.2 Behavioural measures Six behavioural measures were adopted in the study (see §5.5.5.2): Six behavioural measures were adopted in the study (see §5.5.5.2): • number of terminological queries performed under the three conditions; • the EVS for the terms searched and searched/found; • the inter-cluster pause duration; • time to first fixation on term AOIs; • average fixation duration on term AOIs and tool AOIs; • fixation time on tool and speaker AOIs. The results of the analysis of the term queries provide insight into the inter- preter-machine interaction and, to a more limited extent, into the technical effort involved in CASI. The two time-lag measures, i.e. EVS and ICPD, are indicators of the speed of processing. The three gaze-related metrics provide information as to the cognitive effort experienced under the three conditions and about attention allocation during SI with digital terminological support. 215 6 Results and discussion 6.2.2.1 Terminological queries In this respect, the statistically significant difference between the percentage of terms correctly displayed un- der the PDF and the CAI condition suggests that using a CAI tool may provide an advantage when manually searching for terminology as compared to a PDF glossary. In other words, an interpreter wishing to conduct manual queries in a 216 6.2 Discussion 6.2 Discussion digital glossary may be better advised to choose a dedicated CAI tool than a non- bespoke solution. The positive results for the CAI tool are in line with Prandi’s (2015a, 2015b) findings that a relatively high percentage of queries performed with InterpretBank were successful. As for the impact of the position and morphological complexity of the terms on the participants’ search behaviour, the results were mixed. The position of the terms within the sentence seemed to influence the number of terms looked-up and found in the speech. The fact that significantly more sentence-final terms were looked up and found supports my hypothesis that participants may be bet- ter able to anticipate a query and prepare for it if the term is at the end of the sentence (see §5.2). This result also supports the experimental design choice to control for the terms position in the speeches (see §5.4.3).f As for the morphological complexity and the postulated effects of length, the findings from the present study did not paint a unitary picture. Slightly more unigrams were searched on average in the sample, which would support my hy- pothesis that participants may be more inclined to look up shorter terms because the process may be more immediate. This result is also in line with the results of the pilot study. These results could, however, also be due to a number of ad- ditional reasons, even to the very terms chosen for the experiment. The small differences observed were not significant, which suggests that the morpholog- ical complexity of the terms did not seem to affect the participants’ search be- haviour. However, significantly more trigrams were found than unigrams. This would support my hypothesis that the more elements a term offers, the more likely it is for a term to be found in the glossary, because the interpreter has more material to exploit for the search. This is in line with evidence from studies on the length effect on auditory processing, as discussed in §3.1. 6.2.2.1 Terminological queries These results are also in line with Van Cauwenberghe’s (2020) findings that unigrams tended to be interpreted with the glossary equivalent significantly more often than other categories of morphological complexity (bigrams – pentagrams). Although the distribution of terms was not balanced in his speeches (only two pentagrams were present), he also reached the conclusion that especially short and unknown unigrams may pose a challenge to the interpreter (ibid., p. 119–120). This is con- sistent with the significantly lower number of unigrams than trigrams found by the participants of the present study. To sum up, these findings suggests that an ASR-CAI tool may provide valuable support especially for short and unknown terms.ii If the findings discussed above are confirmed by examinations on larger sam- ples, the analysis of how the morphological complexity affects the search be- haviour may provide useful insights as to the strategies to adopt when using 217 6 Results and discussion digital tools with manual look-up, as suggested in Prandi (2018). However, the strategic aspect of CASI appears relevant not only when manual look-up is in- volved, but also when terms are automatically displayed in an ASR-CAI tool, as literature has shown that interpreters find processing multi-word terms challeng- ing, as discussed in §5.4.3.3. It would be interesting to explore whether the ability of interpreters to correctly integrate a suggestion provided by an ASR-CAI tool varies as a function of the morphological complexity of the terms adopted as stim- uli. If it does, this would further motivate the choice to control for the variable “morphological complexity”, as was done in the present study. Overall, controlling for the position and morphological complexity of the stim- ulus terms included in the speeches appears a valuable design choice to consider also in future studies. 6.2.2.2 EVS As a traditional metric used to explore cognitive effort, the EVS (see §4.4.1.2) was also included as a behavioural measure. Two categories of terms were analysed: the terms searched and the terms searched/found, i.e. first all glossary queries and then only successful queries. Note that also in this case, the two categories of terms coincided for the simASR condition (see §6.1.2.2). Additionally, in order to calculate the EVS, only those queries for which the term had been translated could be taken into consideration. For both categories, I expected the average EVS to be shorter for the simASR condition, followed by the CAI tool and the PDF glossary (see §5.2). The results presented in §6.1.2.2 would lend support to my hypothesis of lower cognitive effort when participants interpreted with the support of an ASR-CAI integration: once the terms have been visualised, they may be integrated more quickly into the rendition. This would speak in favour of the integration of ASR technology into CAI tools for live terminology support during interpreting. It should also be noted that when the ASR-CAI mock-up was used, the average EVS for terms was in the lower range of EVS observed in interpreters (around 2s, see §4.4.1). This is rather unsurprising, as the latency for the ASR-CAI mock-up had been intentionally kept short, in line with the findings by Montecchio (2021) that a high latency for an ASR tool may hinder a successful interpretation, and with Van Cauwenberghe’s (2020) observation that the latency was sufficiently short when a real ASR tool was used, rather than a mock-up such as in the present experiment. More interesting is the fact that the EVS was close to the higher end of average EVS values for the PDF condition. The five-second pause between sentence clus- 218 6.2 Discussion ters was an artefact introduced for the purpose of the experiment, as detailed in §5.4.3. It therefore appears that in a real-life situation, an EVS of around five seconds, such as that observed for the PDF condition, may result in an even higher number of severe errors and complete omissions than were observed in the present experiment. For the CAI tool, the median EVS was also rather long (3.3s). Therefore, queries might be possible with this type of support, but nonethe- less lead to errors or omissions in the rendition. 6.2.2.2 EVS Overall, this suggests that using an ASR tool may help prevent issues related to a long EVS and help keep décalage short. Using the EVS as a measure of cognitive effort with ASR support, however, presents an evident limitation: in virtue of the fact that no manual query is re- quired, the ASR tool is inherently faster. However, the EVS length may also be influenced by the syntactic structure chosen by the interpreter. It is possible that, under the simASR condition, the interpreter might choose a structure closer to that of the ST because the target language equivalent immediately becomes avail- able. When a PDF glossary or a CAI tool with manual look-up is used, however, the interpreter may choose to restructure the target sentence to gain precious time during which to conduct the search. A longer EVS for the PDF or the CAI condition may therefore also suggest that the term was integrated into the in- terpreted sentence later down the line, and not necessarily that it took partici- pants more time to find the term equivalent in the glossary. Differences in EVS may therefore be due not only to how fast the term can be found, but also to strategic choices made by the interpreter. Which of the two hypotheses applies remains unclear and should be further investigated. It is possible that a combi- nation of both factors may contribute to the EVS length. Investigating whether interpreters choose specific syntactic structures to accommodate for a glossary query may also provide useful indications as to the best strategic approach to adopt when performing a query during SI. As for the postulated advantage provided by the CAI tool over the PDF glos- sary, the significantly shorter EVS for the stimulus terms (see §6.1.2.2) shows that the interpreting process may indeed be quicker with the CAI tool. This result, in addition to the significantly lower percentage of term omissions or misinterpre- tations for the CAI tool as compared to the PDF glossary (see §6.1.1.2), represents an additional argument in favour of the use of a bespoke tool for interpreters. Es- pecially in the case of a very fast and/or dense speech for which shortening the décalage may be a useful coping tactic, the CAI tool may therefore offer a non- negligible advantage compared to the PDF tool, while an ASR-CAI tool may be the best choice of all. 6.2.2.2 EVS Although subject to the limitations discussed above, especially as concerns the ASR tool, the EVS may indeed be seen as a valuable indicator of how the use of 219 6 Results and discussion a digital glossary, of a CAI tool with manual look-up and of an ASR-CAI tool affects the speed of processing in SI, which in turn reflects the effort exerted by the interpreter. The two hypotheses formulated above as to the differences in EVS length also highlight the different dimensions of the cognitive effort experienced in CASI and reflect the make-up of the cognitive load imposed by the CASI task. As dis- cussed in §3.4, there is an intrinsic load imposed by the interpreting task which is compounded by the extraneous load imposed by the individual tools. Similarly to the post-editing effort (see Krings 2001), the effort exerted by the interpreter during CASI may therefore be described as CAI effort, comprising both the tech- nical effort required to interact with the tool, the cognitive effort required to adjust one’s own rendition to the interaction with the tool, and the temporal ef- fort deriving from the time required to find the term equivalent in the glossary. Considering all three sub-efforts, the CAI effort may reasonably be expected to be lower with the support of an ASR tool than with a CAI tool or a non-bespoke digital glossary. 6.2.2.3 ICPD The inter-cluster pause duration (ICPD, see §6.1.2.3), was used in the analysis as an additional indicator of cognitive effort. Since the silent pause introduced between two sentence clusters was always 5s long in the source text, I consider an ICPD of 5 or more seconds in the participants’ delivery as indicative of a faster interpreting process. On the other hand, cluster-to-cluster spans shorter than 5s should be indicative of higher cognitive effort, or, at least, of a slower interpreting process. The closer to 0, the more complex the task and, presumably, the higher the cognitive effort. If the search requires a lot of time, it is possible that the subject will have to use most of the 5s silent pause to finish interpreting the passage. A longer pause would therefore indicate a more seamless integration of the term into the interpreted sentence, i.e. a more efficient interaction with the tool, and a shorter décalage. On this basis, my hypothesis was that pauses between clusters in the rendi- tions would be longer for the simASR condition than in the CAI and the PDF conditions (see §5.2). For the other two conditions, on the other hand, I expected mean pause duration to be shorter, which would indicate a longer décalage, i.e. the need to use all or part of the silence between clusters to finish interpreting. I however expected the ICPD to be shorter for the PDF condition than for the CAI condition. 220 6.2 Discussion 6.2 Discussion Overall, data show that, when using the simASR tool, the interpreting process was quicker, i.e. the décalage was shorter, and the participants could keep up better with the speaker than when working with a PDF glossary. On the other hand, the hypothesis that the CAI tool may interfere less than the PDF glossary with the production effort in SI, but more than in the simASR condition, did not find support from the analysis of inter-cluster pause duration. As reported in §6.1.2.3, significant differences were found only between the ASR- CAI mock-up and the PDF condition, but not between the CAI tool and the PDF glossary and between the CAI tool and the simASR tool. On the basis of these results, I may only affirm that the simASR seems to slow down the interpreting process less than a traditional digital glossary. 6.2.2.3 ICPD It also appears that the CAI tool with manual look-up did not significantly slow down the interpreting process more than under the simASR condition. I do not have an explanation for the discrepancy between the results for the EVS and the ICPD. As observed for the analysis of errors and omissions, these discrepancies may indeed be due to the small sample size. A qualitative analysis of the TT or a quantitative analysis of partial omissions in the TT may help qualify these findings. This goes beyond the scope of the present contribution. It would also be interesting to use other pause metrics, such as silent pauses as in Gieshoff (2021), as an indicator of cognitive effort, either in place or in addition to the metrics used in the present experiment. 6.2.2.4 Time to first fixation on term AOIs As predicted by Seeber’s CLM (see §3.6.3), the simASR tool should facilitate the fast individuation of the term on the monitor, since the terms were displayed in isolation in the mock-up. The ASR-CAI integrations already available exper- imentally, such as the one offered by InterpretBank (Fantinuoli 2017a), display the results in progressive order of appearance, with the most recently found term at the top of the list. If the interpreters are aware of this, they will probably be prepared to only look in that area of the tool window. Therefore, I suppose that results would be similar if a real tool of this kind was used rather than a mock- up like in my experiment. This remains, however, a hypothesis which should be tested experimentally. The CAI tool presents the advantage of showing results while the query is in progress, which should promote a faster individuation of the target term than in the PDF condition. Additionally, in the version used for the experiment, the CAI tool InterpretBank by default shows a maximum of nine results on the screen, 221 6 Results and discussion and the search stops after at least five terms have been found for the query.3 This was expected to further facilitate the individuation of the term on the screen and to reduce negative effects deriving from a visual search for the target term equivalent, as discussed in §3.6.2. When using the PDF glossary, the participants could visualise the entire glos- sary page. In some cases, for instance for the terms on the last page of the glos- sary, fewer terms were shown, which would make it easier to identify the term searched for. This was a limitation which did not emerge in the pilot study. How- ever, only a limited number of stimuli terms were present on the last page. In most cases, I expected participants to require more time to identify the term on the page due to having to sort through the large amount of visual material (see §5.2). The results presented in §6.1.2.4, i.e. the significantly shorter times to first fixation for the simASR condition as compared to the PDF and CAI condition, are rather unsurprising. However, contrary to my expectation, the supposed ad- vantage provided by the CAI tool was not confirmed by statistical testing. 3This option was removed in later InterpretBank versions. 6.2.2.5 Average fixation duration on term and tool AOIs Due to the lack of a manual query with the ASR-CAI mock-up, I hypothesised that the simASR condition would require lower (technical) effort than the PDF and the CAI condition (see §5.2). In turn, I also expected that using the CAI tool would result in more limited additional effort than the PDF due to the progres- sive search. The progressive search was expected to facilitate the interpreter-tool interaction, resulting in an easier integration of the querying process into the in- terpreting process. The metrics of average fixation duration on the term and tool AOIs were expected to provide insight as to these hypotheses, specifically with reference to the cognitive effort experienced in processing the terms under the three conditions during SI and in interacting with the different tools. The results for the average fixation duration on the term and tool AOIs par- tially support these hypotheses. The analysis of fixation durations on the term AOIs (see §6.1.2.5) showed significant differences between the simASR and the other two conditions and non-significant differences between the PDF glossary and the CAI tool. The significantly shorter fixation durations on the term AOIs found for the simASR condition as compared to the two other conditions lend support to my hypothesis that processing the term may be easier (and faster) when terminology is automatically presented on the screen.if However, the fact that no significant difference was found between the PDF glossary and the CAI tool for this metric may point to a higher degree of re- cruitment of cognitive resources by the CAI tool than expected, specifically of visual-spatial resources (see §3.6.2). My hypothesis is that this may be due to the need to visually identify the term in a list of results also when working with the CAI tool, despite the number of results being smaller than in the PDF condition.ifi The significant difference in the duration of fixations on the tool area (see §6.1.2.5) also supports the hypothesis that the ASR tool may recruit fewer atten- tional resources than the PDF glossary. 6.2.2.4 Time to first fixation on term AOIs This is interesting, especially because many participants had identified as the most problematic aspects for the PDF glossary the lack of filtering among the results and the need to navigate in the document to find the correct equivalent. Only one participant had commented that the CAI tool had offered too many results after a query (see §6.1.1). Caution should however be exercised in interpreting these results as an indi- cation that the CAI tool may not provide an advantage over the PDF glossary. When a CAI tool is used, it is possible that the term may be displayed on the monitor earlier than for the PDF glossary. This might explain why the time to first fixation was longer than expected under the CAI condition. This hypothesis seems justified considering that the EVS for the CAI condition was significantly shorter than for the PDF condition (see §6.2.2.2). It is therefore possible that the CAI tool was indeed able to find the target term faster than the PDF glossary, which would be an encouraging result. The participants who had identified the term on the screen with greater delay when working with the CAI tool may simply have continued typing even though the target term was already visible on the monitor. This may indicate that inter- preters may actually need to type less than they think is necessary when work- ing with a CAI tool. At the same time, these results may also suggest that, in the phase between the start and the end of the query, the intermediate results may be more difficult to process. The search behaviour of the individual participants 222 6.2 Discussion certainly is an important factor: for instance, touch typing may promote greater focus on the monitor and a quicker identification of the term equivalents. 6.2.2.5 Average fixation duration on term and tool AOIs The shorter fixations on the simASR area may be seen as an indication of a lower level of cognitive engagement with the tool.if However, the fact that no statistically significant differences were found for the fixations on the CAI tool area as compared to the other two conditions sug- gests that the CAI tool may not always be less distracting than the PDF tool, and not always more distracting than the ASR tool. It is possible that this may de- pend on the number of results displayed on the screen after a CAI tool query: 223 6 Results and discussion for some terms, the query may yield only one result (similarly to the simASR condition); for others, there may be a rather long list of terms (similarly to the PDF condition). Overall, a lower degree of attentional resources recruited by a terminology support solution may be expected to promote focused attention on the speaker (see §3.4.2). In other words, the interpreter might have felt the need to monitor the simASR tool significantly less than the PDF glossary. Using an ASR tool in particular may help monitor (e.g. Schaeffer et al. 2019) the primary source of information (the speaker video). The metric fixation time, discussed below, may help gain further insight into this hypothesis. 6.2.2.6 Fixation time on tool and speaker AOIs The metric fixation time, i.e. the total duration of fixations, on the tool and speak- er AOIs (see §6.1.2.6), was expected to provide additional information about par- ticipants’ monitoring of the tools as well as about the distribution of attention (see §3.4.2) between the speaker and the support tool. Due to the postulated lower degree of engagement with the tool in the simASR condition, I expected the par- ticipants to spend significantly more time processing the video stimulus in this condition than in the CAI and PDF condition, and, in turn, to spend more time looking at the speaker in the CAI condition than in the PDF condition (see §5.2). In other words, in the simASR condition, they would be better able to monitor the speaker and focus their attention on the speaker video, while they would need to monitor the tool the least. The results of the analysis for this metric (see §6.1.2.6) support my hypoth- esis. Specifically, these results may be interpreted as evidence that, in the PDF condition, attention was more equally distributed between the speaker and the tool, which suggests that the tool required a considerable amount of attentional resources from the participants which could not be devoted to the speaker video. In turn, in the simASR and CAI conditions, it was probably easier for the partic- ipants to intentionally divide their attention between the speaker and the tool and to focus more on the speaker, particularly in the simASR condition. Considering that the tool area only provided term equivalents, participants should ideally only attend to the tool when such information is provided or when they actively seek it during SI. In the PDF condition, however, the participants split their attention more between the speaker and the tool than in the CAI con- dition, and in some cases fixated more on the tool than on the speaker. The signif- icant differences observed when comparing the duration of fixations on the tool 224 6.2 Discussion 6.2 Discussion AOIs and the total fixation time on the speaker AOI and the tool AOI support the choice of an ASR tool over a solution requiring manual look-up. Despite the advantage which the ASR-CAI mock-up seemed to provide over the other solutions, it should be noted that all terminology support tools may cause a negative split attention effect during SI. 6.2.2.6 Fixation time on tool and speaker AOIs Yet, thus far potential split at- tention effects do not seem to have been taken into consideration in the design of CAI tools, with and without integrated ASR. It still remains to ascertain how the placement of the information on the screen affects attention allocation dur- ing SI. Both the Booth Mode of InterpretBank (Defrancq & Fantinuoli 2021) and the SmarTerp prototype (Frittella 2022) present terms, numbers and named enti- ties in separate areas or columns. If a negative split attention effect is identified which can be traced back to the current tool interfaces, it would be interesting to assess whether alternative interfaces can be designed in order to reduce said effect. For instance, solutions along the lines of what has been proposed through integrated subtitles (Fox 2018) may be explored: improving not only the temporal but also the spatial contiguity (see §3.4.2) for ASR tools might help reduce their extraneous load (see §3.4.1). This, however, remains a hypothesis. Nonetheless, for the sake of CAI tool usability, it seems necessary to test different modes of display and determine their impact on the interpreting process and product. 6.2.3 Subjective measure: The debriefing questionnaire In addition to the quantitative measures discussed above, a questionnaire was administered to the participants at the end of the experiment to collect qualitative data. The answers to the debriefing questionnaire (see §6.1.3) allow to further qualify and disambiguate the results of the quantitative analysis discussed above. g The participants were asked to rank the tools according to how useful and how distracting they proved during the experiment. I expected the ASR-CAI tool to be ranked as most useful and least distracting, and the PDF glossary to be ranked as least useful and most distracting (see §5.2). My hypothesis was supported. Overall, the participants’ answers support the hypothesis that the PDF glossary may require more attentional resources than the CAI tool, and that the ASR tool may be preferable both to the PDF glossary and to the CAI tool. It is however interesting to note that two out of nine participants would bring the CAI tool and not the ASR tool into the booth. Let us therefore discuss the most useful and most problematic aspects pointed out by the participants, as they may represent useful indications for the optimisation of CAI tools and ASR tools. With reference to the PDF glossary, participants appreciated being able to see several entries at once. This would explain why the fixation time was particularly 225 6 Results and discussion high on the PDF area: they may have looked at the glossary also when they were not looking for a term, for example during the breaks between sentence clusters. It could be argued that this aspect depends on the way the glossary is organised, on the height of the rows and on the font size, but it seems logical that the presentation of vocabulary one page at a time favours an overview of the terminology. This might also explain why there was no significant difference in the number of queries between the PDF and the CAI tool, as discussed in §6.2.2.1: some participants may not have needed to query the glossary for certain stimuli because they could already see the target term on the monitor. This is however likely to have occurred in a limited number of cases, since the number of queries was overall relatively high. Something similar applies to the alphabetical order, a feature which partici- pants also appreciated as it helped them identify the terms on the screen. 6.2.3 Subjective measure: The debriefing questionnaire The alphabetical order is, however, not a prerogative of a PDF, Word or Excel glos- sary. Therefore, the alphabetical order may be considered as a useful feature per se. It may, however, be relevant mostly when little to no preparation is possible, or in general when the interpreters have not had time to study the glossary in detail, such as in the present experiment. Other glossary structures may be more effective under different conditions. The participants also appreciated being able to scroll the page down to look for terms as an alternative to using the search bar. However, this is likely to require a conspicuous amount of time and might only prove useful in emergency situations. This is supported by the fact that most participants highlighted that the PDF might be distracting and cumbersome: the number of terms displayed on the screen at the same time may indeed prove overwhelming. As for the CAI tool, most participants underlined that the dynamic search and the automatic clearing of the search field make the querying process less time-consuming than in the PDF glossary. In this respect, it is necessary to remind the reader that, although the progres- sive (dynamic) search, the automatic clearance of the search field and the fuzzy search are specific to InterpretBank and not available in the PDF tool, a manual query without automatic clearance of the search field and without fuzzy results can also be performed with the CAI tool. If those settings had been selected for In- terpretBank, the participants’ commentaries for the two conditions would have probably been very similar. Since the aim was to compare three solutions which allow for different ways for the human interpreter to interact with the machine, the search settings which correspond to the most commonly selected features for the CAI tool InterpretBank and which constitute its main element of distinction were selected. 226 6.2 Discussion 6.2 Discussion I found it interesting that, even though only several terms are shown in the CAI tool after a query, they may nonetheless be too numerous for the user to be able to identify the target term under the time pressure inherent to SI, as pointed out by some participants. Finally, the simASR tool was often described as not particularly distracting, specifically because no typing was required, unlike in the CAI and in the PDF condition. 6.2.3 Subjective measure: The debriefing questionnaire Replacing typing with ASR has after all been found beneficial for the trans- lation process (e.g. Carl, Lacruz, et al. 2016a). The fact that the attitude of most participants towards the simASR tool was overall positive might seem to con- trast with a certain adversity to MT observed among translators (see §2.3.4). The ASR-CAI tool mocked-up in the present study, however, is not based on the use of MT, but rather on a combination of an ASR module with an extraction mod- ule. The results presented on the monitor had therefore already been validated through the creation of an ad-hoc glossary. In theory, however, it would be pos- sible to use MT without having prepared a glossary beforehand, similarly to the creation of automatic interlingual subtitles (e.g. Dessloch et al. 2018). If a tool of this kind was used to provide support to interpreters during SI, it is possible that similar negative attitudes to the ones found for translators may also be found for interpreters. Indeed, a certain level of distrust has already been suggested in studies using real tools instead of mock-ups, which, although a glossary was used as the basis for the terminology extraction, may not always display the expected result (see Van Cauwenberghe 2020). Furthermore, while the fact that the human-machine interaction is minimal in the simASR condition and is therefore expected to improve concentration and reduce time loss, it results in an important limitation. When interpreters work with an ASR tool, they are unable to influence the queries performed by the tool and to actively query the glossary when an unknown term is not suggested automatically, as pointed out by some participants. This unfulfilled expectation may also lead to stress or distraction, as it may require reformulating the TT or re- evaluating the strategy chosen to deal with the lack of an available equivalent for the term in question. Interpreters may therefore experience the same “cognitive friction” (O’Brien et al. 2017, Cooper 2004: 19) observed in translators working with CAT tools and MT outputs. It should be noted that the differences in cognitive engagement and distraction potential observed between the tools may also be due to the user interface. The simASR tool was relatively bare, while more elements were present in the CAI tool and especially in the PDF glossary. 6.2.3 Subjective measure: The debriefing questionnaire This aspect could have been accounted 227 6 Results and discussion for by developing three tools ad-hoc for the experiment only differing in the search algorithm (not progressive, progressive, and automatic) and in the dis- play of results, with no difference in terms of user interface, brightness (through the use of grey tones in the UI, for instance), etc. Such a solution, while more time consuming in the experiment preparation phase, would have allowed to pin-point the source of difference between the tools with greater certainty, and to address the limitations discussed for the metric pupil size (see §4.5.3). Nonetheless, even if the PDF tool attracted more attention and thus proved more distracting due to its UI, this would represent an additional argument in favour of the simplification of the UI of CAI tools, whether with or without ASR integration, to reduce the distraction potential of digital support tools for inter- preters. TPR has addressed the issue of the distraction caused by the interface of CAT tools and MT (e.g. O’Brien et al. 2017). Studies have been dedicated to the op- timisation of the UI of CAT tools and PE environments (e.g. Moorkens & O’Brien 2016). It appears fundamental to also address the ergonomics of CAI tools, both with and without ASR integration, to maximise their support potential while reducing the extraneous load (see §3.4.1) deriving from suboptimal interfaces. A first step in this direction has been taken through the EABM (2021) project, which seeks to explore potential user preferences through questionnaires (see also §3.4). It appears necessary to explore this research question further to determine the impact of the tools’ interface on the CASI process and at the same time improve the tools’ usability. By the same token, future experiments may also address the type of informa- tion presented by ASR tools. At present, most prototypes of an ASR integration into a CAI tool present the terms in isolation, devoid of any additional informa- tion. While this is easily modifiable and does not constitute an insurmountable obstacle in the use of the tool, it is also true that both traditional digital glossaries and most CAI tools easily allow the user to add columns or fields for accessory information, such as the column “Booth Info” in InterpretBank, which the user can choose to make visible in the conference modality. 6.2.3 Subjective measure: The debriefing questionnaire This feature may also be integrated into ASR-supported CAI tools, as suggested by one participant in the questionnaire. Unsurprisingly, participants appreciated the low latency of the simASR tool, as the mock-up had been specifically prepared with this characteristic. As such, students experienced an ideal ASR-CAI integration, while real solutions may not perform as well as in the experimental setting. Although the results for the sys- tems currently available are encouraging, especially for numbers (e.g. Defrancq & Fantinuoli 2021), latency remains an essential factor, not to be underestimated in the design and use of ASR tools (see Montecchio 2021). It was surprising to see that one participant mentioned “high latency” as a negative side of the simASR 228 6.2 Discussion 6.2 Discussion tool, as the latency was the same for all terms. My hypothesis is that in some cases, the participant in question may have preferred to know the equivalent faster in order to effectively integrate it into the syntactic structure chosen to formulate the TT. One aspect in which the ASR tool is undoubtedly superior to the other two solutions is the assistance provided by the tool when a term has been misheard or not heard by the interpreter, due to the phonological interference experienced during SI (Díaz-Galaz & Torres 2019), or simply because the term is unknown. This had emerged also during the analysis of how the morphological structure of the terms influenced the number of queries (see §6.2.2.1). An ASR engine should provide useful support in this cases, provided that the SR performs well, which may not always be the case due to the characteristics inherent to oral speech. 6.2.4 Validation of the CLM applied to computer-assisted SI The present study was motivated by the intention to explore how various types of in-booth technological support for terminology affect the process of simultane- ous interpreting. The hypotheses on the differences in human-machine interac- tion and the resulting cognitive effort between traditional digital glossaries, CAI tools with manual terminological lookup and ASR-enhanced CAI tools were for- mulated using Seeber’s Cognitive Load Model of SI illustrated in §3.6. The present section validates the model in light of the results of the experiment discussed in the previous sections. Based on Seeber’s CLM of SI (2011, 2017), the three solutions were expected to differ mainly in the degree of recruitment of manual-spatial resources at the response stage (i.e. while interacting with the tool to find the stimulus term) and of visual-spatial resources at the perceptive-cognitive stage (for the identifica- tion of the equivalent on the monitor). In particular, it was hypothesised that the PDF glossary might recruit a higher degree of resources during the query (manual-spatial resources at the response stage during the production and mon- itoring task), as well as for the identification of the term equivalent on the mon- itor (visual-spatial resources, also during the production and monitoring task) (see §3.6.2 and §5.2). CAI tools were also expected to recruit manual-spatial and visual-spatial re- sources, but also to facilitate the querying process due to the progressive search and to support the identification of the term equivalent by displaying only the relevant terms on the screen (see §3.6.2). Finally, it was speculated that the simASR tool would recruit no additional attentional resources at the response stage during production and monitoring, 229 6 Results and discussion as the terms were displayed automatically on the monitor. The simASR tool was also not expected to recruit visual-spatial resources for the identification of the term on the screen during the production and monitoring phase.if The findings of the experiment support the assignment of different demand vectors to model the differences in the level of resource recruitment for the dif- ferent tools. In particular, the significantly lower number of terms found for the PDF tool as compared to the CAI tool, lends support to the hypothesis that using a traditional digital glossary may be more cognitively taxing than a bespoke solu- tion. If the querying process is successful, it stands to reason that the additional effort interferes less with the production effort in the interpretation. 6.2.4 Validation of the CLM applied to computer-assisted SI Hence, as- signing a lower demand vector (1) to the CAI tool than to the PDF tool (2) for the manual-spatial resources recruited during the response stage appears justified.ifii Due to the lack of significant differences in the time to first fixation between the PDF and the CAI condition (see §6.2.2.4), I may assign a demand vector of 2 to the CAI tool for the recruitment of visual-spatial resources in the receptive- cognitive stage, thus equating it to the PDF glossary (the ASR tool would receive a value of 1). However, these results contrast with the significant difference be- tween the PDF and the CAI condition found for the EVS, as discussed above (§6.2.2.2), which would justify a demand vector of 1 for the CAI tool. As observed in §6.2.2.4, the findings for the CAI condition may be biased by the fact that I used the timestamp of the first moment in which the term AOI becomes visible on screen as a reference to calculate the time to first fixation. A significant dif- ference may indeed be found if the end of the query was taken as a reference point instead. From an operational point of view, this might, however, prove problematic for the CAI condition, because participants may continue to type af- ter having already found the term on the screen. The first fixation may therefore fall outside of the time span considered for the analysis. Due to the inconsistency between these findings, the assignment of a demand vector of 1 for visual-spatial resources in the CAI condition remains but a tentative conclusion, which may be further explored in future studies. The facilitation effect postulated for the multimodal presentation of the terms (visual-verbal and auditory-verbal modality) for the simASR condition is sup- ported by the responses to the questionnaire. These findings also corroborate Seeber’s proposal to model facilitation effects in the CLM. In particular, they support the choice to model the interaction with the ASR-CAI mock-up together with the listening and comprehension task (Seeber 2017: 472), which the multi- modal presentation facilitated in the experiment. Here, a parallel may be drawn with the role of the visual trace of the TT in translation (see Schaeffer et al. 2019). In translation, being able to see the TT may help monitoring the writing process and the translation choices. 6.2.4 Validation of the CLM applied to computer-assisted SI In interpreting, it may support the transcoding of the 230 6.3 Limitations ST as well as the monitoring of the rendition of the TT, albeit with the limitation of individual terms. The use of different demand vectors to model different levels of resource re- cruitment appears useful within the context of the present experiment and is in line with the role of these components in Wickens’s (2002) original model. See- ber made a different use of the demand vectors both in the early and in the later version of his models. As discussed in the literature review (see §3.5.2), in his application of the model the demand vectors can only take a value of 0 or 1, or of 0.5 to model facilitation effects for multimodal integration. This double role of the demand vectors appears rather difficult to reconcile in a single model. One might therefore follow two approaches. The first would con- sist in modelling multimodal and multicue integration and the level of resource recruitment in two separate models: one CLM of resource integration and in- terference, i.e. Seeber’s application of Wickens’ model as in Seeber (2017), and one CLM of resource recruitment in SI, i.e. my application of the model as in Prandi (2017, 2018) and in the present work. The second approach might consist in expanding Wickens’ model to include an additional layer of vectors: demand vectors as in Wickens’ original model, and integration vectors as in Seeber’s ap- plication of the model. To avoid doubling the demand vectors for multiple cues, which would take a full demand vector (1), it might be possible to sum the inte- gration vectors as proposed by Seeber (2017) and to sum “the average demand, across all resources, within a task (and [to sum] over both tasks)”, as originally intended by Wickens (2002: 171). Figures 6.32, 6.33 and 6.34 represent the CLM applied to SI under the PDF, CAI and simASR condition following the second approach of combining demand and integration vectors. As discussed in the previous sections, there are several unanswered questions as to the level of resource recruitment and sub-task interference in CASI. Future studies conducted on larger samples might help provide answers which may lead to a further refinement of the model. 6.3 Limitations The main contribution of the present work lies in the methodology developed for the study of technology support for specialised terminology in simultaneous interpreting. Nonetheless, the study presents some limitations, which I discuss in this section, before providing my concluding remarks, reflecting on the method- ological, practical and didactic implications of my findings, and addressing po- tential avenues for further research in this relatively new domain. 231 6 Results and discussion Figure 6.32: Conflict matrix for the PDF condition Figure 6.32: Conflict matrix for the PDF condition 6.3.1 Participants and choice of experimental design An important limitation for the generalisability of the experimental findings lies in the small population of participants that could be recruited for the study. As discussed in §5.5.1, despite clear efforts devoted to the recruitment of participants, I was only able to include nine students in the study, mostly due to a lack of time and, probably, interest on their part, and on a series of unforeseen factors. The small sample size clearly limits my ability to generalise the results of the present study to the population of interpreting students. To counteract this limi- tation, a within-subjects design was chosen for the experiment. As my aim was to compare simultaneous interpreting with different types of glossary support, a within-subject, repeated measures design allowed me to work with a smaller 232 6.3 Limitations Figure 6.33: Conflict matrix for the CAI condition Figure 6.33: Conflict matrix for the CAI condition sample wherein each participant acted as his or her own control group. This decision proved particularly helpful, as recruitment of participants was more complex than anticipated. As the participants were randomly selected among advanced students of the master course in conference interpreting of the Univer- sity of Mainz and showed, among other aspects, different levels of command of the English language and different typing speeds, I feel quite confident that the trends observed in the sample were not due to the individual characteristics of the participants, but would probably also be found in a larger sample. Nonetheless, a within-subject design also presents some limitations, in partic- ular due to a potential increase in Type 1 error due to repeated tests in statistical analysis. I took this into account in the choice of inferential statistics and by applying corrections when necessary. 233 6 Results and discussion Figure 6.34: Conflict matrix for the simASR condition Figure 6.34: Conflict matrix for the simASR condition Figure 6.34: Conflict matrix for the simASR condition Additionally, learning, boredom and fatigue effects may also represent issues not to be underestimated in within-subject designs. While boredom was not con- sidered as a potential issue given the difficult task at hand4, even more complex when performed by students, learning and fatigue may have posed important limitations.f In particular, it is possible that a certain practice effect may have ensued within- task and also from the first trial to the last. 4Although students may have been less motivated to perform well compared to an exam situ- ation or to a real interpreting assignment. However, I can reasonably assume that a low level of motivation would have affected all trials equally. 6.3.1 Participants and choice of experimental design Specifically, it is possible that the students may have noticed the recurrence of the 5s pause between each contin- uation and introductory sentence. This may have caused them to intentionally 234 6.3 Limitations slow down or interrupt their rendition of the source speech while looking up terms, as they were aware of the fact that they could catch up with the speaker in the short break between clusters. For this reason, the EVS measured cannot be generalised as the average EVS of students working with PDF, CAI or ASR glossaries during SI, as it is possible that different values may be observed in a more naturalistic setting with less controlled speeches.f I had taken the learning effect into account and attempted to exclude it from the equation by training the participants ahead of data collection. If students had developed strategies to work with the individual tools, this would have already happened before testing and not during the experiment. However, it is possible that the fixed structure of the speeches may have promoted a certain practice ef- fect within-task, as the test speeches were made up of collated sentence clusters, unlike the practice speeches which were more naturalistic in nature. Nonethe- less, if a learning effect occurred, it would have occurred for all three treatments. The fact that there was a significant difference in the EVS between the three conditions seems to suggest that the type of tool used to prompt the participant with specialised terminology did affect the speed of integration of the term into the rendition. It would be interesting to further explore this instance in more naturalistic settings and also while manipulating the speaker’s delivery speed, as presentation rate has been proven to influence the interpreter’s ear-voice lag (see Barik 1973, Lee 2002). As for fatigue, it is inevitable that students may have felt more tired at the end of the testing session. The trials lasted around 20 minutes each, of which 14 were occupied by the very demanding interpreting task. Given the high num- ber of terms looked up on average in the PDF and CAI condition, it would have probably been sufficient to limit the length of each speech to around 10 minutes, in order to replicate the average duration of an interpreting exercise. I had con- trolled for a potential order effect experimentally by randomising the speech-tool combinations. 6.3.1 Participants and choice of experimental design It is interesting to notice that performance evaluations and accu- racy scores were higher for the simASR condition across the board, even though in some cases the simASR treatment had been the last in the experiment. In my opinion, this might suggest that, even in situations of increased fatigue, the type of support provided by an ASR tool may help promote a high quality delivery, at least in terms of terminological accuracy and adherence to the source text in terms of informational content. Finally, it should be stressed that, as in most other studies available on the topic (see §2.4), the experiment was conducted on students. Therefore, my findings cannot be generalised to the population of interpreters, for which the experiment outcomes may (or may not) have been very different. 235 6 Results and discussion 6.3.2 Use of simulated ASR The results for the simASR condition speak in favour of the adoption of CAI tools with integrated ASR as compared to traditional digital glossaries and CAI tool with manual look-up. Particularly, a tool with integrated ASR poses the ad- vantage of reducing the additional cognitive load generated by manually query- ing the glossary for terminology. Nonetheless, it is important to remind the reader that, in the present experiment, the absolute ceiling performance of an ASR-enhanced CAI tools was simulated. This implied mocking-up a system with perfect precision and recall, excluding potential system failures from the experi- mental design. As discussed in §5.5.6, this was a conscious choice. It was taken to avoid further complicating an already highly complex experimental setup within the context of a study which presents an exploratory character and which mostly aims to offer a methodological contribution to the field of CAI research. In light of these considerations, it is not possible to extrapolate the findings of the experiment to real-life CAI tool without hypothesising that system failures, which are bound to occur, may have a negative impact on interpreters’ cognitive processing and their interaction with CAI tools. In some instances, it may even prove counterproductive to adopt an ASR-enhanced CAI tool during SI, precisely because even high-performing systems may offer sub-optimal support under cer- tain conditions. On the one hand, this will require further research to explore the effects of system failures on SI, both from a process- and from a product-related perspective. On the other hand, increasing interpreters’ AI literacy may assist them in selecting the right tool for the job and in adjusting their expectations towards AI-enhanced tools, thus contributing to making their use of said tools more effective. 6.3.3 Priming When working with a glossary, be it in PDF format, in a CAI tool, or even printed on paper, it is possible that additional terms may be visualised while looking for another term simply because they are on the same page or because they appear in the list of results. In those cases, it is possible that for said terms a priming effect occurs: the interpreter may identify said terms faster when she looks them up after having already viewed them on screen. In the present experiment, a num- ber of terms had already been displayed on the screen during previous queries. Priming may have determined shorter times to first fixation and shorter fixation durations for these terms. Despite this limitation, in the present study I chose not to include this subset of terms in the analysis. This decision was made for reasons of practicality, as 236 6.3 Limitations discussed in §5.5.7, but it would have certainly been interesting to explore this further. For instance, a sample of terms likely to have been primed may have been analysed for all participants. The first step would have consisted in the identification of the terms that had been looked up by all participants, for each of the three speeches and for both the PDF and the CAI condition. If the simASR tool was used, only one term was shown at a time, so no priming could have occurred. The next step would have consisted in reviewing the gaze replays for the participants working with the PDF glossary. When using a PDF glossary, it is possible to visualise terms occurring later in the speech only if they are alpha- betically close to the term currently being looked up. With a CAI tool, it is more probable for terms to be visualised on screen even though they are mentioned later in the speech, for example because one of the elements making up the term contains the letter sequence typed by the test subject. Therefore, the PDF con- dition should be selected as the reference for the identification of the sample of terms (searched/found in the glossary) likely to have been primed. Finally, addi- tional AOIs may be drawn on top of the regular AOIs to keep the two analyses separate. 6.3.3 Priming While I did not consider this aspect in the analysis of gaze data, I may refor- mulate my conclusions as follows: even though the use of a traditional digital glossary or of a standard CAI tool may facilitate the identification of terms on the screen, the ASR integration nonetheless promotes a faster identification of the term on the monitor. However, it is possible that some of the differences ob- served between the PDF and the CAI tool may be explained by the priming effect rather than only by the different search mechanism. 6.3.4 Eyetracking methodology The present study was the first to explore the cognitive implications of SI with CAI support with the eyetracking methodology in combination with more estab- lished methods focusing on the product of technology-supported SI, such as the analysis of terminological accuracy and errors and omissions in the rendition. In addition to the priming effect discussed above (§6.3.3), the collection of gaze data presented some methodological challenges, which have in part limited my interpretation of the experiment outcomes.i The first challenge is linked with the activation and deactivation of the areas of interest on the stimulus terms. In the PDF and CAI condition, in several occasions, the previous stimulus term was still visible on screen when the next term was pronounced by the speaker. To facilitate my analysis, I decided to deactivate the previous term as soon as the query for the new stimulus had been started by the 237 6 Results and discussion participant. This was a necessary, albeit arbitrary decision, but which was applied to all stimuli in the same way. It would certainly be interesting to explore the gaze behaviour of the participants in situations in which both terms are visible on screen. However, this would have exceeded the scope of the present study, which is why I chose to exclude this aspect from my analysis. It should be noted, however, that in some cases the terms disappeared from the monitor soon after the delivery of the sentence cluster in question had been rendered in the target language. This was due either to the participant actively clearing the search field immediately after the query to prepare for the next search, or to the participant having looked up another term unknown to them, which I had not selected as stimulus (see §6.1.2.1.7). This may explain shorter total fixation times on the term AOIs and even shorter fixation durations for some participants (see §6.1.2.5). The second limitation is linked with the fact that, in the PDF and CAI con- ditions, the AOIs could only be placed on the terms which had been found in the glossary. This aspect of my experimental design made it impossible to use eyetracking to explore the cognitive impact of a failed query, which may have yielded interesting results. 6.3.4 Eyetracking methodology Additionally, the metric “time to first fixation”, which I used to contrast the recruitment of visual-spatial resources in the three conditions, is limited in its ability to compare tools with different query mechanisms, such as in this case. The first fixation on the term in the simASR condition is likely to coincide with processing, while the first fixation on terms in the other two conditions may only reflect the individuation of the term on the screen (§6.1.2.4). For this reason, I did not adopt the metric “first fixation duration”, as identifying the first duration reflecting cognitive processing of the term would have probably resulted in mere guesswork. Despite some limitations, the use of the eyetracking methodology in this field of interpreting studies seems promising and may reveal aspects of human-ma- chine interaction with CAI tools still unexplored with different methods. 238 7 Conclusion In Chapter 5, I defined the research gap addressed by the experimental contri- bution and formulated the hypotheses on CL in CASI. Additionally, I motivated the selection of the metrics analysed in the experimental contribution and illus- trated the methodology of the experiment conducted to verify my hypotheses. I first presented the method and objectives of the pilot study carried out to validate the stimuli and methodology adopted in the main experiment. I then discussed the main experiment, going into the details of the participants’ training, of the materials and apparatus used and of data collection and preparation. Chapter 6 presents and discusses the results of the analyses of the experimen- tal data, which overall support the research hypotheses. The chapter also illus- trates the main lacunae of the study.ifl In this final chapter, I offer some reflections on the methodological, didactic and practical implications of the findings of the present study and suggest poten- tial areas of exploration to be addressed in future studies on the relatively new field of computer-assisted (simultaneous) interpreting. §7.1 presents the method- ological implications for future studies on this topic. In §7.2, I address the im- plications for the inclusion of CAI tools with and without integrated ASR in the interpreting curriculum. §7.3 discusses practical implications for the devel- opment and improvement of bespoke support tools for conference interpreters. §7.4 concludes the work by suggesting potential avenues for future CAI research. 7 Conclusion The present study set out to explore the impact of digital terminology support tools on the SI process. Through an experimental contribution comparing tradi- tional digital glossaries, CAI tools with manual look-up and ASR-enhanced CAI tools, the study addressed two main limitations in current CAI research: the al- most exclusive focus on the product of CASI and the lack of a validated method- ology for the exploration of the CASI process. Building upon previous research conducted on the topic, I explored the impact of the different support solutions on the SI process through product-related measures, but I also analysed the use- fulness of additional process-based measures. The present study was therefore the first to adopt a mixed-method approach for the exploration of the cognitive processes of CASI, deriving its methods from TPR, and exploring assumptions on the impact of CAI tools on the process of SI through a systematic approach. Chapter 1 introduced the topic of terminology in conference interpreting and described the relevance of terminology as a quality factor. Additionally, it re- viewed conference interpreters’ habits for terminology preparation pre-assign- ment and for terminology look-up during the assignment emerging from several surveys conducted on the topic. After reviewing the main technologies applied to interpreting, Chapter 2 broadened the analysis to supporting technologies for terminology work, with a focus on the most recent development in this field: CAI tools with and without ASR support. The chapter reviewed current CAI research and discussed open questions not yet addressed in scholarship. As the main aspect of innovation of the present doctoral work consisted in the exploration of technology-supported SI cognition, Chapter 3 discussed interpret- ing from a cognitive standpoint. The chapter illustrated and contrasted relevant theoretical models derived from cognitive psychology and Interpreting Studies and previously applied to experimental research on the interpreting process with CAI tools and multimodal input. Seeber’s CLM of SI with text was applied to the present research object to illustrate the hypothesised differences in the cognitive impact of tools under the three conditions contrasted during the experiment. Chapter 4 discussed the various approaches adopted in Interpreting Studies and TPR to measure CL, discussing the benefits and shortcoming of theoretical approaches as well as subjective, performance, behavioural and physiological methods. 7.1 Methodological implications The main contribution of the present study lies in the development of an exper- imental methodology for the exploration of the CASI process. As described in Chapters 5 and 6, the research object was investigated through the combination of multiple methods for the collection of primarily quantitative data. Building on a previous study which had highlighted the importance of training for effective in-process interaction with CAI tools (Prandi 2015a,b), the pilot study and the main experiment were preceded by a training phase. The choice to switch to an online training may prove particularly useful also in future studies, as it has the potential to reach a higher number of participants while at the same time allowing for a more personalised and flexible, self-paced training process, both in terms of the frequency and of the integration of the training into the subjects’ schedule. The online format may also allow for an easier integration of the participants’ training into the study design as compared to face-to-face training. As for the stimulus material used in the experiment, the preparation of the speeches turned out to be highly time-consuming. Nonetheless, the result is a set 240 7.1 Methodological implications 7.1 Methodological implications of highly controlled, highly comparable speeches which may be used to further explore the cognitive implications of human-machine interaction during inter- preting, for instance in terms of coping tactics adopted during interpreting. In light of the current lack of uniform methods and validated materials for the in- vestigation of the CASI process and product, it is my hope that this material will prove useful also for future studies in the field. With reference to the study design, the adoption of a mixed-method approach appears promising. This approach allowed me to link observations derived from the product-based measures (term accuracy, errors and omissions) with the find- ings from process-related measures derived from TPR. This represents a step forward as compared to previous studies which had adopted only product-based metrics. In addition, the choice to combine several metrics allowed for a deeper understanding of the process of SI with digital terminological support, bringing to light a series of phenomena which would have otherwise remained unknown. The inclusion of multiple measures and methods commonly adopted in TPR to study the translation and interpreting process proved beneficial. 7 Conclusion Since a few institutions have already introduced CAI tools into their curricula, studies involving larger populations of student participants would represent a valuable contribution. However, an important limitation of the present study and of previous CAI research lies in the almost exclusive involvement of trainee in- terpreters in the experiments, as discussed in §6.3.1. Hence, it appears necessary to also include professional conference interpreters in studies on technology- enhanced interpreting. 7.1 Methodological implications Further work is however necessary to refine the measures used and identify additional useful metrics for the investigation of CASI. Following this approach is however likely to produce a large amount of data which requires extensive preparation, thus complicating the analysis. Future studies may therefore choose to focus on individual aspects which the present work has touched upon. The choice of a within-subject design proved valuable to cope with the limited number of participants available. In particular, it allowed for the use of inferential statistics even on a limited population of participants, improving the reliability of the results presented. Larger samples, however, would provide greater confi- dence in the results. A between-subject design, despite requiring accurate participant selection to ensure sample homogeneity, may ensure even greater comparability in the stim- uli used and lend further support to some tentative observations formulated in the present contribution. However, until the use of CAI tools becomes main- stream, it is possible that between-subject designs with sufficiently large samples may be difficult to adopt and that within-subject designs may constitute a more practical choice. Additionally, it may be considered unethical to only provide training to one group, so this should be taken into consideration especially for studies involving students. Furthermore, if we are to establish which tools may best support interpreters during interpreting while interfering the least with the other subprocesses involved in SI, intra-subject explorations may indeed repre- sent a more adequate design choice to address our research questions. 241 7.2 Didactic implications The outcome of the present study presents a series of implications for training, especially because the sample population involved in the experiment and in the pilot study was made up of student interpreters. To start, the study explored the possibility to train students on the tool func- tionalities through distance-learning. I see potential for the development of e- learning modules on this topic. In light of the still limited inclusion of CAI tools into the training of 21st century interpreters, online training may represent an ef- ficient way to promote the development of new digital skills despite bureaucratic or logistic limitations. Especially the logistic limitations which have emerged sig- nificantly in the wake of the COVID-19 pandemic may be partly addressed by distance training. However, it should be remarked that, for the present experi- ment, no didactic intervention aimed to promote the students’ effective use of CAI tools was included in the training module: the participants’ training mostly included self-directed practice sessions. Our currently still limited understanding of the CAI process and of the impact of CAI tool use on the product of SI rep- resents an important limitation for CAI tool training. Nonetheless, as not only training, but also research has had to address the logistic limitations imposed by the pandemic, using online modules may be useful if participants are only to gain sufficient practice in the experimental task ahead of data collection. Anecdotal observations conducted on the students’ interaction with the dif- ferent support tools during the experiment (see §6.1.2.1.7) highlighted a some- times ineffective approach to the use of the tools, emerging for instance as re- peated queries or as queries performed for terms of which the spelling was un- known. These phenomena, combined with a certain over-reliance on the tools observed in the present and previous studies (e.g. Prandi 2015a,b, Van Cauwen- berghe 2020), would suggest that training may not only be necessary to develop the required practical skills in operating the tools in the booth, but also as con- cerns the strategic dimension of the in-booth interaction with CAI tools. In partic- ular, it appears important that students and professionals know the technologies 242 7.2 Didactic implications 7.2 Didactic implications behind ASR-enhanced CAI tools and be able to adjust their expectations on the systems and adopt effective coping tactics in case of system failure. 7.2 Didactic implications f In light of the above, I believe that training should also address the benefits and shortcomings of each type of solution also from a strategic standpoint, i.e., in terms of how the tool selected may affect the process and product of interpreting. An example is the positive, although unsurprising, effect on the EVS observed for the ASR mock-up (§6.2.2.2). As IS literature has found that adjusting the décalage may be an effective tactic to cope with very dense or fast speeches (e.g. Gile 2009), the choice between an ASR-enhanced and a CAI tool with manual look- up may also be motivated by reasons of this kind. At the same time, in §6.2.2.2, I had formulated the hypothesis, which remains to be tested in future studies, that a faster integration of the term into the rendition may also coincide with a syntactic structure closer to the original. For some language pairs, this may result in a less fluent or idiomatic delivery. This, however, also remains to be explored in future research. Overall, I believe that it would be beneficial to discuss the implications of the choice of one tool over other solutions not only in terms of its potential impact on the delivery, but also of its interference with the other cognitive subprocesses of interpreting. The strategic dimension of human-tool interaction may also be addressed at a deeper level, for instance in terms of the query strategy to adopt during interpret- ing for manual look-up. In my study, trigrams were the category of terms that were found more easily in the PDF and CAI condition, presumably because par- ticipants had more linguistic material to use for the query (see §6.2.2.1). On the other hand, participants had more difficulties in finding unigrams. Trainers may therefore draw the students’ attention to this aspect by pointing out effective querying strategies also oriented to the morphology of the terms. Despite the potential for accuracy improvement with support tools, the pres- ence of errors and omissions highlighted the crucial role of preparation in SI. Therefore, guided in-booth experience with the tool may also prove useful to stress the importance of the pre-assignment phase and to ensure a more effec- tive interaction with the tool. 7.2 Didactic implications Finally, in light of the severe errors and complete omissions observed despite the high level of terminological accuracy achieved, trainers may want to raise awareness on the trade-off between the support provided by the machine and the additional effort inherent to human-machine interaction during SI. The suggestions outlined above are not intended to be exhaustive, but were presented as a means of illustration of the potential implications of CAI research for training. At present, these suggestions remain hypotheses based on the ob- servations conducted in the present study. For all its limitations, however, the 243 7 Conclusion present contribution highlighted several issues which, if corroborated by further studies, may offer important implications not only for professional practice, but also for training. If we are to prepare interpreting trainees to effectively use CAI tools in their professional career, it appears necessary to conduct further research to gain a deeper understanding on the impact of CAI tools on the process and product of SI, and to align training with the findings of CAI research. 7.4 Future work In addition to the tentative conclusions formulated in the previous sections, sev- eral aspects have emerged from the present study which deserve to be further explored in future research. One area of research may focus on the impact of the performance of the ASR engine on the product and process of SI. Previous studies have identified the risk of over-reliance on the tool (Prandi 2015a,b, Defrancq & Fantinuoli 2021, Van Cauwenberghe 2020). At the same time, Defrancq & Fantinuoli (2021: 87) re- ported a potentially beneficial psychological effect deriving from the perception of the ASR tool as a “safety net”. Both aspects also emerged from the debrief- ing questionnaire administered to my participants post-study. However, in my experiment, no system failures were simulated for the simASR tool. Explicitly introducing this variable in future studies may provide further insight into the psychological and practical impact of system failures when ASR is provided in the booth. The questionnaire adopted in the study only explored the reception of the tools at a broad level. Nonetheless, it was useful to generate a series of hypotheses, which may be further tested in future explorations of the research object. If CAI tools become a staple in the standard toolkit of conference interpreters, large- scale questionnaires may be used to further explore the reception of such tools and identify potential shortcomings to be addressed by developers and trainers. First steps in this direction have been taken1, but there are still a lot of aspects which deserve further exploration, and which may be investigated in future stud- ies focused on the CAI process. A third avenue, only briefly touched upon in the pilot study, may investigate how the provision of CAI support in the booth affects the tactics used to cope with specialised terminology. In particular, future studies may explore the impact of preparation on the interaction with the tool in the booth, also with the goal to establish which strategies may lead to the best outcome in terms of the trade-off between terminological accuracy and overall quality of the rendition. Additionally, it should be noted that CAI research has thus far mostly focused on the simultaneous mode. With a few exceptions (e.g. Wang & Wang 2019), con- secutive interpreting has not yet been in focus. However, it may be hypothesised that the technologies deployed to support the SI process may also prove valuable for the consecutive mode. 1See for instance the EABM (2021) project. 7.3 Practical implications The findings from the present study, which are in line with similar experiments conducted on ASR-enhanced CAI tools for the in-booth prompting of numbers, may have important implications also for the future development of support tech- nologies for interpreters. If the ASR module does not fail, as was tested in the context of the present experiment, it seems to provide a superior type of support to the trainee interpreter, with the lowest additional cognitive effort and the highest degree of accuracy observed. Against this background and in light of the considerable advances in AI, in particular in the performance of ASR engines, it is safe to assume that AI-enhanced, third generation CAI tools may represent the in-booth support tool of choice for a large number of interpreters in the future. In response to a question in the debriefing questionnaire, some students had highlighted the inability to perform manual queries when needed in the simASR condition as a shortcoming of the system. Therefore, while it seems reasonable to focus on the development of ASR-enhanced CAI tools in the near future, i. e. requiring no active search in the glossary by the interpreter, hybrid systems may also be explored as an option to reconcile the benefits of automatic support with the need to actively query the glossary under certain circumstances. However, some of the benefits observed for the ASR-CAI mock-up may have originated from the specific user interface and terminology presentation mode chosen for the experiment. Both the present study and previous research have highlighted potential pitfalls in interpreter-tool interaction, such as irritating or distracting features, which may be addressed by interventions on the tools’ UI design. Valuable steps in this direction have been taken in TPR to explore the origins of the cognitive friction experienced by translators using CAT tools. CAI research may derive useful methods and insight from this area of TPR, thus in- forming the design of future tools. 244 7.4 Future work 7.4 Future work 7.4 Future work Further research appears necessary to determine the impact of supporting technologies on the process and product of consecutive in- 245 7 Conclusion terpreting, and to establish validated methodologies for the exploration of this research object. Finally, future studies may choose to explore the adoption of other metrics for the investigation of the product and process of SI with CAI tools, with and with- out integrated ASR. For instance, the analysis of the quality of the interpreters’ output may be conducted using a more comprehensive framework going beyond the broad categories of grave error and complete omission chosen for the present study. As discussed in §6.2.1.2, the adaptation of established and standardised quality evaluation frameworks, as has been done for instance in research on hu- man translation and post-editing, may offer a more nuanced picture of the impact of live in-booth support on the quality of SI. Furthermore, it would certainly be interesting to include metrics centred around the presentation of the interpreted texts, such as prosody and pause pat- terns, and to include final users in the evaluation. In addition to cognitive effort, the stress experienced in working with in-process support tools may also be in- vestigated, ideally with a combination of physiological measures and subjective measures. Despite having produced a growing amount of research in recent years, the area of Interpreting Studies focusing on CAI is still in its infancy. The present study was conceived at a time when ASR support for interpreters was just start- ing to emerge. Since then, the application of technology to interpreting has ex- perienced a quantum leap, and an increasing technologisation of the profession may reasonably be expected in light of the increasing uptake of technology by interpreters over the past few years, further accelerated by the recent pandemic. The increasing sophistication and capabilities of technology thanks to the ad- vances in AI may equally be expected to further promote the permeation of tech- nology into the interpreting profession. Despite its focus on a specific set of tech- nological solutions, it is my hope that the methodological contribution offered by the present study may prove valuable for further explorations on technology applied to interpreting, a research subject which is bound to become increasingly relevant for the interpreting profession in the near future, but which currently remains still largely unexplored. 246 Appendix A: Speech A They have the potential to help us respond to the challenges we’re facing. (7) There are other options that make our transport sector less harmful to the environment. By focusing, for instance, on woody biomass fuels, we can truly make a difference. They have the potential to help us respond to the challenges we’re facing. (8) These fuels can be used in road, railroad and also aviation transportation. A real breakthrough in the transport sector could be represented by water splitting. Luckily, there are already many different types of clean fuels that can be used in transport. (8) These fuels can be used in road, railroad and also aviation transportation. A real breakthrough in the transport sector could be represented by water splitting. Luckily, there are already many different types of clean fuels that can be used in transport. (9) As you can imagine, the production of these fuels involves very advanced technologies. New technologies are replacing plug flow reactors, which present some issues. And I think that producing clean fuel shouldn’t harm the environment. (9) As you can imagine, the production of these fuels involves very advanced technologies. New technologies are replacing plug flow reactors, which present some issues. And I think that producing clean fuel shouldn’t harm the environment. (10) I don’t want to bore you too much, so I’m not going to name all kinds of clean fuel. However, I would like to provide you with a good example: rapeseed methyl ester. I know this name may sound intimidating, but it’s actually just fuel. (10) I don’t want to bore you too much, so I’m not going to name all kinds of clean fuel. However, I would like to provide you with a good example: rapeseed methyl ester. I know this name may sound intimidating, but it’s actually just fuel. (10) (11) Clean fuels are obtained through different processes and with different methods. Of the two main ways to obtain them, the most innovative one is transesterification. It’s however a rather complicated chemical process, so I won’t go into too much detail. (11) Clean fuels are obtained through different processes and with different methods. Of the two main ways to obtain them, the most innovative one is transesterification. It’s however a rather complicated chemical process, so I won’t go into too much detail. Appendix A: Speech A (1) Commissioner, thank you once again for inviting me to today’s conference. I would like to start my speech by providing some information about biocrude. At present, the efforts to produce it are still limited, but things are changing. (2) Before talking about its production, let me address some of its chemical properties. Perhaps the most interesting one is the fact that this kind of fuel has a low flash point. This is very important because it means that less energy is required to produce it. (3) I believe we should support research in this field, because the advantages don’t end here. One should also mention that its pour point is also much lower than for other fuels. But this conference is not about chemistry, so let’s move on to the next point. (3) I believe we should support research in this field, because the advantages don’t end here. One should also mention that its pour point is also much lower than for other fuels. But this conference is not about chemistry, so let’s move on to the next point. (4) Now, of course the production of this fuel is quite complex, as is often the case. It first undergoes what is known as hydrotreating, which is quite common in this industry. After that, the fuel is ready to be used in transport instead of more harmful ones. (4) Now, of course the production of this fuel is quite complex, as is often the case. It first undergoes what is known as hydrotreating, which is quite common in this industry. After that, the fuel is ready to be used in transport instead of more harmful ones. (5) To make the production of this fuel possible, various devices have been developed. A good example is the bubbling fluidized bed, which has a pretty long history. It is quite clear, however, that we need to pursue many options in terms of technology. A Speech A (6) These kinds of devices I have just talked about come with some disadvantages. The biggest one is that they require an inert carrier gas to operate, unlike other devices. This might lead to a series of issues in terms of quality of the end product. (7) There are other options that make our transport sector less harmful to the environment. By focusing, for instance, on woody biomass fuels, we can truly make a difference. Appendix A: Speech A (12) I would like to mention, however, that this process presents a lot of advantages. For instance, it makes it possible to reuse materials such as spent bleaching earths. This shows that simple materials can be used to produce clean fuel. (12) 248 (13) Fuel of this kind can be obtained, for example, through a process that extracts oil from seeds. At the end of this process, we have a residue cake, which is what is left after seeds are crushed. Depending on its characteristics, this residue can then be used as a fertiliser or to feed animals. (13) (14) But what are the resources we can use to produce these kinds of fuels? The first category involves, for instance, short rotation coppice. Let me give you some more information before moving on to the other two categories. (14) But what are the resources we can use to produce these kinds of fuels? The first category involves, for instance, short rotation coppice. Let me give you some more information before moving on to the other two categories. (15) These resources are used to produce more efficient and less harmful fuels. Fuels of this sort are still quite new and can be derived, for example, from switchgrass. The new technologies used in this field are expected to help expand production as well. (16) The second category involves resources that are derived from the resources in the first category. These resources can be obtained chemically, as is the case for black liquor. They can, however, also be produced using physical or biological processes. (17) This means that we have a lot of different methods at our disposal. Let me just briefly mention pulping, which will be further discussed this afternoon. I am sure my colleagues will be able to tell you something more about it. (17) This means that we have a lot of different methods at our disposal. Let me just briefly mention pulping, which will be further discussed this afternoon. I am sure my colleagues will be able to tell you something more about it. (18) Unfortunately, we don’t have much time to address this topic in particular. I would like, however, to mention liquid rosin, which I know a bit more about. Unlike the resource I mentioned before, it is the result of a natural process. (18) Unfortunately, we don’t have much time to address this topic in particular. Appendix A: Speech A I would like, however, to mention liquid rosin, which I know a bit more about. Unlike the resource I mentioned before, it is the result of a natural process. (19) The third category is nothing special as it involves all kinds of waste from human activities. Another good example I forgot to give you for the first category is corn stover. As you can see, it is possible to use very common materials to produce fuel. (19) 249 A Speech A (20) There are, however, also other sources from which clean fuels can be derived, such as wheat. To this aim, wheat must undergo two different processes, the first of which is called milling. This simply means that the cereal is cut into much smaller pieces. (21) (21) After undergoing this first process, the second phase begins. An important stage of this process is kilning, which occurs under controlled conditions. So far, I’ve been talking about clean fuels mainly used in the transport system. (22) There is however another field I would like to describe in more detail. A good alternative to coal, which can be used for heating purposes, is lignocellulosic solid biomass. The gas obtained from it can also be used to generate electricity. (23) There are two main processes used to produce the necessary resources for clean heating. The first one is called gasification and what we obtain through it can also be used for cooking. The gas can, however, also be converted to electricity or used for other applications. (24) To perform this process, various devices can be used, although some are preferable. Among these systems, the entrained flow gasifier is an interesting device. It is preferred because it has a lower impact on the environment. (25) (25) After the process, the gas is filtered and can be used to generate power more efficiently. Sometimes, however, the efficiency of the process can be lower due to elutriation. This is why it is very important to focus on devices that limit this issue. (26) Many African countries are already trying to find clean alternatives to traditional coal. One of these is the production of green charcoal, which looks to be very promising. It is an environmentally friendly alternative to traditional charcoal, derived from vegetation. (26) 250 (27) There has been a lot of media coverage about the negative consequences on the environment. Appendix A: Speech A One of the first things that can be done to prevent them is investing in soil amendment. The negative impact of uncontrolled production is, unfortunately, quite wide-ranging. (28) Luckily, there are various methods and tools that we can use to counteract such effects. Let me briefly talk about biochar – not a new invention, but a very useful one. It can endure in soil for many years and has many potential benefits. (28) (29) On the one hand, it can be used to improve the fertility of our soils. On the other hand, however, it may also play an important part in carbon sequestration. This is true especially when coupled with other processes and measures. (30) We have witnessed improvements in many sectors and many countries. In the transportation sector there has been an increase in the use of carbon sequestration. This technology makes trains more efficient and is good for the environment. (30) We have witnessed improvements in many sectors and many countries. In the transportation sector there has been an increase in the use of carbon sequestration. This technology makes trains more efficient and is good for the environment. (31) To produce gas, research and development is exploring new possibilities. In order to maximise production, researchers are working on steam methane reforming. This process comes with a lot of advantages, even though it is quite expensive. (31) To produce gas, research and development is exploring new possibilities. In order to maximise production, researchers are working on steam methane reforming. This process comes with a lot of advantages, even though it is quite expensive. (32) Despite the advantages, the process I just mentioned comes with some challenges. One of these is something called coking, which might have negative effects on production. Additionally, the whole process is still considered quite costly and impractical. (33) We all know that we must also start looking elsewhere if we want to protect the environment. We need to keep investing in tidal barrages, for example, which are proving to be a valid alternative. By investing in this new sector, my country had an annual export of several billion euros a year. (33) 251 A Speech A (34) Let me conclude my speech by describing how we can harvest the energy of ocean waters. The most famous system used to obtain this kind of energy is probably ebb generation. Appendix A: Speech A I guess all of us have heard about it before, so I don’t need to explain this further. (35) We can, however, also capture the energy derived from ocean waves. For instance, we can use point absorber buoys in order to generate electricity. This is, however, only one of the four most common approaches and technologies. (35) (35) We can, however, also capture the energy derived from ocean waves. For instance, we can use point absorber buoys in order to generate electricity. This is, however, only one of the four most common approaches and technologies. (36) Over the past few years, new technological solutions have entered the market. Experts have already been working to develop oscillating water columns. This is, however, still not enough, so Europe must keep investing to bring about real change. Thank you. (36) (36) Over the past few years, new technological solutions have entered the market. Experts have already been working to develop oscillating water columns. This is, however, still not enough, so Europe must keep investing to bring about real change. Thank you. 252 Appendix B: Speech B (1) Ladies and gentlemen, today I will talk about a topic that is very dear to me. I will be talking about transmutation, a process used to generate nuclear energy. It is rather complicated, so I promise not to go into too much detail. (2) Some of you may have already heard this term before, although in a different context. It described the transformation of base metals into gold attempted by alchemists. I find the history of this word very fascinating, but let’s not digress. (2) Some of you may have already heard this term before, although in a different context. It described the transformation of base metals into gold attempted by alchemists. I find the history of this word very fascinating, but let’s not digress. (3) I’m sure you will agree with me that nothing in life is without consequences. Let’s think about what happens when we build a storage dam. It is for sure useful to generate electricity, but also negatively impacts the environment. (4) This is true also for nuclear energy, which generates a lot of waste of different types. To start, I’d like to mention depletalloy, which was long considered unusable. It is only mildly radioactive, but is not, however, the only type of waste we get. (5) After its extraction, the raw material containing uranium is crushed into sand. The useful material is then removed and we are left with mill tailings. They can be carried by the wind and enter our waters, which is not a desirable outcome. (6) This is a complicated topic, because even apparently safe materials can be radioactive. Even though we’re talking about small amounts, this is true also for shale. But this is something that will be addressed by other speakers later today. B Speech B (7) Luckily, however, in this industry there are also less problematic types of waste. You might have guessed that I’m referring to spent nuclear fuel. Even though it can no longer be used directly, it can be repurposed. (7) Luckily, however, in this industry there are also less problematic types of waste. You might have guessed that I’m referring to spent nuclear fuel. Even though it can no longer be used directly, it can be repurposed. (8) So as you can see, even in this controversial sector there’s a positive side of the coin. Appendix B: Speech B Industry is focusing on mixed oxide fuel, which is already widely used. Who would have thought that recycling was possible in this sector, too? (8) So as you can see, even in this controversial sector there’s a positive side of the coin. Industry is focusing on mixed oxide fuel, which is already widely used. Who would have thought that recycling was possible in this sector, too? (9) One of the main advantages of this fuel is that it’s pretty easy to manufacture. It is not used all over the world, but in some countries it powers breeder reactors. This certainly doesn’t solve the problem, but at least reduces the amount of waste. (9) One of the main advantages of this fuel is that it’s pretty easy to manufacture. It is not used all over the world, but in some countries it powers breeder reactors. This certainly doesn’t solve the problem, but at least reduces the amount of waste. (10) Not all waste is equally dangerous and can be classified into three groups. After initial treatment, some waste can be mixed with blast furnace slag. This is, however, not a solution feasible for the most dangerous material. (10) Not all waste is equally dangerous and can be classified into three groups. After initial treatment, some waste can be mixed with blast furnace slag. This is, however, not a solution feasible for the most dangerous material. (11) Let’s take a look at the other two categories that I haven’t discussed yet. The first one isn’t very interesting, while an example for the second is cladding. A problem, however, still remains: at some point you have to get rid of the waste. (11) Let’s take a look at the other two categories that I haven’t discussed yet. The first one isn’t very interesting, while an example for the second is cladding. A problem, however, still remains: at some point you have to get rid of the waste. (12) And this is where things get tricky; this is what causes concern. What makes people think twice is especially the issue of nuclear waste disposal. The waste can indeed remain radioactive for thousands of years. (13) There are, however, solutions to this problem, but not all of them are good. The first one is burying the waste in so-called deep geological repositories. But who would want to have these near where they live? Appendix B: Speech B (14) I am quite sure no one would, and that’s why so many people are against this type of energy. And the waste has to go through vitrification beforehand, so the process is actually not so easy. So as you can see, things are actually not quite as easy as they might seem at first. 254 (15) What I just mentioned isn’t, however, the only solution we could adopt. One additional potential solution is storing the waste in boreholes. But because of its costs and impact on the environment, this is not implemented. (16) Some scientists have already come up with very creative solutions. It may actually be possible to transport the waste into space using mass drivers. The necessary technology is not there yet, but it might be available in the future. (16) Some scientists have already come up with very creative solutions. It may actually be possible to transport the waste into space using mass drivers. The necessary technology is not there yet, but it might be available in the future. (17) As you all know, the waste cannot be eliminated completely. This is the reason why dry cask storage is often used in this field. This system comes in various designs, some of which can also be used for transportation. (17) As you all know, the waste cannot be eliminated completely. This is the reason why dry cask storage is often used in this field. This system comes in various designs, some of which can also be used for transportation. (18) Research is constantly looking for new solutions to the issues I’m discussing today. One such solution is simply storing the waste products in salt domes. Some, however, criticise this solution because they consider it dangerous. (19) (19) Another problem is what to do with the plants once they are no longer active. Among the solutions to this problem, entombment is just one of the many possibilities. It is the least used option because it is very complex and requires continuous surveillance. (20) After a plant has been shut down, some issues must still be solved. Even with no reactions, some decay heat remains, which is very dangerous. For this reason, many people tend to discard the subject very quickly. (20) After a plant has been shut down, some issues must still be solved. Even with no reactions, some decay heat remains, which is very dangerous. Appendix B: Speech B For this reason, many people tend to discard the subject very quickly. (21) The answer to all these issues lies in the process I’m talking about today. It should not be confused with the process called fission, which produces the waste. That process is the one used to create the atomic bomb, such as that of Hiroshima. (21) The answer to all these issues lies in the process I’m talking about today. It should not be confused with the process called fission, which produces the waste. That process is the one used to create the atomic bomb, such as that of Hiroshima. (22) And then of course there’s another process that we all know and learnt about at school. The reaction I’m talking about naturally occurs in stars and is called fusion. (22) (22) And then of course there’s another process that we all know and learnt about at school. The reaction I’m talking about naturally occurs in stars and is called fusion. 255 B Speech B It is the process that takes place in the Sun, at extremely high temperatures. (23) This process can produce energy with a virtually unlimited supply of fuel. For this and other reasons, tritium breeding is already being tested and demonstrated. As you can see, the process I’m discussing is feasible and can help get rid of the waste. (23) (24) It cannot, however, solve the problem completely, so we must keep looking for alternatives. Let me just mention, for instance, airborne wind turbines, which have proven to be very useful. It may not sound like it, but I am in favour of this kind of technology, too. (25) (25) And I particularly appreciate the investments that have been made into it. This has made possible to develop nacelles made from more sustainable materials. This would also reduce costs and make the technology accessible to more countries. (26) I was reading about this the other day and I have to say things look very promising. The reduction in the costs for onshore wind is just one factor in favour of this source of energy. Nonetheless, huge progress has been made in the sector we’re discussing today. (26) (27) Part of this progress is linked with the possibility of using it for other applications. Many countries today are discussing the need to promote cogeneration. Appendix B: Speech B Though not very widespread, this concept has already been applied in various industry sectors. (27) (27) Part of this progress is linked with the possibility of using it for other applications. Many countries today are discussing the need to promote cogeneration. Though not very widespread, this concept has already been applied in various industry sectors. (28) It has significant potential and could open up many new markets in the mid and long term. Partnerships between the public and private sector are underway in the field of desalination. Through this process, the same plant can be used to produce fresh water as well as energy. (28) (29) (29) This is not only good for our communities, but especially for the environment. 256 In fact, we can make full use of brackish water thanks to this process. This is relevant for the Middle East, which will probably invest in the sector we’re discussing today. (30) I believe that if we truly want to protect the environment, we must be flexible and open. We cannot get rid of peaker plants because they are still needed. But I think focusing on new approaches can really help us take the next step. (30)l (31) And this is true not only for clean energy, but also for traditional solutions. Governments should start setting aside funds for fast burst reactors. The technology is already there, but developing it further would help cut costs too. (32) On the other hand, technology is also needed to make sure the plants can last years. That is why research is essentially focused on long-lived components, such as the reactor vessel. This component is not replaced regularly, so it’s essential to ensure it can stand the test of time. (33) (33) I really believe that in the energy sector technology is the key and can help solve many current issues. The industry is working on organically moderated reactors to reach isolated markets. The first prototypes are already available, but there’s still much to be done in this sector. (34) Technology will really be essential to shape the future of the sector that we’re talking about today. At the moment, for example, pressurised water reactors are the most widespread kind. There is only a handful of countries where this is not the case. (35) They are very stable because they tend to produce less power as temperatures increase. Appendix B: Speech B The second most widespread kind is the boiling water reactor. These reactors tend to have uniform designs and are very similar to one another. (35) 257 B Speech B (36) In order to improve sustainability and safety, research is trying to offer new solutions as well. Among the designs currently available, the molten salt reactor is one of the most promising. I won’t go into detail about this, but I hope my speech helped clarify any doubts. Thank you. 258 Appendix C: Speech C (1) Ladies and gentlemen, let me give you some information about the coal industry. The vast majority of the world’s electricity is generated using the vapour power cycle. The main source of heat used in this process is coal, used to generate steam. (2) Its use is likely to continue in certain regions of our planet, even if it’s very polluting. Organic material derived, for instance, from peatlands, can also be used as fuel. We should, however, be very careful when we use natural resources. (2) Its use is likely to continue in certain regions of our planet, even if it’s very polluting. Organic material derived, for instance, from peatlands, can also be used as fuel. We should, however, be very careful when we use natural resources. (3) Mother nature is truly very generous, even though we often forget it. When we think of the existence of carbon sinks, this becomes immediately clear. But let’s move on to explaining how the cycle I’m talking about works. (3) Mother nature is truly very generous, even though we often forget it. When we think of the existence of carbon sinks, this becomes immediately clear. But let’s move on to explaining how the cycle I’m talking about works. (4) I guess part of the audience will already be familiar with this, but I’ll just describe it briefly. First of all, a large grinder turns the coal that has been transported to the plant into a very fine powder. This step is necessary to make sure that all the coal is burnt to maximize the production of heat. (5) Unfortunately, this cycle requires a lot of water, so it’s not good for the environment. Before moving on to the next step, the water can be transported to a feedwater heater. This additional step in the cycle increases the efficiency of the system. (6) It can, however, also lead to a pretty big issue, which shouldn’t be forgotten. Serious damage can be caused to the plant if the water doesn’t go through a deaerator. It comes in very different types, but they all serve the same purpose. C Speech C (7) Let’s now move on to describing the other steps and elements of the cycle. After the first step, the coal ends up in a boiler, where it is burnt. Appendix C: Speech C This way, the coal provides heat to the power plant, which is used to produce steam. (8) I know the audience isn’t made up of engineers, so I won’t be too technical. However, I think I should also mention the importance played by the boiler steam drum. It’s exciting to see so much technical progress being made in this respect (8) I know the audience isn’t made up of engineers, so I won’t be too technical. However, I think I should also mention the importance played by the boiler steam drum. It’s exciting to see so much technical progress being made in this respec (9) In the cycle I’m describing, there are also optional steps the steam can go through. For instance, the steam can enter a superheater, which can be pretty small or very big. The main advantage of using it is that it reduces the consumption of water. (9) In the cycle I’m describing, there are also optional steps the steam can go through. For instance, the steam can enter a superheater, which can be pretty small or very big. The main advantage of using it is that it reduces the consumption of water. (10) There are, of course, also clean ways of generating energy using steam. This requires using different technologies, such as the one involving flash steam. Nonetheless, coal is still widely used and this comes with huge environmental issues. (11) So focusing on coal to generate energy is not a feasible solution, and let me tell you why. Coal has the most substantial carbon footprint of all fuels. For this reason, this source of energy will have to be used less and less. (12) As you can see, what this industry is promoting is not a real solution. I cannot deny that there are ways to reduce its impact, like using baghouses. But it still remains a messy business and a pretty complex one. (13) So before discussing the desirable alternatives, let’s try to understand it better. Among the different methods used to extract coal, we can mention longwall mining. It is the one that makes best use of the natural resources present underground. (13) 260 (14) As we all know, coal is not readily available on the surface. In order to reach it, the overburden must first be removed. This is done mechanically nowadays, but this wasn’t the case in the past. Appendix C: Speech C (15) Not all of the coal is used to produce heat and, ultimately, electricity. Part of it can be used in other industries, for example in the case of middlings. This is, however, a totally different topic, so let’s move on. (16) The coal doesn’t reach the plant immediately; its journey is a bit longer. In order to remove all impurities, the coal must first reach a coal handling plant. This increases its value and at the same time lowers its transportation costs. (17) To achieve better quality, the coal must undergo a series of processes to be purified. One such process requires dense medium separation to discard useless material. There are many different types of it, but we don’t need to name them. (17) To achieve better quality, the coal must undergo a series of processes to be purified. One such process requires dense medium separation to discard useless material. There are many different types of it, but we don’t need to name them. (18) I’m really just trying to give you an idea of the complexity of this industry. Coal contains water, so a screen bowl centrifuge can be used to remove it. The good thing about this process is that water can also be recycled. (18) I’m really just trying to give you an idea of the complexity of this industry. Coal contains water, so a screen bowl centrifuge can be used to remove it. The good thing about this process is that water can also be recycled. (19) (19) When it comes to this industry, however, there are really a lot of negative aspects. Let’s think, for instance, of the many lethal accidents caused by firedamp. This industry is extremely dangerous for the workers, as I’m sure you all know. (19) When it comes to this industry, however, there are really a lot of negative aspects. Let’s think, for instance, of the many lethal accidents caused by firedamp. This industry is extremely dangerous for the workers, as I’m sure you all know. (20) Using coal to generate electricity also has a devastating impact on the environment. Let me just mention the amount of flue ash that contaminates our waters. And we all know how precious water is, especially in today’s world. (20) Using coal to generate electricity also has a devastating impact on the environment. Appendix C: Speech C Let me just mention the amount of flue ash that contaminates our waters. And we all know how precious water is, especially in today’s world. (21) Plants using coal as fuel are also the biggest water polluters, at least in my country. Every day, these plants dump millions of litres of sullage into rivers and lakes. So it’s quite clear that we need to use clean sources of energy, like that of the sun. 261 C Speech C (22) Solar energy is within the reach of individuals and small businesses. It is expected to play a major role in the future power generation mix. There have been a lot of developments in this field, so let’s discuss them. (23) (23) First of all, research is focusing on improving some components of the system. For instance, industry seems to be increasingly in favour of negatively doped wafers. This is by far the best choice available on the market, because it’s the most efficient. (24) They are, however, still quite expensive, because there are only a few manufacturers. The recent introduction of interdigitated back contacts has made things easier. They make the cells more efficient and easier to place closer together. (24) (25) It is also very important that storage is efficient and nothing goes to waste. This has been made possible with the creation of batteries with better float life. There is a wide range of devices used to harvest the energy of the sun. (26) (26) They are usually classified into three generations, depending on the basic materials used. The first generation uses crystalline silicon, the material most commonly used in this industry. This is a mature technology that currently dominates the market and is in mass production. (26) They are usually classified into three generations, depending on the basic materials used. The first generation uses crystalline silicon, the material most commonly used in this industry. This is a mature technology that currently dominates the market and is in mass production. (27) (27) Despite being an advanced technology, its costs can be reduced through improvements in materials. Manufacturing also plays a role, for example when it comes to the production of boules. This is however not the only technology available on this market at the moment. (27) Despite being an advanced technology, its costs can be reduced through improvements in materials. Appendix C: Speech C Manufacturing also plays a role, for example when it comes to the production of boules. This is however not the only technology available on this market at the moment. (28) After many years of research and development, a new technology is beginning to be deployed. The second generation focuses on another kind of technology called thin film. It could potentially provide lower-cost electricity than first generation cells. (28) After many years of research and development, a new technology is beginning to be deployed. The second generation focuses on another kind of technology called thin film. It could potentially provide lower-cost electricity than first generation cells. (28) 262 (29) Three primary types of this kind of cells have been commercially developed, using different materials. The most widely known type is based on amorphous silicon, which has pretty good efficiency. Then there are other technologies, some of which are still in the early stage of development. (30) I’m going to talk about three of the four technologies of the third generation. The first of these new technologies are the so-called concentrating photovoltaic systems. They offer a pretty high level of efficiency, which is even higher in experimental settings. (30) (31) Let’s now move on to another solution that is part of third generation technology. I’m talking about evacuated tube collectors, which use low-cost materials. They are also simple to manufacture, but their performance can degrade over time. (32) Then there are a number of innovative technologies, all very promising. Good examples of those technologies are the ones that rely on superlattices. Now I’m not going to into detail about this, because it’s rather complicated. (33) It is beside the point I want to make, but I think it’s always interesting to know what’s out there. Another type of these new, third generation technologies is based on quantum dots. These are very small particles that can conduct electricity and are used in other fields too. (33) It is beside the point I want to make, but I think it’s always interesting to know what’s out there. Another type of these new, third generation technologies is based on quantum dots. These are very small particles that can conduct electricity and are used in other fields too. (34) Finally, there are some ideas that haven’t yet been developed fully. Some of those have a huge potential, like solar updraft towers. Appendix C: Speech C They have been criticised in the past, but could become increasingly relevant. (35) A very positive thing about this technology is that it is very versatile. Let’s think for example of floating solar arrays, which can be installed on water. They represent a new development and make this technology more accessible. 263 C Speech C (36) There are many positive aspects when it comes to these technologies. The most evident one is that they allow us to easily reach socket parity. So as you can see this is a good alternative to coal, which we should focus on in the future. Thank you. 264 References Ahrens, Barbara. 2005. Prosodic phenomena in simultaneous interpreting: A con- ceptual approach and its practical application. Interpreting: International Jour- nal of Research and Practice in Interpreting 7(1). 51–76. DOI: 10.1075/intp.7.1. 04ahr. Ahrens, Barbara. 2007. Pauses (and other prosodic features) in simultaneous in- terpreting. FORUM: Revue internationale d’interprétation et de traduction / In- ternational Journal of Interpretation and Translation 5(1). 1–18. DOI: 10.1075/ forum.5.1.01ahr. AIIC. 2002. Interpreter workload study: Full report. https://aiic.org/document/ 468/AIICWebzine_FebMar2002_7_AIIC_Interpreter_workload_study_full_ report_WLS_Full_Report.pdf. Altman, Janet. 1994. Error analysis in the teaching of simultaneous interpretation: A pilot study. In Sylvie Lambert & Barbara Moser-Mercer (eds.), Bridging the gap: Empirical research in simultaneous interpretation, 25–48. Amsterdam: John Benjamins. DOI: 10.1075/btl.3.05alt. Alves, Fábio, Adriana Pagano & Igor da Silva. 2012. Towards an investigation of reading modalities in/for translation: An exploratory study using eye-tracking data. In Sharon O’Brien (ed.), Cognitive explorations of translation, 175–196. London: Bloomsbury. Alves, Fábio, Karina Sarto Szpak, José Luiz Gonçalves, Kyoko Sekino, Marceli Aquino, Rodrigo Araújo e Castro, Arlene Koglin, Norma B. de Lima Fonseca & Bartolomé Mesa-Lao. 2016. Investigating cognitive effort in post-editing: A relevance-theoretical approach. In Silvia Hansen-Schirra & Sambor Grucza (eds.), Eyetracking and applied linguistics, 109–142. Berlin: Language Science Press. DOI: 10.17169/langsci.b108.296. Ammon, Ulrich. 2011. The dominance of English as a language of science. Berlin: De Gruyter Mouton. Andersson, Bodil, Johan Dahl, Kenneth Holmqvist, Jana Holsanova, Victoria Jo- hansson, Henrik Karlsson, Sven Strömqvist & Åsa Wengelin. 2006. Combin- ing keystroke logging with eye-tracking. In Luuk Van Waes, Marielle Leiten & Christophe Neuwirth (eds.), Writing and digital media, 45–72. Leiden: Brill. References Apfelbaum, Birgit & Cecilia Wadensjö. 1997. How does a verbmobil affect con- versation? Discourse analysis and machine-supported translatory interaction. In Christa Hauenschild & Susanne Heizmann (eds.), Machine translation and translation theory, 93–122. Berlin: De Gruyter Mouton. Arcara, Giorgio, Carlo Semenza & Valentina Bambini. 2014. Word structure and decomposition effects in reading. Cognitive Neuropsychology 31(1-2). 184–218. DOI: 10.1080/02643294.2014.903915. Arntz, Reiner & Heribert Picht. 1982. Einführung in die übersetzungsbezogene Ter- minologiearbeit. Hildesheim: Georg Olms Verlag.f Atkinson, Richard C. & Richard M. Shiffrin. 1968. Human memory: A proposed system and its control processes. In Kenneth W. Spence & Janet Taylor Spence (eds.), The psychology of learning and motivation: Advances in research and the- ory, vol. 2, 89–195. New York: Academic Press. Atkinson, Richard C. & Richard M. Shiffrin. 1971. The control of short-term mem- ory. Scientific American 225(2). 82–90. DOI: 10.1038/scientificamerican0871-82. i Audacity Team. 2021. References Audacity®: Free audio editor and recorder. https : / / audacityteam.org/ (17 March, 2021). Baddeley, Alan D. 1975. Word length and the structure of short-term memory. Journal of Verbal Learning and Verbal Behavior 14(6). 575–589. DOI: 10.4324/ 9781315111261-11. Baddeley, Alan D. 1990. Human memory: Theory and practice. Needham Heights: Allyn & Bacon. Baddeley, Alan D. 2000. Short-term and working memory. In Endel Tulving & Fergus I. M. Craik (eds.), The Oxford handbook of memory, 77–92. New York: Oxford University Press. Baddeley, Alan D. 2012. Working memory: Theories, models, and controversies. Annual Review of Psychology 63(1). 1–29. DOI: 10.1146/annurev-psych-120710- 100422. Baddeley, Alan D. & Graham Hitch. 1974. Working memory. In Gordon H. Bower (ed.), Psychology of learning and motivation, vol. 8, 47–89. New York: Academic Press. Baddeley, Alan D., Graham Hitch & Philip T. Quinlan. 2018. Is the phonological similarity effect in working memory due to proactive interference? Journal of Experimental Psychology: Learning, Memory, and Cognition 44(8). 1312–1316. DOI: 10.1037/xlm0000509. Baddeley, Alan D., Giuseppe Vallar & Barbara Wilson. 1987. Sentence comprehen- sion and phonological memory: Some neuropsychological evidence. In Max Coltheart (ed.), Attention and performance 12: The psychology of reading, 509– 529. Hillsdale: Lawrence Erlbaum Associates, Inc. 266 Baigorri-Jalón, Jesús. 1999. Conference interpreting: From modern times to space technology. Interpreting: International Journal of Research and Practice in Inter- preting 4(1). 29–40. DOI: 10.1075/intp.4.1.05bai. Barik, Henri C. 1971. A description of various types of omissions, additions and errors of translation encountered in simultaneous interpretation. Meta: Jour- nal des Traducteurs/Meta: Translators’ Journal 16(4). 199–210. DOI: 10.7202/ 001972ar. Barik, Henri C. 1973. Simultaneous interpretation: Temporal and quantitative data. Language and Speech 16(3). 237–270. DOI: 10.1177/002383097301600307. Barrouillet, Pierre, Sophie Bernardin & Valérie Camos. 2004. Time constraints and resource sharing in adults’ working memory spans. Journal of Experimen- tal Psychology: General 133(1). 83–100. DOI: 10.1037/0096-3445.133.1.83. Barton, Jason J. S., Hashim M. Hanif, Laura Eklinder Björnström & Charlotte Hills. 2014. The word-length effect in reading: A review. Cognitive Neuropsychology 31(5-6). 378–412. DOI: 10.1080/02643294.2014.895314. Beatty, Jackson. 1982. Task-evoked pupillary responses, processing load, and the structure of processing resources. Psychological Bulletin 91(2). 276–292. DOI: 10.1037/0033-2909.91.2.276. Behl, Holly. 2013a. The paperless interpreter experiment: Part I. http : / / www . precisolanguage.com/2013/01/paperless-interpreter-part-i/ (3 July, 2021). Behl, Holly. 2013b. The paperless interpreter experiment: Part II. http : / / www . precisolanguage.com/tag/consecutive-interpreting/ (3 July, 2021). Berber-Irabien, Diana. 2010. Information and communication technologies in con- ference interpreting. Tarragona: Universitat Rovira i Virgili. References (Doctoral disserta- tion). Biagini, Giulio. 2015. Glossario cartaceo e glossario elettronico durante l’interpretazione. Trieste: Università di Trieste. (MA thesis). Bilgen, Baris. 2009. Investigating terminology management for conference inter- preters. Ottawa: University of Ottawa. (MA thesis). Blancafort, Helena, Béatrice Daille, Tatiana Gornostay, Ulrich Heid, Claude Mé- choulam & Serge Sharoff. 2010. TTC: Terminology extraction, translation tools and comparable corpora. In 14th EURALEX International Congress, 263–268. Leeuwarden. Blancafort, Helena, Ulrich Heid, Tatiana Gornostay, Claude Méchoulam, Béatrice Daille & Serge Sharoff. 2011. User-centred views on terminology extraction tools: Usage scenarios and integration into MT and CAT tools. In Tralogy I. Métiers et technologies de la traduction: Quelles convergences pour l’avenir?, 1– 10. 267 References Blumenthal, Pamela, Thomas W. Britt, Jason A. Cohen, James McCubbin, Nathan Maxfield, Erica B. Michael, Philip Moore, Loraine K. Obler, Petra Scheck, Teresa M. Signorelli & Thomas S. Wallsten. 2006. Stress effects on bilingual lan- guage professionals’ performance. International Journal of Bilingualism 10(4). 477–497. DOI: 10.1177/13670069060100040501. Braun, Peter, Burkhard Schaeder & Johannes Volmert (eds.). 2014. International- ismen II: Studien zur interlingualen Lexikologie und Lexikographie. Berlin: Max Niemeyer Verlag. DOI: 10.1515/9783110911978. Braun, Sabine. 2019. Technology and interpreting. In Minako O’Hagan (ed.), The Routledge handbook of translation and technology, 271–288. London: Routledge. Broadbent, Donald E. 1958. Perception and communication. Oxford: Pergamon Press Braun, Peter, Burkhard Schaeder & Johannes Volmert (eds.). 2014. International- ismen II: Studien zur interlingualen Lexikologie und Lexikographie. Berlin: Max Niemeyer Verlag. DOI: 10.1515/9783110911978. Braun Sabine 2019 Technology and interpreting In Minako O’Hagan (ed ) The Routledge handbook of translation and technology, 271–288. London: Routledge. Broadbent, Donald E. 1958. Perception and communication. Oxford: Pergamon Press. Broadbent, Donald E. 1958. Perception and communication. Oxford: Pergamon Press. Brüsewitz, Nora. 2019. Simultandolmetschen 4.0: Ist automatische Spracherken- nung der nächste Schritt? In Wolfram Bauer & Felix Mayer (eds.), Proceedings of Übersetzen und Dolmetschen 4.0.: Neue Wege im digitalen Zeitalter. Berlin: BDÜ Fachverlag. Bruya, Brian & Yi-Yuan Tang. 2018. Is attention really effort? Revisiting Daniel Kahneman’s influential 1973 book Attention and effort. Frontiers in Psychology 9. 1–10. DOI: 10.3389/fpsyg.2018.01133. Bühler, Hildegund. 1986. Linguistic (semantic) and extra-linguistic (pragmatic) criteria for the evaluation of conference interpretation and interpreters. Mul- tilingua 5(4). 231–235. DOI: 10.1515/mult.1986.5.4.231. Bundgaard, Kristine, Tina Paulsen Christensen & Anne Schjoldager. 2016. Translator-computer interaction in action: An observational process study of computer-aided translation. The Journal of Specialised Translation 25. 106–130. Cabré Castellví, Maria Teresa. 1998. Do we need an autonomous theory of terms? Terminology 5(1). References 4–19. DOI: 10.1075/term.5.1.03cab. Cabré Castellví, Maria Teresa. 2003. Theories of terminology: Their description, prescription and explanation. Terminology: International Journal of Theoretical and Applied Issues in Specialized Communication 9(2). 163–199. DOI: 10.1075/ term.9.2.03cab. Campo, Ángela. 2005. The politics of scientific and technical translation. Newslet- ter of the Canadian Association for Translation Studies 18(1). Canali, Sara. 2018. Utilizzo del riconoscimento vocale come supporto durante l’interpretazione simultanea dei numeri. Roma: Università degli Studi Inter- nazionali di Roma. (MA thesis). Carl, Michael, Akiko Aizawa & Masaru Yamada. 2016. English-to-Japanese trans- lation vs. dictation vs. post-editing: Comparing translation modes in a multi- lingual setting. In Proceedings of the tenth international conference on language 268 resources and evaluation (LREC’16), 4024–4031. Portorož: European Language Resources Association (ELRA). https://aclanthology.org/L16- 1635 (25 May, 2021). Carl, Michael & Matthias Buch-Kromann. 2010. Correlating translation product and translation process data of professional and student translators. In Pro- ceedings of the 14th annual conference of the European Association for Machine Translation, 1–8. Saint Raphaël: European Association for Machine Translation. https://aclanthology.org/2010.eamt-1.14 (24 July, 2021). Carl, Michael & Gyde Hansen. 2011. Digital humanities and empirical human translation process research: Past, present and future perspectives. In Bente Maegaard (ed.), Conference proceedings: Supporting digital humanities, Copen- hagen 17–18 November 2011, 1–7. Croatia: Croatian National Corpus. Carl, Michael, Isabel Lacruz, Masaru Yamada & Akiko Aizawa. 2016a. Compar- ing spoken and written translation with post-editing in the ENJA15 English à Japanese Translation Corpus. Paper presented at the 22nd Annual Meeting of the Association for Natural Language Processing, 1–4. Sendai: NLP 2016. Carl, Michael, Isabel Lacruz, Masaru Yamada & Akiko Aizawa. 2016b. Measuring the translation process. Paper presented at the 22nd Annual Meeting of the Association for Natural Language Processing, 1–4. Sendai: NLP 2016. Cavallo, Patrizia & Luis Eduardo Schild Ortiz. 2018. Computer-assisted interpret- ing tools (CAI) and options for automation with automatic speech recognition. Tradterm 32. 9–31. DOI: 10.11606/issn.2317-9511.v32i0p9-31. Cecot, Michela. 2001. Pauses in simultaneous interpretation: A contrastive anal- ysis of professional interpreters’ performances. The Interpreters’ Newsletter 11. 63–85. http://hdl.handle.net/10077/2448 (24 July, 2021). Center for Reading Research, Ghent University. 2014. Wordornot. http : / / vocabulary.ugent.be/ (14 September, 2021). Chabasse, Catherine. 2009. Gibt es eine Begabung für das Simultandolmetschen? Erstellung eines Dolmetscheignungstests mit Schwerpunkt Simultandolmetschen (Beiträge zur Translationswissenschaft 4). Berlin: SAXA Verlag. Chandler, Paul & John Sweller. 1991. Cognitive load theory and the format of instruction. Cognition and Instruction 8(4). 293–332. DOI: 10 . 1207 / s1532690xci0804_2. References Chang, Chieh Ying. 2009. Testing applicability of eye-tracking and fMRI to trans- lation and interpreting studies: An investigation into directionality. London: Im- perial College London. (PhD Thesis). http://hdl.handle.net/10044/1/11409 (25 June, 2021). Chen, Sijia. 2017. The construct of cognitive load in interpreting and its measure- ment. Perspectives 25(4). 640–657. DOI: 10.1080/0907676X.2016.1278026. 269 References Chernov, Ghelly V. 1979. Semantic aspects of psycholinguistic research in si- multaneous interpretation. Language and Speech 22(3). 277–295. DOI: 10.1177/ 002383097902200308. Chernov, Ghelly V. 1994. Message redundancy and message anticipation in simul- taneous interpreting. In Sylvie Lambert & Barbara Moser-Mercer (eds.), Bridg- ing the gap: Empirical research in simultaneous interpretation, 139–153. Amster- dam: John Benjamins. DOI: 10.1075/btl.3.13che. Chesterman, Andrew. 2016. Memes of translation: The spread of ideas in translation theory. Amsterdam: John Benjamins. DOI: 10.1075/btl.123. Chiaro, Delia & Giuseppe Nocella. 2004. Interpreters’ perception of linguistic and non-linguistic factors affecting quality: A survey through the world wide web. Meta: Journal des Traducteurs/Meta: Translators’ Journal 49(2). 278–293. DOI: 10.7202/009351ar. Chmiel, Agnieszka, Przemysław Janikowski & Anna Cieślewicz. 2020. The eye or the ear? Source language interference in sight translation and simultane- ous interpreting. Interpreting: International Journal of Research and Practice in Interpreting 22(2). 187–210. DOI: 10.1075/intp.00043.chm. Chmiel, Agnieszka, Przemysław Janikowski & Agnieszka Lijewska. 2020. Mul- timodal processing in simultaneous interpreting with text: Interpreters focus more on the visual than the auditory modality. Target: International Journal of Translation Studies 32(1). 37–58. DOI: 10.1075/target.18157.chm. Chmiel, Agnieszka & Iwona Mazur. 2013. Eye tracking sight translation per- formed by trainee interpreters. In Catherine Way, Sonia Vandepitte, Reine Meylaerts & Magdalena Bartłomiejczyk (eds.), Tracks and treks in translation studies, 189–205. Amsterdam: John Benjamins. DOI: 10.1075/btl.108.10chm. Cho, Eunah, Christian Fügen, Teresa Herrmann, Kevin Kilgour, Mohammed Me- diani, Christian Mohr, Jan Niehues, Kay Rottmann, Christian Saam, Sebastian Stüker & Alex Waibel. 2013. A real-world system for simultaneous transla- tion of German lectures. In Proc. Interspeech 2013, 3473–3477. DOI: 10.21437/ Interspeech.2013-612. Christoffels, Ingrid K. & Annette M. B. De Groot. 2004. Components of si- multaneous interpreting: Comparing interpreting with shadowing and para- phrasing. Bilingualism: Language and Cognition 7(3). 227–240. DOI: 10.1017/ S1366728904001609. Conklin, Kathy, Ana Pellicer-Sánchez & Gareth Carrol. 2018. Eye-tracking: A guide for applied linguistics research. Cambridge: Cambridge University Press. DOI: 10.1017/9781108233279. Cooper, Alan. 2004. The inmates are running the asylum: Why high tech products drive us crazy and how to restore the sanity. Indianapolis: Sams. 270 Corpas Pastor, Gloria & Lily May Fern. 2016. References A survey of interpreters’ needs and practices related to language technology. Technical paper. https://www. researchgate.net/publication/303685153_A_survey_of_interpreters%5C%27_ needs_and_practices_related_to_language_technology (24 July, 2021). Costa, Albert, Alfonso Caramazza & Nuria Sebastian-Galles. 2000. The cognate facilitation effect: Implications for models of lexical access. Journal of Exper- imental Psychology: Learning, Memory, and Cognition 26(5). 1283–1296. DOI: 10.1037/0278-7393.26.5.1283. Costa, Hernani, Gloria Corpas Pastor & Isabel Durán Muñoz. 2014a. Technology- assisted interpreting. MultiLingual 143 25(3). 27–32. https://multilingual.com/ articles/technology-assisted-interpreting/ (24 July, 2021). Costa, Hernani, Gloria Corpas Pastor & Isabel Durán Muñoz. 2014b. A compara- tive user evaluation of terminology management tools for interpreters. In Pro- ceedings of the 4th international workshop on computational terminology (Com- puterm), 68–76. Dublin: Association for Computational Linguistics & Dublin City University. DOI: 10.3115/v1/W14-4809. Costa, Hernani, Gloria Corpas Pastor & Isabel Durán Muñoz. 2015. An inter- preters’ guide to selecting terminology management tools. In NATO Confer- ence on Terminology Management. Brussels. Cowan, Nelson. 1988. Evolving conceptions of memory storage, selective atten- tion, and their mutual constraints within the human information-processing system. Psychological Bulletin 104(2). 163–191. DOI: 10.1037/0033-2909.104.2. 163. Cowan, Nelson. 2009. Sensory and immediate memory. In William P. Banks (ed.), Encyclopedia of consciousness, 327–339. Amsterdam: Elsevier. DOI: 10 . 1016 / B978-012373873-8.00048-7. Čulo, Oliver, Silvia Hansen-Schirra & Jean Nitzke. 2017. Contrasting terminologi- cal variation in post-editing and human translation of texts from the technical and medical domain. In Gert De Sutter, Marie-Aude Lefer & Isabelle Delaere (eds.), Empirical translation studies: New methodological and theoretical tradi- tions, 183–206. Berlin: De Gruyter Mouton. Darling, Stephen, Richard J. Allen & Jelena Havelka. 2017. Visuospatial bootstrap- ping: When visuospatial and verbal memory work together. Current Directions in Psychological Science 26(1). 3–9. DOI: 10.1177/0963721416665342. Darling, Stephen & Jelena Havelka. 2010. Visuospatial bootstrapping: Evidence for binding of verbal and spatial information in working memory. Quar- terly Journal of Experimental Psychology 63(2). 239–245. DOI: 10 . 1080 / 17470210903348605. 271 References Darò, Valeria & Franco Fabbro. 1994. Verbal memory during simultaneous inter- pretation: Effects of phonological interference. Applied Linguistics 15(4). 365– 381. DOI: 10.1093/applin/15.4.365. Darò, Valeria. 1994. Non-linguistic factors influencing simultaneous interpreta- tion. In Sylvie Lambert & Barbara Moser-Mercer (eds.), Bridging the gap: Em- pirical research in simultaneous interpretation, 249–271. Amsterdam: John Ben- jamins. DOI: 10.1075/btl.3.20dar. Darò, Valeria. 1997. Experimental studies on memory in conference interpreta- tion. Meta: Journal des Traducteurs/Meta: Translators’ Journal 42(4). 622–628. DOI: 10.7202/002484ar. De Merulis, Gianpiero. 2013. References L’uso di InterpretBank per la preparazione di una con- ferenza sul trattamento delle acque reflue: Glossario terminologico e contributo sperimentale. Forlì: Università di Bologna. (MA thesis). Defrancq, Bart. 2015. Corpus-based research into the presumed effects of short EVS. Interpreting: International Journal of Research and Practice in Interpreting 17(1). 26–45. DOI: 10.1075/intp.17.1.02def. Defrancq, Bart & Claudio Fantinuoli. 2021. Automatic speech recognition in the booth: Assessment of system performance, interpreters’ performances and in- teractions in the context of numbers. Target 33(1). 73–102. DOI: 10.1075/target. 19166.def. Déjean Le Féal, Karla. 1980. Die Satzsegmentierung beim freien Vortrag bzw. beim Verlesen von Texten und ihr Einfluss auf das Sprachverstehen. Sprache und Verstehen 1. 161–168. Déjean Le Féal, Karla. 1981. Lectures et improvisations: Incidences de la forme de l’énonciation sur la traduction simultanée. Fremdsprachen 1(81). 29–32. Déjean Le Féal, Karla. 1990. Some thoughts on the evaluation of simultaneous interpretation. In David Bowen & Margareta Bowen (eds.), Interpreting: Yester- day, today, and tomorrow, 154–160. Binghamton: John Benjamins. DOI: 10.1075/ ata.iv.27lef. de Saussure, Ferdinand. 1959. Course in general linguistics. New York: Philosoph- ical Library. Desblache, Lucile. 2001. Aspects of specialised translation. Paris: La Maison du Dictionnaire. Desmet, Bart, Mieke Vandierendonck & Bart Defrancq. 2018. Simultaneous in- terpretation of numbers and the impact of technological support. In Claudio Fantinuoli (ed.), Interpreting and technology, 13–27. Berlin: Language Science Press. DOI: 10.5281/zenodo.1493291. 272 Dessloch, Florian, Thanh-Le Ha, Markus Müller, Jan Niehues, Thai-Son Nguyen, Ngoc-Quan Pham, Elizabeth Salesky, Matthias Sperber, Sebastian Stüker, Thomas Zenkel & Alexander Waibel. 2018. Kit lecture translator: Multilingual speech translation with one-shot learning. In Zhao Dongyan (ed.), Proceed- ings of the 27th International Conference on Computational Linguistics: System Demonstrations, 89–93. Santa Fe: Association for Computational Linguistics. Deutsch, J. Anthony & Diana Deutsch. 1963. Attention: Some theoretical consid- erations. Psychological Review 70(1). 80–90. DOI: 10.1037/h0039515. Díaz-Galaz, Stephanie & Alejandro Torres. 2019. Comprehension in interpreting and translation: Testing the phonological interference hypothesis. Perspectives 27(4). 622–638. DOI: 10.1080/0907676X.2019.1569699. Donovan, Clare. 1990. La fidélité en interprétation. Paris: Université de la Sor- bonne Nouvelle/Paris III. (Doctoral dissertation). Dragsted, Barbara. 2004. Segmentation in translation and translation memory sys- tems: An empirical investigation of cognitive segmentation and effects of integrat- ing a TM system into the translation process (Ph.D. series 2004). Copenhagen: Samfundslitteratur. Dragsted, Barbara. 2010. Coordination of reading and writing processes in trans- lation: An eye on uncharted territory. In Gregory M. Shreve & Erik Angelone (eds.), Translation and cognition, 41–62. Amsterdam: John Benjamins. References DOI: 10. 1075/ata.xv.04dra. Dragsted, Barbara. 2012. Indicators of difficulty in translation: Correlating prod- uct and process data. Across Languages and Cultures 13(1). 81–98. DOI: 10.1556/ Acr.13.2012.1.5. Dragsted, Barbara & Inge Gorm Hansen. 2008. Comprehension and production in translation: A pilot study on segmentation and the coordination of reading and writing processes. In Susanne Göpferich, Arnt Lykke Jakobsen & Inger M. Mees (eds.), Looking at eyes: Eye-tracking studies of reading and translation processing (Copenhagen Studies in Language), 9–29. Frederiksberg: Samfund- slitteratur. Dragsted, Barbara & Inge Gorm Hansen. 2009. Exploring translation and in- terpreting hybrids. The case of sight translation. Meta: Journal des Traduc- teurs/Meta: Translators’ Journal 54(3). 588–604. DOI: 10.7202/038317ar. Drechsel, Alexander. 2004. Computereinsatz beim Dolmetschen in Ausbildung und Praxis: Ein Beitrag zur Dolmetschdidaktik. Leipzig: Universität Leipzig. (MA thesis). Drechsel, Alexander & Joshua Goldsmith. 2016. Tablet interpreting: The evolu- tion and uses of mobile devices in interpreting. In Hannelore Lee-Jahnke & Martin Forstner (eds.), CIUTI-forum 2016. Geneva. 273 References Drewer, Petra & Donatella Pulitano (eds.). 2019. Terminologie: Epochen– Schwerpunkte–Umsetzungen. Berlin: Springer. Duchowski, Andrew T. 2017. Eye tracking methodology: Theory and practice. Lon- don: Springer. EABM. 2021. Survey - Ergonomics for the artificial boothmate. https://www.eabm. ugent.be/survey/ (13 August, 2022). Ehrensberger-Dow, Maureen. 2014. Challenges of translation process research at the workplace. MonTI: Monografías de Traducción e Interpretación. 355–383. DOI: 10.6035/MonTI.2014.ne1.12. Ehrensberger-Dow, Maureen, Michaela Albl-Mikasa, Katrin Andermatt, Andrea Hunziker Heeb & Caroline Lehr. 2020. Cognitive load in processing ELF: Trans- lators, interpreters, and other multilinguals. Journal of English as a Lingua Franca 9(2). 217–238. DOI: 10.1515/jelf-2020-2039. Ehlich, Konrad & Jochen Rehbein. 1976. Halbinterpretative Arbeitstranskriptio- nen (HIAT). Linguistische Berichte 45. 21–41. Ehrensberger-Dow, Maureen & Sharon O’Brien. 2015. Ergonomics of the transla- tion workplace: Potential for cognitive friction. Translation Spaces 4(1). 98–118. DOI: 10.5922/2225-5346-2019-1-3. Eisenrieth Dokumentations GmbH. 2010. Flashterm. https://www.flashterm.eu/ en/ (13 August, 2021). ELAN. 2020. Version 5.9. https://archive.mpi.nl/tla/elan (13 August, 2022).f Eriksen, Barbara A. & Charles W. Eriksen. 1974. Effects of noise letters upon the identification of a target letter in a nonsearch task. Perception & Psychophysics 16(1). 143–149. DOI: 10.3758/BF03203267. European Parliament, DG Translation. 2019. Live speech to text and machine trans- lation tool for 24 languages. https://etendering.ted.europa.eu/cft/cft-display. html?cftId=5249 (14 September, 2021). Fabbro, Franco & Laura Gran. 1994. Neurological and neuropsychological aspects of polyglossia and simultaneous interpretation. In Sylvie Lambert & Barbara Moser-Mercer (eds.), Bridging the gap: Empirical research in simultaneous inter- pretation, 273–318. Amsterdam: John Benjamins. DOI: 10.1075/btl.3.21fab. Fantinuoli, Claudio. 2006. p g 0.5922/2225-5346-2019-1-3. References Specialized corpora from the web and term extrac- tion for simultaneous interpreters. In Marco Baroni & Silvia Bernardini (eds.), Wacky! Working papers on the web as corpus, 173–190. Bologna: CEDIB. Fantinuoli, Claudio. 2012. InterpretBank: Design and implementation of a termi- nology and knowledge management software for conference interpreters. Germ- ersheim: University of Mainz. (Doctoral dissertation). 274 Fantinuoli, Claudio. 2016. InterpretBank. Redefining computer-assisted interpret- ing tools. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the Translating and the Computer 38 Conference in London, 42–52. Geneva: Editions Tradulex. Fantinuoli, Claudio. 2017a. Computer-assisted preparation in conference inter- preting. Translation & Interpreting 9(2). 24–37. DOI: 10.12807/ti.109202.2017. a02. Fantinuoli, Claudio. 2017b. Speech recognition in the interpreter workstation. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 39th Conference Translating and the Computer, 25–34. Geneva: Editions Tradulex. Fantinuoli, Claudio. 2018a. Computer-assisted interpreting: Challenges and fu- ture perspectives. In Gloria Corpas Pastor & Isabel Durán-Muñoz (eds.), Trends in e-tools and resources for translators and interpreters, 153–174. Leiden: Brill. DOI: 10.1163/9789004351790_009. Fantinuoli, Claudio. 2018b. Interpreting and technology: The upcoming tech- nological turn. In Claudio Fantinuoli (ed.), Interpreting and technology, 1–12. Berlin: Language Science Press. DOI: 10.5281/zenodo.1493289. Fantinuoli, Claudio. 2019. The technological turn in interpreting: The challenges that lie ahead. In Wolfram Baur & Felix Mayer (eds.), Übersetzen und Dol- metschen 4.0: Neue Wege im Digitalen Zeitalter, 334–354. Berlin: BDÜ Fachver- lag. Fantinuoli, Claudio, Giulia Marchesini, David Landan & Lukas Horak. 2022. KUDO Interpreter Assist: Automated real-time support for remote interpre- tation. In João Esteves-Ferreira, Ruslan Mitkov, Maria Recort Ruiz, Olaf- Michael Stefanov, David Chambers, Juliet Margaret Macan & Vilelmini Sosoni (eds.), Proceedings of the 43th Conference Translating and the Computer, 68–77. Geneva: Editions Tradulex. Fantinuoli, Claudio & Maddalena Montecchio. 2022. Defining maximum accept- able latency of AI-enhanced CAI tools. arXiv preprint. DOI: 10.48550/arXiv. 2201.02792. Fantinuoli, Claudio & Bianca Prandi. 2018. Teaching information and communi- cation technologies: A proposal for the interpreting classroom. Transkom 11(2). 162–182. Fantinuoli, Claudio & Bianca Prandi. 2021. Towards the evaluation of automatic simultaneous speech translation from a communicative perspective. In Pro- ceedings of the 18th International Conference on Spoken Language Translation, 245–254. Bangkok: Association for Computational Linguistics. 275 References Ferreira, Aline, John W. Schwieter & Daniel Gile. 2015. The position of psycholin- guistic and cognitive science in translation and interpreting. In Aline Ferreira & John W. References Schwieter (eds.), Psycholinguistic and cognitive inquiries into trans- lation and interpreting, 3–15. Amsterdam: John Benjamins. DOI: 10.1075/btl.115. 01fer. Flesch, Rudolph. 1948. A new readability yardstick. Journal of Applied Psychology 32(3). 221–233. DOI: 10.1037/h0057532. Floros, Georgios. 2003. Kulturelle Konstellationen in Texten: Zur Beschreibung und Übersetzung von Kultur in Texten. Tübingen: Gunter Narr. Folaron, Debbie. 2019. Technology, technical translation and localization. In Mi- nako O’Hagan (ed.), The Routledge handbook of translation and technology, 203–219. London: Routledge. Fox, Wendy. 2018. Can integrated titles improve the viewing experience? Investi- gating the impact of subtitling on the reception and enjoyment of film using eye tracking and questionnaire data (Translation and Multilingual Natural Lan- guage Processing 9). Berlin: Language Science Press. DOI: 10.5281/zenodo. 1180721. Frittella, Francesca Maria. 2022. CAI-tool supported SI of numbers: A theoretical and methodological contribution. International Journal of Interpreter Education 14(1). 32–56. DOI: 10.34068/ijie.14.01.05. Gacek, Michael. 2015. Softwarelösungen für DolmetscherInnen. Wien: Universität Wien. (MA thesis). García, Adolfo M. 2013. Brain activity during translation: A review of the neu- roimaging evidence as a testing ground for clinically-based hypotheses. Jour- nal of Neurolinguistics 26(3). 370–383. DOI: 10.1016/j.jneuroling.2012.12.002. García, Adolfo M. 2015. Translating with an injured brain: Neurolinguistic as- pects of translation as revealed by bilinguals with cerebral lesions. Meta: Jour- nal des Traducteurs/Meta: Translators’ Journal 60(1). 112–134. DOI: 10 . 7202 / 1032402ar. García, Adolfo M. 2019. The neurocognition of translation and interpreting. Ams- terdam: John Benjamins. DOI: 10.1075/btl.147. García Becerra, Olalla. 2016. Survey research on quality expectations in inter- preting: The effect of method of administration on subjects’ response rate. Meta: Journal des Traducteurs/Meta: Translators’ Journal 60(3). 542–556. DOI: 10.7202/1036142ar. García de Quesada, Mercedes. 2011. Terminologie. In Angela Collados Aís, Emilia Iglesias Fernández, E. Macarena Pradas Macías & Elisabeth Stévaux (eds.), Qualitätsparameter beim Simultandolmetschen: Interdisziplinäre Perspektiven, 219–251. Tübingen: Gunter Narr. 276 Garzone, Giuliana. 2006. Perspectives on ESP and popularization. Milano: CUEM. Garzone, Giuliana. 2006. Perspectives on ESP and popularization. Milano: CUEM. Gerver, David. 1976. Empirical studies of simultaneous interpretation: A review and a model. In Richard W. Brislin (ed.), Translation: Applications and research, Gerver, David. 1976. Empirical studies of simultaneous interpretation: A review and a model. In Richard W. Brislin (ed.), Translation: Applications and research, 165–207. New York: Gardner Press. Gerzymisch-Arbogast, Heidrun. 1996. Termini im Kontext: Verfahren zur Er- schließung und Übersetzung der textspezifischen Bedeutung von fachlichen Aus- drücken. Tübingen: Gunter Narr. Ghiselli, Serena. 2015. References Le sfide traduttive dei sintagmi nominali con modificatori in posizione prenominale nell’interpretazione simultanea dall’inglese in italiano: Uno studio sul corpus EPIC. Forlì: Università di Bologna. (MA thesis). Gieshoff, Anne Catherine. 2012. Aus 92 wird zwölf: Zahlen im Simultandol- metschen. Germersheim: University of Mainz. (MA thesis).f Gieshoff, Anne Catherine. 2018. The impact of audio-visual speech input on work-load in simultaneous interpreting. Germersheim: Johannes Gutenberg- Universität Mainz. (Doctoral dissertation). DOI: 10.25358/openscience-2180. Gieshoff, Anne Catherine. 2021. The impact of visible lip movements on silent pauses in simultaneous interpreting. Interpreting: International Journal of Re- search and Practice in Interpreting 23(2). 168–191. DOI: 10.1075/intp.00061.gie. Gile, Daniel. 1988. Le partage de l’attention et le “Modèle d’efforts” en interpré- tation simultanée. The Interpreters’ Newsletter 1. 4–22. http://hdl.handle.net/ 10077/2132 (24 July, 2021). Gile, Daniel. 1990. L’évaluation de la qualité de l’interprétation par les délégués: Une étude de cas. The Interpreters’ Newsletter 3. 66–71. http://hdl.handle.net/ 10077/2156 (24 July, 2021). Gile, Daniel. 1991a. A communicative-oriented analysis of quality in nonliter- ary translation and interpretation. In Mildred L. Larson (ed.), Translation: The- ory and practice, tension and interdependence, 188–200. Binghamton: John Ben- jamins. DOI: 10.1075/ata.v.19gil. Gile, Daniel. 1991b. The processing capacity issue in conference interpretation. Babel: Revue internationale de la traduction / International Journal of Transla- tion 37(1). 15–27. DOI: 10.1075/babel.37.1.04gil. Gile, Daniel. 1995. Regards sur la recherche en interprétation de conférence. Lille: Presses universitaires de Lille. Gile, Daniel. 1997. Conference interpreting as a cognitive management problem. In Franz Pöchhacker & Miriam Shlesinger (eds.), The interpreting studies reader, 163–176. London: Routledge. Gile, Daniel. 1999. Testing the effort models’ tightrope hypothesis in simultane- ous interpreting: A contribution. HERMES: Journal of Language and Commu- nication in Business 23. 153–172. DOI: 10.7146/hjlcb.v12i23.25553. 277 References Gile, Daniel. 2008. Local cognitive load in simultaneous interpreting and its impli- cations for empirical research. FORUM: Revue internationale d’interprétation et de traduction / International Journal of Interpretation and Translation 6(2). 59– 77. DOI: 10.1075/forum.6.2.04gil. Gile, Daniel. 2009. Basic concepts and models for interpreter and translator training. Amsterdam: John Benjamins. Gile, Daniel. 2011. Errors, omissions and infelicities in broadcast interpreting: Pre- liminary findings from a case study. In Cecilia Alvstad, Adelina Hild & Elisabet Tiselius (eds.), Methods and strategies of process research, 201–218. Amsterdam: John Benjamins. DOI: 10.1075/btl.94.15gil.f Gile, Daniel. 2015. Effort models. In Franz Pöchhacker (ed.), Routledge encyclope- dia of interpreting studies, 135–137. Oxon: Routledge.fi Gile, Daniel. 2016. The effort models: Clarifications and update. (Presentation). DOI: 10.13140/RG.2.1.4221.7849. Gile, Daniel. References 2017a. Biases in critical reading of TIS literature. Tradução em Revista 23. 1–19. DOI: 10.17771/PUCRio.TradRev.32213. Gile, Daniel. 2017b. The effort models clarifications. (Presentation). DOI: 10.13140/ RG.2.2.21975.27049. Gile, Daniel. 2020. 2020 update of the effort models and gravitational model. (Pre- sentation). DOI: 10.13140/RG.2.2.24895.94889.f Gile, Daniel & Victoria Lei. 2020. Translation, effort and cognition. In Fábio Alves & Arnt Lykke Jakobsen (eds.), The Routledge handbook of translation and cog- nition, 263–278. London: Routledge. Ginns, Paul. 2005. Meta-analysis of the modality effect. Learning and Instruction 15(4). 313–331. DOI: 10.1016/j.learninstruc.2005.07.001. Godefroy, Olivier, Philippe Azouvi, Philippe Robert, Martine Roussel, Didier LeGall, Thierry Meulemans & Groupe de Réflexion sur l’Evaluation des Fonc- tions Exécutives Study Group. 2010. Dysexecutive syndrome: Diagnostic crite- ria and validation study. Annals of Neurology 68(6). 855–864. DOI: 10.1002/ana. 22117. Goldman-Eisler, Frieda. 1972. Segmentation of input in simultaneous translation. Journal of Psycholinguistic Research 1(2). 127–140. DOI: 10.1007/BF01068102. Goldsmith, Joshua. 2017. A comparative user evaluation of tablets and tools for consecutive interpreters. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 39th Conference Translating and the Computer, 40–50. Geneva: Editions Tradulex. Goldsmith, Joshua & Josephine Christine Holley. 2015. Consecutive interpreting 2.0: The tablet interpreting experience. Geneva: University of Geneva. (MA the- sis). 278 Görög, Attila. 2014a. Quality evaluation today: The dynamic quality framework. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 36th Conference Translating and the Computer, 155–164. Geneva: Tradulex. Görög, Attila. 2014b. Quantifying and benchmarking quality: The TAUS Dynamic Quality Framework. Tradumàtica: tecnologies de la traducció 12. 443–454.f Gumul, Ewa. 2018. Searching for evidence of Gile’s effort models in retrospective protocols of trainee simultaneous interpreters. Między oryginałem a przekła- dem 24(4(42)). 17–39. DOI: 10.12797/MOaP.24.2018.42.02. Halliday, Michael A. K. & James R. Martin. 2003. Writing science: Literacy and discursive power. London: Routledge. Hamidi, Miriam & Franz Pöchhacker. 2007. Simultaneous consecutive interpret- ing: A new technique put to the test. Meta: Journal des Traducteurs/Meta: Trans- lators’ Journal 52(2). 276–289. DOI: 10.7202/016070ar. Hansen, Gyde. 2007. Ein Fehler ist ein Fehler ... oder?: Der Bewertungsprozess in der Übersetzungsprozessforschung. In Gerd Wotjak (ed.), Quo vadis Trans- latologie?, 115–131. Berlin: Frank & Timme. Hansen, Gyde. 2008. The dialogue in translation process research. In Translation and cultural diversity: Selected proceedings of the XVIII FIT world congress 2008, 386–397. Shanghai: Foreign Languages Press. Hansen, Gyde. 2009. Some thoughts about the evaluation of translation products in translation process research. References In Inger M. Mees, Susanne Göpferich & Fábio Alves (eds.), Methodology, technology and innovation in translation process re- search: A tribute to Arnt Lykke Jakobsen (Copenhagen Studies in Language), 389–402. Frederiksberg: Samfundslitteratur. Hansen, Inge Gorm & Barbara Dragsted. 2007. Speaking your translation: Ex- ploiting synergies between translation and interpreting. In Franz Pöchhacker, Arnt Lykke Jakobsen & Inger M. Mees (eds.), Interpreting studies and beyond: A tribute to Miriam Shlesinger (Copenhagen Studies in Language 35), 251–274. Frederiksberg: Copenhagen Business School Press. Hansen-Schirra, Silvia. 2012. Nutzbarkeit von Sprachtechnologien für die Trans- lation. Transkom 5(2). 211–226. (24 July, 2021). Hansen-Schirra, Silvia & Jean Nitzke. 2020. Translation, the process–product in- terface and cognition. In Fábio Alves & Arnt Lykke Jakobsen (eds.), The Rout- ledge handbook of translation and cognition, 415–432. London: Routledge. Hansen-Schirra, Silvia, Jean Nitzke & Katharina Oster. 2017. Predicting cognate translation. In Silvia Hansen-Schirra, Oliver Čulo & Sascha Hofmann (eds.), Empirical modelling of translation and interpreting, 23–39. Berlin: Language Science Press. DOI: 10.5281/zenodo.1090944. 279 References Hart, Sandra G. & Lowell E. Staveland. 1988. Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research. In Peter A. Hancock & Najmedin Meshkati (eds.), Human mental workload (Advances in Psychol- ogy 52), 139–183. Amsterdam: Elsevier. DOI: 10.1016/S0166-4115(08)62386-9. Heid, Ulrich & Anita Gojun. 2012. Term candidate extraction for terminography and CAT: An overview of TTC. In Ruth Vatvedt Fjeld & Julie Matilde Torjusen (eds.), Proceedings of the 15th Euralex International Congress, 585–594. Oslo: Re- prosentralen, UiO. Herbig, Nico, Santanu Pala, Antonio Krügera & Josef van Genabitha. 2021. Multi- modal estimation of cognitive load in post-editing of machine translation. In Tra&Co Group (ed.), Translation, interpreting, cognition: The way out of the box, 1–32. Berlin: Language Science Press. DOI: 10.5281/zenodo.4545029. Hervais-Adelman, Alexis Georges, Barbara Moser-Mercer & Narly Golestani. 2011. Executive control of language in the bilingual brain: Integrating the evi- dence from neuroimaging to neuropsychology. Frontiers in Psychology 2. 1–8. DOI: 10.3389/fpsyg.2011.00234. Hervais-Adelman, Alexis Georges, Barbara Moser-Mercer, Christoph M. Michel & Narly Golestani. 2015. fMRI of simultaneous interpretation reveals the neu- ral basis of extreme language control. Cerebral Cortex 25(12). 4727–4739. DOI: 10.1093/cercor/bhu158. Hess, Eckhard H. & James M. Polt. 1964. Pupil size in relation to mental activ- ity during simple problem-solving. Science 143(3611). 1190–1192. DOI: 10.1126/ science.143.3611.1190. Hessels, Roy S., Chantal Kemner, Carlijn van den Boomen & Ignace T. C. Hooge. 2016. References The area-of-interest problem in eyetracking research: A noise-robust so- lution for face and sparse stimuli. Behavior Research Methods 48(4). 1694–1712. DOI: 10.3758/s13428-015-0676-y. Hoberg, Felix. 2021. Dialogue-oriented evaluation of Microsoft’s skype transla- tor in the language pair Catalan-German. In Tra&Co Group (ed.), Translation, interpreting, cognition: The way out of the box, 67–77. Berlin: Language Science Press. DOI: 10.5281/zenodo.4545035. Hoberg, Felix. 2022. Informationsintegration in mehrsprachigen Textchats: Der Skype Translator im Sprachenpaar Katalanisch-Deutsch (Translation and Multi- lingual Natural Language Processing 17). Berlin: Language Science Press. DOI: 10.5281/zenodo.5902971. Hoeks, Bert & Willem J. M. Levelt. 1993. Pupillary dilation as a measure of atten- tion: A quantitative system analysis. Behavior Research Methods, Instruments, & Computers 25(1). 16–26. DOI: 10.3758/BF03204445. 280 Holmqvist, Kenneth. 2011. Eye tracking: A comprehensive guide to methods and measures. Oxford: Oxford University Press. Hvelplund, Kristian Tangsgaard. 2011. Allocation of cognitive resources in transla- tion: An eye-tracking and key-logging study (PhD series 10.2011). Frederiksberg: Samfundslitteratur. Hvelplund, Kristian Tangsgaard. 2014. Eye tracking and the translation process: Reflections on the analysis and interpretation of eye-tracking data. MonTI: Monografías de Traducción e Interpretación. 201–223. DOI: 10.6035/MonTI.2014. ne1.6. Hvelplund, Kristian Tangsgaard. 2017. Eye tracking in translation process re- search. In John W. Schwieter & Aline Ferreira (eds.), The handbook of trans- lation and cognition, 248–264. Hoboken: Wiley Online Library. Hyönä, Jukka, Jorma Tommola & Anna-Mari Alaja. 1995. Pupil dilation as a mea- sure of processing load in simultaneous interpretation and other language tasks. The Quarterly Journal of Experimental Psychology 48(3). 598–612. DOI: 10.1080/14640749508401407. Jacquemot, Charlotte, Emmanuel Dupoux & Anne-Catherine Bachoud-Lévi. 2011. Is the word-length effect linked to subvocal rehearsal? Cortex 47(4). 484–493. DOI: 10.1016/j.cortex.2010.07.007. Jakobsen, Arnt Lykke. 2003. Effects of think aloud on translation speed, revision, and segmentation. In Fábio Alves (ed.), Triangulating translation: Perspectives in process oriented research, 69–95. Amsterdam: John Benjamins. Jakobsen, Arnt Lykke. 2006. Research methods in translation: Translog. In Eva Lindgren & Kirk P. H. Sullivan (eds.), Computer keystroke logging and writing (Studies in writing 18), 95–105. Oxford: Pergamon Press. Jakobsen, Arnt Lykke. 2017. Translation process research. In John W. Schwieter & Aline Ferreira (eds.), The handbook of translation and cognition, 19–49. Hobo- ken: John Wiley & Sons, Ltd. DOI: 10.1002/9781119241485.ch2. Jakobsen, Arnt Lykke. 2019. Translation technology research with eye tracking. In Minako O’Hagan (ed.), The Routledge handbook of translation and technology, 398–416. London: Routledge. Jakobsen, Arnt Lykke & Kristian Tangsgaard Hvelplund. 2008. Eye movement behaviour across four different types of reading task. References Copenhagen Studies in Language 36. 103–124. Jia, Ye, Michelle Tadmor Ramanovich, Tal Remez & Roi Pomerantz. 2022. Trans- latotron 2: High-quality direct speech-to-speech translation with voice preser- vation. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu & Sivan Sabato (eds.), Proceedings of the 39th international confer- ence on machine learning, 10120–10134. Baltimore: PMLR. 281 References Jia, Ye, Ron J. Weiss, Fadi Biadsy, Wolfgang Macherey, Melvin Johnson, Zhifeng Chen & Yonghui Wu. 2019. Direct speech-to-speech translation with a sequence-to-sequence model. In Proc. interspeech 2019, 1123–1127. Jiang, Hong. 2013. The interpreter’s glossary in simultaneous interpreting: A sur- vey. Interpreting: International Journal of Research and Practice in Interpreting 15(1). 74–93. DOI: 10.1075/intp.15.1.04jia. Jiang, Hong. 2015. A survey of glossary practice of conference interpreters. https: //aiic.org/document/950/AIICWebzine_2015_Issue66_5_JIANG_A_survey_ of_glossary_practice_of_conference_interpreters_EN.pdf (24 July, 2021). Just, Marcel A. & Patricia A. Carpenter. 1980. A theory of reading: From eye fixations to comprehension. Psychological Review 87(4). 329–354. DOI: 10.1037/ 0033-295X.87.4.329. Just, Marcel A. & Patricia A. Carpenter. 1993. The intensity dimension of thought: Pupillometric indices of sentence processing. Canadian Journal of Experimen- tal Psychology/Revue canadienne de psychologie expérimentale 47(2). 310–339. DOI: 10.1037/h0078820. Kageura, Kyo & Elizabeth Marshman. 2019. Terminology extraction and man- agement. In Minako O’Hagan (ed.), The Routledge handbook of translation and technology, 61–77. London: Routledge. ahneman, Daniel. 1973. Attention and effort. Englewood Cliffs: Prentice-Hall. Kahneman, Daniel & Jackson Beatty. 1966. Pupil diameter and load on memory. Science 154(3756). 1583–1585. DOI: 10.1126/science.154.3756.1583. Kalderon, Eliza. 2017. Neurophysiologie des Simultandolmetschens: Eine fMRI- Studie mit Konferenzdolmetschern. Germersheim: Johannes Gutenberg- Universität Mainz. (Doctoral dissertation). DOI: 10.25358/openscience-4538. Kalina, Sylvia. 1998. Strategische Prozesse beim Dolmetschen: Theoretische Grundla- gen, empirische Fallstudien, didaktische Konsequenzen. Tübingen: Gunter Narr. Kalina, Sylvia. 2005. Quality assurance for interpreting processes. Meta: Jour- nal des Traducteurs/Meta: Translators’ Journal 50(2). 768–784. DOI: 10.7202/ Kalina, Sylvia. 1998. Strategische Prozesse beim Dolmetschen: Theoretische Grundla- gen, empirische Fallstudien, didaktische Konsequenzen. Tübingen: Gunter Narr. Kalina, Sylvia. 2005. Quality assurance for interpreting processes. Meta: Jour- nal des Traducteurs/Meta: Translators’ Journal 50(2). 768–784. DOI: 10.7202/ 011017ar. Kalina, Sylvia. 2006. Zur Dokumentation von Maßnahmen der Qualitätssiche- rung beim Konferenzdolmetschen. In Carmen Heine, Klaus Schubert & Heid- run Gerzymisch-Arbogast (eds.), Text and translation: Theory and methodology of translation, 253–268. Tübingen: Gunter Narr. Kantowitz, Barry H. & James L. Knight. 1976. On experimenter-limited processes. Psychological Review 83(6). 502–507. DOI: 10.1037/0033-295X.83.6.502. Karsh, Robert & Francis W. Breitenbach. 1983. Looking at looking: The amor- phous fixation measure. References In Rudolph Groner, Christine Menz, Dennis F. Fisher 282 & Richard A. Monty (eds.), Eye movements and psychological functions: Inter- national views, 53–64. Hillsdale: Erlbaum. & Richard A. Monty (eds.), Eye movements and psychological functions: Inter- national views, 53–64. Hillsdale: Erlbaum. Keating, Gregory D. & Jill Jegerski. 2015. Experimental designs in sentence pro- cessing research: A methodological review and user’s guide. Studies in Second Language Acquisition 37(1). 1–32. DOI: 10.1017/S0272263114000187. Kilgarriff, Adam, Vít Baisa, Jan Bušta, Miloš Jakubíček, Vojtěch Kovář, Jan Michelfeit, Pavel Rychlỳ& Vít Suchomel. 2014. The Sketch Engine: Ten years on. Lexicography 1. 7–36. DOI: 10.1007/s40607-014-0009-9. Kirchhoff, Hella. 1976. Simultaneous interpreting: Interdependence of variables in the interpreting process, interpreting models and interpreting strategies. In Franz Pöchhacker & Miriam Shlesinger (eds.), The interpreting studies reader, 110–119. London: Routledge. Klonowicz, Tatiana. 1994. Putting one’s heart into simultaneous interpretation. In Sylvie Lambert & Barbara Moser-Mercer (eds.), Bridging the gap: Empirical research in simultaneous interpretation, 213–224. Amsterdam: John Benjamins. DOI: 10.1075/btl.3.16klo. Kopczynski, Andrzej. 1994. Quality in conference interpreting: Some pragmatic problems. In Mary Snell-Hornby, Franz Pöchhacker & Klaus Kaindl (eds.), Translation studies: An interdiscipline, 189–198. Amsterdam: John Benjamins. Köpke, Barbara & Teresa M. Signorelli. 2012. Methodological aspects of work- ing memory assessment in simultaneous interpreters. International Journal of Bilingualism 16(2). 183–197. DOI: 10.1177/1367006911402981. Korpal, Paweł. 2016. Interpreting as a stressful activity: Physiological measures of stress in simultaneous interpreting. Poznań Studies in Contemporary Linguis- tics 52(2). 297–316. DOI: 10.1515/psicl-2016-0011. Korpal, Paweł. 2017. Linguistic and psychological indicators of stress in simulta- neous interpreting (Outstanding WA dissertations 12). Poznań: Wydawnictwo Naukowe UAM. Koshkin, Roman, Yury Shtyrov, Andriy Myachykov & Alex Ossadtchi. 2018. Test- ing the efforts model of simultaneous interpreting: An ERP study. PLOS ONE 13(10). 1–18. DOI: 10.1371/journal.pone.0206129. Kosma, Alexandra. 2007. Le fonctionnement spécifique de la mémoire de travail en traduction. Meta: Journal des Traducteurs/Meta: Translators’ Journal 52(1). 22–28. DOI: 10.7202/014716ar. Krings, Hans P. 2001. Repairing texts: Empirical investigations of machine transla- tion post-editing processes. Kent: Kent State University Press. Kuang, Huolingxiao. 2019. Computerized note-taking in consecutive interpret- ing: A pen-voice integrated approach towards omissions, additions and recon- structions in notes. In Michael Carl & Silvia Hansen-Schirra (eds.), Proceedings 283 References of the second MEMENTO workshop on Modelling Parameters of Cognitive Effort in Translation Production, 18–19. Dublin: European Association for Machine Translation. of the second MEMENTO workshop on Modelling Parameters of Cognitive Effort in Translation Production, 18–19. Dublin: European Association for Machine Translation. References Kumpulainen, Minna. 2015. On the operationalisation of ‘pauses’ in translation process research. The International Journal for Translation & Interpreting Re- search 7(1). 47–58. DOI: ti.106201.2015.a04. Kurz, Ingrid. 1989. Conference interpreting: User expectations. In Deanna L. Hammond (ed.), Coming of age: Proceedings of the 30th annual conference of the American Translators Association, 143–148. Medford: Learned Information.f Kurz, Ingrid. 1993. Conference interpretation: Expectations of different user groups. The Interpreters’ Newsletter 5. 13–21. http://hdl.handle.net/10077/4908 (24 July, 2021). Kurz, Ingrid. 1994. A look into the “black box”: EEG probability mapping during mental simultaneous interpreting. In Mary Snell-Hornby, Franz Pöchhacker & Klaus Kaindl (eds.), Translation studies: An interdiscipline. Selected papers from the Translation Studies Congress, Vienna, 1992, 199–208. Amsterdam: John Benjamins. DOI: 10.1075/btl.2.25kur. Kurz, Ingrid. 1996. Simultandolmetschen als Gegenstand der interdisziplinären Forschung. Wien: WUV-Universitätsverlag. Kurz, Ingrid. 1996. Simultandolmetschen als Gegenstand der interdisziplinären Forschung. Wien: WUV-Universitätsverlag. Kurz, Ingrid. 2002. Physiological stress responses during media and conference interpreting. In Giuliana Garzone & Maurizio Viezzi (eds.), Interpreting in the 21st century: Challenges and opportunities, 195–202. Amsterdam: John Ben- jamins. DOI: 10.1075/btl.43.19kur. Kurz, Ingrid. 2003. Physiological stress during simultaneous interpreting: A com- parison of experts and novices. The Interpreters’ Newsletter 12. 51–67. http:// hdl.handle.net/10077/2472. Lacruz, Isabel, Michael Carl & Masaru Yamada. 2018. Literality and cognitive ef- fort: Japanese and Spanish. In Proceedings of the Eleventh International Confer- ence on Language Resources and Evaluation (LREC 2018), 3818–3821. Lacruz, Isabel, Michael Carl, Masaru Yamada & Akiko Aizawa. 2016. Pause metrics and machine translation utility. Paper presented at the 22nd Annual Meeting of the Association for Natural Language Processing, 1–4. Sendai: NLP 2016. Lacruz, Isabel, Michael Denkowski & Alon Lavie. 2014. Cognitive demand and cognitive effort in post-editing. In Proceedings of the 11th Conference of the Asso- ciation for Machine Translation in the Americas, 73–84. Vancouver: Association for Machine Translation in the Americas. https://aclanthology.org/2014.amta- wptp.6 (24 July, 2021). 284 Lacruz, Isabel, Gregory M. Shreve & Erik Angelone. 2012. Average pause ratio as an indicator of cognitive effort in post-editing: A case study. In Sharon O’Brien, Michel Simard & Lucia Specia (eds.), Workshop on post-editing technology and practice, 1–10. San Diego: Association for Machine Translation in the Americas. https://aclanthology.org/2012.amta-wptp.3 (24 July, 2021). Lambert, Sylvie. 1988. Information processing among conference interpreters: A test of the depth-of-processing hypothesis. Meta: Journal des Traducteurs/Meta: Translators’ Journal 33(3). 377–387. DOI: 10.7202/003380ar. Landis, J. Richard & Gary G. Koch. 1977. The measurement of observer agreement for categorical data. References Biometrics 33(1). 159–174. DOI: 10.2307/2529310. Laplace, Colette. 1994. Théorie du langage et théorie de la traduction. Les concepts- clefs de trois auteurs: Kade (Leipzig), Coseriu (Tübingen), Seleskovitch (Paris). Paris: Didier Érudition. Läubli, Samuel & David Orrego-Carmona. 2017. When Google Translate is bet- ter than some human colleagues, those people are no longer colleagues. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 39th Conference Translating and the Computer, 59–69. Geneva: Editions Tradulex. Läubli, Samuel & David Orrego-Carmona. 2017. When Google Translate is bet- ter than some human colleagues, those people are no longer colleagues. In Lavie, Nilli. 1995. Perceptual load as a necessary condition for selective attention. Journal of Experimental Psychology: Human Perception and Performance 21(3). 451–468. DOI: 10.1037/0096-1523.21.3.451. Lavie, Nilli. 2000. Selective attention and cognitive control: Dissociating atten- tional functions through different types of load. In Stephen Monsell & Jon Driver (eds.), Control of cognitive processes: Attention and performance XVIII, 175–194. Cambridge: MIT Press. Lavie, Nilli. 2005. Distracted and confused? Selective attention under load. Trends in Cognitive Sciences 9(2). 75–82. DOI: 10.1016/j.tics.2004.12.004. Lavie, Nilli. 2010. Attention, distraction, and cognitive control under load. Current Directions in Psychological Science 19(3). 143–148. DOI: 10 . 1177 / 0963721410370295. Lavie, Nilli, Aleksandra Hirst, Jan W. De Fockert & Essi Viding. 2004. Load the- ory of selective attention and cognitive control. Journal of Experimental Psy- chology: General 133(3). 339–354. DOI: 10.1037/0096-3445.133.3.339. Lazar, Jonathan. 2017. Research methods in human computer interaction. Cam- bridge: Elsevier. Lederer, Marianne. 1978. Simultaneous interpretation: Units of meaning and other features. In David Gerver & H. Wallace Sinaiko (eds.), Language inter- pretation and communication, 323–332. Boston: Springer. 285 References Lederer, Marianne. 1981. La traduction simultanée: Expérience et théorie. Paris: Mi- nard Lettres Modernes. Lee, Tae-Hyung. 2002. Ear voice span in English into Korean simultaneous inter- pretation. Meta: Journal des Traducteurs/Meta: Translators’ Journal 47(4). 596– 606. DOI: 10.7202/008039ar. Lee, Tae-Hyung. 2003. Tail-to-tail span: A new variable in conference interpret- ing research. FORUM: Revue internationale d’interprétation et de traduction / International Journal of Interpretation and Translation 1(1). 41–62. DOI: 10.1075/ forum.1.1.03lee. Leijten, Mariëlle & Luuk Van Waes. 2013. Keystroke logging in writing research: Using Inputlog to analyze and visualize writing processes. Written Communi- cation 30(3). 358–392. DOI: 10.1177/0741088313491692. Leipzig Corpora Collection. 2012. English web text corpus (United Kingdom) based on material from 2012. In Leipzig corpora collection. Dataset. https:// corpora.uni-leipzig.de/?corpusId=eng-uk_web_2012 (24 May, 2022). Leipzig Corpora Collection. 2016. References English news corpus based on material from 2016. In Leipzig corpora collection. Dataset. https://corpora.uni- leipzig.de? corpusId=eng_news_2016 (26 May, 2022). Leipzig Corpora Collection. 2021. English Wikipedia corpus based on material from 2021. In Leipzig corpora collection. Dataset. https://corpora.uni-leipzig. de?corpusId=eng_wikipedia_2021 (26 May, 2022). Levenshtein, Vladimir I. 1966. Binary codes capable of correcting deletions, inser- tions and reversals. Soviet Physics Doklady 10(8). 707–710. https://nymity.ch/ sybilhunting/pdf/Levenshtein1966a.pdf (24 August, 2021). ogie, Robert H. 1995. Visuo-spatial working memory. London: Psychology Pres Logie, Robert H. 1995. Visuo-spatial working memory. London: Psychology Press. Logie, Robert H. 2016. Retiring the central executive. Quarterly Journal of Exper- imental Psychology 69(10). 2093–2109. DOI: 10.1080/17470218.2015.1136657. Logie, Robert H. 2016. Retiring the central executive. Quarterly Journal of Exper- imental Psychology 69(10). 2093–2109. DOI: 10.1080/17470218.2015.1136657.t Lommel, Arle, Aljoscha Burchardt, Maja Popović, Kim Harris, Eleftherios Avramidis & Hans Uszkoreit. 2014. Using a new analytic measure for the an- notation and analysis of MT errors on real data. In Proceedings of the 17th annual conference of the European Association for Machine Translation, 165– 172. Dubrovnik: European Association for Machine Translation. https : / / aclanthology.org/2014.eamt-1.38 (24 July, 2022). Lommel, Arle, Hans Uszkoreit & Aljoscha Burchardt. 2014. Multidimensional quality metrics (MQM) definition. http : / / www . qt21 . eu / mqm - definition/ (24 May, 2021). Low, Renae & John Sweller. 2014. The modality principle in multimedia learning. In Richard Mayer (ed.), The Cambridge handbook of multimedia learning, 227– Low, Renae & John Sweller. 2014. The modality principle in multimedia learning. In Richard Mayer (ed.), The Cambridge handbook of multimedia learning, 227– Low, Renae & John Sweller. 2014. The modality principle in multimedia learning. In Richard Mayer (ed.), The Cambridge handbook of multimedia learning, 227– 286 246. Cambridge: Cambridge University Press. DOI: 10.1017/CBO9781139547369. 012. 246. Cambridge: Cambridge University Press. DOI: 10.1017/CBO9781139547369. 012. Lu, Sara A., Christopher D. Wickens, Julie C. Prinet, Shaun D. Hutchins, Nadine Sarter & Angelia Sebok. 2013. Supporting interruption management and multi- modal interface design: Three meta-analyses of task performance as a function of interrupting task modality. Human Factors: The Journal of the Human Factors and Ergonomics Society 55(4). 697–724. DOI: 10.1177/0018720813476298. Mack, Gabriele & Lorella Cattaruzza. 1995. User surveys in simultaneous inter- pretation: A means of learning about quality and/or raising some reasonable doubts. In Jorma Tommola (ed.), Topics in interpreting research, 37–49. Turku: Centre for Translation & Interpreting, University of Turku (Finland). Marrone, Stefano. 1993. Quality: A shared objective. References The Interpreters’ Newsletter 5. 35–41. http://hdl.handle.net/10077/14664 (5 August, 2021). p g Martin, Reg. 2014. Glossary assistant. http://swiss32.com (13 August, 2021). Martínez, Silvia Montero & Pamela Faber. 2009. Terminological competence in translation. Terminology: International Journal of Theoretical and Applied Issues in Specialized Communication 15(1). 88–104. DOI: 10.1075/term.15.1.05mon. Marzouk, Shaimaa. 2021. German light verb construction in the course of the development of Machine Translation. In Tra&Co Group (ed.), Translation, in- terpreting, cognition: The way out of the box, 47–66. Berlin: Language Science Press. DOI: 10.5281/zenodo.4545033. Massaro, Dominic W. 1978. An information-processing model of understanding speech. In David Gerver & H. Wallace Sinaiko (eds.), Language interpretation and communication, 299–314. Boston: Springer US. DOI: 10.1007/978-1-4615- 9077-4_26. Matysiak, Anna. 2001. Controlled processing in simultaneous interpretation: A study based on Daniel Gile’s Effort Models. Poznan: University of Poznan. (MA thesis). Mauri, Daniele. 2021. QEMI-C: An ad hoc corpus for quality evaluation of machine interpreting. Germersheim: University of Mainz. (MA thesis). Mayer, Felix. 2019. Terminologiearbeit und Terminographie. In Petra Drewer & Donatella Pulitano (eds.), Terminologie: Epochen–Schwerpunkte–Umsetzungen, 83–93. Berlin: Springer. Mayer, Richard E. & Logan Fiorella. 2014. Principles for reducing extraneous pro- cessing in multimedia learning: Coherence, signaling, redundancy, spatial con- tiguity, and temporal contiguity principles. In Richard Mayer (ed.), The Cam- bridge handbook of multimedia learning, 279–315. Cambridge: Cambridge Uni- versity Press. DOI: 10.1017/CBO9781139547369.015. 287 References McDonald, Janet L. & Patricia A. Carpenter. 1981. Simultaneous translation: Idiom interpretation and parsing heuristics. Journal of Verbal Learning and Verbal Behavior 20(2). 231–247. DOI: 10.1016/S0022-5371(81)90397-2. McLeod, Peter. 1977. A dual task response modality effect: Support for multipro- cessor models of attention. Quarterly Journal of Experimental Psychology 29(4). 651–667. DOI: 10.1080/14640747708400639. McNamara, Danielle S., Arthur C. Graesser, Philip M. McCarthy & Zhiqiang Cai. 2014. Automated evaluation of text and discourse with Coh-Metrix. New York: Cambridge University Press. Mead, Peter. 2002. Exploring hesitation in consecutive interpreting: An empirical study. In Giuliana Garzone & Maurizio Viezzi (eds.), Interpreting in the 21st cen- tury: Challenges and opportunities, 73–82. Amsterdam: John Benjamins. DOI: 10.1075/btl.43.08mea. Mead, Peter. 2005. Methodological issues in the study of interpreters’ fluency. The Interpreters’ Newsletter 13. 39–63. http://hdl.handle.net/10077/2469 (24 August, 2021). Meak, Lidia. 1990. Interprétation simultanée et congrès médical: Attentes et com- mentaires. The Interpreters’ Newsletter 3. 8–13. http://hdl.handle.net/10077/ 2148 (24 August, 2021). Mellinger, Christopher D. 2019. Computer-assisted interpreting technologies and interpreter cognition: A product and process-oriented perspective. Trad- umàtica: tecnologies de la traducció 17. 33–44. DOI: 10.5565/rev/tradumatica. 228. Mellinger, Christopher D. References & Thomas A. Hanson. 2017. Quantitative research meth- ods in translation and interpreting studies. London: Routledge. Mertin, Elvira. 2006. Prozessorientiertes Qualitätsmanagement im Dienstleistungs- bereich Übersetzen. Berlin: Peter Lang. Mikkelsen, Hans Kristian. 1991. Arntz, Reiner/Heribert Picht: Einführung in die Terminologiearbeit. Hildesheim, Zürich, New York: Georg Olms Verlag 1989. 344 Seiten. Hermes 6. 161–173. Miller, George A. 1956. The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review 63(2). 81–97. DOI: 10.1037/h0043158. Mizuno, Akira. 2005. Process model for simultaneous interpreting and working memory. Meta: Journal des Traducteurs/Meta: Translators’ Journal 50(2). 739– 752. DOI: 10.7202/011015ar. Montecchio, Maddalena. 2021. Defining maximum acceptable latency of ASR- enhanced CAI tools. Quantitative and qualitative assessment of the impact of 288 ASR latency on interpreters’ performance. Germersheim: University of Mainz. (MA thesis). ASR latency on interpreters’ performance. Germersheim: University of Mainz. (MA thesis). Moorkens, Joss & Sharon O’Brien. 2015. Post-editing evaluations: Trade-offs be- tween novice and professional participants. In Proceedings of the 18th annual conference of the European Association for Machine Translation, 75–81. Moorkens, Joss & Sharon O’Brien. 2016. Assessing user interface needs of post- editors of machine translation. In Dorothy Kenny (ed.), Human issues in trans- lation technology, 109–130. London: Taylor & Francis. Moser, Barbara. 1978. Simultaneous interpretation: A hypothetical model and its practical application. In David Gerver & H. Wallace Sinaiko (eds.), Language interpretation and communication, 353–368. Boston: Springer. Moser, Peter. 1995. Survey on expectations of users of conference interpretation. Vi- enna: SRZ Stadt + Regionalforschung. Moser, Peter. 1996. Expectations of users of conference interpretation. Interpret- ing: International Journal of Research and Practice in Interpreting 1(2). 145–178. DOI: 10.1075/intp.1.2.01mos. Moser-Mercer, Barbara. 1992. Banking on terminology: Conference interpreters in the electronic age. Meta: Journal des Traducteurs/Meta: Translators’ Journal 37(3). 507–522. DOI: 10.7202/003634ar. Moser-Mercer, Barbara. 1997. Process models in simultaneous interpretation. In Christa Hauenschild & Susanne Heizmann (eds.), Machine translation and translation theory, 3–18. Berlin: De Gruyter Mouton. DOI: 10 . 1515 / 9783110802474.3. Moser-Mercer, Barbara. 2005. Remote interpreting: The crucial role of presence. Bulletin Suisse de Linguistique Appliquée 81. 73–97. https : / / core . ac . uk / download/20649880.pdf (24 August, 2021). Moser-Mercer, Barbara, Ulrich Frauenfelder, Beatriz Casado & Alexander Künzli. 2000. Searching to define expertise in interpreting. In Birgitta Englund Dim- itrova & Kenneth Hyltenstam (eds.), Language processing and simultaneous in- terpreting: Interdisciplinary perspectives, 107–132. Amsterdam: John Benjamins. DOI: 10.1075/btl.40.09mos. Mudersbach, Klaus. 1999. References Die holistische Betrachtung von Fachtexten und deren Übersetzung. In Heidrun Gerzymisch-Arbogast, Daniel Gile, Juliane House & Annely Rothkegel (eds.), Wege der Übersetzungs- und Dolmetschforschung, 13– 42. Tübingen: Gunter Narr. Muñoz Martín, Ricardo & Celia Martín de León. 2020. Translation and cogni- tive science. In Fábio Alves & Arnt Lykke Jakobsen (eds.), The Routledge hand- book of translation and cognition, 52–68. London: Routledge. DOI: 10.4324/ 9781315178127-5. 289 References Napier, Jemina. 2004. Interpreting omissions: A new perspective. Interpreting: In- ternational Journal of Research and Practice in Interpreting 6(2). 117–142. DOI: 10.1075/intp.6.2.02nap. Napier, Jemina. 2016. Linguistic coping strategies in sign language interpreting. Washington: Gallaudet University Press. Navigli, Roberto, Michele Bevilacqua, Simone Conia, Dario Montagnini & Francesco Cecconi. 2021. Ten years of BabelNet: A survey. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, 4559–4567. Montreal: International Joint Conferences on Artificial Intelligence Organiza- tion. DOI: 10.24963/ijcai.2021/620. Navigli, Roberto & Simone Paolo Ponzetto. 2012. BabelNet: The automatic con- struction, evaluation and application of a wide-coverage multilingual semantic network. Artificial Intelligence 193. 217–250. DOI: 10.1016/j.artint.2012.07.001. Norman, Donald A. 1968. Toward a theory of memory and attention. Psychologi- cal Review 75(6). 522–536. DOI: 10.1037/h0026699. O’Brien, Sharon. 2006. Pauses as indicators of cognitive effort in post-editing machine translation output. Across Languages and Cultures 7(1). 1–21. DOI: 10. 1556/Acr.7.2006.1.1. O’Brien, Sharon. 2008. Processing fuzzy matches in translation memory tools: An eye-tracking analysis. In Susanne Göpferich, Arnt Lykke Jakobsen & Inger M. Mees (eds.), Looking at eyes: Eye-tracking studies of reading and translation pro- cessing (Copenhagen Studies in Language), 79–102. Frederiksberg: Samfund- slitteratur. O’Brien, Sharon. 2009. Eye tracking in translation process research: Method- ological challenges and solutions. In Inger M. Mees, Fábio Alves & Susanne Göpferich (eds.), Methodology, technology and innovation in translation process research, 251–266. Copenhagen: Samfundslitteratur. O’Brien, Sharon. 2012. Translation as human–computer interaction. Translation Spaces 1. 101–122. DOI: 10.1075/ts.1.05obr. O’Brien, Sharon, Laura Winther Balling, Michael Carl, Michel Simard & Lucia Specia (eds.). 2014. Post-editing of machine translation: Processes and applica- tions. United Kingdom: Cambridge Scholars Publishing. O’Brien, Sharon, Maureen Ehrensberger-Dow, Megan Connolly & Marcel Hasler. 2017. Irritating CAT tool features that matter to translators. HERMES: Journal of Language and Communication in Business 56. 145–162. DOI: 10.7146/hjlcb. v0i56.97229. Ogden, Charles Kay & Ivor Armstrong Richards. 1923. The meaning of meaning: A study of the influence of language upon thought and of the science of symbolism. New York: Harcourt, Brace & World. 290 Oléron, Pierre & Hubert Nanpon. References 2002. Research into simultaneous translation. In Franz Pöchhacker & Miriam Shlesinger (eds.), The interpreting studies reader, 43–50. London: Routledge. Olohan, Maeve. 2008. Scientific and technical translation. In Mona Baker & Gabriela Saldanha (eds.), Routledge encyclopedia of translation studies, 246–249. London: Routledge Olohan, Maeve. 2013. Scientific and technical translation. In Carmen Millán & Francesca Bartrina (eds.), The Routledge handbook of translation studies, 425– 437. London: Routledge. Orlando, Marc. 2010. Digital pen technology and consecutive interpreting: An- other dimension in notetaking training and assessment. The Interpreters’ Newsletter 15. 71–86. http://hdl.handle.net/10077/4750 (24 August, 2021). Orlando, Marc. 2014. A study on the amenability of digital pen technology in a hybrid mode of interpreting: Consec-simul with notes. Translation & Interpret- ing 6(2). 39–54. DOI: 10.12807/ti.106202.2014.a03. Oster, Katharina. 2019. Lexical activation and inhibition of cognates among transla- tion students. Germersheim: Johannes Gutenberg-Universität Mainz. (Doctoral dissertation). DOI: 10.25358/openscience-3748. Paas, Fred, Juhani E. Tuovinen, Huib Tabbers & Pascal W. M. Van Gerven. 2003. Cognitive load measurement as a means to advance cognitive load theory. Ed- ucational Psychologist 38(1). 63–71. DOI: 10.1207/S15326985EP3801_8. Padilla, Francisca, Maria Teresa Bajo & Pedro Macizo. 2005. Articulatory sup- pression in language interpretation: Working memory capacity, dual tasking and word knowledge. Bilingualism: Language and Cognition 8(3). 207–219. DOI: 10.1017/S1366728905002269. Paone, Matteo Domenico. 2016. Mobile Geräte beim Simultandolmetschen mit besonderem Bezug auf Tablets. Vienna: University of Vienna. (MA thesis). Papagno, Costanza, Alessandro Comi, Marco Riva, Alberto Bizzi, Mirta Vernice, Alessandra Casarotti, Enrica Fava & Lorenzo Bello. 2017. Mapping the brain network of the phonological loop. Human Brain Mapping 38(6). 3011–3024. DOI: 10.1002/hbm.23569. Pashler, Harold E. 1998. The psychology of attention. Cambridge: MIT Press. Pavel, Silvia & Diane Nolet. 2001. Handbook of terminology. Gatineau: Minister of Public Works & Government Services Canada. Petsche, Hellmuth, Susan C. Etlinger & Oliver Filz. 1993. Brain electrical mech- anisms of bilingual speech management: An initial investigation. Electroen- cephalography and Clinical Neurophysiology 86(6). 385–394. DOI: 10.1016/0013- 4694(93)90134-H. 291 References Pignataro, Clara. 2012. Terminology and interpreting in LSP conferences: A computer-aided vs. empirical-based approach. In Cynthia J. Kellett Bidoli (ed.), Interpreting across genres: Multiple research perspectives, 125–140. Trieste: EUT Edizioni Università di Trieste. Pio, Sonia. 2003. The relation between ST delivery rate and quality in simultane- ous interpretation. The Interpreters’ Newsletter 12. 69–100. http://hdl.handle. net/10077/2475 (3 April, 2022). Pisani, Elisabetta & Claudio Fantinuoli. 2021. Measuring the impact of automatic speech recognition on number rendition in simultaneous interpreting. References In Cai- wen Wang & Binghan Zheng (eds.), Empirical studies of translation and in- terpreting: The post-structuralist approach, 181–197. London: Routledge. DOI: 10.4324/9781003017400. Plevoets, Koen & Bart Defrancq. 2016. The effect of informational load on dis- fluencies in interpreting: A corpus-based regression analysis. Translation and Interpreting Studies 11(2). 202–224. DOI: 10.1075/tis.11.2.04ple. Plevoets, Koen & Bart Defrancq. 2018. The cognitive load of interpreters in the European Parliament: A corpus-based study of predictors for the disfluency uh(m). Interpreting: International Journal of Research and Practice in Interpret- ing 20(1). 1–28. DOI: 10.1075/intp.00001.ple. Pöchhacker, Franz. 1994. Simultandolmetschen als komplexes Handeln. Tübingen: Gunter Narr. Pöchhacker, Franz. 2001. Quality assessment in conference and community in- terpreting. Meta: Journal des Traducteurs/Meta: Translators’ Journal 46(2). 410– 425. DOI: 10.7202/003847ar. Pöchhacker, Franz. 2004. Introducing interpreting studies. 1st ed. Amsterdam: John Benjamins. Pöchhacker, Franz. 2011. Simultaneous interpreting. In Franz Pöchhacker (ed.), Routledge encyclopaedia of interpreting studies, 381–382. London: Routledge. Pöchhacker, Franz. 2015. Routledge encyclopedia of interpreting studies. London: Routledge. Pöchhacker, Franz. 2016. Introducing interpreting studies. 2nd ed. London: Rout- ledge. Poole, Alex & Linden J. Ball. 2005. Eye tracking in human-computer interac- tion and usability research. In Claude Ghaoui (ed.), Encyclopedia of human- computer interaction, 211–219. Pennsylvania: Idea Group. Popovic, Maja. 2011. Hjerson: An open source tool for automatic error classifica- tion of machine translation output. The Prague Bulletin of Math and Linguistics 96. 59–68. DOI: 10.2478/v10108-011-0011-4. 292 Posner, Michael I. 1980. Orienting of attention. Quarterly Journal of Experimental Psychology 32(1). 3–25. DOI: 10.1080/00335558008248231. Postle, Bradley R. 2006. Working memory as an emergent property of the mind and brain. Neuroscience 139(1). 23–38. DOI: 10.1016/j.neuroscience.2005.06.005. Prandi, Bianca. 2015a. L’uso di Interpretbank nella didattica dell’interpretazione: Uno studio esplorativo. Forlì: Università di Bologna. (MA thesis). https : / / amslaurea.unibo.it/id/eprint/8206 (24 August, 2021). Prandi, Bianca. 2015b. The use of CAI tools in interpreters’ training: A pilot study. In Proceedings of the 37th Conference Translating and the Computer, 48–57. Lon- don: AsLing. Prandi, Bianca. 2017. Designing a multimethod study on the use of CAI tools during simultaneous interpreting. In João Esteves-Ferreira, Juliet Macan, Rus- lan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 39th Conference Translating and the Computer, 76–88. Geneva: Editions Tradulex. Prandi, Bianca. 2018. An exploratory study on CAI tools in simultaneous inter- preting: Theoretical framework and stimulus validation. In Claudio Fantinuoli (ed.), Interpreting and technology, 29–59. Berlin: Language Science Press. DOI: 10.5281/zenodo.1493293. Prandi, Bianca. 2020. References The use of CAI tools in interpreter training: Where are we now and where do we go from here? inTRAlinea Special Issue: Technology in Interpreter Education and Practice. 1–10. Price, Cathy J., David W. Green & Roswitha von Studnitz. 1999. A functional imaging study of translation and language switching. Brain 122(12). 2221–2235. DOI: 10.1093/brain/122.12.2221. Pym, Anthony. 2008. On omission in simultaneous interpreting: Risk analysis of a hidden effort. In Gyde Hansen, Andrew Chesterman & Heidrun Gerzymisch- Arbogast (eds.), Efforts and models in interpreting and translation research: A tribute to Daniel Gile, 83–105. Amsterdam: John Benjamins. DOI: 10.1075/btl. 80.08pym. Pym, Anthony. 2011. What technology does to translating. Translation & Inter- preting 3(1). 1–9. Rayner, Keith. 1998. Eye movements in reading and information processing: 20 years of research. Psychological Bulletin 124(3). 372. DOI: 10.1037/0033-2909. 124.3.372. Rayner, Keith & Susan A. Duffy. 1986. Lexical complexity and fixation times in reading: Effects of word frequency, verb complexity, and lexical ambiguity. Memory & Cognition 14(3). 191–201. DOI: 10.3758/BF03197692. 293 References Rehbein, Jochen, Thomas Schmidt, Franziska Watzke & Annette Herkenrath. 2004. Handbuch für das computergestützte Transkribieren nach HIAT. Ar- beiten zur Mehrsprachigkeit B(56). 1–79. Riccardi, Alessandra. 2005. On the evolution of interpreting strategies in simul- taneous interpreting. Meta: Journal des Traducteurs/Meta: Translators’ Journal 50(2). 753–767. DOI: 10.7202/011016ar. Riccardi, Alessandra. 2015. Speech rate. In Franz Pöchhacker (ed.), Routledge en- cyclopaedia of interpreting studies, 397–399. London: Routledge. Rinne, Juha O., Jorma Tommola, Matti Laine, Bernd J. Krause, D. Schmidt, Valtteri Kaasinen, Mika Teräs, Hanni Sipilä & Marianna Sunnari. 2000. The translat- ing brain: Cerebral activation patterns during simultaneous interpreting. Neu- roscience Letters 294(2). 85–88. DOI: 10.1016/S0304-3940(00)01540-8. Rodríguez, Nadia & Bettina Schnell. 2009. A look at terminology adapted to the requirements of interpretation. Language Update 6(1). 1–5. Rodríguez, Susana, Roberto Gretter, Marco Matassoni, Daniele Falavigna, Ál- varo Alonso, Oscar Corcho & Mariano Rico. 2021. SmarTerp: A CAI System to Support Simultaneous Interpreters in Real-Time. In Proceedings of the Trans- lation and Interpreting Technology Online Conference TRITON 2021, 102–109. Shoumen: INCOMA Ltd. DOI: 10.26615/978-954-452-071-7_012. Rohrer, Doug, Harold Pashler & Jason Etchegaray. 1998. When two memories can and cannot be retrieved concurrently. Memory & Cognition 26(4). 731–739. DOI: 10.3758/BF03211393. Rosado, Tony. 2013. Note taking with iPad: Making our life easier. (Blog article). http://rpstranslations.wordpress.com/2013/05/28/note-taking-with-ipad- making-our-life- (3 July, 2021). Rösener, Christoph. 2016. Eye tracking and beyond: The dos and don’ts of creat- ing a contemporary usability lab. References In Silvia Hansen-Schirra & Sambor Grucza (eds.), Eyetracking and applied linguistics, 143–163. Berlin: Language Science Press. DOI: 10.17169/langsci.b108.297. Roy, Cynthia B. 1996. An interactional sociolinguistic analysis of turn-taking in an interpreted event. Interpreting: International Journal of Research and Practice in Interpreting 1(1). 39–67. DOI: 10.1075/intp.1.1.04roy. Roy, Cynthia B. 2000. Interpreting as a discourse process. Oxford: Oxford Univer- sity Press. Rütten, Anja. 2000. Terminologieprogramme für Dolmetscher. Saarbrücken: Uni- versität des Saarlandes. (MA thesis). Rütten, Anja. 2004. Why and in what sense do conference interpreters need spe- cial software? Linguistica Antverpiensia, New Series – Themes in Translation Studies 3. 167–178. DOI: 10.52034/lanstts.v3i.110. 294 Rütten, Anja. 2007. Informations- und Wissensmanagement im Konferenzdol- metschen (Sabest: Saarbrücker Beiträge zur Sprach- und Translationswis- senschaft 14). Frankfurt am Main: Peter Lang. Rütten, Anja. 2008. Zielgerichtet und effizient Wissen managen. MDÜ 1. 22–25. Rütten, Anja. 2008. Zielgerichtet und effizient Wissen managen. MDÜ 1. 22–25. Rütten, Anja. 2011. Translation-Memory-Systeme: Auch für Dolmetscher? MDÜ Rütten, Anja. 2011. Translation-Memory-Systeme: Auch für Dolmetscher? MDÜ 5. 42–43. Rütten, Anja. 2012a. From desk to booth: TMs for interpreters. MultiLingual 23. 43–47. Rütten, Anja. 2012b. Pragmatik in der Informations- und Wissensarbeit von Konferenzdolmetschern. In Barbara Ahrens, Silvia Hansen-Schirra, Monika Krein-Kühle, Michael Schreiber & Ursula Wienen (eds.), Translationswis- senschaftliches Kolloquium II, Beiträge zur Übersetzungs- und Dolmetschwis- senschaft, Köln/Germersheim, 261–271. Frankfurt/Main: Peter Lang. Rütten, Anja. 2013. Terminologiesysteme im Dolmetscheinsatz. eDITion 1. 25–28. Rütten, Anja. 2017. Terminology management tools for conference interpreters: Current tools and how they address the specific needs of interpreters. In João Esteves-Ferreira, Juliet Macan, Ruslan Mitkov & Olaf-Michael Stefanov (eds.), Proceedings of the 39th Conference Translating and the Computer, 98–103. Geneva: Editions Tradulex. Rütten, Anja. 2018. Can interpreters’ booth notes tell us what really matters in terms of information and terminology management? In Proceedings of the 40th Conference Translating and the Computer, 132–144. London: AsLing. SAE International. 2001. SAE J2450: Translation quality metric. Warrendale: Soci- ety of Automotive Engineers. g Salevsky, Heidemarie. 1987. Probleme des Simultandolmetschens. Eine Studie zur Handlungsspezifik. Berlin: Akademie der Wissenschaften der DDR. Sand, Peter. 2003. Manage your terminology with Interplex. https : / / aiic . org / document / 541 / AIICWebzine _ Dec2003 _ 3 _ SAND _ Manage _ your _ terminology_with_Interplex_EN.pdf (24 June, 2021). Sand, Peter. 2010. The new Interplex: Glossaries made fast and easy. https://aiic. org/document/813/AIICWebzine_Winter2011_5_SAND_The_new_Interplex_ glossaries_made_fast_and_easy_EN.pdf (25 June, 2022). Scarpa, Federica. 2010. La traduction spécialisée: Une approche professionnelle à l’enseignement de la traduction. References Ottawa: University of Ottawa Press.ff Schaeffer, Moritz & Michael Carl. 2014. Measuring the cognitive effort of literal translation processes. In Proceedings of the EACL 2014 workshop on humans and computer-assisted translation, 29–37. Gothenburg: Association for Compu- tational Linguistics. DOI: 10.3115/v1/W14-0306. 295 References Schaeffer, Moritz, Sandra L. Halverson & Silvia Hansen-Schirra. 2019. “Monitor- ing” in translation: The role of visual feedback. Translation, Cognition & Be- havior 2(1). 1–34. DOI: 10.1075/tcb.00017.sch. Schaeffer, Moritz, David Huepe, Silvia Hansen-Schirra, Sascha Hofmann, Edin- son Muñoz, Boris Kogan, Eduar Herrera, Agustín Ibáñez & Adolfo M. Gar- cía. 2020. The Translation and Interpreting Competence Questionnaire: An online tool for research on translators and interpreters. Perspectives 28(1). 90– 108. DOI: 10.1080/0907676X.2019.1629468. Schepens, Job, Ton Dijkstra & Franc Grootjen. 2012. Distributions of cognates in Europe as based on Levenshtein distance. Bilingualism: Language and Cogni- tion 15(1). 157–166. DOI: 10.1017/S1366728910000623. Schjoldager, Anne. 1995. An exploratory study of translational norms in simulta- neous interpreting. Methodological reflections. In Franz Pöchhacker & Miriam Shlesinger (eds.), The interpreting studies reader, 301–311. London: Routledge. Schmidt, Thomas & Kai Wörner. 2014. EXMARaLDA. In Jacques Durand, Ulrike Gut & Gjert Kristoffersen (eds.), The Oxford handbook of corpus phonology, 402– 419. Oxford: Oxford University Press. Schultheis, Holger & Anthony Jameson. 2004. Assessing cognitive load in adap- tive hypermedia systems: Physiological and behavioral methods. In Paul M. E. De Bra & Wolfgang Nejdl (eds.), Adaptive hypermedia and adaptive web-based systems, 225–234. Berlin: Springer. Schumacher, Eric H., Travis L. Seymour, Jennifer M. Glass, David E. Fencsik, Er- ick J. Lauber, David E. Kieras & David E. Meyer. 2001. Virtually perfect time sharing in dual-task performance: Uncorking the central cognitive bottleneck. Psychological Science 12(2). 101–108. DOI: 10.1111/1467-9280.00318. SDI München. 2007. Detaillierte Angaben zur Projektarbeit von Studierenden des SDI München. MDÜ 3. 26–36. SDL. 2008. Terminology: An end-to-end perspective: SDL research paper. (Blog ar- ticle). http://www.sdl.com/en/globalization-knowledge-centre/research_ results/terminology-an-end-to-end-%20perspective.asp (1 November, 2021). Seeber, Kilian G. 2005. Temporale Aspekte der Antizipation beim Simultandol- metschen komplexer SOV-Strukturen aus dem Deutschen. Bulletin Suisse de Linguistique Appliquée 81. 123–140. Seeber, Kilian G. 2007. Thinking outside the cube: Modeling language processing tasks in a multiple resource paradigm. In Interspeech 2007. Antwerp. Seeber, Kilian G. 2011. Cognitive load in simultaneous interpreting: Existing theories—new models. Interpreting: International Journal of Research and Prac- tice in Interpreting 13(2). 176–204. DOI: 10.1075/intp.13.2.02see. 296 Seeber, Kilian G. 2012. Multimodal input in simultaneous interpreting: An eye- tracking experiment. In Lew. N. References Zybatow, Alena Petrova & Michael Us- taszewski (eds.), Proceedings of the 1st International Conference TRANSLATA, Translation & Interpreting Research: Yesterday - today - tomorrow, May, 2011, Innsbruck, 341–347. Frankfurt am Main: Peter Lang. Seeber, Kilian G. 2013. Cognitive load in simultaneous interpreting: Measures and methods. Target: International Journal of Translation Studies 25(1). 18–32. DOI: 10.1075/target.25.1.03see. Seeber, Kilian G. 2017. Multimodal processing in simultaneous interpreting. In John W. Schwieter & Aline Ferreira (eds.), The handbook of translation and cognition, 461–475. Hoboken: Wiley. DOI: 10.1002/9781119241485.ch25. Seeber, Kilian G., Laura Keller & Alexis Georges Hervais-Adelman. 2020. When the ear leads the eye: The use of text during simultaneous interpretation. Lan- guage, Cognition and Neuroscience 35(10). 1480–1494. DOI: 10.1080/23273798. 2020.1799045. Seeber, Kilian G. & Dirk Kerzel. 2012. Cognitive load in simultaneous interpreting: Model meets data. International Journal of Bilingualism 16(2). 228–242. DOI: 10.1177/1367006911402982. Seel, Norbert M. (ed.). 2012. Encyclopedia of the sciences of learning. New York: Springer. Seleskovitch, Danica. 1976. Interpretation, a psychological approach to transla- tion. In Richard W. Brislin (ed.), Translation: Applications and research, 92–116. New York: Gardner Press. Seleskovitch, Danica. 1978. Language and cognition. In David Gerver & H. Wal- lace Sinaiko (eds.), Language interpretation and communication, 333–341. Berlin: Springer. Semenza, Carlo & Claudio Luzzatti. 2014. Combining words in the brain: The processing of compound words. Introduction to the special issue. Cognitive Neuropsychology 31(1-2). 1–7. DOI: 10.1080/02643294.2014.898922. SensoMotoric Instruments. 2017a. Begaze manual. Version 3.7. http://www.humre. vu.lt/files/doc/Instrukcijos/SMI/BeGaze2.pdf (17 March, 2021). SensoMotoric Instruments. 2017b. Experiment center Version 3.7. Setton, Robin. 1999. Simultaneous interpretation: A cognitive-pragmatic analysis. Amsterdam: John Benjamins. Setton, Robin. 2003. Models of the interpreting process. In Angela Collados Aís & José Antonio Sabio Panilla (eds.), Avances en la investigación sobre inter- pretación, 29–89. Granada: Editorial Comares. Seubert, Sabine. 2019. Visuelle Informationen beim Simultandolmetschen: Eine Eyetracking-Studie. Berlin: Frank & Timme. 297 References Shannon, Claude E. & Warren Weaver. 1949. The mathematical theory of commu- nication. Champaign: University of Illinois Press. Shlesinger, Miriam. 1998. Corpus-based interpreting studies as an offshoot of corpus-based translation studies. Meta: Journal des Traducteurs/Meta: Trans- lators’ Journal 43(4). 486–493. DOI: 10.7202/004136ar. Shlesinger, Miriam. 2000. Strategic allocation of working memory and other atten- tional resources in simultaneous interpreting. Ramat Gan: Bar-Ilan University. (Doctoral dissertation). Shreve, Gregory M., Isabel Lacruz & Erik Angelone. 2010. Cognitive effort, syn- tactic disruption, and visual interference in a sight translation task. In Gregory M. Shreve & Erik Angelone (eds.), Translation and cognition, 63–84. References Amster- dam: John Benjamins. DOI: 10.1075/ata.xv.05shr. Simons, Daniel J. & Christopher F. Chabris. 1999. Gorillas in our midst: Sustained inattentional blindness for dynamic events. Perception 28(9). 1059–1074. DOI: 10.1068/p281059. Smallwood, Jonathan & Jonathan W. Schooler. 2006. The restless mind. Psycho- logical Bulletin 132(6). 946–958. DOI: 10.1037/0033-2909.132.6.946. Smith, Philip L. & Daniel R. Little. 2018. Small is beautiful: In defense of the small- N design. Psychonomic Bulletin & Review 25(6). 2083–2101. DOI: 10.3758/s13423- 018-1451-8. Snover, Matthew, Bonnie Dorr, Rich Schwartz, Linnea Micciulla & John Makhoul. 2006. A study of translation edit rate with targeted human annotation. In Pro- ceedings of the 7th Conference of the Association for Machine Translation in the Americas: Technical Papers, 223–231. Cambridge: Association for Machine Translation in the Americas. Spinner, Patti, Susan M. Gass & Jennifer Behney. 2013. Ecological validity in eye-tracking: An empirical study. Studies in Second Language Acquisition 35(2). 389–415. DOI: 10.1017/S0272263112000927. Staub, Adrian & Keith Rayner. 2007. Eye movements and on-line comprehen- sion processes. In M. Gareth Gaskell (ed.), The Oxford handbook of psycholin- guistics, 326–342. Oxford: Oxford University Press. DOI: 10.1093/oxfordhb/ 9780198568971.013.0019. Steurs, Frieda, Iulianna van der Lek-Ciudin & Tom Vanallemeersch. 2017. How translators work in real-life: SCATE observations. In Translating for Europe forum. Brussels. http://ccl.kuleuven.be/scate/TEF.pdf (10 November, 2021). Stewart, Craig, Nikolai Vogler, Junjie Hu, Jordan Boyd-Graber & Graham Neu- big. 2018. Automatic estimation of simultaneous interpreter performance. In Proceedings of the 56th annual meeting of the Association for Computational Lin- 298 guistics (Volume 2: Short Papers), 662–666. Melbourne: Association for Compu- tational Linguistics. Stoll, Christoph. 2009. Jenseits simultanfähiger Terminologiesysteme: Methoden der Vorverlagerung und Fixierung von Kognition im Arbeitsablauf professioneller Konferenzdolmetscher. Trier: Wissenschaftlicher Verlag Trier. Stoll, Christoph. 2010. The Heidelberg model of simultaneouus interpreting. T21N: Translation in Transition. 1–26. Stroop, John Ridley. 1935. Studies of interference in serial verbal reactions. Jour- nal of Experimental Psychology 18(6). 643–662. DOI: 10.1037/h0054651. Stuss, Donald T. & Michael P. Alexander. 2007. Is there a dysexecutive syndrome? Philosophical Transactions of the Royal Society B: Biological Sciences 362(1481). 901–915. DOI: 10.1098/rstb.2007.2096. Sun, Sanjun & Gregory M. Shreve. 2014. Measuring translation difficulty: An em- pirical study. Target: International Journal of Translation Studies 26(1). 98–127. DOI: 10.1075/target.26.1.04sun. Sweller, John. 2005. The redundancy principle in multimedia learning. In Richard Mayer (ed.), The Cambridge handbook of multimedia learning, 159–168. Cam- bridge: Cambridge University Press. DOI: 10.1017/CBO9780511816819.011. Taravella, Anne Marie & Alain O. Vielleneuve. 2013. research, 81 110. 030-69777-8_4. ansen-Schirra S References Acknowledging the needs of computer-assisted translation tools users: The human perspective in human- machine translation. The Journal of Specialised Translation 19. 62–74.f Tardel, Anke. 2021. Measuring effort in subprocesses of subtitling: The case of post-editing via pivot language. In Michael Carl (ed.), Explorations in empirical translation process research, 81–110. Cham: Springer International Publishing. DOI: 10.1007/978-3-030-69777-8_4. Tardel, Anke, Silvia Hansen-Schirra, Silke Gutermuth & Moritz Schaeffer. 2019. Automatization of subprocesses in subtitling. In Proceedings of the second MEMENTO workshop on Modelling Parameters of Cognitive Effort in Transla- tion Production, 19–20. Dublin: European Association for Machine Translation. https://aclanthology.org/W19-7010 (24 June, 2021). Tardel, Anke, Silvia Hansen-Schirra, Moritz Schaeffer, Silke Gutermuth, Volker Denkel & Miriam Hagmann-Schlatterbeck. 2021. Attention distribution and monitoring during intralingual subtitling. In Tra&Co Group (ed.), Translation, interpreting, cognition: The way out of the box, 145–162. Berlin: Language Sci- ence Press. DOI: 10.5281/zenodo.4545043. Tarmizi, Rohani A. & John Sweller. 1988. Guidance during mathematical problem solving. Journal of Educational Psychology 80(4). 424–436. DOI: 10.1037/0022- 0663.80.4.424. 299 References Teixeira, Carlos S. C. & Sharon O’Brien. 2019. Investigating the cognitive er- gonomic aspects of translation tools in a workplace setting. In Hanna Risku, Regina Rogl & Jelena Milosevic (eds.), Translation practice in the field: Current research on socio-cognitive processes, 79–103. Amsterdam: John Benjamins. DOI: 10.1075/bct.105.05tei. Thayer, Julian F., Fredrik Åhs, Mats Fredrikson, John J. Sollers & Tor D. Wager. 2012. A meta-analysis of heart rate variability and neuroimaging studies: Impli- cations for heart rate variability as a marker of stress and health. Neuroscience & Biobehavioral Reviews 36(2). 747–756. DOI: 10.1016/j.neubiorev.2011.11.009. Timarová, Šárka. 2015. Time lag. In Franz Pöchhacker (ed.), Routledge encyclope- dia of interpreting studies, 418–420. Oxford: Routledge. Timarová, Šárka, Ivana Čeňková, Reine Meylaerts, Erik Hertog, Arnaud Szmalec & Wouter Duyck. 2014. Simultaneous interpreting and working memory ex- ecutive control. Interpreting: International Journal of Research and Practice in Interpreting 16(2). 139–168. DOI: 10.1075/intp.16.2.01tim. Timarová, Šárka, Ivana Čeňková, Reine Meylaerts, Erik Hertog, Arnaud Szmalec & Wouter Duyck. 2015. Simultaneous interpreting and working memory ca- pacity. In Aline Ferreira & John W. Schwieter (eds.), Psycholinguistic and cog- nitive inquiries into translation and interpreting, 101–126. Amsterdam: John Ben- jamins. Timarová, Šárka, Barbara Dragsted & Inge Gorm Hansen. 2011. Time lag in translation and interpreting: A methodological exploration. In Cecilia Alvs- tad, Adelina Hild & Elisabet Tiselius (eds.), Methods and strategies of process re- search: Integrative approaches in translation studies, 121–146. Amsterdam: John Benjamins. DOI: 10.1075/btl.94.10tim. Tiselius, Elisabet. 2015. Accuracy. References In Franz Pöchhacker (ed.), Routledge encyclo- pedia of interpreting studies, 3–4. London: Routledge. Tiselius, Elisabet & Kayle Sneed. 2020. Gaze and eye movement in dialogue in- terpreting: An eye-tracking study. Bilingualism: Language and Cognition 23(4). 780–787. DOI: 10.1017/S1366728920000309. Tissi, Benedetta. 2001. Silent pauses and disfluencies in simultaneous interpreta- tion. The Interpreter’s Newsletter 10. 103–127. http://hdl.handle.net/10077/2455 (11 May, 2022). Tommola, Jorma & Jukka Hyönä. 1990. Mental load in listening, speech shadowing and simultaneous interpreting: A pupillometric study. Paper presented at the 9th Meeting of the World Congress of Applied Linguistics, 2–11. Thessaloniki: ERIC. 300 Tommola, Jorma, Matti Laine, Marianna Sunnari & Juha O. Rinne. 2000. Images of shadowing and interpreting. Interpreting: International Journal of Research and Practice in Interpreting 5(2). 147–167. DOI: 10.1075/intp.5.2.06tom. Tommola, Jorma & Pekka Niemi. 1986. Mental load in simultaneous interpreting: An on-line pilot study. In Lars S. Evensen (ed.), Nordic research in text linguistics and discourse analysis, 171–184. Norway: Trondheim. Towse, John N. & Graham Hitch. 1995. Is there a relationship between task de- mand and storage space in tests of working memory capacity? The Quarterly Journal of Experimental Psychology Section A 48(1). 108–124. DOI: 10 . 1080 / 14640749508401379. Treisman, Anne M. 1964. Selective attention in man. British Medical Bulletin 20(1). 12–16. DOI: 10.1093/oxfordjournals.bmb.a070274. Treisman, Anne M. 1965. The effects of redundancy and familiarity on translating and repeating back a foreign and a native language. British Journal of Psychol- ogy 56(4). 369–379. DOI: 10.1111/j.2044-8295.1965.tb00979.x. Treisman, Anne M. & Alison Davies. 2012. Divided attention to ear and eye. In Jeremy Wolfe & Lynn Robertson (eds.), From perception to consciousness, 24– 31. Oxford: Oxford University Press. DOI: 10.1093/acprof:osobl/9780199734337. 003.0005. Tripepi Winteringham, Sarah. 2010. The usefulness of ICTs in interpreting prac- tice. The Interpreters’ Newsletter 15. 87–99. http://hdl.handle.net/10077/4751 (11 April, 2022). Valentini, Cristina. 2001. Uso del computer in cabina di interpretazione. Forlì: Uni- versità di Bologna. (MA thesis). Valentini, Cristina. 2002. Uso del computer in cabina di interpretazione: Inchiesta sui bisogni terminologici degli interpreti prima e durante la simultanea. https: //aiic.org/document/470/AIICWebzine_FebMar2002_9_VALENTINI_Uso_ del_computer_in_cabina_de_interpretazione_FULL_IT.pdf (22 June, 2022). Van Cauwenberghe, Goran. 2020. La reconnaissance automatique de la parole en interprétation simultanée. Gent: Universiteit Gent. (MA thesis). Van Merrienboer, Jeroen J. G. & John Sweller. 2005. Cognitive load theory and complex learning: Recent developments and future directions. Educational Psy- chology Review 17(2). 147–177. DOI: 10.1007/s10648-005-3951-0. Vardaro, Jennifer, Moritz Schaeffer & Silvia Hansen-Schirra. 2019. References Translation quality and error recognition in professional neural machine translation post- editing. Informatics 6(41). 1–29. DOI: 10.3390/informatics6030041. Veisbergs, Andrejs. 2007. Terminology issues in interpreter training. In Proceed- ings of the Baltic Sea Region university network “quality and qualifications Veisbergs, Andrejs. 2007. Terminology issues in interpreter training. In Proceed- ings of the Baltic Sea Region university network “quality and qualifications 301 References in translation and interpreting II”. http : / / www . tlu . ee / files / arts / 645 / Quali698bb7e395eob88d73d603e33f5b153f.pdf (22 May, 2020). Velásquez, Gonzalo. 2002. La traducción y la terminología en la comunicación bil- ingüe mediada. Meta: Journal des Traducteurs/Meta: Translators’ Journal 47(3). 444–459. DOI: 10.7202/008030ar. Vintar, Špela. 2001. Using parallel corpora for translation-oriented term extrac- tion. Babel 47(2). 121–132. DOI: 10.1075/babel.47.2.04vin. Vogler, Nikolai, Craig Stewart & Graham Neubig. 2019. Lost in interpretation: Pre- dicting untranslated terminology in simultaneous interpretation. In Proceed- ings of the 2019 conference of the North American chapter of the Association for Computational Linguistics: Human language technologies, 109–118. Minneapo- lis: Association for Computational Linguistics. Vuorikoski, Anna-Riitta. 1993. Simultaneous interpretation: User experience and expectations. In Catriona Picken (ed.), Translation: The vital link. Proceedings of the 13th world congress of FIT, 317–327. London: Institute of Translation & Interpreting. Vuorikoski, Anna-Riitta. 1998. User responses to simultaneous interpreting. In Lynne Bowker, Michael Cronin, Dorothy Kenny & Jennifer Pearson (eds.), Unity in diversity? Current trends in translation studies, 184–197. Manchester: St. Jerome. Wadensjö, Cecilia. 1993. The double role of a dialogue interpreter. Perspectives 1(1). 105–121. DOI: 10.1080/0907676X.1993.9961204. Wadensjö, Cecilia. 1998. Interpreting as interaction. London: Routledge. Wadensjö, Cecilia. 2005. Interpreting as interaction. 2nd edn. London: Longman. Wagener, Leonie. 2012. Vorbereitende Terminologiearbeit im Konferenzdolmet- schen unter besonderer Berücksichtigung der Zusammenarbeit im Dolmetsch- team. Köln: Fachhochschule Köln. (MA thesis). Wahlster, Wolfgang. 2000. Verbmobil: Foundations of speech-to-speech translation. Heidelberg: Springer. Wang, Xinyu & Caiwen Wang. 2019. Can computer-assisted interpreting tools assist interpreting? Transletters: International Journal of Translation and Inter- preting 2. 109–139. Warren, Tessa, Sarah J. White & Erik D. Reichle. 2009. Investigating the causes of wrap-up effects: Evidence from eye movements and E–Z reader. Cognition 111(1). 132–137. DOI: 10.1016/j.cognition.2008.12.011. Webex Help Center. 2022. Real time translation in Webex meetings and events. https://help.webex.com/en-US/article/7dzhht/Real-Time-Translation-in- Webex-Meetings-and-Events (13 August, 2022). 302 Weller, G. & P. Yanez. 1998. The audience’s views on simultaneous conference interpreters performance in Mexico. In Muriel M. References Jerome-O’Keefe (ed.), Pro- ceedings of the 38th annual conference of the American Translators Association, 69–85. San Francisco: American Translators Association. Wickens, Christopher D. 1976. The effects of divided attention on information processing in manual tracking. Journal of Experimental Psychology: Human Perception and Performance 2(1). 1–13. DOI: 10.1037/0096-1523.2.1.1. Wickens, Christopher D. 1984. Processing resources in attention. In Raja Para- sumaran & David R. Davies (eds.), Varieties of attention, 63–101. New York: Academic Press. Wickens, Christopher D. 2002. Multiple resources and performance predic- tion. Theoretical Issues in Ergonomics Science 3(2). 159–177. DOI: 10 . 1080 / 14639220210123806. Will, Martin. 2000. Bemerkungen zum Computereinsatz beim Simultandol- metschen. In Sylvia Kalina, Silke Buhl & Heidrun Gerzymisch Arbogast (eds.), Dolmetschen: Theorie, Praxis, Didaktik – mit ausgewählten Beiträgen der Saar- brücker Symposien, 125–135. Sankt Ingbert: Röhrig Universitätsverlag. Will, Martin. 2007. Terminology work for simultaneous interpreters in LSP con- ferences: Model and method. In Heidrun Gerzymisch-Arbogast & Gerhard Budin (eds.), Proceedings of the Marie Curie Euroconferences MuTra: LSP Trans- lation Scenarios, 65–99. Vienna: MuTra. Will, Martin. 2015. Zur Eignung simultanfähiger Terminologiesysteme für das Konferenzdolmetschen. trans-kom 8(1). 179–201. Will, Martin. 2020. Computer Aided Interpreting (CAI) for conference inter- preters. Concepts, content and prospects. ESSACHESS-Journal for Communi- cation Studies 13(25). 37–71. Wintringham, Nils. 2009. Terminus. http://www.wintringham.ch/cgi/ayawp.pl? T=terminus (13 August, 2022). Wright, Sue Ellen & Leland D. Wright. 1997. Terminology management for tech- nical translation. In Sue Ellen Wright & Gerhard Budin (eds.), Handbook of terminology management: Volume 1: Basic aspects of terminology management, 147–159. Amsterdam: John Benjamins. Wüster, Eugen. 1931. Internationale Sprachnormung in der Technik, besonders in der Elektrotechnik: Die nationale Sprachnormung und ihre Verallgemeinerung (Sprachforum: Zeitschrift fur angewandte Sprachwissenschaft Beiheft Nr. 2). Bonn: Bouvier. Wüster, Eugen. 1979. Einführung in die Allgemeine Terminologielehre und Termi- nologische Lexikographie. Wien: Romanistischer Verlag. 303 References Xu, Ran. 2015. Terminology preparation for simultaneous interpreters. Leeds: Uni- versity of Leeds. (Doctoral dissertation). Young, Laurence R. & David Sheena. 1975. Survey of eye movement recording methods. Behavior Research Methods & Instrumentation 7(5). 397–429. DOI: 10. 3758/BF03201553. Žganec-Gros, Jerneja, France Mihelič, Tomaž Erjavec & Špela Vintar. 2005. The VoiceTran speech-to-speech communicator. In Václav Matoušek, Pavel Mautner & Tomáš Pavelka (eds.), Text, speech and dialogue, 379–384. Berlin: Springer. Ziegler, Klaus & Sebastiano Gigliobianco. 2018. Present? Remote? Remotely present! New technological approaches to remote simultaneous conference interpreting. In Claudio Fantinuoli (ed.), Interpreting and technology, 119–139. Berlin: Language Science Press. DOI: 10.5281/zenodo.1493299. Zielinski, Daniel & Yamile Ramírez-Safar. 2005. References Research meets practice: t-survey 2005: An online survey on terminology extraction and terminology manage- ment. In Translating and the computer 27, 1–27. London: Aslib. Zwischenberger, Cornelia & Franz Pöchhacker. 2010. Survey on quality and role: Conference interpreters’ expectations and self-perceptions. https : / / aiic . org / document/9646/ (24 July, 2021). 304 Name index Bruya, Brian, 71 Buch-Kromann, Matthias, 109 Bühler, Hildegund, 12 Bundgaard, Kristine, 77 Burchardt, Aljoscha, 10 Bruya, Brian, 71 Buch-Kromann, Matthias, 109 Bühler, Hildegund, 12 Bundgaard, Kristine, 77 Burchardt, Aljoscha, 10 Ahrens, Barbara, 110 Aizawa, Akiko, 121 Alexander, Michael P., 65 Altman, Janet, 109 Alves, Fábio, 119, 121 Ammon, Ulrich, 8 Andersson, Bodil, 112 Apfelbaum, Birgit, 31 Arcara, Giorgio, 136 Arntz, Reiner, 13, 14 Atkinson, Richard C., 64 Ahrens, Barbara, 110 Aizawa, Akiko, 121 Alexander, Michael P., 65 Altman, Janet, 109 Alves, Fábio, 119, 121 Ammon, Ulrich, 8 Andersson, Bodil, 112 Cabré Castellví, Maria Teresa, 13, 14 Campo, Ángela, 8 Canali, Sara, 2 Arcara, Giorgio, 136 Carl, Michael, 27, 106, 109, 111, 113, Arntz, Reiner, 13, 14 Atkinson, Richard C., 64 Carpenter, Patricia A., 115, 120, 124 Baddeley, Alan D., 64–66, 79, 84, 85 Cattaruzza, Lorella, 11 Cavallo, Patrizia, 34, 139 Baigorri-Jalón, Jesús, 26, 27 Cecot, Michela, 110 Ball, Linden J., 117 Ball, Linden J., 117 Barik, Henri C., 10, 109, 113, 114, 145, Center for Reading Research, Ghent University, 152 Chabasse, Catherine, 123 Chabris, Christopher F., 70 Chandler, Paul, 75, 78 Chang, Chieh Ying, 120, 123 Chen, Sijia, 105–107, 122 Chernov, Ghelly V., 4 Chesterman, Andrew, 3 Chiaro, Delia, 12 Chmiel, Agnieszka, 91, 92, 96, 104, 107, 113, 120, 129 Cho, Eunah, 28 Christoffels, Ingrid K., 114, 157 Cieślewicz, Anna, 91, 92, 96, 107, 113 Conklin, Kathy, 118, 135, 140, 158, 164, Center for Reading Research, Ghent University, 152 157, 235 157, 235 Chabasse, Catherine, 123 Barrouillet, Pierre, 80 Chabris, Christopher F., 70 Barton, Jason J. S., 136 Chandler, Paul, 75, 78 Beatty, Jackson, 124, 125 Behl, Holly, 31 Chang, Chieh Ying, 120, 123 Berber-Irabien, Diana, 44, 46, 47, 50 Chen, Sijia, 105–107, 122 Chen, Sijia, 105–107, 122 Biagini, Giulio, 2, 55, 112, 127, 137, 143, 146, 159, 161, 214–216 Bilgen, Baris, 44, 46, 49 Bilgen, Baris, 44, 46, 49 Blancafort, Helena, 15, 16 113, 120, 129 Blumenthal, Pamela, 123 Cho, Eunah, 28 Braun, Peter, 8 Braun, Peter, 8 Braun, Peter, 8 Braun, Sabine, 26, 28, 30 EABM, 75, 228, 245 EABM, 75, 228, 245 Gile, Daniel, 1, 4, 10, 11, 18, 41, 52, 57, 82–86, 91, 95, 96, 98, 99, 107, 135, 137, 155, 212, 243 Gile, Daniel, 1, 4, 10, 11, 18, 41, 52, 57, 82–86, 91, 95, 96, 98, 99, 107, 135, 137, 155, 212, 243 Ehrensberger-Dow, Maureen, 51, 105, Ehrensberger-Dow, Maureen, 51, 105, 107, 108, 116, 118, 121 135, 137, 155, 212, 243 Ginns, Paul, 78 Ginns, Paul, 78 Ehlich, Konrad, 168 Godefroy, Olivier, 65 Godefroy, Olivier, 65 Eisenrieth Dokumentations GmbH, 36 Gojun, Anita, 15, 16 Eriksen, Barbara A., 70 Gojun, Anita, 15, 16 Eriksen, Charles W., 70 Goldman-Eisler, Frieda, 109 Ghiselli, Serena, 136f Ghiselli, Serena, 136f Drechsel, Alexander, 31, 32, 42, 46 Gieshoff, Anne Catherine, 65, 74, 90, 96, 100, 107, 110, 125, 126, 164, 169, 221 Gieshoff, Anne Catherine, 65, 74, 90, Gieshoff, Anne Catherine, 65, 74, 90, 96, 100, 107, 110, 125, 126, 164, 169, 221 Gigliobianco, Sebastiano, 26, 27 Drewer, Petra, 13, 14 Drewer, Petra, 13, 14 Duchowski, Andrew T., 116, 118f Duffy, Susan A., 135 Gigliobianco, Sebastiano, 26, 27 Christoffels, Ingrid K., 114, 157 Breitenbach, Francis W., 118 Broadbent, Donald E., 69 Brüsewitz, Nora, 57, 104 167 Name index Cooper, Alan, 76, 227 Corpas Pastor, Gloria, 44, 45, 47 Costa, Albert, 9, 155 Costa, Hernani, 34, 52, 53 Cowan, Nelson, 64, 66 Čulo, Oliver, 212 Cooper, Alan, 76, 227 Corpas Pastor, Gloria, 44, 45, 47 Costa, Albert, 9, 155 Costa, Hernani, 34, 52, 53 Cowan, Nelson, 64, 66 Čulo, Oliver, 212 Fabbro, Franco, 4, 67 Faber, Pamela, 9, 15 Fabbro, Franco, 4, 67 Faber, Pamela, 9, 15 Fantinuoli, Claudio, 1, 2, 20, 25, 27, 29–31, 34–38, 44, 46, 50, 52, 54–59, 127, 128, 139, 146, 148, 211, 214, 215, 221, 225, 228, 245 Fantinuoli, Claudio, 1, 2, 20, 25, 27, 29–31, 34–38, 44, 46, 50, 52, 54–59, 127, 128, 139, 146, 148, 211, 214, 215, 221, 225, 228, 245 29–31, 34–38, 44, 46, 50, 52, 54–59, 127, 128, 139, 146, 148, 211, 214, 215, 221, 225, 228, Fern, Lily May, 44, 45, 47 Fern, Lily May, 44, 45, 47 Darò, Valeria, 4, 67 Darò, Valeria, 4, 67 Ferreira, Aline, 80 Davies, Alison, 74 Fiorella, Logan, 104 De Groot, Annette M. B., 114, 157 Flesch, Rudolph, 157 De Merulis, Gianpiero, 53 Floros, Georgios, 19 de Saussure, Ferdinand, 13 Folaron, Debbie, 8 Defrancq, Bart, 2, 57, 58, 110, 114, 127, 148, 157, 214, 225, 228, 245 Fox, Wendy, 78, 225 Frittella, Francesca Maria, 2, 60, 127, 147, 215, 225 Déjean Le Féal, Karla, 4, 10 Desblache, Lucile, 8 Desblache, Lucile, 8 Gacek, Michael, 53, 146 Gacek, Michael, 53, 146 Desmet, Bart, 57, 164 García Becerra, Olalla, 11 García Becerra, Olalla, 11 Dessloch, Florian, 29, 227 Dessloch, Florian, 29, 227 García de Quesada, Mercedes, 10, 12 Deutsch, Diana, 69 Deutsch, Diana, 69 García, Adolfo M., 122 García, Adolfo M., 122 Garzone, Giuliana, 8 Deutsch, J. Anthony, 69 Deutsch, J. Anthony, 69 Garzone, Giuliana, 8 Garzone, Giuliana, 8 Díaz-Galaz, Stephanie, 229 Gerver, David, 4, 134 Gerver, David, 4, 134 Donovan, Clare, 4 Donovan, Clare, 4 Gerzymisch-Arbogast, Heidrun, 14, 18, Dragsted, Barbara, 67, 110, 111, 113, 115, Lavie, Nilli, 70 Lavie, Nilli, 70 Jacquemot, Charlotte, 65 Läubli, Samuel, 31 Jakobsen, Arnt Lykke, 75, 107, 111– 119, 121 Jakobsen, Arnt Lykke, 75, 107, 111– Lavie, Nilli, 70 131 Hansen, Inge Gorm, 110, 111, 113, 120 135, 136, 147, 163, 164f Kilgarriff, Adam, 16f Hansen-Schirra, Silvia, 8, 15, 36, 51, 56, 130, 149 , , , , , , 56, 130, 149 Hanson, Thomas A., 173 Hart, Sandra G., 107 Havelka, Jelena, 66 Kirchhoff, Hella, 81, 82 Hanson, Thomas A., 173 Klonowicz, Tatiana, 123 Klonowicz, Tatiana, 123 Knight, James L., 71 Klonowicz, Tatiana, 123 Knight, James L., 71 Koch, Gary G., 174 Hart, Sandra G., 107 Havelka, Jelena, 66 Koch, Gary G., 174 Havelka, Jelena, 66 Heid, Ulrich, 15, 16 Kopczynski, Andrzej, 11 Köpke, Barbara, 67 Herbig, Nico, 124 Hervais-Adelman, Alexis Georges, 123 Korpal, Paweł, 123 Koshkin, Roman, 84, 85 Hess, Eckhard H., 124 Kosma, Alexandra, 67 Hessels, Roy S., 169 Hitch, Graham, 64, 80, 84, 85 Krings, Hans P, 220 Kuang, Huolingxiao, 121 Hoberg, Felix, 29, 31 Kumpulainen, Minna, 110 Kurz, Ingrid, 4, 11, 123 Kumpulainen, Minna, 110 Hoeks, Bert, 125 Holley, Josephine Christine, 31, 32 Holmqvist, Kenneth, 118, 124, 158, 164 Lacruz, Isabel, 27, 110, 111, 113, 118, 121, 227 Lacruz, Isabel, 27, 110, 111, 113, 118, 121, 227 Hvelplund, Kristian Tangsgaard, 67, 113, 115–119, 124, 158, 164, 167 Lambert, Sylvie, 4 Lambert, Sylvie, 4 Hyönä, Jukka, 4, 124–126, 163, 164 Hyönä, Jukka, 4, 124–126, 163, 164 Landis, J. Richard, 174 Landis, J. Richard, 174 Laplace, Colette, 4 Jacquemot, Charlotte, 65 Hansen, Gyde, 106, 108, 117, 128, 129, 131 Hansen, Gyde, 106, 108, 117, 128, 129, 131 Kerzel, Dirk, 86, 88, 91, 98, 106, 125, 119, 121 Lazar, Jonathan, 130, 131, 167 Jameson, Anthony, 126 Lederer, Marianne, 4, 81, 114, 134, 157 Janikowski, Przemysław, 91, 92, 96, 104, 107, 113, 129 Janikowski, Przemysław, 91, 92, 96, 104 107 113 129 Lee, Tae-Hyung, 114, 235 Lei, Victoria, 107 131 Kerzel, Dirk, 86, 88, 91, 98, 106, 125, Lei, Victoria, 107 Jegerski, Jill, 135, 138, 140, 167 Leijten, Mariëlle, 112 Jia, Ye, 28 Leipzig Corpora Collection, 139 Jiang, Hong, 1, 44–46, 50 Levelt, Willem J. M., 125 Eriksen, Charles W., 70 Goldsmith, Joshua, 31, 32 European Parliament, DG Translation, Görög, Attila, 10 30 306 Name index Kahneman, Daniel, 71, 84, 124, 125 Kalderon, Eliza, 123 Kalina, Sylvia, 4, 8, 17, 19 Kantowitz, Barry H., 71 Karsh, Robert, 118 Keating Gregory D 135 138 140 167 Kahneman, Daniel, 71, 84, 124, 125 Gran, Laura, 4 Kalderon, Eliza, 123 Gumul, Ewa, 84, 85 Kalina, Sylvia, 4, 8, 17, 19 Halliday, Michael A. K., 9 Kantowitz, Barry H., 71 Karsh, Robert, 118 Hamidi, Miriam, 33 Keating, Gregory D., 135, 138, 140, 167 Keating, Gregory D., 135, 138, 140, 167 Hansen, Gyde, 106, 108, 117, 128, 129, Levelt, Willem J. M., 125 Just, Marcel A., 115, 124 Levenshtein, Vladimir I., 156 Polt, James M., 124 Moser, Peter, 12 O’Brien, Sharon, 31, 51, 75, 77, 110, 111, Martin, Reg, 36 113, 121, 164, 227, 228 Martínez, Silvia Montero, 9, 15 Ogden, Charles Kay, 13 Marzouk, Shaimaa, 108 Oléron, Pierre, 114, 157 Oléron, Pierre, 114, 157 Massaro, Dominic W., 4 Olohan, Maeve, 8, 9 Matysiak, Anna, 84 Matysiak, Anna, 84 Orlando, Marc, 32, 33 Orlando, Marc, 32, 33 Mauri, Daniele, 31 Mauri, Daniele, 31 Orrego-Carmona, David, 31 Orrego-Carmona, David, 31 Mayer, Felix, 14 Ortiz, Luis Eduardo Schild, 34, 139 Ortiz, Luis Eduardo Schild, 34, 139 Mayer, Richard E., 104 Oster, Katharina, 122 Oster, Katharina, 122 Mazur, Iwona, 120 Paas, Fred, 105, 106 Paas, Fred, 105, 106 McDonald, Janet L., 120 Padilla, Francisca, 67 Padilla, Francisca, 67 McLeod, Peter, 74 Paone, Matteo Domenico, 32 McNamara, Danielle S., 157 Papagno, Costanza, 65 Mead, Peter, 110 Pashler, Harold E., 80 Meak, Lidia, 11 Pavel, Silvia, 15, 16 Mellinger, Christopher D., 129, 173 Petsche, Hellmuth, 4, 122 Petsche, Hellmuth, 4, 122 Mertin, Elvira, 10 Mikkelsen, Hans Kristian, 13, 19 Picht, Heribert, 13, 14 Picht, Heribert, 13, 14 Pignataro, Clara, 8, 9 Pignataro, Clara, 8, 9 Miller, George A., 67 Mizuno, Akira, 67 Pio, Sonia, 134 Pisani, Elisabetta, 2, 127, 128 Montecchio, Maddalena, 58, 78, 104, 127, 178, 218, 228 Plevoets, Koen, 110 Moorkens, Joss, 31, 51, 75, 228 Pöchhacker, Franz, 3, 4, 10–12, 18, 33, 80, 84, 109, 113, 129 Moser, Barbara, 4, 52 Polt, James M., 124 Kageura, Kyo, 15 Kageura, Kyo, 15 Lijewska, Agnieszka, 104, 129 307 Name index Little, Daniel R., 130 Logie, Robert H., 65 Lommel, Arle, 10 Low, Renae, 79 Lu, Sara A., 74 Luzzatti, Claudio, 136 Little, Daniel R., 130 Muñoz Martín, Ricardo, 81 Muñoz Martín, Ricardo, 81 Little, Daniel R., 130 Logie, Robert H., 65 Lommel, Arle, 10 Low, Renae, 79 Lu, Sara A., 74 Luzzatti, Claudio, 136 Logie, Robert H., 65 Logie, Robert H., 65 Lommel, Arle, 10 Low, Renae, 79 Lu, Sara A., 74 Nanpon, Hubert, 114, 157 Napier, Jemina, 109 Navigli, Roberto, 35 Niemi, Pekka, 125 Nitzke, Jean, 130, 149 Nocella, Giuseppe, 12 Nolet, Diane, 15, 16 Norman, Donald A., 69 Luzzatti, Claudio, 136 Nocella, Giuseppe, 12 Marrone, Stefano, 11 Nolet, Diane, 15, 16 Marshman, Elizabeth, 15 Norman, Donald A., 69 O’Brien, Sharon, 31, 51, 75, 77, 110, 111, 113, 121, 164, 227, 228 Ogden, Charles Kay, 13 Oléron, Pierre, 114, 157 Olohan, Maeve, 8, 9 Orlando, Marc, 32, 33 Orrego-Carmona, David, 31 Ortiz, Luis Eduardo Schild, 34, 139 Oster, Katharina, 122 Paas, Fred, 105, 106 Padilla, Francisca, 67 Paone, Matteo Domenico, 32 Papagno, Costanza, 65 Pashler, Harold E., 80 Pavel, Silvia, 15, 16 Petsche, Hellmuth, 4, 122 Picht, Heribert, 13, 14 Pignataro, Clara, 8, 9 Pio, Sonia, 134 Pisani, Elisabetta, 2, 127, 128 Plevoets, Koen, 110 Pöchhacker, Franz, 3, 4, 10–12, 18, 33, 80, 84, 109, 113, 129 Polt, James M., 124 Ponzetto, Simone Paolo, 35 Poole, Alex, 117 Popovic Maja 10 Norman, Donald A., 69 Martín de León, Celia, 81 O’Brien, Sharon, 31, 51, 75, 77, 110, 111, O’Brien, Sharon, 31, 51, 75, 77, 110, 111, Martin, James R., 9 Moser, Peter, 12 Ponzetto, Simone Paolo, 35 Moser-Mercer, Barbara, 4, 42, 43, 45, 49, 52, 114, 123 Moser-Mercer, Barbara, 4, 42, 43, 45, 49, 52, 114, 123 Mudersbach, Klaus, 19 Moser-Mercer, Barbara, 4, 42, 43, 45, Poole, Alex, 117 Mudersbach, Klaus, 19 Popovic, Maja, 10 308 Name index SDI München, 47, 49 Posner, Michael I., 115 Postle, Bradley R., 67, 68 Seeber, Kilian G., 4, 6, 69, 73, 79, 84, 86–89, 91–98, 105, 106, 108, 120, 121, 123, 125, 126, 129, 134–136, 147, 157, 163, 164, 169, 230, 231 Seeber, Kilian G., 4, 6, 69, 73, 79, 84, 86–89, 91–98, 105, 106, 108, 120, 121, 123, 125, 126, 129, 134–136, 147, 157, 163, 164, 169, 230, 231 Prandi, Bianca, 2, 31, 34, 53–55, 58, 98, 101, 102, 109, 112, 127, 131– 133, 142, 143, 146–149, 159, 161, 163, 187, 215, 216, 218, 231, 240, 242, 245 Prandi, Bianca, 2, 31, 34, 53–55, 58, 98, 101, 102, 109, 112, 127, 131– 133, 142, 143, 146–149, 159, 161, 163, 187, 215, 216, 218, 231, 240, 242, 245 120, 121, 123, 125, 126, 129, 134–136, 147, 157, 163, 164, 169, 230, 231 240, 242, 245 Seel, Norbert M., 64 Price, Cathy J., 123 Price, Cathy J., 123 Seleskovitch, Danica, 4, 134 Pulitano, Donatella, 13, 14 Pulitano, Donatella, 13, 14 Semenza, Carlo, 136 Pym, Anthony, 84, 102 Pym, Anthony, 84, 102 SensoMotoric Instruments, 162, 167 SensoMotoric Instruments, 162, 167 Ramírez-Safar, Yamile, 43, 44, 46 Ramírez-Safar, Yamile, 43, 44, 46 Setton, Robin, 4, 83 Setton, Robin, 4, 83 Rayner, Keith, 118, 135 Rayner, Keith, 118, 135 Seubert, Sabine, 87, 90, 94, 95, 99, 120, Rehbein, Jochen, 141, 168 SAE International, 10 SAE International, 10 p Staub, Adrian, 118 Staveland, Lowell E., 107 Steurs, Frieda, 51 Stewart, Craig, 59, 60 Stoll, Christoph, 18, 21, 23, 34, 47, 51 Stroop, John Ridley, 70 Stuss, Donald T., 65 Sun, Sanjun, 107 Sweller, John, 67, 75–79 Tang, Yi-Yuan, 71 Taravella, Anne Marie, 77 Tardel, Anke, 2, 28, 29, 121 Staub, Adrian, 118 Salevsky, Heidemarie, 4 Staveland, Lowell E., 107 Sand, Peter, 36 Steurs, Frieda, 51 Scarpa, Federica, 8f Stewart, Craig, 59, 60 Schaeffer, Moritz, 119, 149, 150, 224, 230 Stoll, Christoph, 18, 21, 23, 34, 47, 51 Stoll, Christoph, 18, 21, 23, 34, 47, 51 Schepens, Job, 156 Schjoldager, Anne, 4 Simons, Daniel J., 70 Smallwood, Jonathan, 115 Smith, Philip L., 130 Sneed, Kayle, 120 169 Riccardi, Alessandra, 67, 134 Shannon, Claude E., 82 Richards, Ivor Armstrong, 13 Sheena, David, 118 Rinne, Juha O., 4, 123 Shiffrin, Richard M., 64 Rinne, Juha O., 4, 123 Rodríguez, Nadia, 8 Rodríguez, Susana, 58, 59 Rohrer, Doug, 80 Rosado, Tony, 31 Rösener, Christoph, 117 Roy, Cynthia B., 4 Rütten, Anja, 1, 8, 17–19, 21–23, 35, 37, 38, 47, 51, 211 Rodríguez, Nadia, 8 Shlesinger, Miriam, 4, 113 Rodríguez, Susana, 58, 59 Rodríguez, Susana, 58, 59 Shreve, Gregory M., 107, 120 Rohrer, Doug, 80 Signorelli, Teresa M., 67 Rosado, Tony, 31 Simons, Daniel J., 70 Snover, Matthew, 111 Snover, Matthew, 111 Spinner, Patti, 131, 140, 167 SAE International, 10 Xu, Ran, 109 Xu, Ran, 109 Torres, Alejandro, 229 Towse, John N., 80 Yanez, P., 11 Young, Laurence R., 118 Yanez, P., 11 Young, Laurence R., 118 Yanez, P., 11 Treisman, Anne M., 69, 74, 113 Young, Laurence R., 118 Tripepi Winteringham, Sarah, 35, 50 Žganec-Gros, Jerneja, 29 Ziegler, Klaus, 26, 27 Zielinski, Daniel, 43, 44, 46 Zwischenberger, Cornelia, 12 Uszkoreit, Hans, 10 Uszkoreit, Hans, 10 Ziegler, Klaus, 26, 27 Valentini, Cristina, 43, 46 Zielinski, Daniel, 43, 44, 46 Van Cauwenberghe, Goran, 2, 58, 104, 127, 147, 210, 215, 217, 218, 227, 242, 245 Zwischenberger, Cornelia, 12 Sun, Sanjun, 107 Schmidt, Thomas, 141, 169 Schnell, Bettina, 8 Schooler, Jonathan W., 115 Schultheis, Holger, 126 Schumacher, Eric H., 86 309 Name index Will, Martin, 1, 14, 17, 19–21, 26, 34, 35, 51, 52, 211 Wintringham, Nils, 36 Wörner, Kai, 141, 169 Wright, Leland D., 14 Wright, Sue Ellen, 14 Wüster, Eugen, 13, 52 Xu, Ran, 109 Yanez, P., 11 Young, Laurence R., 118 Žganec-Gros, Jerneja, 29 Ziegler, Klaus, 26, 27 Zielinski, Daniel, 43, 44, 46 Zwischenberger, Cornelia, 12 Tarmizi, Rohani A., 77 Will, Martin, 1, 14, 17, 19–21, 26, 34, 35, 51, 52, 211 Wintringham, Nils, 36 Wörner, Kai, 141, 169 Wright, Leland D., 14 Wright, Sue Ellen, 14 Wüster, Eugen, 13, 52 Will, Martin, 1, 14, 17, 19–21, 26, 34, 35, 51, 52, 211 Teixeira, Carlos S. C., 77 Thayer, Julian F., 123 Wintringham, Nils, 36 Wörner, Kai, 141, 169 Wright, Leland D., 14 Wright, Sue Ellen, 14 Wüster, Eugen, 13, 52 Timarová, Šárka, 67, 113, 114, 157, 161, Tiselius, Elisabet, 108, 120 Tissi, Benedetta, 110 Tommola, Jorma, 4, 122, 123, 125 227, 242, 245 Van Merrienboer, Jeroen J. G., 67, 76 Van Waes, Luuk, 112 Vardaro, Jennifer, 7, 108, 121, 215 Veisbergs, Andrejs, 50 Velásquez, Gonzalo, 9 Vielleneuve, Alain O., 77 Vintar, Špela, 15 Vintar, Špela, 15 Vogler, Nikolai, 77, 78 Vuorikoski, Anna-Riitta, 11 Vuorikoski, Anna-Riitta, 11 Wadensjö, Cecilia, 4, 10, 31, 109, 143 Wagener, Leonie, 18, 22, 47, 48 Wahlster, Wolfgang, 28 Wang, Caiwen, 2, 31, 245 Wang, Caiwen, 2, 31, 245 Wang, Xinyu, 2, 31, 245 Wang, Xinyu, 2, 31, 245 Warren, Tessa, 136 Warren, Tessa, 136 Weaver, Warren, 82 Weaver, Warren, 82 Webex Help Center, 30 Webex Help Center, 30 Weller, G., 11 Wickens, Christopher D., 71–73, 86, Wickens, Christopher D., 71–73, 86, 92, 96, 101, 231 310 Computer-assisted simultaneous interpreting The present work explores computer-assisted simultaneous interpreting (CASI) from a primarily cognitive perspective. Despite concerns over the potentially negative impact of computer-assisted interpreting (CAI) tools on interpreters’ cognitive load (CL), this hy- pothesis remains untested. Previous research is restricted to the evaluation of the CASI product and a methodology for the process-oriented evaluation of CASI and the em- pirical evidence for its cognitive modelling are missing. Overcoming these limitations appears essential to advance CAI research, particularly to foster a deeper understanding of the cognitive aspects of CAI through a validated research methodology and to deter- mine the feasibility of the integration of CAI tools into the interpreting process. This book tests and validates a methodology for the combined exploration of the product and process of CASI. Quantitative and qualitative data were collected during an eyetracking experiment at the Translation & Cognition Centre of the University of Mainz. The study followed a convergent mixed-method and multi-method approach and involved nine interpret- ing students. Prior to the experimental task, the informants were trained in the use of three terminology support tools: a digital glossary in PDF format, a CAI tool with man- ual look-up, and a mock-up CAI tool with integrated automatic speech recognition (ASR) for terminology. After several pre-tests, the participants interpreted three speeches from English into their native German using a different tool each time. To increase comparabil- ity between the three conditions and control for potentially confounding variables, the speeches were validated during a pilot study. The students’ gaze data and deliveries were recorded and analysed. Qualitative data on the informants’ perception of the tools were collected post-hoc. In the study, several performance, behavioural, and subjective mea- sures were analysed: terminological accuracy and errors and omissions; glossary queries, ear-voice span, inter-cluster pause duration, time to first fixation, average fixation dura- tion, and fixation time; qualitative questionnaire responses. The findings provide insights into the effects of CAI tools on CL and attention allocation in interpreter-CAI tool inter- action during simultaneous interpreting.i As this is the first study on in-process CAI tool use with a markedly cognitive ori- entation, it entails significant implications for the methodological development of CAI research and the design of future studies on cognitive aspects of CASI, while raising additional questions in need of further investigation.
https://openalex.org/W4285085050	https://www.frontiersin.org/articles/10.3389/fpsyg.2022.817960/pdf	English	null	A Study on Multimedia Integrated Pre-service Education to Learning Behavior and Competitiveness in Workplace of Employees in Hospitality	Frontiers in psychology	2,022	cc-by	5,127	A Study on Multimedia Integrated Pre-service Education to Learning Behavior and Competitiveness in Workplace of Employees in Hospitality Chih-Hung Pai 1, Yu-Lan Wang 2, Yunfeng Shang 1 and Ta-Kuang Hsu 3 1 School of Hospitality Management, Zhejiang Yuexiu University, Shaoxing, China, 2 Teacher Education Institute, Weifang University of Science and Technology, Shandong, China, 3 Department of Restaurant, Hotel and Institutional Management, Jen Catholic University, New Taipei City, Taiwan Chih-Hung Pai 1, Yu-Lan Wang 2, Yunfeng Shang 1 and Ta-Kuang Hsu 3 The domestic situation of the past few years shows the practices of employees’ unpaid leave and layoffs and the constant drain on capital, talent, and technologies in hospitality. Owners expect to reduce the losses to as low as possible by saving on human costs. Nevertheless, in face of such a changing environment, hospitality has to accumulate high-quality human capital through systematic investment, sensitive development, and continuous learning and growth to discover competitive advantages through the cultivation of human capital. The pre-service education of new employees could accelerate their familiarity with the operations of the company and their understanding of their job role and duties. More importantly, with good planning, it could make employees feel emphasized with and respected with the result of largely changing their thoughts and working habits. Aiming at supervisors and employees in hospitality in Zhejiang as the research objects, a total of 420 copies of our questionnaire are distributed, and 357 valid copies were retrieved, with a retrieval rate of 85%. According to the results to propose discussions, it is expected to generate systematic and proper education methods for the pre-service education in hospitality, promote the education effectiveness, and promote employees’ capability and organizational performance. Edited by: Michael A. Talias, Open University of Cyprus, Cyprus Open University of Cyprus, Cyprus Reviewed by: Bang-Lee Chang, Chinese Culture University, Taiwan Jorge Gomes, Universidade Lusófona, Portugal Correspondence: Yu-Lan Wang wylan@wfust.edu.cn Reviewed by: Bang-Lee Chang, Chinese Culture University, Taiwan Jorge Gomes, g g Chinese Culture University, Taiwan Jorge Gomes, Universidade Lusófona, Portugal Chinese Culture University, Taiwan Jorge Gomes, Universidade Lusófona, Portugal Correspondence: Yu-Lan Wang wylan@wfust.edu.cn Correspondence: Yu-Lan Wang wylan@wfust.edu.cn Specialty section: This article was submitted to Educational Psychology, a section of the journal Frontiers in Psychology Received: 18 November 2021 Accepted: 27 May 2022 Published: 13 July 2022 Keywords: multimedia, pre-service education, learning behavior, competitiveness in workplace, hospitality, delicacy BRIEF RESEARCH REPORT published: 13 July 2022 doi: 10.3389/fpsyg.2022.817960 BRIEF RESEARCH REPORT published: 13 July 2022 doi: 10.3389/fpsyg.2022.817960 LITERATURE REVIEW AND HYPOTHESIS Chen et al. (2017) revealed that a hospitality organization could enhance employees’ learning behavior by providing them with pre-service education or encouraging further study. Aiming at staff receiving pre-service education within a year, Kimzey et al. (2019) revealed that those participating in pre-service education presented higher learning behavior after the education and regarded the education as the welfare and being able to enhance the competency. Deci et al. (2017) proposed significant correlations between personality traits and learning behavior that managers in hospitality could arrange different pre-service education according to employees’ personality traits to enhance the learning behavior. Looking at social workers of non-profit organizations in Kaohsiung, Wijma et al. (2018) found positive effects of pre-service education on social workers’ learning behavior. present excellent performance, education aims to have the employees master the competencies required for the work in education courses and apply these abilities to the work (Aarabi et al., 2013). Under the development trends of globalization, technology, and a knowledge-based economy, hospitality puts a lot of stress on education-related issues in order to enhance the quality of human capital and promote employees’ professional competency. For this reason, hospitality invests lots of resources in education, as pertains to whether pre-service education would develop the business’ educational function and result in positive benefits. As a result, it could promote employees’ competency by understanding and evaluating employees’ competency gaps and designing and planning courses aiming bridging these insufficiencies before education begins. The pre-service education of new employees could accelerate their familiarity with company operations and understanding of their job role and work duties. More importantly, through good planning, it could make employees feel emphasized with and respected, with the result of largely changing their thoughts and working habits. Pre-service education demonstrates the meanings of positions and directions, i.e., assisting new staff in knowing their position and work direction. The content contains the points of the philosophy, objective, organization, policy, environment, and attendance regulations of the company to prevent new employees from reality shock when it comes to executing their actual work (Zumrah et al., 2013). Education is an activity of human resource development. Effective pre-service education could enhance personnel’s ability to do better at work and properly develop to promote work performance. Along with the development wave of the hospitality industry, talent cultivation presents long-term development in the human resource strategy for the hospitality industry. LITERATURE REVIEW AND HYPOTHESIS As a result, long-term human resource planning and cultivation are particularly emphasized in the hospitality industry. Meanwhile, the human resource level is promoted from the operational and administrative level to the strategic level to enhance human resource managerial effectiveness and organizational performance. The development, utilization, and management of human resources become critical factors in human and socio-economic development (Paulsen et al., 2015). The possession of talent allows the hospitality industry to present a competitive ability to control the market. In order to retain and acquire talent, human capital and development should be emphasized (Saseen et al., 2017). Multimedia integrated pre-service education to teach behavior and competitiveness in the workplace of employees in hospitality is therefore studied, it is expected to clarify the effect of multimedia integrated orientation training on learning behavior and workplace competitiveness as well as to show systematic and proper education methods for the orientation training in the hospitality g Lee et al. (2018) regarded remarkably positive correlations between learning behavior and competitiveness in the workplace of employees in the iron and steel industry with high explanatory power and notably positive effects. Looking at employees in Taiwan and China, Mirani et al. (2019) pointed out the positive effects of employees’ learning behavior on the organizational commitment to affect the competitiveness in the workplace. Gholamzadeh et al. (2018) revealed that employees with better outcomes of on-the-job education showed higher learning behavior and better competitiveness in the workplace. Gehrman et al. (2018) discovered that grass-root employees with learning behavior, in finance, would enhance the competitiveness in the workplace at work, i.e., there was a notably positive relationship between learning behavior and competitiveness in the workplace of grass-root employees in finance. p p g p yi Rickard et al. (2009) stated that employees receiving pre-service education would enhance work efficiency and quality of employees’ work performance and further promote the competitiveness in the workplace. Annear et al. (2017) discovered that most people could learn knowledge and skills in pre-service education for the application to the work as well as to enhance personal ability. Mastel-Smith et al. (2019) found a difference in pre-service education between Taiwanese companies and American companies; however, pre-service education showed positive effects on competitiveness in the workplace in both Taiwanese and American companies. Mavounza et al. Citation: Pai C-H, Wang Y-L, Shang Y and Hsu T-K (2022) A Study on Multimedia Integrated Pre-service Education to Learning Behavior and Competitiveness in Workplace of Employees in Hospitality. Front. Psychol. 13:817960. doi: 10.3389/fpsyg.2022.817960 Employees are the essential capital for hospitality or any organization where the managers have tried to reinforce human capital. Nonetheless, the domestic situation reveals the practice of unpaid employee leave, layoffs, and the constant drainage of capital, talent, and technologies in hospitality. Owners expect to reduce organizational losses to as low as possible by saving on human costs; however, hospitality should accumulate high-quality human capital through systematic investment, sensitive development, and constant learning and growth and find competitive advantages through Employees are the essential capital for hospitality or any organization where the managers have tried to reinforce human capital. Nonetheless, the domestic situation reveals the practice of unpaid employee leave, layoffs, and the constant drainage of capital, talent, and technologies in hospitality. Owners expect to reduce organizational losses to as low as possible by saving on human costs; however, hospitality should accumulate high-quality human capital through systematic investment, sensitive development, and constant learning and growth and find competitive advantages through July 2022 \| Volume 13 \| Article 817960 Frontiers in Psychology \| www.frontiersin.org Pre-service Education to Hospitality Employees Pai et al. industry in order to enhance the educational effectiveness and promote employee ability and organizational performance. the cultivation of human capital. It is the primary issue for managers to transform knowledgeable employees into organizational assets. A talent shortage in an organization would affect the long-term management and even lose good business opportunities. In hospitality, the practice of education is the commonly used and broadly accepted method to reinforce human capital. In order to have the employees learn the skills required for their work, including the knowledge, skills, and behavior necessary to present excellent performance, education aims to have the employees master the competencies required for the work in education courses and apply these abilities to the work (Aarabi et al., 2013). LITERATURE REVIEW AND HYPOTHESIS (2020) considered that securities specialists, in order to cater to environmental changes, would enhance their competitiveness in the workplace after participating in cross-business pre-service education to show remarkably positive correlations between pre-service education and competitiveness in the workplace. METHODOLOGY Research Framework Frontiers in Psychology \| www.frontiersin.org Reliability and Validity Testi Reliability and Validity Test Confirmatory Factor Analysis (CFA) is an important part of SEM that the measurement model of the structural model should be tested before the two-stage model modification. When the model fit is acceptable, the second step SEM model evaluation is preceded. The CFA reveals the factor loadings of dimensions in 0.60~0.90, the composite reliability in 0.70~0.90, and the average variance extracted in 0.60~0.80, conforming to the standards of (1) factor loadings > 0.5, (2) composite reliability > 0.6, and (3) average variance extracted > 0.5. The dimensions, therefore, present convergent validity. multimedia integrated orientation training, learning behavior, and competitiveness in the workplace. Structural Model Analysish The structural model analysis contains the goodness-of-fit analysis of the research model and the explanatory power of the overall research model. In this case, seven numerical indices are used for testing the goodness-of-fit of the overall model, including the chi-square (χ2) test, χ2-degree of freedom ratio, the goodness of fit index, adjusted goodness of fit index, root- mean-square error, comparative fit index, comparative hypothesis model, and chi-square of the independent model. The overall results are organized in Table 1. H3: (Hypothesis 3): Multimedia integrated pre-service education shows positive and notable effects on competitiveness in the workplace. Research Questions According to the above literature review and conceptual framework, the following research issues are acquired. H1: (Hypothesis 1): Multimedia integrated pre-service education shows positive and significant effects on learning behavior. H2: (Hypothesis 2): Learning behavior reveals positive and remarkable effects on competitiveness in the workplace. H2: (Hypothesis 2): Learning behavior reveals positive and remarkable effects on competitiveness in the workplace. Learning Behavior Referring to the learning behavior scale proposed by Bi et al. (2019), learning behavior is divided into “learning method,” “learning habit,” and “learning attitude” in this study. Operational Definition Multimedia Integrated Pre-service Education Referring to the education scale developed by Chan and Leung (2020), three dimensions of educational perception, educational support, and satisfaction with education are contained in this study. g From the above indices, when using the χ2-degree of freedom ratio for testing the model, the ratio is better smaller; this research model shows the χ2-degree of freedom ratio <3 (1.94). GFI and AGFI are better close to 1, have no absolute standards to judge the fit, and GFI >0.9 and AGFI >0.8 are acceptable; this research model reveals GFI and AGFI 0.96 and 0.83, respectively. RMSEA being in 0.05–0.08 stands for a good model and reasonable fit; this research model shows RMSEA of 0.05. The allowable standard of CFI is >0.9, and this research model shows CFI as 0.92. NFI should be at least higher than 0.9, and this research model reveals the NFI is 0.91. Overall speaking, the goodness-of-fit indices conform to the standards, revealing the model to be acceptable. The research data therefore could explain the actual observed data. Research Framework Summing up the above literature review, the conceptual framework (Figure 1) is drawn to discuss the relations between Frontiers in Psychology \| www.frontiersin.org July 2022 \| Volume 13 \| Article 817960 2 Pre-service Education to Hospitality Employees Pai et al. FIGURE 1 \| Conceptual framework. Research Sample and Objecth Research Sample and Objecth Research Sample and Object The Zhejiang Food and Beverage Industry Association implements the catering and cooking business policy of “inheritance, development, development, and innovation,” which positively develops the construction of “three fame engineering” (famous restaurants, famous cooks, and famous dishes), organizes provincial and municipal cooking technology competitions in a planned way, participates in national cooking technology competitions and world Chinese cooking competitions, and develops activities of Zhejiang cuisine skills training and exchanges inside and outside the province to develop the important function for the prosperity of hospitality market and the development of Zhejiang dishes. Aiming at supervisors and employees in hospitality in Zhejiang, a total of 420 copies of our questionnaire were distributed, and 357 valid copies retrieved, with a retrieval rate of 85%. FIGURE 1 \| Conceptual framework. Competitiveness in Workplace Evaluation item Parameter/evaluation standard Result Preliminary fit Multimedia integrated pre- service education Educational perception 0.70 Educational support 0.67* Satisfaction with education 0.75** Learning behavior Learning method 0.69* Learning habit 0.73 Learning attitude 0.71 Internal fit Multimedia integrated pre-service education → learning behavior 0.81* Learning behavior → competitiveness in 0.86* TABLE 1 \| Research model fit analysis. Fit indices Allowable range This research model Model fit judgment χ2 (Chi-square) The smaller the better 17.68 χ2-degree of freedom ratio <3 1.94 Match GFI >0.9 0.96 Match AGFI >0.8 0.83 Match RMSEA <0.08 0.05 Match CFI >0.9 0.92 Match NFI >0.9 0.91 Match TABLE 2 \| Overall linear structural model analysis result. From the above-mentioned overall model fit indices, the model structured in this study and the observed data present a favorable fit, revealing that the theoretical model could fully explain the observed data. Consequently, after the model fit test, the correlation coefficient and coefficient estimate of multimedia integrated pre-service education to learning behavior and competitiveness in the workplace could be understood. h The research hypothesis test results are shown in Table 3. The research data are organized in Table 2. The overall model analysis results show that three dimensions of multimedia integrated pre-service education (educational perception, educational support, and satisfaction with education) could significantly explain multimedia integrated pre-service education (t > 1.96, p < 0.05), three dimensions of learning behavior (learning method, learning habit, and learning attitude) could remarkably explain learning behavior (t > 1.96, p < 0.05), and competitiveness in the workplace presents the notable explanation (t > 1.96, p < 0.05). The overall model appears to present a good preliminary fit.i orientation training reveals positive and notable effects on workplace competitiveness (0.83, p < 0.01), showing that H3 is supported. Apparently, in a time with multiple knowledge-based economies, the hospitality industry promotes multimedia integrated orientation training to enhance employees’ professional knowledge and skills or encourages employees to cultivate learning behavior in order to promote workplace competitiveness. Regarding the internal fit, multimedia integrated pre-service education shows positive and significant correlations with learning behavior (0.81, p < 0.01), learning behavior reveals positive and remarkable correlations with competitiveness in the workplace (0.86, p < 0.01), and multimedia integrated pre-service education shows positive and notable correlations with competitiveness in workplace (0.83, p < 0.01). H1, H2, and H3 are therefore supported. The research hypothesis test results are shown in Table 3. Frontiers in Psychology \| www.frontiersin.org Competitiveness in Workplace Referring to the competitiveness in the workplace scale proposed by Wang et al. (2018), competitiveness in the workplace is measured with a single dimension in this study. July 2022 \| Volume 13 \| Article 817960 Frontiers in Psychology \| www.frontiersin.org 3 Pre-service Education to Hospitality Employees Pai et al. Pai et al. TABLE 2 \| Overall linear structural model analysis result. Evaluation item Parameter/evaluation standard Result Preliminary fit Multimedia integrated pre- service education Educational perception 0.70* Educational support 0.67* Satisfaction with education 0.75** Learning behavior Learning method 0.69* Learning habit 0.73 Learning attitude 0.71 Internal fit Multimedia integrated pre-service education → learning behavior 0.81* Learning behavior → competitiveness in workplace 0.86* Multimedia integrated pre-service education → competitiveness in workplace 0.83*** p < 0.05, p < 0.01, **p < 0.001. From the above-mentioned overall model fit indices, the model structured in this study and the observed data present a favorable fit, revealing that the theoretical model could fully explain the observed data. Consequently, after the model fit test, the correlation coefficient and coefficient estimate of multimedia integrated pre-service education to learning behavior and competitiveness in the workplace could be understood. The research hypothesis test results are shown in Table 3. The research data are organized in Table 2. The overall model analysis results show that three dimensions of multimedia integrated pre-service education (educational perception, educational support, and satisfaction with education) could significantly explain multimedia integrated pre-service education (t > 1.96, p < 0.05), three dimensions of learning behavior (learning method, learning habit, and learning attitude) could remarkably explain learning behavior (t > 1.96, p < 0.05), and competitiveness in the workplace presents the notable explanation (t > 1.96, h orientation training reveals positive and notable effects on workplace competitiveness (0.83, p < 0.01), showing that H3 is supported. Apparently, in a time with multiple knowledge-based economies, the hospitality industry promotes multimedia TABLE 1 \| Research model fit analysis. Fit indices Allowable range This research model Model fit judgment χ2 (Chi-square) The smaller the better 17.68 χ2-degree of freedom ratio <3 1.94 Match GFI >0.9 0.96 Match AGFI >0.8 0.83 Match RMSEA <0.08 0.05 Match CFI >0.9 0.92 Match NFI >0.9 0.91 Match TABLE 3 \| Hypothesis test. Research hypothesis C Empirical result p Result H1 + 0.81 <0.01 Supported H2 + 0.86 <0.01 Supported H3 + 0.83 <0.01 Supported TABLE 2 \| Overall linear structural model analysis result. DISCUSSION Under the premise of satisfying customer requirements, hospitality should present sufficient and excellent professional technology and develop selectively. The enhancement of customer satisfaction should result in a repeat visit and even a recommendation to friends. On the contrary, dissatisfied customers might not visit again and might urge others not to visit. Since professional technology and sensitive development are so important, multimedia integrated pre-service education should be particularly emphasized. With existing bases, enhancing personal professional technology would promote the competitiveness in the workplace. Businesses in the hospitality industry are suggested to positively and effectively establish a training program and system for trainers as well as understand orientation-training related instructional methods and communication skills with orientation training lecturers to further promote the teaching quality of internal orientation training (Hyndman and Pill, 2017). As a result, experts and scholars could be invited to participate in the practical lectures and professional technology demonstration of multimedia integrated pre-service education to instruct and correct the employees’ From the research results, multimedia integrated orientation training presents positive and significant effects on learning behavior (0.81, p < 0.01), showing that H1 is supported. Apparently, in order to cope with environmental changes, the hospitality industry applies multimedia integrated orientation training to promote employees’ learning behavior, i.e., remarkably positive effects of multimedia integrated orientation training were found on learning behavior. Learning behavior shows positive and remarkable effects on workplace competitiveness (0.86, p < 0.01), meaning that H2 is supported. In this case, when the hospitality industry is able to provide relevant multimedia integrated orientation training courses, the employees feel that the organization is taking care of and is responsible for them. Consequently, employees would present more rigorous learning behavior for the affirmation of self-capability to further promote workplace competitiveness. Finally, multimedia integrated July 2022 \| Volume 13 \| Article 817960 Frontiers in Psychology \| www.frontiersin.org 4 Pre-service Education to Hospitality Employees Pai et al. Pai et al. professional technology and sensitivity development and promote competitiveness. In order to have their personal ability be affirmed, the employees in hospitality would make efforts and learn professional technology and related professional knowledge from multimedia integrated pre-service education. In this case, preceding learning behavior for the application to customer needs and enhancing customer satisfaction to promote work performance and competitiveness in the workplace would enhance the career development of the employees in hospitality as well as the coherence. DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author. ETHICS STATEMENT The research results reveal positive effects of multimedia integrated orientation training on learning behavior and workplace competitiveness of employees in the hospitality industry. Employees could learn better and utilize more professional technology in pre-service education; once the professional technology is promoted, the learning behavior would be enhanced. The research results reveal the greater the learning behavior of employees in hospitality, the larger the competitiveness in the workplace. Employees being willing to participate in learning behavior and further make extra efforts for the learning behavior would enhance competitiveness in the workplace. The research results reveal that employees in hospitality attending more multimedia integrated pre-service education would present stronger competitiveness in the workplace. Hospitality requires professional technology and skills that the employees have to regularly participate in relevant technology study or on-the-job education courses for discussion and exchange with each other. It could enhance professional skills and further promote customer satisfaction and loyalty This study was reviewed and approved by the ethics committee of the Zhejiang Yuexiu University. Written informed consent was obtained from all participants for their participation in this study. DISCUSSION Employees in the hospitality industry should regularly check and promote hardware facilities and attempt to apply interactive multimedia interactive teaching and materials or other relevant educational equipment or teaching media to stimulate the employees’ learning willingness and interests as well as achieve the training objective and further enhance training effectiveness (Appleby et al., 2016). For instance, they can utilize audiovisual teaching for self-learning, i.e., using electronic equipment of films, video files, and recorders for delivering teaching content. It is suggested that the hospitality industry, in addition to stressing employees’ educational training, should list constant expenses for purchasing educational training-related equipment to promote educational effectiveness (Cullen, 2016). to help the promotion of competitiveness in the workplace. Roetert and Jefferies (2014) mentioned that there was not “the best” training method, but merely “the most suitable” method. Any training courses to achieve the training objective with multiple training methods would show higher effectiveness than those with merely one method. With similar research results to this study, French et al. (2015) revealed that multimedia integrated orientation training presented better assistance to the effectiveness of employees in the hospitality industry. Accordingly, the orientation training objects and the required skills, knowledge, or attitude should be carefully considered in the construction of orientation training content and planning of orientation training programs before the practice of orientation training courses. Proper orientation training methods are then selected with appropriate orientation training materials and relevant teaching to achieve the expected orientation training effectiveness (Kilcommins and Spain, 2016). FUNDING This work was supported by 2020 Zhejiang Social Science Project, Zhejiang Province, China (Grant Number: 20XXJC04ZD). This work was supported by 2020 Zhejiang Social Science Project, Zhejiang Province, China (Grant Number: 20XXJC04ZD). AUTHOR CONTRIBUTIONS C-HP and Y-LW: collected and analyzed the data and drafted the manuscript. YS: designed the research protocol and contributed to the literature review. T-KH: reviewed the manuscript and revised the manuscript. All authors revised and approved the submitted version of the manuscript. C-HP and Y-LW: collected and analyzed the data and drafted the manuscript. YS: designed the research protocol and contributed to the literature review. T-KH: reviewed the manuscript and revised the manuscript. All authors revised and approved the submitted version of the manuscript. J. Phys. Educ. Recreat. Dance 87, 16–21. doi: 10.1080/07303084.2015. 1131214 J. Phys. Educ. Recreat. Dance 87, 16–21. doi: 10.1080/07303084.2015. 1131214 Bi, J. W., Liu, Y., Fan, Z. P., and Zhang, J. (2019). Conducting importance- performance analysis (IPA) through online reviews. Tour. Manag. 70, 460–478. doi: 10.1016/j.tourman.2018.09.010 Chan, W. S., and Leung, A. Y. (2020). Facebook as a novel tool for continuous professional education on dementia: pilot randomized controlled trial. J. Medical Internet Res. 22:e16772. doi: 10.2196/16772 Chen, T. B., Yiao, S. Y., Sun, Y., Lee, H. J., Yang, S. C., Chiu, M. J., et al. (2017). Comorbidity and dementia: a nationwide survey in Taiwan. PLoS One 12:e0175475. doi: 10.1371/journal.pone.0175475 REFERENCES Aarabi, M. S., Subramamiam, I. D., and Akeel, A. B. (2013). Relationship between motivational factors and job performance of employees in Malaysian service industry. Asian Soc. Sci. 9, 301–310. doi: 10.5539/ass.v9n9p301 Bi, J. W., Liu, Y., Fan, Z. P., and Zhang, J. (2019). Conducting importance- performance analysis (IPA) through online reviews. Tour. Manag. 70, 460–478. doi: 10.1016/j.tourman.2018.09.010 Annear, M. J., Toye, C., Elliott, K. J., McInerney, F., Eccleston, C., and Robinson, A. (2017). Dementia knowledge assessment scale (DKAS): confirmatory factor analysis and comparative subscale scores among an international cohort. Biomed. Cent. Geriatr. 17:168. doi: 10.1186/s12877-017-0552-y Chan, W. S., and Leung, A. Y. (2020). Facebook as a novel tool for continuous professional education on dementia: pilot randomized controlled trial. J. Medical Internet Res. 22:e16772. doi: 10.2196/16772 Chen, T. B., Yiao, S. Y., Sun, Y., Lee, H. J., Yang, S. C., Chiu, M. J., et al. (2017). Comorbidity and dementia: a nationwide survey in Taiwan. PLoS One 12:e0175475. doi: 10.1371/journal.pone.0175475 Appleby, K. M., Foster, E., and Kamusoko, S. (2016). Full speed ahead: using a senior capstone course to facilitate students’ professional transition. Frontiers in Psychology \| www.frontiersin.org July 2022 \| Volume 13 \| Article 817960 5 Pre-service Education to Hospitality Employees Pai et al. Pai et al. community of practice. J. Agric. Educ. 56, 210–226. doi: 10.5032/jae. 2015.03210h Cullen, T. (2016). Designing journalism capstone units that demonstrate student skills. J. Mass Commun. Educ. 71, 360–370. doi: 10.1177/1077695816666077 community of practice. J. Agric. Educ. 56, 210–226. doi: 10.5032/jae. 2015.03210h Deci, E. L., Olafsen, A. H., and Ryan, R. M. (2017). Self-determination theory in work organizations: the state of a science. Annu. Rev. Organ. Psychol. Organ. Behav. 4, 19–43. doi: 10.1146/annurev-orgpsych-032516-113108 Rickard, A., McAvinia, C., and Quirke-Bolt, N. (2009). The challenge of change: digital video analysis and constructivist teaching approaches on a one year preservice teacher education program in Ireland. J. Technol. Teach. Educ. 17, 349–367. French, E., Bailey, J., van Acker, E., and Wood, L. (2015). From mountaintop to corporate ladder – what new professionals really really want in a capstone experience! Teach. High. Educ. 20, 767–782. doi: 10.1080/13562517.2015. 1085852 Roetert, E. P., and Jefferies, S. C. (2014). Embracing physical literacy. J. Phys. Educ. Recreat. Dance 85, 38–40. doi: 10.1080/07303084.2014.948353 Saseen, J. J., Linnebur, S. A., Borgelt, L. M., Trujillo, J. M., Fish, D. N., and Mueller, S. W. (2017). A pharmacotherapy capstone course to target student learning and programmatic curricular assessment. Am. REFERENCES J. Pharm. Educ. 81, 1–12. doi: 10.5688/ajpe81345 Gehrman, P., Gooneratne, N. S., Brewster, G. S., Richards, K. C., and Karlawish, J. (2018). Impact of alzheimer disease patients’ sleep disturbances on their caregivers. Geriatr. Nurs. 39, 60–65. doi: 10.1016/j.gerinurse.2017.06.003 Gholamzadeh, S., Khastavaneh, M., Khademian, Z., and Ghadakpour, S. (2018). The effects of empathy skills training on nursing students’ empathy and attitudes toward elderly people. BMC Med. Educ. 18:198. doi: 10.1186/ s12909-018-1297-9 Wang, Y., Xiao, L. D., Luo, Y., Xiao, S. Y., Whitehead, C., and Davies, O. (2018). Community health professionals’ dementia knowledge, attitudes and care approach: a cross-sectional survey in Changsha, China. BMC Geriatrics 18:122. doi: 10.1186/s12877-018-0821-4 Hyndman, B., and Pill, S. (2017). What’s in a concept? A Leximancer text mining analysis of physical literacy across the international literature. Eur. Phys. Educ. Rev. 24, 292–313. doi: 10.1177/1356336x17690312 Wijma, E. M., Veerbeek, M. A., Prins, M., Pot, A. M., and Willemse, B. M. (2018). A virtual reality intervention to improve the understanding and empathy for people with dementia in informal caregivers: results of a pilot study. Aging Ment. Health 22, 1121–1129. doi: 10.1080/13607863.2017.1348470h Kilcommins, S., and Spain, E. (2016). “Deaths in prison custody” capstone course: engaging final-year law students in service learning and public value. J. Crim. Justice Educ. 27, 285–298. doi: 10.1080/10511253.2015.1117118 Zumrah, A. R., Boyle, S., and Fein, E. C. (2013). The consequences of transfer of training for service quality and job satisfaction: an empirical study in the Malaysian public sector. Int. J. Train. Dev. 17, 279–294. doi: 10.1111/ ijtd.12017 Kimzey, M., Mastel-Smith, B., and Seale, A. (2019). Effects of dementia-specific education for nursing students. Nurse Educ. 44, 338–341. doi: 10.1177/ 0969733015622060 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 potential conflicts of interest. Lee, L. A., Chao, Y. P., Huang, C. G., Fang, J. T., Wang, S. L., Chuang, C. K., et al. (2018). Cognitive style and mobile e-learning in emergent otorhinolaryngo- logy head and neck surgery disorders for millennial undergraduate medical students: randomized controlled trial. J. Med. Internet Res. 20:e56. doi: 10.2196/jmir.8987 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. Frontiers in Psychology \| www.frontiersin.org REFERENCES 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. Mastel-Smith, B., Kimzey, M., and He, Z. (2019). Dementia care education for nursing students. J. Nurs. Educ. 58, 136–143. doi: 10.3928/01484834- 20190221-03 Mavounza, C., Ouellet, M. C., and Hudon, C. (2020). Caregivers’ emotional distress due to neuropsychiatric symptoms of persons with amnestic mild cognitive impairment or Alzheimer’s disease. Aging Ment. Health 24, 423–430. doi: 10.3928/01484834-20190221-03 Copyright © 2022 Pai, Wang, Shang and Hsu. 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. Mirani, S. H., Shaikh, N. A., and Tahir, A. (2019). Assessment of clinical empathy among medical students using the Jefferson Scale of Empathy- Student Version. Cureus 11:e4160. doi: 10.7759/cureus.4160 Paulsen, T. H., Anderson, R. G., and Tweeten, J. F. (2015). Concerns expressed by agricultural education preservice teachers in a Twitter-based electronic July 2022 \| Volume 13 \| Article 817960 Frontiers in Psychology \| www.frontiersin.org 6
https://openalex.org/W4366606128	https://www.researchsquare.com/article/rs-2722241/latest.pdf	English	null	Diagnosis of Pulmonary Scedosporium apiospermum Infection from Bronchoalveolar Lavage Fluid by Metagenomic Next-Generation Sequencing in an Immunocompetent Female Patient with Normal Lung Structure: A Case Report and Literature Review	Research Square (Research Square)	2,023	cc-by	7,959	Diagnosis of Pulmonary Scedosporium apiospermum Infection from Bronchoalveolar Lavage Fluid by Metagenomic Next-Generation Sequencing in an Immunocompetent Female Patient with Normal Lung Structure: A Case Report and Literature Review Diagnosis of Pulmonary Scedosporium apiospermum Infection from Bronchoalveolar Lavage Fluid by Metagenomic Next-Generation Sequencing in an Immunocompetent Female Patient with Normal Lung Structure: A Case Report and Literature Review Jingru Han The First Affiliated Hospital of Chongqing Medical University Lifang Liang Chongqing Medical University Qingshu Li Chongqing Medical University Ruihang Deng Chongqing Medical University Chenyang Liu Chongqing Medical University Xuekai Wu Chongqing Medical University Yuxin Zhang Chongqing Medical University Ruowen Zhang Chongqing Medical University Haiyun Dai The First Affiliated Hospital of Chongqing Medical University Case Report Keywords: Metagenomic next-generation sequencing, pulmonary Posted Date: April 21st, 2023 DOI: https://doi.org/10.21203/rs.3.rs-2722241/v1 License:   This work is licensed under a Creative Commons A Additi l D l ti N ti i t t t d Jingru Han The First Affiliated Hospital of Chon Lifang Liang Chongqing Medical University Qingshu Li Chongqing Medical University Ruihang Deng Chongqing Medical University Chenyang Liu Chongqing Medical University Xuekai Wu Chongqing Medical University Yuxin Zhang Chongqing Medical University Ruowen Zhang Chongqing Medical University Haiyun Dai Jingru Han The First Affiliated Hospital of Chongqing Medical Universi Lifang Liang Chongqing Medical University Qingshu Li Chongqing Medical University Ruihang Deng Chongqing Medical University Chenyang Liu Chongqing Medical University Xuekai Wu Chongqing Medical University Yuxin Zhang Chongqing Medical University Ruowen Zhang Chongqing Medical University Haiyun Dai The First Affiliated Hospital of Chongqing Medical University Diagnosis of Pulmonary Scedosporium apiospermum Infection from Bronchoalveolar Lavage Fluid by Metagenomic Next-Generation Sequencing in an Immunocompetent Female Patient with Normal Lung Structure: A Case Report and Literature Review Jingru Han Th Fi t Affili t d H it l f Ch i M di l U i it Diagnosis of Pulmonary Scedosporium apiospermum Infection from Bronchoalveolar Lavage Fluid by Metagenomic Next-Generation Sequencing in an Immunocompetent Female Patient with Normal Lung Structure: A Case Report and Literature Review Jingru Han Th Fi t Affili t d H it l f Ch i M di l U i it Case Presentation A 54-year-old Chinese non-smoker female, who worked in a chicken processing factory, experienced intermittent right chest pain with occasional dry coughing for 8 months with no apparent trigger. Although the patient's sputum anti-acid staining was negative, she was receiving empirical anti-tuberculosis therapy (2021.04.10) (the specific drug is unknown) in another hospital since her chest computed tomography (CT), dated April 2, 2021, suggested pulmonary tuberculosis. Further, the patient developed skin itching and systemic redness, prompting the anti-tuberculosis drugs to be changed (the specific drug is unknown), however, shortness of breath and shivering occurred following 3 days of medication, and the patient was eventually switched to isoniazid, rifampicin, ethambutol, and levofloxacin (HRE + Lfx) for anti-tuberculosis therapy. Unfortunately, her symptoms and imaging manifestation did not improve, and the CT at Hechuan People's Hospital on June 16, 2021, indicated a thick-walled cavity in the upper lobe of the right lung, with irregular morphology, uneven wall thickness, and mild soft tissue enhancement within it. On June 22nd, 2021, she was admitted to our department for further evaluation and treatment. During the investigation, the patient denied experiencing symptoms such as dyspnea, chest tightness, hot flashes, night sweats, hemoptysis, chills, or a high fever. No significant abnormality was observed in the physical examination, and auxiliary inspection results are demonstrated in Table 1. The patient was initially diagnosed with bacteriologically negative pulmonary tuberculosis and continued with anti-tuberculosis therapy with HRE and Lfx. The BALF was submitted for examination with mNGS, and pulmonary S. apiospermum infection was considered, followed by antifungal therapy with voriconazole (300 mg iv q12h d1; 200 mg q12h iv d2-d20). Chest enhanced CT considers the possibility of lung cancer (Fig. 1A, 1B), and positron emission tomography/CT (PET-CT) indicated that peripheral lung adenocarcinoma was not excluded (SUVmax 2.8) (Fig. 2). No significant improvement was observed in her imaging manifestation after the post-treatment review (Fig. 1C, 1D), and the possibility of fungal infection along with pulmonary neoplasms was not completely excluded. Then a CT-guided percutaneous lung biopsy was performed, whose pathological report suggested fibroproliferation with chronic inflammatory cell infiltration (Fig. 3). Although no evidence of pulmonary neoplasms was observed during the lung biopsy, a thoracoscopic right upper lobectomy and lymph node dissection were performed. The postoperative pathological results revealed visible silver-stained (Fig. 4A) and PAS-positive (Fig. 4B) oval spores in the alveolar septum, bronchiole wall, and alveolar cavity, thus, indicating fungal infection. Background Scedosporium apiospermum (S. apiospermum) belongs to the asexual form of Pseudallescheria boydii, which is widely distributed in various environments. S. apiospermum is one of the most common causes of invasive fungal infection in patients with immune deficiencies, particularly after organ transplantation, acquired immune deficiency syndrome (AIDS), cystic fibrosis lung disease, structural lung diseases, and long-term use of immunosuppressants or glucocorticoids. The most common infection site of S. apiospermum is the lung, and the clinical symptoms are usually cough, expectoration, hemoptysis, fever, dyspnea, and pleuritic chest pain. Imaging changes associated with pulmonary S. apiospermum infection can be similar to those observed in pulmonary aspergillosis, such as typical fungal balls or non-specific, such as single or multiple nodular lesions with or without cavities, focal infiltration, phyllode infiltration, and bilateral diffuse infiltration. The key to effective treatment is an accurate and timely etiological diagnosis. Otherwise, delayed diagnosis may cause fatal consequences, especially for patients with hypoimmunity. However, it is noteworthy that S. apiospermum can also rarely infect people with normal immune function, similar to our case. Surgical resection has become an essential part of treatment. [1–3] We presented a rare case of a 54-year-old non- immunocompromised female patient who developed pulmonary S. apiospermum infection and was diagnosed with pulmonary S. apiospermum infection by metagenomic next-generation sequencing (mNGS) of bronchoalveolar lavage fluid (BALF), as well as the first literature review of pulmonary S. apiospermum infection in immunocompetent patients. Case Report Keywords: Metagenomic next-generation sequencing, pulmonary infection, Scedosporium apiospermum DOI: https://doi.org/10.21203/rs.3.rs-2722241/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License cense:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full Lic Version of Record: A version of this preprint was published at BMC Infectious Diseases on March 13th, 2024. See the published version at https://doi.org/10.1186/s12879-024-09140-3. Page 1/17 Page 1/17 Page 1/17 Abstract Background: Scedosporium apiospermum (S.apiospermum) belongs to the asexual form of Pseudallescheria boydiiand is widely distributed in various environments. S. apiospermumis the most common cause of pulmonary infection; however, invasive diseases are usually limited to patients with immunodeficiency. Case presentation: A 54-year-old Chinese non-smoker female patient with normal lung structure and function was diagnosed with pulmonary S. apiospermum infection by metagenomic next-generation sequencing (mNGS) of bronchoalveolar lavage fluid (BALF). The patient was admitted to the hospital after experiencing intermittent right chest pain for 8 months. Chest computed tomography revealed a thick-walled cavity in the upper lobe of the right lung with mild soft tissue enhancement. S. apiospermum was observed by the mNGS of BALF, sequence number 283. Following treatment with voriconazole (300 mg q12h d1; 200 mg q12h d2-d20), there was no improvement in chest imaging, and a thoracoscopic right upper lobectomy was performed. Postoperative pathological results observed silver staining and PAS-positive oval spores in the alveolar septum, bronchiolar wall, and alveolar cavity, and fungal infection was considered. The patient’s symptoms improved; the patient continued voriconazole for 2 months after surgery. No signs of radiological progression or recurrence were observed at the 10-month postoperative follow-up. Conclusion: This case report indicates that S. apiospermum infection can occur in immunocompetent individuals and that the mNGS of BALF can assist in its diagnosis and treatment. Additionally, the combined therapy of antifungal drugs and surgery exhibits a potent effect on the disease. Case Presentation Lung biopsy tissue from the upper lobe of the right lung revealed metaplasia from alveolar to bronchial, along with partial bronchiectasis. In and around the cavity, there was a large amount of inflammatory cell infiltration and foam cell aggregation, accompanied by lymphoid tissue hyperplasia. Fiber hyperplasia was observed in some regions, and alveolar epithelial hyperplasia was also visible (Fig. 5). The patient continued to consume voriconazole (200 mg po bid) for 2 months after surgery, and the diagnosis and treatment process are indicated in Fig. 6. Chest imaging was followed up at 1, 2, and 10 months after surgery, and no signs of recurrence were observed (Fig. 7). Page 2/17 Page 2/17 Table 1 Detailed auxiliary inspection results Parameter Result/Value CBC normal Hepatic function normal Coagulation function normal Scr normal BUN normal Stool routine normal Urine routine normal Tuberculosis related PPD negative TB-Ab negative T-SPOT negative X-PERT negative Acid-fast bacilli in sputum negative Non-tuberculous mycobacteria negative Infection related CRP normal PCT normal ESR normal Nine respiratory pathogens negative Tumor related CYFRA21-1 3.4ng/ml(0-3.3) ProGRP 42.6pg/ml(25.3–77.8) SCC 1.2ng/ml(0-2.7) CEA 1.7ng/ml(0.2–10.0) NSE 12.1pg/ml(0-16.3) Fungi related GM(plasma) test negative G(plasma) test negative GM(BALF) test 1.24 Immune status HIV negative Literature Review We searched the keywords “Pulmonary” or “Lung” and “Scedosporium” or “Scedosporiums” or “Scedosporiumapiospermum” on the PubMed database, and 25 medical records with complete case report data were retrospectively analyzed. The flow chart of the screening process is indicated in Fig. 8. The included patients were elaboratively summarized per the age, sex, major clinical manifestations, presence or absence of pre-existing disease, diagnostic methods, imaging manifestations, extensive or limited lesions, presence or absence of delay in diagnosis and treatment, treatment plan, and treatment outcome. (Table 2). Page 3/17 Brief summarization of included patients Brief summarization of included patients Characteristics 25 patients Sex (male / female) 12/13 Median age (year) 55 (7–83) Underlying disease 21 (0.84) Pulmonary tuberculosis 11 (0.44) Pulmonary cystic fibrosis 4 (0.16) Bronchiectasis 3 (0.12) Previous diagnosis of S. apiospermum 1 (0.04) Pulmonary hypertension 1 (0.04) Diabetes 1 (0.04) Previous tumor history 1 (0.04) Oral abscess 1 (0.04) No underlying disease 4 (0.16) Clinical manifestations Cough, Expectoration 16 (0.64) Hemoptysis 12 (0.48) Fever 10 (0.40) Dyspnea 9 (0.36) Night sweats 3 (0.12) Blood in sputum 2 (0.08) Weight loss 3 (0.12) Loss of appetite 2 (0.08) Chest pain 2 (0.08) Fatigue 2 (0.08) Pneumothorax 1 (0.04) Diagnosis time Misdiagnosed 9 (0.36) No misdiagnosis 16 (0.64) Diagnostic methods Bronchoalveolar lavage fluid culture 13 (0.52) Sputum culture 7 (0.28) Blood culture 1 (0.04) Tissue smear after lung biopsy 1 (0.04) Specimen culture after lung puncture biopsy and transbronchial lung biopsy 2 (0.08) Specimen culture after operation 4 (0.16) Metagenomic sequencing in bronchoalveolar lavage fluid 2 (0.08) Lesion range Limited 12 (0.48) Extensive 13 (0.52) Treatment scheme Antifungal therapy 15 (0.60) Surgical treatment 4 (0 16) Page 4/17 Characteristics 25 patients Antifungal + Surgical 6 (0.24) Prognosis Cure 19 (0.76) Improvement 2 (0.08) Death 2 (0.08) No mention 2 (0.08) A total of 25 immunocompetent patients with pulmonary S. apiospermum infection were reported on PubMed, and the basic characteristics of these patients are summarized in Table 3. In total, 12 males (48%) and 13 females (52%) were included; the average age of the patients was 50.96 years, ranging from 7 to 83 years. Literature Review Of the included patients, 84% had the following underlying diseases: 16 with cough and expectoration (64%), 12 with hemoptysis (48%), 10 with fever (40%), nine with dyspnea (36%), and other symptoms including night sweats (12%), weight loss (12%), chest pain (8%), blood in the sputum (8%), anorexia (8%), fatigue (8%), and pneumothorax (4%). Coughing and expectoration were the most common symptoms, followed by hemoptysis and fever. The severity of symptoms also varied, from inconspicuous pulmonary symptoms (three cases) to dyspnea (nine cases). Previously, pulmonary tuberculosis was the most common underlying disease (11/25, 44%), followed by pulmonary cystic fibrosis and bronchiectasis (3/25, 12%). The main diagnostic method was BALF culture in 13 cases (52%), followed by sputum culture in seven cases (28%), postoperative tissue culture in four cases (16%), lung biopsy and transbronchial lung biopsy in two cases (8%), gene sequencing of alveolar lavage fluid in two cases (8%), blood culture in one case (4%), and lung biopsy smear in one case (4%). Among the 25 patients included, four (16%) were treated with surgery, 15 (60%) with antifungal therapy (one with nebulized alpha- streptase in association with 7% hypertonic saline), and six (24%) with a combination of both therapies. Nine patients (36%) delayed treatment, of which four (16%) were misdiagnosed with Aspergillus infection, one (8%) had empirical anti-tuberculosis treatment, one (8%) whose BALF culture was misdiagnosed with contamination, one (8%) had cultured fungi in lung biopsy but the type was not further determined and treatment was delayed, one (8%) had a sputum culture of S. apiospermum from 3 years before but it was overlooked, and one (8%) was treated with antibiotics without etiological evidence. Fortunately, the nine patients received effective treatment after diagnosis. Two of the 25 reported patients did not mention the prognosis. Among the remaining 23 reported, two died (8.7%), 19 were cured (82.6%), and two (8.7%) exhibited improved conditions. One patient died after receiving only antifungal treatment (1/15, 6.7%). All four patients who only received surgical treatment were cured, and five (83.3%, 5/6) responded favorably to surgery combined with antifungal therapy. The majority of hosts with normal immune function had a favorable prognosis, however, factors such as prolonged disease duration, combined underlying diseases, and delayed diagnosis and treatment may cause the death of some patients. Literature Review Page 5/17 Table 3 Basic characteristics of included patients Case load Age /Gender Symptom Past medical history and associated risk factors Diagnosis Imaging performance Limit /extensive Delay diagnosis or not Therapy Outcome 1[16] 44/female Hemoptysis Cough Blood in sputum Weight loss Anorexia None BALF culture Hollow lesion in the left upper lobe, Bronchiectasia Limit Yes, Anti-TB Antibiotic treatment Voriconazole → Surgery→ Voriconazole Cure 2[17] 72/male Fever Hemoptysis TB at the age of 30 years Sputum culture Hollow lesion pulmonary infiltration, Air crescent sign Extensive Yes, Misdiagnosed as aspergillus infection Miconazole Not mention 3[18] 24 /male Chronic cough Expectoration Intermittent Hemoptysis Tooth decay recurrent oral abscesses Puncture biopsy Hollow lesion typical of a fungal ball Limit Yes, Antibiotic treatment Itraconazole →Surgery Not mention 4[19] 47 /male Hemoptysis Cough Expectoration Dyspnea TB for 6 years Sputum culture Fungal bulb, Bilateral uneven infiltrating foci Limit No Itraconazole →Surgery Cure 5[20] 59 /female Fever None BALF culture and DNA sequence Infiltrates and nodular lesions on both sides of the lungs Extensive Yes, Misdiagnosed as aspergillus infection, treated with micafengin, in parallel with empiric antimicrobial therapy Voriconazole, Liposomal Amphotericin Improve 6[21] 26 /male Cough Expectoration Fever Spontaneous Pneumothora Fungal empyema S. Discussion And Conclusions S. apiospermum is widely distributed in various environments, such as contaminated water, wetlands, sewage, and saprobic heritage [4]. Most of the infections occur in patients with immune deficiency, such as those with AIDS, malignant tumors, long-term use of immunosuppressants or glucocorticoids, and organ transplantation, which can cause fatal disseminated infection [5-8]. Additionally, it can occur in patients with normal immune function. A literature review revealed that 11 patients (44%) had a history of pulmonary tuberculosis infection, which was consistent with Kantarcioglu et al's hypothesis that pulmonary tuberculosis infection was the main risk factor for S. apiospermum pulmonary infection [9]. It has been reported that the risk factors for S. apiospermum infection in immunocompromised patients include lymphopenia, neutropenia, and serum albumin levels of <3 mg/dL. [6]. In immunocompetent patients, the main risk factors for S. apiospermum infection are surgery or trauma [10], and the lung and upper respiratory tract are the most infected sites. These infections can fall into the following categories: Transient local colonization, bronchopulmonary saprobic involvement, fungus ball formation, and invasive S. apiospermum pneumonia [1]. Among the clinical features of S. apiospermum pulmonary infection, fever is the most common clinical sign and symptom in most cases, and other common symptoms are cough, expectoration, hemoptysis, dyspnea, and pleuritic chest pain [11]. The imaging manifestations of S. apiospermum pulmonary infection are similar to those of other infections, such as the formation of a fungal ball in the original cavity, which is difficult to differentiate from an Aspergillus ball using radiograms. It may also exhibit solitary or multiple nodular lesions with or without cavitation, focal, lobar, or bilateral diffused infiltration [1]. Consistent with this review, S. apiospermum infection is frequently misdiagnosed as pulmonary aspergillosis or tuberculosis using non-specific imaging. The imaging of this patient presented thick-walled cavities in the right upper lobe, with uneven thickness, an irregular shape, and adjacent pleural adhesion. It was consistent with the lack of specificity in imaging of S. apiospermum-caused pulmonary infection, which is the main reason for using empirical anti-tuberculosis treatment before antifungal therapy. S. apiospermum infection can be diagnosed by microbiology (including direct staining and culture), histopathology, and polymerase chain reaction to identify fungal DNA. Additionally, serology can also assist in the diagnosis as S. apiospermum infection can be detected by antigen detection using S. Literature Review apoispermum infection BALF culture Bronchiectasia Multiple cavities with nodules Enlarged mediastinal lymph nodes Extensive No Posaconazole →Surgery→ Posaconazole Cure 7[15] 73/female None None BALF and TBLB sample culture Single bossing Limit No Surgery Cure 8[22] 40/male Cough Hemoptysis TB for 15 years Postoperative specimens culture Typical fungal balls Air crescent sign Extensive Yes, Misdiagnosed as TB and Aspergillus infection Voriconazole Cure 9[23] 51/female Dry cough Night sweats None BALF culture Hollow lesion Airway dilation Limit Yes, considered as contaminant Voriconazole → Surgery Cure Page 6/17 Case load Age /Gender Symptom Past medical history and associated risk factors Diagnosis Imaging performance Limit /extensive Delay diagnosis or not Therapy Outcome 10[24] 83/female Cough Blood in sputum Fatigue Dyspnea Bronchiectasia COPD Chronic atrial fibrillation BALF culture Bronchiectasia Tree bud sign Limit No Voriconazole Cure 11[25] 72/female Hemoptysis Fever Polypnea None Boold culture Both lungs are scattered in blurred patches Extensive No Voriconazole and Amphotericin B →Terbinafine Cure 12[26] 67/male Hemoptysis Fever Non- tuberculous Mycobacte for 15 years BALF culture Fungal sphere cavular lesions Limit No Voriconazole Cure 13[27] 67/male Cough Hemoptysis Dyspnea Bronchiectasia TB BALF culture Hollow lesions Bronchiectasia Tree bud sign Extensive Yes, Antibacterial therapy Itraconazole Voriconazole Cure 14[28] 71/male Fever Cough Expectoration TB Hypertension BALF culture and lung tissue biopsy culture Hollow lesions Fungal sphere- like shadows limit Yes, Misdiagnosed as Aspergillus Voriconazole Cure 15[29] 74/female None Mycobacterium tuberculosis avium infection BALF culture Bronchiectasia, Cavity, Nodules Extensive No Voriconazole Cure 16[30] 54/female Fever Dry cough Dyspnoea Weight loss TB Sputum culture Left lower lung infiltration and diffuse small nodular infiltration in the right lung Extensive No Miconazole nitrate Ketoconazole Cure 17[31] 68/male Cough Purulent sputum Hemoptysis TB 40 years before Sputum culture A thick walled cavity with necrosis Extensive No Voriconazole →Surgery Death Page 7/17 Blood in sputum Fatigue Dyspnea COPD Chronic atrial fibrillation Tree bud sign 11[25] 72/female Hemoptysis Fever Polypnea None Boold culture Both lungs are scattered in blurred patches Extensive No Voriconazole and Amphotericin B →Terbinafine Cure 12[26] 67/male Hemoptysis Fever Non- tuberculous Mycobacte for 15 years BALF culture Fungal sphere cavular lesions Limit No Voriconazole Cure 13[27] 67/male Cough Hemoptysis Dyspnea Bronchiectasia TB BALF culture Hollow lesions Bronchiectasia Tree bud sign Extensive Yes, Antibacterial therapy Itraconazole Voriconazole Cure 14[28] 71/male Fever Cough Expectoration TB Hypertension BALF culture and lung tissue biopsy culture Hollow lesions Fungal sphere- like shadows limit Yes, Misdiagnosed as Aspergillus Voriconazole Cure 15[29] 74/female None Mycobacterium tuberculosis avium infection BALF culture Bronchiectasia, Cavity, Nodules Extensive No Voriconazole Cure 16[30] 54/female Fever Dry cough Dyspnoea Weight loss TB Sputum culture Left lower lung infiltration and diffuse small nodular infiltration in the right lung Extensive No Miconazole nitrate Ketoconazole Cure 17[31] 68/male Cough Purulent sputum Hemoptysis Night sweats Fever Dyspnea Weight loss Fatigue Anorexia TB 40 years before Sputum culture A thick walled cavity with necrosis Extensive No Voriconazole →Surgery Death 18[32] 36/female Chest pain Fever Cough Purulent sputum Dyspnea DM Postoperative tissue culture A nodular mass with meniscus sign in the right lower lobe with undefined border Limit No Surgery Cure 19[32] 57/male Right-side chest pain Hemoptysis TB Postoperative tissue culture Partial fibroatelectasic retraction of the left upper lobe and a thin- walled cavity Limit No Surgery Cure Page 7/17 Case load Age /Gender Symptom Past medical history and associated risk factors Diagnosis Imaging performance Limit /extensive Delay diagnosis or not Therapy Outcome 20[33] 61/female Cough Hemoptysis TB Sputum culture and BLAF culture Numerous cavities with indwelling fungal balls Bronchiectasis Extensive No Voriconazole and Bronchial artery embolism Cure 21[34] 55/male none History of bladder cancer Post- operative tissue culture Air crescent sign Limit No Surgery Cure 22[35] 17/male Hemoptysis Respiratory failure Pulmonary cystic fibrosis BALF Gene sequencing Severe spongous lung destruction Extensive Yes, Delayed treatment Venous Voriconazole and Liposomal Amphotericin B→nebulidized Voriconazole and intravenous Voriconazole Improve 23[36] 37/male Cough Expectoration Pulmonary cystic fibrosis Bronchiectasia Sputum culture Bronchiectasis Air crescent sign Hollow lesions Extensive No Intravenous Voriconazole → nebulidized Voriconazole and Amphotericin B Death 24[37] 7/female Fever Dyspnea Cough Pulmonary cystic fibrosis BALF culture and sputum culture Multiple bronchiectasis and bronchial thickening Extensive No Amphotericin B Itraconazole Cure 25[38] 12/male Dry cough Dyspnea Pulmonary cystic fibrosis BALF culture Peribronchial thickening in lower lobes Limit No Voriconazole → nebulidized dornase Alfa and 7% hypertonic saline Inprove Discussion And Conclusions apiospermum infection in patients with the normal immune function was 12.5%, 9.09% (2/22) in the operation group and 16.67% (3/18) in the non-operation group [16]. Our literature review revealed that the total mortality, curability, and improvement rates of 25 patients with normal immune function were 8.7%, 82.6%, and 8.7%, respectively. One patient who received antifungal treatment alone died (6.7%, 1/15) whereas four patients who received surgical treatment were cured, and five patients (83.3%, 5/6) responded favorably to surgery combined with antifungal therapy. In this case, the patient's immune function and lung structure were normal. The reasons for surgical treatment are as follows: 1. Based on the PET-CT report, it is suggested that the local metabolic activity of the upper lobe of the right lung was increased (SUVmax =2.8), and peripheral gonadal carcinoma was suspected; 2. after antifungal treatment, the foci were not significantly absorbed, and the possibility of pulmonary fungal infection complicated with lung cancer could not be excluded. This surgery aimed to remove the foci and actively resect the pulmonary tumor simultaneously based on a reported case of pulmonary S. apiospermum infection with pulmonary tumorlets in an immunocompetent patient [15]. The patient was followed up for 10 months after surgery, and the therapeutic effect was effective. In immunocompromised patients, clinical infections caused by S. apiospermum are difficult to treat and usually fatal whereas immunocompetent hosts had a better prognosis [1]. S. apiospermum infection is difficult to treat as it has been reported to be resistant to many antifungal agents, such as fluconazole, ketoconazole, flucytosine, terbinafine, itraconazole, and liposomal amphotericin B. Nevertheless, it is susceptible to voriconazole, and a few studies have reported its efficacy in the treatment of S. apiospermum infection [5, 12, 13, 15]. According to the literature, surgical resection is an effective treatment for localized lesions, and even in immunocompetent individuals, the infection usually necessitates complete surgical resection [1]. According to Liu et al.'s a meta- analysis and systematic review of pulmonary S. apiospermum infection, more than half of the immunocompetent patients with pulmonary infection received surgical treatment, however, this did not cause a better overall survival rate. However, since pharmacological antifungal therapy failure is more common in immunocompromised patients, surgical resection may help to improve survival rates whereas immunocompetent patients treated with antifungal therapy alone can have a good prognosis. Total mortality from pulmonary S. Discussion And Conclusions apiospermum infection in patients with the normal immune function was 12.5%, 9.09% (2/22) in the operation group and 16.67% (3/18) in the non-operation group [16]. Our literature review revealed that the total mortality, curability, and improvement rates of 25 patients with normal immune function were 8.7%, 82.6%, and 8.7%, respectively. One patient who received antifungal treatment alone died (6.7%, 1/15) whereas four patients who received surgical treatment were cured, and five patients (83.3%, 5/6) responded favorably to surgery combined with antifungal therapy. In this case, the patient's immune function and lung structure were normal. The reasons for surgical treatment are as follows: 1. Based on the PET-CT report, it is suggested that the local metabolic activity of the upper lobe of the right lung was increased (SUVmax =2.8), and peripheral gonadal carcinoma was suspected; 2. after antifungal treatment, the foci were not significantly absorbed, and the possibility of pulmonary fungal infection complicated with lung cancer could not be excluded. This surgery aimed to remove the foci and actively resect the pulmonary tumor simultaneously based on a reported case of pulmonary S. apiospermum infection with pulmonary tumorlets in an immunocompetent patient [15]. The patient was followed up for 10 months after surgery, and the therapeutic effect was effective. Despite the absence of trauma or surgery, people with normal immune function and lung structure can also be infected with S. apiospermum. This case demonstrated how mNGS of BALF can be used to diagnose pathogen infection when etiological evidence cannot be confirmed by routine auxiliary examination. It is a rapid, accurate, and effective detection method that can avoid the adverse consequences of delayed diagnosis and treatment. The combination of antifungal therapy and surgery is effective in the treatment of local lesions of pulmonary infection caused by S. apiospermum in hosts with normal immune function, especially when the possibility of a combined tumor cannot be overlooked. Abbreviations S. apiospermum: Scedosporium apiospermum; mNGS: metagenomic next-generation sequencing; BALF: bronchoalveolar lavage fluid; AIDS: acquired immune deficiency syndrome; CT: computed tomography; PET-CT: positron emission tomography/computedtomography; S. apiospermum: Scedosporium apiospermum; S. apiospermum: Scedosporium apiospermum; mNGS: metagenomic next-generation sequencing; BALF: bronchoalveolar lavage fluid; AIDS: acquired immune deficiency syndrome; CT: computed tomography; PET-CT: positron emission tomography/computedtomography; PET-CT: positron emission tomography/computedtomography; Discussion And Conclusions apiospermum infection can be diagnosed by microbiology (including direct staining and culture), histopathology, and polymerase chain reaction to identify fungal DNA. Additionally, serology can also assist in the diagnosis as S. apiospermum infection can be detected by antigen detection using counterimmunoelectrophoresis. However, owing to the cross-reactions with antigens from other fungi such as Aspergillus spp, this method was not reported in the case [1]. To the best of our knowledge, this is the first reported case to use mNGS of BALF in the diagnosis of pulmonary S. apiospermum infection. It can provide accurate and timely etiological evidence and effective diagnostic assistance. Furthermore, it reduces the time between diagnosis and treatment, which is critical because it can be fatal, especially in patients with low immunity. counterimmunoelectrophoresis. However, owing to the cross-reactions with antigens from other fungi such as Aspergillus spp, this method was not reported in the case [1]. To the best of our knowledge, this is the first reported case to use mNGS of BALF in the diagnosis of pulmonary S. apiospermum infection. It can provide accurate and timely etiological evidence and effective diagnostic assistance. Furthermore, it reduces the time between diagnosis and treatment, which is critical because it can be fatal, especially in patients with low immunity. Page 8/17 Page 8/17 In immunocompromised patients, clinical infections caused by S. apiospermum are difficult to treat and usually fatal whereas immunocompetent hosts had a better prognosis [1]. S. apiospermum infection is difficult to treat as it has been reported to be resistant to many antifungal agents, such as fluconazole, ketoconazole, flucytosine, terbinafine, itraconazole, and liposomal amphotericin B. Nevertheless, it is susceptible to voriconazole, and a few studies have reported its efficacy in the treatment of S. apiospermum infection [5, 12, 13, 15]. According to the literature, surgical resection is an effective treatment for localized lesions, and even in immunocompetent individuals, the infection usually necessitates complete surgical resection [1]. According to Liu et al.'s a meta- analysis and systematic review of pulmonary S. apiospermum infection, more than half of the immunocompetent patients with pulmonary infection received surgical treatment, however, this did not cause a better overall survival rate. However, since pharmacological antifungal therapy failure is more common in immunocompromised patients, surgical resection may help to improve survival rates whereas immunocompetent patients treated with antifungal therapy alone can have a good prognosis. Total mortality from pulmonary S. Acknowledgements JH, LL and RD conducted the literature review and edited the case presentation. QL and CL collected clinical data. XW, YZ and RZ provided valuable feedback for the report. HD reviewed and revised the manuscript. JH and LL contributed equally to this work. All the authors read and approved the final manuscript. Funding Availability of data and materials sets used and analysed during the current study available from the corresponding author on reasonable request. Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor of this journal. Ethics approval and consent to participate Page 9/17 The present study was conducted in line with the Declaration of Helsinki and approved by the Ethics Committee of The First Affiliated Hospital of Chongqing Medical University(Approval number K2023-107). Informed written and signed consent for participation from the patient was acquired prior to the submission. Author details 1Department of Oncology, The First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuan Jiagang, Yuzhong District, 400010, Chongqing, China. 2The First College of Clinical Medicine, Chongqing Medical University, Chongqing, 400016, China. 2The First College of Clinical Medicine, Chongqing Medical University, Chongqing, 400016, China. 2The First College of Clinical Medicine, Chongqing Medical University, Chongqing, 400016, China. 3Department of Pathology, School of Basic Medicine, Chongqing Medical University, No.1 Medical College Road 3Department of Pathology, School of Basic Medicine, Chongqing Medical University, No.1 Medical College Road, Yuzhong District, Chongqing 400016, China. 3Department of Pathology, School of Basic Medicine, Chongqing Medical University, No.1 Medical College Road, Yuzhong District, Chongqing 400016, China. 4The Second College of Clinical Medicine, Chongqing Medical University, Chongqing, 400016, China. 4The Second College of Clinical Medicine, Chongqing Medical University, Chongqing, 400016, China. 5Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuan Jiagang, Yuzhong District, 400010, Chongqing, China. 5Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuan Jiagang, Yuzhong District, 400010, Chongqing, China. Corresponding author: Haiyun Dai, Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuan Jiagang, Yuzhong District, 400010, Chongqing, China. Phone number: +8615086608108; fax number: +86 23-89012097; E-mail: auroredai@163 com Corresponding author: Haiyun Dai, Department of Respiratory and Critical Care Medicine, The First Affiliated Hospital of Chongqing Medical University, No.1 Youyi Road, Yuan Jiagang, Yuzhong District, 400010, Chongqing, China. Phone number: +8615086608108; fax number: +86 23-89012097; E-mail: auroredai@163.com. Xuekai Wu: 2079585503@qq.com Yuxin Zhang: 1902821108@qq.com Ruowen Zhang: 1066835222@qq.com Haiyun Dai: auroredai@163.com Haiyun Dai: auroredai@163.com References 1. Cortez KJ, Roilides E, Quiroz-Telles F, et al. Infections caused by Scedosporium spp. Clin Microbiol Rev. 2008;21(1):157-97. Doi: 10.1128/CMR.00039-07. 2. Douglas AP , Chen SC, Slavin MA: Emerging infections caused by non-Aspergillus filamentous fungi. Clin Microbiol Infect. 2016; 22(8):670-80. Doi:10.1016/j.cmi.2016.01.011 1. Cortez KJ, Roilides E, Quiroz-Telles F, et al. Infections caused by Scedosporium spp. Clin Microbiol Rev. 2008;21(1):157-97. Doi: 10.1128/CMR.00039-07. 1. Cortez KJ, Roilides E, Quiroz-Telles F, et al. Infections caused by Scedosporium spp. Clin Microbiol Rev. 200 1. Cortez KJ, Roilides E, Quiroz-Telles F, et al. Infections caused by Scedosporium spp. Clin Microbiol Rev. 2008;21(1):157-97. Doi: 10.1128/CMR.00039-07 1. Cortez KJ, Roilides E, Quiroz-Telles F, et al. Infections caused by Scedosporium spp. Clin Microbiol Rev. 2008;21(1):157-97. Doi: 10.1128/CMR.00039-07. 2. Douglas AP , Chen SC, Slavin MA: Emerging infections caused by non-Aspergillus filamentous fungi. Clin Microbiol Infect. 2016; 22(8):670-80. Doi:10.1016/j.cmi.2016.01.011 2. Douglas AP , Chen SC, Slavin MA: Emerging infections caused by non-Aspergillus filamentous fungi. Clin Microbiol Infect. 2016; 22(8):670-80. Doi:10.1016/j.cmi.2016.01.011 3. Ramirez-Garcia A, Pellon A, Rementeria A, et al.: Scedosporium and Lomentospora: an updated overview of underrated opportunists. Med Mycol. 2018;56:102-25.Doi:10.1093/mmy/myx113. 4. Kaltseis J, Rainer J, De Hoog GS: Ecology of Pseudallescheria and Scedosporium species in human- dominated and natural environments and their distribution in clinical samples. Med Mycol. 2009;47(4):398- 405. Doi: 10.1080/13693780802585317. 5. Husain S, Munoz P, Forrest G, et a1. Infecfions due to Scedosporium apiospermum and Scedosporium prolificans in transplant recipients：clinical characterstics and impact of alltlfungal agent therapy on outcome．Clin Infect Dis. 2005;40(1):89-99. Doi: 10.1086/426445. 6. Lamaris GA, Chamilos G, Lewis RE, et a1．Scedospodum infection in atertiary care cancer center：a review of 25 caes from 1989-2006. Clin Infect Dis. 2006;43(12):1580-1584. Doi: 10.1086/509579. 7. Ridde UJ, Chenoweth CE, Kauffman CA．Disseminated Scedosporium apiospermum infection in a previously healthy woman with HELLP syndrome. Mycoses. 2004;47(9):442-446. Doi: 10.1111/j.1439-0507.2004.01015.x. 8. Ortoneda, M, Pastor FA, Mayayo E, et a1. Comparison of the virulence of Scedosporium prolificans strains from different origins in a murine model. J Med Microbiol. 2002; 51(11):924-928. Doi: 10.1099/0022-1317-51-11-924. 9. Kantarcioglu AS, de Hoog GS, Guarro J. Clinical characteristic and epidemiology of pulmonary pseudallescheriasis. Rev Iberoam Micol. 2012;29(1):1-13. Doi: 10.1016/j.riam.2011.04.002. Competing interests The authors declare that they have no conficts of interest. Page 10/17 Doi: 10.1177/0300060520931620. 17. Koga T, Kitajima T, Tanaka R, et a1. Chronic pulmonary scedosporiosis simulating aspergillosis. Respirology. 2005;10(5):682-4. Doi: 10.1111/j.1440- 1843.2005.00769.x. 17. Koga T, Kitajima T, Tanaka R, et a1. Chronic pulmonary scedosporiosis simulating aspergillosis. Respirology. 2005;10(5):682-4. Doi: 10.1111/j.1440- 1843.2005.00769.x. 18. Fernando SSE, Jones P, Vaz R. Fine needle aspiration of a pulmonary mycetoma. A case report and review of the literature. Pathology. 2005;37(4):322-4 Doi: 10.1080/00313020500168786. 18. Fernando SSE, Jones P, Vaz R. Fine needle aspiration of a pulmonary mycetoma. A case report and review of the literature. Pathology. 2005;37(4):322-4. Doi: 10.1080/00313020500168786. 19. Agatha D, Krishnan KU, Dillirani VA, et a1. Invasive lung infection by Scedosporium apiospermum in an immunocompetent individual. Indian J Pathol Microbiol. 2014;57(4):635-7. Doi: 10.4103/0377-4929.142716. 19. Agatha D, Krishnan KU, Dillirani VA, et a1. Invasive lung infection by Scedosporium apiospermum in an immunocompetent individual. Indian J Pathol Microbiol. 2014;57(4):635-7. Doi: 10.4103/0377-4929.142716. 20. Nakamura Y, Utsumi Y, Suzuki N, et a1. Multiple Scedosporium apiospermum abscesses in a woman survivor of a tsunami in northeastern Japan: a case report. J Med Case Rep. 2011;25;5:526. Doi: 10.1186/1752-1947-5-526. 20. Nakamura Y, Utsumi Y, Suzuki N, et a1. Multiple Scedosporium apiospermum abscesses in a woman survivor of a tsunami in northeastern Japan: a cas report. J Med Case Rep. 2011;25;5:526. Doi: 10.1186/1752-1947-5-526. 21. Hassan T, Nicholson S, Fahy R. Pneumothorax and empyema complicating Scedosporium apiospermum mycetoma: not just a problem in the immunocompromised patients. Ir J Med Sci. 2011; 180(4):931-2. Doi: 10.1007/s11845-010-0621-0. 22. Rahman FU, Irfan M, Fasih N. Pulmonary scedosporiosis mimicking aspergilloma in an immunocompetent host: a case report and review of the literature. Infection. 2016;44(1):127-32. Doi: 10.1007/s15010-015-0840-4. 23. Ma KC, Pino A, Narula N, et a1. Scedosporium apiospermum mycetoma in an immunocompetent patient without prior lung disease. Ann Am Thorac Soc. 2017;14(1):145-147. Doi: 10.1513/AnnalsATS.201609-697LE. 24. Mir MAY, Shrestha DB, Suheb MZK. Scedosporium apiospermum pneumonia in an immunocompetent host.Cureus. 2021; 4;13(8):e16891. Doi: 10.7759/cureus.16891. 25. Jabr R, Hammoud K. Scedosporium apiospermum fungemia successfully treated with voriconazole and terbinafine. IDCases. 2020;8;22:e00928. Doi: 10.1016/j.idcr.2020.e00928. 26. Ogata H, Harada E, Okamoto I. Scedosporium apiospermum lung disease in a patient with nontuberculous mycobacteria. Respirol Case Rep. 2020; 10;9(1):e00691. Doi: 10.1002/rcr2.691. 27. Cruz R, Barros M, Reyes M. Pulmonary non invasive infection by Scedosporium apiospermum. Rev Chilena Infectol. 2015; 32(4):472-5. Doi: 10 4067/S0716 10182015000500018 27. Cruz R, Barros M, Reyes M. Pulmonary non invasive infection by Scedosporium apiospermum. Rev Chilena 10.4067/S0716-10182015000500018. 27. Page 10/17 Cruz R, Barros M, Reyes M. Pulmonary non invasive infection by Scedosporium apiospermum. Rev Chilena Infectol. 2015; 32(4):472-5. Doi: 10.4067/S0716-10182015000500018. 8. Ogata R, Hagiwara E, Shiihara J. A case of lung scedosporiosis successfully treated with monitoring of plasma Kokyuki Gakkai Zasshi. 2011; 49(5):388-92. 28. Ogata R, Hagiwara E, Shiihara J. A case of lung scedosporiosis successfully treated with monitoring of plasma voriconazole concentration level. Nihon Kokyuki Gakkai Zasshi. 2011; 49(5):388-92. 29. Jimeno VM, Muñoz EC. Diagnosis of A typical Mycobacterial and Fungal Coinfection. Int J Mycobacteriol. 2020;9(4):435-437. Doi: 10 4103/ijmy ijmy 98 20 ñoz EC. Diagnosis of A typical Mycobacterial and Fungal Coinfection. Int J Mycobacteriol. 2020;9(4):435-437. Do my_98_20. 9. Jimeno VM, Muñoz EC. Diagnosis of A typical Mycobacterial and Fungal Coinfection. Int J Mycobacteriol. 202 10.4103/ijmy.ijmy_98_20. 29. Jimeno VM, Muñoz EC. Diagnosis of A typical Mycobacterial and Fungal Coinfection. Int J Mycobacteriol. 2020;9(4):435-437. Doi: 10.4103/ijmy.ijmy_98_20. 0. Tekavec J, Mlinarić-Missoni E, Babic-Vazic V. Pulmonary Tuberculosis Associated With Invasive Pseudallesche 10.1378/chest.111.2.508. 30. Tekavec J, Mlinarić-Missoni E, Babic-Vazic V. Pulmonary Tuberculosis Associated With Invasive Pseudallescheriasis. Chest. 1997;111(2):508-11. Doi: 10.1378/chest.111.2.508. 31. Abgrall S, Pizzocolo C, Bouges-Michel C. Scedosporium apiospermum Lung Infection with Fatal Subsequen Immunocompetent Host. Clin Infect Dis. 2007;15;45(4):524-5. Doi: 10.1086/520009. 31. Abgrall S, Pizzocolo C, Bouges-Michel C. Scedosporium apiospermum Lung Infection with Fatal Subsequent Postoperative Outcome in an Immunocompetent Host. Clin Infect Dis. 2007;15;45(4):524-5. Doi: 10.1086/520009. 32. Severo LC, Oliveira FM, Irion K. Respiratory tract intracavitary colonization due to Scedosporium apiospermum: report of four cases. Rev Inst Med Trop Sao Paulo. 2004;46(1):43-6. Doi: 10.1590/s0036-46652004000100009. 32. Severo LC, Oliveira FM, Irion K. Respiratory tract intracavitary colonization due to Scedosporium apiosperm Sao Paulo. 2004;46(1):43-6. Doi: 10.1590/s0036-46652004000100009. 32. Severo LC, Oliveira FM, Irion K. Respiratory tract intracavitary colonization due to Scedosporium apiospermum: report of four cases. Rev Inst Med Trop Sao Paulo. 2004;46(1):43-6. Doi: 10.1590/s0036-46652004000100009. 33. Durand CM, Durand DJ, Lee R. A 61 year-old female with a prior history of tuberculosis presenting with hemoptysis. Clin Infect Dis. 2011;1;52(7):910, 957- 9. Doi: 10.1093/cid/cir009. 3. Durand CM, Durand DJ, Lee R. A 61 year-old female with a prior history of tuberculosis presenting with hemopt 9. Doi: 10.1093/cid/cir009. 33. Durand CM, Durand DJ, Lee R. A 61 year-old female with a prior history of tuberculosis presenting with hemoptysis. Clin Infect Dis. 2011;1;52(7):910, 957- 9. Doi: 10.1093/cid/cir009. 34. Refaï MA, Duhamel C, Rochais GPL. Lung scedosporiosis: a differential diagnosis of aspergillosis. Eur J Cardiothorac Surg. 2002;21(5):938-9. Page 10/17 Doi: 10.1016/s1010-7940(02)00068-4. 34. Refaï MA, Duhamel C, Rochais GPL. Lung scedosporiosis: a differential diagnosis of aspergillosis. Eur J Cardiothorac Surg. 2002;21(5):938-9. Doi: 10.1016/s1010-7940(02)00068-4. 4. Refaï MA, Duhamel C, Rochais GPL. Lung scedosporiosis: a differential diagnosis of aspergillosis. Eur J Cardio 10.1016/s1010-7940(02)00068-4. 35. Holle J, Leichsenring M,Meissner PE. Nebulized voriconazole in infections with Scedosporium apiospermum — case report and review of the literature. J Cyst Fibros. 2014;13(4):400-2. Doi: 10.1016/j.jcf.2013.10.014. 35. Holle J, Leichsenring M,Meissner PE. Nebulized voriconazole in infections with Scedosporium apiospermum — case report and review of the literature. J Cyst Fibros. 2014;13(4):400-2. Doi: 10.1016/j.jcf.2013.10.014. 5. Holle J, Leichsenring M,Meissner PE. Nebulized voriconazole in infections with Scedosporium apiospermum — Cyst Fibros. 2014;13(4):400-2. Doi: 10.1016/j.jcf.2013.10.014. 36. Borghi E, Iatta R, Manca A. Chronic airway colonization by Scedosporium apiospermum with a fatal outcome in a patient with cystic fibrosis. Med Mycol,.2010; 48 Suppl 1:S108-13. Doi: 10.3109/13693786.2010.504239. 36. Borghi E, Iatta R, Manca A. Chronic airway colonization by Scedosporium apiospermum with a fatal outcome in a patient with cystic fibrosis. Med Mycol,.2010; 48 Suppl 1:S108-13. Doi: 10.3109/13693786.2010.504239. 37. Vázquez-Tsuji O, Rivera TC, Zárate AR. Endobronchitis by Scedosporium apiospermum in a child with cystic fibrosis. Rev Iberoam Micol. 2006;23(4):245- 8. Doi: 10.1016/s1130-1406(06)70054-7. 37. Vázquez-Tsuji O, Rivera TC, Zárate AR. Endobronchitis by Scedosporium apiospermum in a child with cystic fibrosis. Rev Iberoam Micol. 2006;23(4):245- 8. Doi: 10.1016/s1130-1406(06)70054-7. Page 10/17 10. Castiglioni B, Sutton DA, Rinaldi MG. Pseudallescheria boydii (Anamorph Scedosporium apiospermum). Infection in solid organ transplant recipients in a tertiary medical center and review of the literature.Medicine (Baltimore). 2002;81(5):333-48. Doi: 10.1097/00005792-200209000-00001. 10. Castiglioni B, Sutton DA, Rinaldi MG. Pseudallescheria boydii (Anamorph Scedosporium apiospermum). Infection in solid organ transplant recipients in a tertiary medical center and review of the literature.Medicine (Baltimore). 2002;81(5):333-48. Doi: 10.1097/00005792-200209000-00001. 10. Castiglioni B, Sutton DA, Rinaldi MG. Pseudallescheria boydii (Anamorph Scedosporium apiospermum). Infection in solid organ transplant recipients in a tertiary medical center and review of the literature.Medicine (Baltimore). 2002;81(5):333-48. Doi: 10.1097/00005792-200209000-00001. 11. Lackner M, de Hoog GS, Verweij PE, et a1. Species-specific antifungal susceptibility patterns of Scedosporium and Pseudallescheria species. Antimicrob Agents Chemother. 2012; 56(5):2635-42. Doi:10.1128/AAC.05910-11. 11. Lackner M, de Hoog GS, Verweij PE, et a1. Species-specific antifungal susceptibility patterns of Scedosporium and Pseudallescheria species. Antimicrob Agents Chemother. 2012; 56(5):2635-42. Doi:10.1128/AAC.05910-11. 12. Troke P, Aguirrebengoa K, Arteaga C, et al. Treatment of scedosporiosis with voriconazole: clinical experience with 107 patients. Antimicrob Agents Chemother. 2008;52(5):1743-50. Doi: 10.1128/AAC.01388-07. 12. Troke P, Aguirrebengoa K, Arteaga C, et al. Treatment of scedosporiosis with voriconazole: clinical experience with 107 patients. Antimicrob Agents Chemother. 2008;52(5):1743-50. Doi: 10.1128/AAC.01388-07. 13. Capilla J, Guarro J. Correlation between in vitro susceptibility of Scedosporium apiospermum to voriconazole and in vivo outcome of scedosporiosis in guinea pigs. Antimicrob Agents Chemother.2004;48(10):4009-11.Doi:10.1128/AAC.48.10.4009-4011.2004. 13. Capilla J, Guarro J. Correlation between in vitro susceptibility of Scedosporium apiospermum to voriconazole and in vivo outcome of scedosporiosis in guinea pigs. Antimicrob Agents Chemother.2004;48(10):4009-11.Doi:10.1128/AAC.48.10.4009-4011.2004. 14. Liu W, Feng RZ, Jiang HL. Scedosporium spp. lung infection in immunocompetent patients: a systematic review and MOOSE-compliant meta- analysis.Medicine. 2019; 98(41):e17535. Doi: 10.1097/MD.0000000000017535. 14. Liu W, Feng RZ, Jiang HL. Scedosporium spp. lung infection in immunocompetent patients: a systematic review and MOOSE-compliant meta- analysis.Medicine. 2019; 98(41):e17535. Doi: 10.1097/MD.0000000000017535. 15. Motokawa N, Miyazaki T, Hara A. Pulmonary Scedosporium apiospermum infection with pulmonary tumorlet in an immunocompetent patient. Intern Med. 2018;1;57(23):3485-3490. Doi: 10.2169/internalmedicine.1239-18. 15. Motokawa N, Miyazaki T, Hara A. Pulmonary Scedosporium apiospermum infection with pulmonary tumorlet in an immunocompetent patient. Intern Me 2018;1;57(23):3485-3490. Doi: 10.2169/internalmedicine.1239-18. 16. Liu W, Feng RZ, Jiang HL. Management of pulmonary Scedosporium apiospermum infection by thoracoscopic surgery in an immunocompetent woman. J Int Med Res. 2020; 48(7):300060520931620. Doi: 10.1177/0300060520931620. 16. Liu W, Feng RZ, Jiang HL. Management of pulmonary Scedosporium apiospermum infection by thoracoscopic surgery in an immunocompetent woman J Int Med Res. 2020; 48(7):300060520931620. Figures Figures Figures Figure 1 Chest CT at different time points July 1st, 2021 Chest enhanced CT showed irregular soft-tissue density mass in the upper lobe of the right lung, and the bronchial branch of the upper lobe of the right lung was invaded and narrowed, which suggested a high possibility of lung cancer. There were also several small punctate calcification foci in the left lung(A). Slight calcification in mediastinum and left hilar lymph nodes(B). July 17th, 2021 After antifungal therapy, CT showed irregular soft tissue density shadow in the upper lobe of the right lung with cavity formation, which was considered to be lung cancer, with little change from the CT result before treatment(C,D). Page 11/17 38. Padoan R, Poli P, Colombrita D. Acute Scedosporium apiospermum endobronchial infection in cystic fibrosis. Pediatr Infect Dis J.2016;35(6):701-2. Doi: 10.1097/INF.0000000000001130. 38. Padoan R, Poli P, Colombrita D. Acute Scedosporium apiospermum endobronchial infection in cystic fibrosis. Pediatr Infect Dis J.2016;35(6):701-2. Doi: 10 1097/INF0000000000001130 38. Padoan R, Poli P, Colombrita D. Acute Scedosporium apiospermum endobronchial infection in cystic fibrosis. Pediatr Infect Dis J.2016;35(6):701-2. Doi: 10.1097/INF.0000000000001130. cedosporium apiospermum endobronchial infection in cystic fibrosis. Pediatr Infect Dis J.2016;35(6):701-2. Doi: Figure 1 Chest CT at different time points July 1st, 2021 Chest enhanced CT showed irregular soft-tissue density mass in the upper lobe of the right lung, and the bronchial branch of the upper lobe of the right lung was invaded and narrowed, which suggested a high possibility of lung cancer. There were also several small punctate calcification foci in the left lung(A). Slight calcification in mediastinum and left hilar lymph nodes(B). July 1st, 2021 Chest enhanced CT showed irregular soft-tissue density mass in the upper lobe of the right lung, and the bronchial branch of the upper lobe of the right lung was invaded and narrowed, which suggested a high possibility of lung cancer. There were also several small punctate calcification foci in the left lung(A). Slight calcification in mediastinum and left hilar lymph nodes(B). July 17th, 2021 After antifungal therapy, CT showed irregular soft tissue density shadow in the upper lobe of the right lung with cavity formation, which was considered to be lung cancer, with little change from the CT result before treatment(C,D). Page 12/17 Page 12/17 Page 12/17 Figure 2 PET-CT June 30th, 2021 PET-CT indicated space-occupying lesions in the upper lobe of the right lung with increased metabolic activity. Peripheral lung cancer was considered. Figure 2 PET-CT June 30th, 2021 PET-CT indicated space-occupying lesions in the upper lobe of the right lung with increased metaboli considered. Figure 2 Figure 2 June 30th, 2021 PET-CT indicated space-occupying lesions in the upper lobe of the right lung with increased metabolic activity. Peripheral lung cancer was considered. June 30th, 2021 PET-CT indicated space-occupying lesions in the upper lobe of the right lung with increased metabolic activity. Peripheral lung cancer was considered. June 30th, 2021 PET-CT indicated space-occupying lesions in the upper lobe of the right lung with increased metabolic activity. Peripheral lung cancer was considered Page 13/17 The pathological report of CT-guided percutaneous lung biopsy suggested fibroproliferation with chronic inflammatory cell infiltration Figure 4 The postoperative pathological results of thoracoscopic right upper lobectomy and lymph node dissection showed that silver staining (A) and PAS positive(B) oval spores were found in alveolar septum, bronchiolar wall and alveolar cavity, suggesting fungal infection. athological report of CT-guided percutaneous lung biopsy suggested fibroproliferation with chronic inflammatory Figure 4 Figure 4 The postoperative pathological results of thoracoscopic right upper lobectomy and lymph node dissection showed that silver staining (A) and PAS positive(B) oval spores were found in alveolar septum, bronchiolar wall and alveolar cavity, suggesting fungal infection. The postoperative pathological results of thoracoscopic right upper lobectomy and lymph node dissection showed that silver staining (A) and PAS positive(B) oval spores were found in alveolar septum, bronchiolar wall and alveolar cavity, suggesting fungal infection. Page 14/17 Figure 5 Lung biopsy tissue from the upper lobe of the right lung revealed metaplasia from alveolar to bronchial, along with partial bronchiectasis. In and around the cavity, there was a large amount of inflammatory cell infiltration and foam cell aggregation, accompanied by lymphoid tissue hyperplasia. Fiber hyperplasia was observed in some regions, and alveolar epithelial hyperplasia was also visible. Figure 5 September 29th, 2021 Two months after operation, CT showed the absence of the upper lobe of the right lung, the linear high-density shadow of the right side of the lung, and the adjacent patchy soft tissue shadow, which was slightly smaller than that of 1 month after operation. There was a little effusion in the right interlobar fissure, which was slightly less than before(C,D). August 31st, 2021 August 31st, 2021 One month after operation, HRCT showed the absence of the upper lobe of the right lung, the linear high-density shadow of the right hilar, and the adjacent patchy soft tissue shadow, considering the possibility of postoperative changes. There is a little effusion in the right interlobar fissure(A,B). One month after operation, HRCT showed the absence of the upper lobe of the right lung, the linear high-density shadow of the right hilar, and the adjacent patchy soft tissue shadow, considering the possibility of postoperative changes. There is a little effusion in the right interlobar fissure(A,B). Figure 5 Lung biopsy tissue from the upper lobe of the right lung revealed metaplasia from alveolar to bronchial, along with partial bronchiectasis. In and around the cavity, there was a large amount of inflammatory cell infiltration and foam cell aggregation, accompanied by lymphoid tissue hyperplasia. Fiber hyperplasia was observed in some regions, and alveolar epithelial hyperplasia was also visible. Page 15/17 Figure 6 Timeline of events A flowchart shows the patient's entire diagnosis and treatment process. Page 15/17 Postoperative reexamination of CT Postoperative reexamination of CT June 24th, 2022 Ten months after operation, CT showed the absence of the upper lobe of the right lung, the linear high-density shadow in the right hilar area and the adjacent cord shadow, and the soft tissue shadow disappeared 2 months after operation(E,F). Page 16/17 Page 16/17 Page 16/17 Page 16/17 Figure 8 Screening process The flow chart shows the process of literature review. Figure 8 Screening process The flow chart shows the process of literature review. The flow chart shows the process of literature review. The flow chart shows the process of literature review. Page 17/17
https://openalex.org/W4220820607	http://real.mtak.hu/140052/1/plants-11-00864.pdf	English	null	Redox and Hormonal Changes in the Transcriptome of Grape (Vitis vinifera) Berries during Natural Noble Rot Development	Plants	2,022	cc-by	16,837	Citation: Pogány, M.; Dankó, T.; Hegyi-Kaló, J.; Kámán-Tóth, E.; Szám, D.R.; Hamow, K.Á.; Kalapos, B.; Kiss, L.; Fodor, J.; Gullner, G.; et al. Redox and Hormonal Changes in the Transcriptome of Grape (Vitis vinifera) Berries during Natural Noble Rot Development. Plants 2022, 11, 864. https://doi.org/10.3390/ plants11070864 Academic Editor: Tika Adhikari Received: 22 February 2022 Accepted: 17 March 2022 Published: 24 March 2022 Citation: Pogány, M.; Dankó, T.; Hegyi-Kaló, J.; Kámán-Tóth, E.; Szám, D.R.; Hamow, K.Á.; Kalapos, B.; Kiss, L.; Fodor, J.; Gullner, G.; et al. Redox and Hormonal Changes in the Transcriptome of Grape (Vitis vinifera) Berries during Natural Noble Rot Development. Plants 2022, 11, 864. https://doi.org/10.3390/ plants11070864 Keywords: grapevine; Botrytis; Furmint; noble rot; Tokaj; redox; hormone; abscisic acid plants plants Article Redox and Hormonal Changes in the Transcriptome of Grape (Vitis vinifera) Berries during Natural Noble Rot Development Miklós Pogány 1,, Tamás Dankó 1, Júlia Hegyi-Kaló 2, Evelin Kámán-Tóth 1, Dorottya Réka Szám 3, Kamirán Áron Hamow 1 , Balázs Kalapos 1 , Levente Kiss 1,4 , József Fodor 1, Gábor Gullner 1, Kálmán Zoltán Váczy 2 and Balázs Barna 1 1 Centre for Agricultural Research, 2462 Martonvásár, Hungary; tamas.danko89@gmail.com (T.D.); toth.evelin@atk.hu (E.K.-T.); hamow.kamiran@atk.hu (K.Á.H.); kalapos.balazs@atk.hu (B.K.); kiss.levente@atk.hu or levente.kiss@usq.edu.au (L.K.); fodor.jozsef@atk.hu (J.F.); gullner.gabor@atk.hu (G.G.); barna.balazs@atk.hu (B.B.) 2 Food and Wine Research Institute, Eszterházy Károly Catholic University, 3300 Eger, Hungary; hegyi-kalo.julia@uni-eszterhazy.hu (J.H.-K.); vaczy.kalman@uni-eszterhazy.hu (K.Z.V.) gy j y y y 3 Georgikon Campus, Hungarian University of Agriculture and Life Sciences, 8360 Keszthely, Hungary; szam.dorottya.reka@gmail.com y g 4 Centre for Crop Health, University of Southern Queensland, Toowoomba, QLD 4350, Australia 4 Centre for Crop Health, University of Southern Queensland, Toowoomba, QLD 4350, Australia Correspondence: poganymiklos@atk hu 4 Centre for Crop Health, University of Southern Queensland, Toowoomba, QLD 4350, Australia * Correspondence: pogany.miklos@atk.hu Ce t e o C op ea t , U ve s ty o Sout e Quee s a d, oowoo ba, Q 350, ust a a * Correspondence: pogany.miklos@atk.hu * Correspondence: pogany.miklos@atk.hu Abstract: Noble rot is a favorable form of the interaction between grape (Vitis spp.) berries and the phytopathogenic fungus Botrytis cinerea. The transcriptome pattern of grapevine cells subject to natural noble rot development in the historic Hungarian Tokaj wine region has not been previously published. Furmint, a traditional white Tokaj variety suited to develop great quality noble rot was used in the experiments. Exploring a subset of the Furmint transcriptome redox and hormonal changes distinguishing between noble rot and bunch rot was revealed. Noble rot is deﬁned by an early spike in abscisic acid (ABA) accumulation and a pronounced remodeling of ABA-related gene expression. Transcription of glutathione S-transferase isoforms is uniquely upregulated, whereas gene expression of some sectors of the antioxidative apparatus (e.g., catalases, carotenoid biosynthesis) is downregulated. These mRNA responses are lacking in berries exposed to bunch rot. Our results help to explain molecular details behind the ﬁne and dynamic balance between noble rot and bunch rot development. plants plants 1. Introduction Botrytis cinerea (gray mold), the ﬁlamentous plant pathogenic fungus, is responsible for losses in crop production of $10 billion to $100 billion worldwide [1]. Grapevine (Vitis vinifera L.) is the economically most important host of this pathogen. With the onset of ripening, grape berries exhibit a striking rise in their susceptibility to gray mold [2]. Cooler, rainy weather during harvest time often results in the development of disastrous bunch rot epidemics in vineyards unless the clusters are properly protected from B. cinerea. The interaction of B. cinerea with mature berries may also develop into a unique direction called noble rot, under the inﬂuence of special mesoclimatic conditions. This includes a short rainy period lasting for a few days followed by an extended period of warmer, windy, dry weather during daytime and cooler, misty, humid nights [3,4]. Berries of white grapevine cultivars affected by noble rot change their color to dark brown. They also shrivel and accumulate charming aroma components, whereas berries exposed to bunch rot develop a texture of soft decay with unappealing taste and smell [5,6]. Noble rotted grape berries are the main sources of ﬂavor and odor in the outstanding botrytized dessert wines, such as the Trockenbeerenauslese of Germany, the Sauternes of France, or the Aszú Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/plants Plants 2022, 11, 864. https://doi.org/10.3390/plants11070864 Plants 2022, 11, 864 2 of 21 nes of of Hungary. Transcriptome studies of grape berry samples affected by natural or artiﬁcial noble rot induction have been published [7,8]. Here we present a subset of our RNAseq results using Vitis vinifera cv. Furmint samples exposed to natural noble rot development that were collected in a Tokaj vineyard in Hungary. Grapevine transcriptomic changes connected to the redox status and hormonal responses of botrytized grape berry cells are in the center of this work. Results of fungal gene expression will be presented in a later paper. Several reactive oxygen species (ROS) producing, scavenging, and signaling grapevine molecules have been connected to noble rot development [7,8]. 1. Introduction The role of plant hormones, especially ethylene, abscisic acid, and auxins, has also been shown to be crucial in the processes of noble rot and berry ripening [7–9]. Redox-related transcriptional responses and transitions in plant hormone-associated gene activity are thoroughly investigated and discussed, focusing speciﬁcally on those factors that discriminate between noble rot and bunch rot. by natural or artificial noble rot induction have been published [7,8]. Here we present a subset of our RNAseq results using Vitis vinifera cv. Furmint samples exposed to natural noble rot development that were collected in a Tokaj vineyard in Hungary. Grapevine transcriptomic changes connected to the redox status and hormonal responses of botrytized grape berry cells are in the center of this work. Results of fungal gene expression will be presented in a later paper. Several reactive oxygen species (ROS) producing, scavenging, and signaling grapevine molecules have been connected to noble rot development [7,8]. The role of plant hormones, especially ethylene, abscisic acid, and auxins, has also been shown to be crucial in the processes of noble rot and berry ripening [7–9]. Redox-related transcriptional responses and transitions in plant hormone- associated gene activity are thoroughly investigated and discussed, focusing specifically on those factors that discriminate between noble rot and bunch rot. Interactions between host plants and pathogens result in striking modifications in the ll l d t t d h l t b li U i RNA d t t bt i d Interactions between host plants and pathogens result in striking modiﬁcations in the cellular redox status and hormonal metabolism. Using our RNAseq datasets obtained from mature grape berries affected by noble rot and published gene expression data, we evalu- ated the mRNA expression patterns of various botrytized samples. The most intriguing question is whether the molecular mechanisms of gray mold or noble rot development could be partly unraveled by comparing hormone-associated and redox transcriptional patterns? cellular redox status and hormonal metabolism. Using our RNAseq datasets obtained from mature grape berries affected by noble rot and published gene expression data, we evaluated the mRNA expression patterns of various botrytized samples. The most intriguing question is whether the molecular mechanisms of gray mold or noble rot development could be partly unraveled by comparing hormone-associated and redox transcriptional patterns? 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon Four different types of Vitis vinifera cv. Furmint samples were collected starting with healthy berries followed by 3 consecutive stages of noble rot. Stage I was characterized by the partly botrytized epidermis and no shriveling; stage II berries showed fully colored epidermis and mild shriveling, while stage III exhibited all features of mature noble rot berries (fully shriveled with only limited presence of B. cinerea material on the surface) (Figure 1). Four different types of Vitis vinifera cv. Furmint samples were collected starting with healthy berries followed by 3 consecutive stages of noble rot. Stage I was characterized by the partly botrytized epidermis and no shriveling; stage II berries showed fully colored epidermis and mild shriveling, while stage III exhibited all features of mature noble rot berries (fully shriveled with only limited presence of B. cinerea material on the surface) (Figure 1). Figure 1. Healthy and botrytized Vitis vinifera cv. Furmint berries authentically represent samples used in this work. Besides healthy berries, three successive stages of noble rot were defined. Figure 1. Healthy and botrytized Vitis vinifera cv. Furmint berries authentically represent samples used in this work. Besides healthy berries, three successive stages of noble rot were deﬁned. Figure 1. Healthy and botrytized Vitis vinifera cv. Furmint berries authentically represent samples used in this work. Besides healthy berries, three successive stages of noble rot were defined. Figure 1. Healthy and botrytized Vitis vinifera cv. Furmint berries authentically represent samples used in this work. Besides healthy berries, three successive stages of noble rot were deﬁned. Dry weight (mg/g) of berries representing individual stages of noble rot were as follows: Healthy—0.228 ± 0.028 SD, Stage I. Noble rot—0.255 ± 0.008, Stage II Noble rot— 0.285 ± 0.065, Stage III Noble rot—0.603 ± 0.065. Dry weight (mg/g) of berries representing individual stages of noble rot were as follows: Healthy—0.228 ± 0.028 SD, Stage I. Noble rot—0.255 ± 0.008, Stage II Noble rot—0.285 ± 0.065, Stage III Noble rot—0.603 ± 0.065. Using the transcriptome data of all grape berry samples, a principal component analysis (PCA) was performed (Figure 2). 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon PCA analysis did not separate collected at different sampling times (end of October or beginning of November) ch other. Therefore data of the two sampling dates were analyzed together in ent statistical processing of transcriptome results. Figure 2. Principal component analysis of transcriptome profiles detected in various Vitis vinifera cv. Furmint berry samples. Healthy, stage I. noble rot, stage II. noble rot and stage III. noble rot samples were collected in October and November of 2016. Figure 2. Principal component analysis of transcriptome proﬁles detected in various Vitis vinifera cv. Furmint berry samples. Healthy, stage I. noble rot, stage II. noble rot and stage III. noble rot samples were collected in October and November of 2016. Principal component analysis of transcriptome profiles detected in various Vitis vinifera int berry samples. Healthy, stage I. noble rot, stage II. noble rot and stage III. noble rot were collected in October and November of 2016. Figure 2. Principal component analysis of transcriptome proﬁles detected in various Vitis vinifera cv. Furmint berry samples. Healthy, stage I. noble rot, stage II. noble rot and stage III. noble rot samples were collected in October and November of 2016. provide an overall view on the transcriptional reprogramming of Furmint grape lls under the influence of noble rot, we compared the pattern of activated and d genes to those reported in two earlier Vitis vinifera-Botrytis transcriptome [2,7]. In one study, Sémillon samples that were subjects of natural noble rot ment were analyzed [7]. In another approach, Marselan grape samples collected neyard were artificially inoculated with B. cinerea and incubated under bunch rot- g conditions [2]. Stage I. Furmint noble rot samples (Figure 1) were contrasted ge II. Sémillon samples as they emerged most similar by macroscopic observation. ning the bunch rot experiment, mRNA expression results of mature Marselan collected 48 h after inoculation with B. cinerea were included. f h 1524 l d i i i F i 1186 (78%) To provide an overall view on the transcriptional reprogramming of Furmint grape berry cells under the inﬂuence of noble rot, we compared the pattern of activated and repressed genes to those reported in two earlier Vitis vinifera-Botrytis transcriptome stud- ies [2,7]. In one study, Sémillon samples that were subjects of natural noble rot development were analyzed [7]. In another approach, Marselan grape samples collected in the vineyard were artiﬁcially inoculated with B. 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon The first principal component (PC1) explaining 24.16% of the variance in our data separates healthy berries and stage III noble rot berries from the two earlier stages of noble rot (but it clusters the two earlier stages together). The second principal component (PC2) responsible for 14.78% of variance separates healthy and stage III noble rot samples from stages I and II samples, but it clusters healthy and stage III samples, and clusters stage I and stage II noble rot samples together. During the Using the transcriptome data of all grape berry samples, a principal component analysis (PCA) was performed (Figure 2). The ﬁrst principal component (PC1) explaining 24.16% of the variance in our data separates healthy berries and stage III noble rot berries from the two earlier stages of noble rot (but it clusters the two earlier stages together). The second principal component (PC2) responsible for 14.78% of variance separates healthy and stage III noble rot samples from stages I and II samples, but it clusters healthy and stage III samples, and clusters stage I and stage II noble rot samples together. During the discussion of our transcriptome data, we are focusing on the transition between healthy berries and stage I–II noble rot berries. We could not distinguish clearly between stage I Plants 2022, 11, 864 3 of 21 3 of 21 and stage II berries. Although the transcriptome pattern observed in stage III noble rot berries represents a unique group according to PCA, our prior berry vitality tests indicated a drastic decline in living berry exo- and mesocarp (skin and pulp) cells in Furmint stage III noble rot tissues. Therefore, transcriptome results of stage III noble rot samples were excluded from the discussion of results. PCA analysis did not separate samples collected at different sampling times (end of October or beginning of November) from each other. Therefore data of the two sampling dates were analyzed together in subsequent statistical processing of transcriptome results. ge II berries. Although the transcriptome pattern observed in stage III noble rot represents a unique group according to PCA, our prior berry vitality tests d a drastic decline in living berry exo- and mesocarp (skin and pulp) cells in stage III noble rot tissues. Therefore, transcriptome results of stage III noble rot were excluded from the discussion of results. (Supplemental Table S3). Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artificial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artiﬁcial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. 2.2. Abscisic Acid igure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble ot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples nalyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artificial inoculation with B. cinerea. (a) U e ulated DEG i bot yti ed healthy a le (b) Do e ulated DEG Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artiﬁcial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. igure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble ot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples nalyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artificial inoculation with B. cinerea. (a) U l t d DEG i b t ti d h lth l (b) D l t d DEG Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artiﬁcial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artiﬁcial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon 2.1. Transcriptional Pattern of Noble Rot in Furmint Correlates Well with Sémillon cinerea and incubated under bunch rot-inducing con- ditions [2]. Stage I. Furmint noble rot samples (Figure 1) were contrasted with stage II. Sémillon samples as they emerged most similar by macroscopic observation. Concerning the bunch rot experiment, mRNA expression results of mature Marselan berries collected 48 h after inoculation with B. cinerea were included. t of the 1524 upregulated grapevine transcripts in Furmint, 1186 (78%) were d in Sémillon. A matching comparison between Marselan and Sémillon indicates 7% overlap (Figure 3a). Examining the list of repressed transcripts, 70% of gulated Furmint grapevine mRNAs showed decreases in Sémillon by noble rot. In t, Marselan berries affected by bunch rot exhibited a lower (45%) overlap with n noble rot samples regarding the list of transcripts with reduced abundance 3b) Out of the 1524 upregulated grapevine transcripts in Furmint, 1186 (78%) were acti- vated in Sémillon. A matching comparison between Marselan and Sémillon indicates only 57% overlap (Figure 3a). Examining the list of repressed transcripts, 70% of downregulated Furmint grapevine mRNAs showed decreases in Sémillon by noble rot. In contrast, Marse- lan berries affected by bunch rot exhibited a lower (45%) overlap with Sémillon noble rot samples regarding the list of transcripts with reduced abundance (Figure 3b). 3b). These results suggest that Furmint and Sémillon samples (both exposed to noble rot-type botrytization) carry rather similar transcriptional patterns compared to bunch rot-affected Marselan samples. Real-time RT-PCR results conﬁrmed the reliability of our RNA-Seq analysis. The two assays provided comparable transcript abundance estimations with a correlation Plants 2022, 11, 864 4 of 21 coefﬁcient of 0.93 for Stage I. noble rot samples and 0.90 for Stage II. noble rot samples (Supplemental Table S3). 4 of 22 coefﬁcient of 0.93 for Stage I. noble rot samples and 0.90 for Stage II. noble rot samples (Supplemental Table S3). Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. natural Furmint noble rot samples analyzed in this study were compared to stage II. natural Sémillon noble rot samples [7] and Marselan berries [2] that were collected 48 h after artificial inoculation with B. cinerea. (a) Upregulated DEGs in botrytized vs. healthy samples. (b) Downregulated DEGs. Figure 3. Venn diagrams show the overlap of differentially expressed genes (DEGs) during noble rot in Furmint and Sémillon and bunch rot in Marselan. Stage I. Up egu ated EGs i 2.2. Abscisic Acid PYL2 and PYL4 are members of the PR10 group of pathogenesis-related proteins [11]. A PYL receptor- mediated crosstalk between ABA, JA, and anthocyanin accumulation was suggested [12,13]. receptor mediated crosstalk between ABA, JA, and anthocyanin accumulation was suggested [12,13]. Noble rot specifically activates mRNA expression of many ABA-responsive, drought tolerance-related transcription factors, such as the orthologs of Arabidopsis Dehydration- Responsive Element-Binding Protein 1E (VIT_02s0025g04460), NAC72 (VIT_19s0014g03290), AZF1 (VIT_07s0129g00240), and Abscisic Acid-Insensitive 5 (ABI5) (VIT_08s0007g03420). Beyond the group of transcription factors, noble rot also specifically upregulates other osmotic stress-associated genes, including two ABI5-targeted Late Embryogenesis Abundant 6 orthologs (VIT_13s0067g01240 and VIT_13s0067g01250), an ortholog of Arabidopsis Late Embryogenesis Abundant 3 (VIT_07s0005g00660) and GSTU17, a drought-stress responsive tau class glutathione S-transferase (VIT_01s0026g01340). Late Embryogenesis Abundant (LEA) proteins have been connected to cellular dehydration tolerance [14,15]. Another conspicuous ABA signaling response detected during noble rot is the repression of two Hypersensitive to ABA1 (HAB1) orthologs (VIT_04s0008g01420 and VIT_11s0016g01780). Gray mold development does not lead to the repression of grapevine HAB1 genes in the berries [2,16]. HAB1 is a type 2C protein phosphatase and an important negative regulator y gg Noble rot speciﬁcally activates mRNA expression of many ABA-responsive, drought tolerance-related transcription factors, such as the orthologs of Arabidopsis Dehydration- Responsive Element-Binding Protein 1E (VIT_02s0025g04460), NAC72 (VIT_19s0014g03290), AZF1 (VIT_07s0129g00240), and Abscisic Acid-Insensitive 5 (ABI5) (VIT_08s0007g03420). Beyond the group of transcription factors, noble rot also speciﬁcally upregulates other osmotic stress-associated genes, including two ABI5-targeted Late Embryogenesis Abundant 6 orthologs (VIT_13s0067g01240 and VIT_13s0067g01250), an ortholog of Arabidopsis Late Embryogenesis Abundant 3 (VIT_07s0005g00660) and GSTU17, a drought-stress responsive tau class glutathione S-transferase (VIT_01s0026g01340). Late Embryogenesis Abundant (LEA) proteins have been connected to cellular dehydration tolerance [14,15]. Another conspicuous ABA signaling response detected during noble rot is the repression of two Hypersensitive to ABA1 (HAB1) orthologs (VIT_04s0008g01420 and VIT_11s0016g01780). Gray mold development does not lead to the repression of grapevine HAB1 genes in the berries [2,16]. HAB1 is a type 2C protein phosphatase and an important negative regulator of ABA signaling [17]. ABA receptors PYL2 and PYL4 were shown to establish a physical interaction with HAB1 [13,18]. Postharvest drying of Malvasia grapes results in elevated ABA content [19]. Whether noble rot was also accompanied by an increasing concentration of cellular ABA needed conﬁrmation. Therefore, a UHPLC-MS/MS analysis was performed with our Furmint grape berry samples to monitor any transition in the ABA content during noble rot. Up egu ated EGs i 2.2. Abscisic Acid Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 4. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 4. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Seventy-five percent of ABA-associated responses detected in our noble-rotted Furmint samples were missing in grape berries affected by gray mold. These noble rot- specific transcriptional responses included the upregulation of transcripts encoding two ABA receptors, PYL2 and PYL4 (VIT_04s0008g00890 and VIT_13s0067g01940). PYL2 and PYL4 are members of the PR10 group of pathogenesis-related proteins [11]. A PYL receptor-mediated crosstalk between ABA, JA, and anthocyanin accumulation was Seventy-ﬁve percent of ABA-associated responses detected in our noble-rotted Furmint samples were missing in grape berries affected by gray mold. These noble rot-speciﬁc transcriptional responses included the upregulation of transcripts encoding two ABA receptors, PYL2 and PYL4 (VIT_04s0008g00890 and VIT_13s0067g01940). Up egu ated EGs i 2.2. Abscisic Acid These results suggest that Furmint and Sémillon samples (both exposed to noble rot- type botrytization) carry rather similar transcriptional patterns compared to bunch rot- affected Marselan samples. Real-time RT-PCR results confirmed the reliability of our RNA-Seq analysis. The two id d bl t i t b d ti ti ith l ti The plant stress and senescence hormone abscisic acid (ABA) is crucial in fruit ripening and Botrytis-induced berry responses [9,10]. Nevertheless, noble rot provoked a much more pronounced remodeling in the ABA-related transcriptional pattern compared to gray mold (Figure 4). Plants 2022, 11, 864 5 of 21 oked a mpared Figure 4. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 4. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 4. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of ABA- associated V. vinifera genes in Furmint, Sémillon, and Marselan. Noble rot affects the expression of ABA-connected transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 4. Up egu ated EGs i 2.2. Abscisic Acid Dihy- drophaseic acid (DPA), a key metabolite of ABA catabolism, was also detected. However, Plants 2022, 11, 864 6 of 21 etected he limits the level of the other key metabolite phaseic acid (PA) did not reach the limits of detection in the examined samples. Noble rot triggered a sharp increase in ABA concentration from the beginning of the ﬁrst visible symptoms (before any signs of shriveling), and its level kept exponentially growing in the berries till the end of the process. DPA content followed a similar pattern (Figure 5). Bunch rot (gray mold) disease of mature grape berries, however, did not seem to result in elevated ABA concentration [10]. Similar results were obtained with nearly mature kiwifruit after infection with B. cinerea [20]. g g y p ( y g shriveling), and its level kept exponentially growing in the berries till the end of the process. DPA content followed a similar pattern (Figure 5). Bunch rot (gray mold) disease of mature grape berries, however, did not seem to result in elevated ABA concentration [10]. Similar results were obtained with nearly mature kiwifruit after infection with B. cinerea [20]. Figure 5. Noble rot-induced remodeling of abscisic acid (ABA) and dihydrophaseic acid (DPA) levels in Furmint samples. Values are means ± SEM, n = 6 for all groups. Data were log-transformed before analysis by ANOVA. Different letters indicate statistically significant differences at p ≤ 0.01 (a) ABA content in berries representing consecutive stages of noble rot; (b) DPA content in berries representing consecutive stages of noble rot. Figure 5. Noble rot-induced remodeling of abscisic acid (ABA) and dihydrophaseic acid (DPA) levels in Furmint samples. Values are means ± SEM, n = 6 for all groups. Data were log-transformed before analysis by ANOVA. Different letters indicate statistically signiﬁcant differences at p ≤0.01. (a) ABA content in berries representing consecutive stages of noble rot; (b) DPA content in berries representing consecutive stages of noble rot. Figure 5. Noble rot-induced remodeling of abscisic acid (ABA) and dihydrophaseic acid (DPA) levels in Furmint samples. Values are means ± SEM, n = 6 for all groups. Data were log-transformed before analysis by ANOVA. Different letters indicate statistically significant differences at p ≤ 0.01 (a) ABA content in berries representing consecutive stages of noble rot; (b) DPA content in berries representing consecutive stages of noble rot. Figure 5. Up egu ated EGs i 2.2. Abscisic Acid Noble rot-induced remodeling of abscisic acid (ABA) and dihydrophaseic acid (DPA) levels in Furmint samples. Values are means ± SEM, n = 6 for all groups. Data were log-transformed before analysis by ANOVA. Different letters indicate statistically signiﬁcant differences at p ≤0.01. (a) ABA content in berries representing consecutive stages of noble rot; (b) DPA content in berries representing consecutive stages of noble rot. Since noble rot is accompanied by a prompt and drastic surge of ABA formation and an obvious remodeling of ABA-connected gene expression, and these responses are Since noble rot is accompanied by a prompt and drastic surge of ABA formation and an obvious remodeling of ABA-connected gene expression, and these responses are lacking in berries affected by bunch rot, ABA seems to possess a central and distinctive role in noble rot development. p ABA seems to be a versatile player in the resistance of tomatoes against Botrytis [21]. On the one hand, it contributes to Botrytis susceptibility in tomato leaves [22]. But on the other hand, its positive role in Botrytis resistance was also suggested through the activity of Abscisic Acid-Induced MYB1 (AIM1) transcription factor [23]. 2.4. Auxin The expression of several genes related to auxin metabolism or signaling was significantly affected by noble rot. Two auxin-responsive protein-encoding transcripts (VIT_08s0058g01160, VIT_09s0002g00650), an ARG7-like and a SAUR50 gene were massively upregulated. Further- more, an IAA-amino acid hydrolase ILR1-like 6 gene (VIT_18s0001g02570) and a type IV inositol polyphosphate 5-phosphatase 11 gene (VIT_18s0001g11790) were also induced during noble rot. Nevertheless, transcriptional changes that were unique to noble rot mostly included down- regulated mRNA responses, such as genes encoding an auxin-induced ubiquitin superfamily protein 22D (VIT_14s0030g02310), an auxin transporter-like protein 2 (VIT_03s0038g02140), or an auxin response factor 18 (VIT_13s0019g04380). This is in accord with Lovato et al. [8], where RNA-Seq and subsequent Mapman analysis revealed the suppression of the auxin pathway in Garganega grape berries exposed to artiﬁcial noble rot. 2.5. Cytokinin Noble rot affected the expression of only a handful of cytokinin-related transcripts. The most notable noble rot-speciﬁc mRNA response (that does not appear during bunch rot) is the repression of the cytokinin receptor transcript histidine kinase 4 (VIT_01s0011g06190). 2.3. Ethylene Response Transcription Factors ERF 109 is an APETALA2/ethylene response transcription factor that activates inhibitors of programmed cell death under salt stress and is involved in jasmonate-induced wound signaling in Arabidopsis [25,26]. ERF109 is also known as Redox Responsive Transcription Factor 1 (RRTF1). It is a component of a core redox signaling network in Arabidopsis [27]. ERF109 also activates transcription of a senescence-associated C2H2-type zinc ﬁnger transcription factor ZAT12 in Arabidopsis [28]. RRTF1 and ZAT12 provide crosslinks between oxidative stress and abiotic/biotic stress responses in plants [29]. Grapevine orthologs of redox-responsive and interconnected Arabidopsis TF genes ERF109 and ZAT12 (VIT_03s0063g00460 and VIT_13s0019g00480) were massively induced in berries subjected to noble rot. During bunch rot, however, RRTF1 transcription remains unaffected, and ZAT12 is only mildly upregulated. A citrus ortholog of ERF17 has been reported to play a role in chlorophyll degradation and degreening of citrus fruits during ripening by establishing a physical interaction with the DRE motif of pheophytin pheophorbide hydrolase [30]. Arabidopsis ERF71 is a major regulator of hypoxia response [31,32]. In grape berries, not only the Vitis vinifera ortholog of ERF71 was acti- vated during noble rot, but also its predicted interacting partners, such as grapevine cysteine oxidase 2 (VIT_08s0056g01300, VIT_14s0108g00090, VIT_17s0000g05790) acting as oxygen sensors [33], the non-symbiotic hemoglobin 1 (VIT_03s0063g01960) acting as a scavenger of nitric oxide [34] or the later discussed alcohol dehydrogenase 1. These interacting factors of ERF 71 have been conclusively associated with hypoxia stress adaptation in plants [35–37]. 2.3. Ethylene Response Transcription Factors A vast majority of noble rot-affected grapevine genes associated with the gaseous plant hormone ethylene belong to the group of ethylene response transcription factors Plants 2022, 11, 864 7 of 21 7 of 21 (ERFs). The expression of ERFs can be ethylene dependent or independent [24]. No- ble rot apparently activated the transcription of grapevine ERF109 (VIT_03s0063g00460), ERF17 (VIT_04s0008g02230), ERF98 (VIT_05s0049g00500), ERF71 (VIT_07s0005g00820), ERF11 (VIT_07s0141g00690), ERF12 (VIT_10s0003g00130), ERF16 (VIT_11s0016g00670), ERF54 (VIT_12s0059g00280), ERF3 (VIT_12s0059g01460 and VIT_19s0090g01080), ERF2 (VIT_16s0013g00890), ERF10 (VIT_18s0001g10150) and ERF23 (VIT_18s0089g01030). These ERFs were not induced in berries subjected to gray mold. ERF 109 is an APETALA2/ethylene response transcription factor that activates inhibitors of programmed cell death under salt stress and is involved in jasmonate-induced wound signaling in Arabidopsis [25,26]. ERF109 is also known as Redox Responsive Transcription Factor 1 (RRTF1). It is a component of a core redox signaling network in Arabidopsis [27]. ERF109 also activates transcription of a senescence-associated C2H2-type zinc ﬁnger transcription factor ZAT12 in Arabidopsis [28]. RRTF1 and ZAT12 provide crosslinks between oxidative stress and abiotic/biotic stress responses in plants [29]. Grapevine orthologs of redox-responsive and interconnected Arabidopsis TF genes ERF109 and ZAT12 (VIT_03s0063g00460 and VIT_13s0019g00480) were massively induced in berries subjected to noble rot. During bunch rot, however, RRTF1 transcription remains unaffected, and ZAT12 is only mildly upregulated. A citrus ortholog of ERF17 has been reported to play a role in chlorophyll degradation and degreening of citrus fruits during ripening by establishing a physical interaction with the DRE motif of pheophytin pheophorbide hydrolase [30]. Arabidopsis ERF71 is a major regulator of hypoxia response [31,32]. In grape berries, not only the Vitis vinifera ortholog of ERF71 was acti- vated during noble rot, but also its predicted interacting partners, such as grapevine cysteine oxidase 2 (VIT_08s0056g01300, VIT_14s0108g00090, VIT_17s0000g05790) acting as oxygen sensors [33], the non-symbiotic hemoglobin 1 (VIT_03s0063g01960) acting as a scavenger of nitric oxide [34] or the later discussed alcohol dehydrogenase 1. These interacting factors of ERF 71 have been conclusively associated with hypoxia stress adaptation in plants [35–37]. (ERFs). The expression of ERFs can be ethylene dependent or independent [24]. No- ble rot apparently activated the transcription of grapevine ERF109 (VIT_03s0063g00460), ERF17 (VIT_04s0008g02230), ERF98 (VIT_05s0049g00500), ERF71 (VIT_07s0005g00820), ERF11 (VIT_07s0141g00690), ERF12 (VIT_10s0003g00130), ERF16 (VIT_11s0016g00670), ERF54 (VIT_12s0059g00280), ERF3 (VIT_12s0059g01460 and VIT_19s0090g01080), ERF2 (VIT_16s0013g00890), ERF10 (VIT_18s0001g10150) and ERF23 (VIT_18s0089g01030). These ERFs were not induced in berries subjected to gray mold. 2.8.1. Glutathione S-Transferases One of the most pronounced redox-related (and global) noble rot-speciﬁc transition in the grape berry transcriptome is the uniform upregulation of the gene expression of glutathione S-transferase (GST) isoforms. In our experiments, several GST genes were markedly induced in grape berries during the progress of noble rot infection. The most signiﬁcant induction was observed in the case of the VIT_06s0004g05700 gene, which encodes a tau class GST. Furthermore, three tau class GST genes (VIT_00s0153g00050, VIT_16s0039g01070, VIT_19s0093g00320) and a phi class GST (VIT_12s0028g00920) were also markedly induced in the noble-rotted berries. Taken together, noble rot development in Furmint berries led to the marked induction of 19 GST isoforms (Figure 6), and in sharp contrast, gray mold infection resulted in a slight activation of only 5 GST genes [2]. The ‘classical’ function of GSTs is the cellular detoxiﬁcation of a wide variety of endobiotic and xenobiotic substrates by conjugating them to glutathione. Generally, tau class GST genes (GSTUs) can be induced by different abiotic stress factors [40] as well as by microbial infections [41], and they have an important role in the protection of plants against oxida- tive damage [42]. A tau class GST (GSTU10) was shown to participate in the transport of trans-resveratrol out of grapevine cells [43]. The expression of the several phi class grapevine GSTs (VvGST2, VvGST3, and VvGST4) was markedly higher in post-veraison berry skins concomitantly with the accumulation of anthocyanins [44]. These GSTs showed a non-enzymatic carrier (ligandin) function, by which they participate in the transport of anthocyanins and proanthocyanidins (tannins) from the cytosol to the plant vacuole [44,45]. Members of the lambda class GSTs (GSTLs) catalyze deglutathionylation reactions via a catalytic cysteine residue [46]. 2.6. Salicylic Acid Two grapevine salicylic acid-binding protein 2 isoforms (VIT_00s0253g00140, VIT_ 00s0253g00150) were upregulated in Furmint in a noble rot-speciﬁc manner. The Ara- bidopsis ortholog of VIT_00s0253g00140 (AtMES1) possessed methyl salicylate esterase activity with some additional preference for methyl jasmonate as a substrate as well. Ex- pression of AtMES1 complemented systemic acquired resistance deﬁciency in salicylic acid-binding protein 2-silenced tobacco [38]. In addition, two salicylate 1-O-methyltransferase Plants 2022, 11, 864 8 of 21 isoforms (VIT_12s0057g01060, VIT_12s0057g01070) were repressed in Furmint berries dur- ing noble rot. isoforms (VIT_12s0057g01060, VIT_12s0057g01070) were repressed in Furmint berries dur- ing noble rot. 2.8. Analysis of Genes Modulating the Redox State during Noble Rot A subset of our transcriptomic data associated with cellular redox status and regulation was also comprehensively evaluated because previous extensive gene expression studies have already identiﬁed some important aspects of redox-related mRNA responses during noble rot [7,8]. 2.7. Jasmonic Acid Concerning transcripts that regulate jasmonic acid metabolism, two jasmonate O- methyltransferase isoforms (VIT_18s0001g12890, VIT_18s0001g12900) were markedly acti- vated as a result of noble rot development in Furmint. This response is not speciﬁc to noble rot because Marselan berries subjects of bunch rot also showed strong induction of these two transcripts [2]. Jasmonate O-methyltransferases catalyze the methylation of jasmonate into methyl jasmonate, a plant volatile that serves as an important cellular regu- lator conveying diverse developmental processes and defense responses. Interestingly, a hydroxyjasmonate sulfotransferase gene (VIT_13s0084g00240) is highly suppressed in Furmint in a noble rot-speciﬁc manner. Sulfonation would inactivate jasmonic acid in plants, and inhibition of this reaction maintains the level of functional jasmonic acid in the berries during noble rot [39]. 2.8.2. Crocetin Glucosyltransferases Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 6. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 6. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 6. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. 2.8.2. Crocetin Glucosyltransferases Five crocetin glucosyltransferase genes were robustly activated during the development of B. cinerea infection. The highest induction was observed in the case of VIT_05s0062g00310 (183-fold induction), while the genes VIT_05s0062g00300, VIT_05s0062g00270, VIT_05s0062g00710, and VIT_05s0062g00700 were also markedly induced. Crocetin and its glycosylated deriva- tives, crocins are antioxidative apocarotenoids, which also provide a red color to saf- fron [47,48]. In addition, crocetin exhibits various health-promoting properties including anti-tumor, neuroprotective, anti-diabetic, anti-inﬂammatory and anti-hyperlipidemia ef- Plants 2022, 11, 864 Plants 2022, 11, x FOR 9 of 21 9 of 22 9 of 21 9 of 22 fects [49,50]. The ﬁnal step in the biosynthesis of the 20-carbon esteriﬁed carotenoid crocin is the transformation of the insoluble crocetin into a soluble and stable storage form by glucosylation [47]. In grapevine berries, the exact biochemical roles of crocetin and crocin are largely unknown. GSTs showed a non-enzymatic carrier (ligandin) function, by which they participate in the transport of anthocyanins and proanthocyanidins (tannins) from the cytosol to the plant vacuole [44,45]. Members of the lambda class GSTs (GSTLs) catalyze deglutathionylation reactions via a catalytic cysteine residue [46]. Figure 6. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 6. Heat map diagram showing the transcript abundance (Log2 fold change) of a set of V. vinifera glutathione S-transferase (GST) genes in Furmint, Sémillon, and Marselan cultivars. Fold induction results are presented after Log2 transformation. Noble rot affects the expression of GST transcripts more robustly than bunch rot does. Transcriptome data for Furmint stage I. noble rot was analyzed in this study and Sémillon stage II. noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. Figure 6. 2.8.2. Crocetin Glucosyltransferases noble rot samples [7] are displayed as they appear most comparable by macroscopic observation of the berries. Concerning the bunch rot dataset, mRNA expression results of mature Marselan berries [2] collected 48 h after inoculation with B. cinerea are presented. 2.8.2. Crocetin Glucosyltransferases 2.8.3. Changes in Transcript Levels of Phenylpropanoid Pathway Components Five crocetin glucosyltransferase genes were robustly activated during the development of B. cinerea infection. The highest induction was observed in the case of VIT_05s0062g00310 (183-fold induction), while the genes VIT_05s0062g00300, VIT_05s0062g00270, VIT_05s0062g00710, and VIT_05s0062g00700 were also markedly induced Crocetin and its glycosylated derivatives crocins are antioxidative Induction of the phenylpropanoid pathway, including the accumulation of ﬂavonoids is a typical metabolic response of plant cells exposed to pathogen infection [51]. Phenyl- propanoid molecules, such as the non-ﬂavonoid stilbenes, lignans, and phenolic acids or the ﬂavonoid ﬂavones, ﬂavonols, ﬂavanones, ﬂavanols, anthocyanins, and chalcones are signiﬁcant constituents of the non-enzymatic antioxidant pool of plant cells [52,53]. induced. Crocetin and its glycosylated derivatives, crocins are antioxidative apocarotenoids, which also provide a red color to saffron [47,48]. In addition, crocetin exhibits various health-promoting properties including anti-tumor, neuroprotective, anti- diabetic, anti-inflammatory and anti-hyperlipidemia effects [49,50]. The final step in the biosynthesis of the 20-carbon esterified carotenoid crocin is the transformation of the insoluble crocetin into a soluble and stable storage form by glucosylation [47]. In grapevine berries, the exact biochemical roles of crocetin and crocin are largely unknown. 2.8.3. Changes in Transcript Levels of Phenylpropanoid Pathway Components Induction of the phenylpropanoid pathway, including the accumulation of flavonoids is a typical metabolic response of plant cells exposed to pathogen infection [51]. Phenylpropanoid molecules, such as the non-flavonoid stilbenes, lignans, and phenolic g y p p Noble rot development triggered the transcription of enzymes responsible for the initial steps of the phenylpropanoid pathway (Figure 7). Several isoforms of phenylalanine ammonia-lyase and trans-cinnamate 4-monooxygenase were activated, contributing to the for- mation of key phenolic compounds, cinnamic acid, and p-coumaric acid. Induction of these two enzymes at the mRNA level is not speciﬁc to noble rot but rather a common transcriptional response of grape berry cells elicited by B. cinerea infection [2,16]. Stilbene synthase (STS) isoforms encoded in two massive clusters on chromosomes 10 and 16 have been uniformly upregulated in our Furmint noble rot samples similar to berries subjected to typical bunch rot disease. Two chalcone synthase (CHS) isoforms (VIT_14s0068g00920 and VIT_14s0068g00930), on the other hand, were inversely regulated between noble rot (repres- sion) and bunch rot (activation). 2.8.2. Crocetin Glucosyltransferases 2.8.3. Changes in Transcript Levels of Phenylpropanoid Pathway Components STS and CHS catalyzed branches of the phenylpropanoid pathway might be competing with each other in berries during noble rot, as it was re- Plants 2022, 11, 864 10 of 21 10 of 21 ported before in downy mildew-infected grapevine leaves [54]. Two chalcone synthase (CHS) genes (VIT_14s0068g00920 and VIT_14s0068g00930) are speciﬁcally downregulated during noble rot (Figure 7). In bunch rot, these two CHS isoforms are activated [16]. It is conceivable that they control branches of the ﬂavonoid biosynthesis pathway such as Flavonol synthase 1 (FLS1) and ﬂavonol synthesis that are responsible for the reduced ﬂavonol synthase activity characteristic of noble-rotted berries [7]. Interestingly, chalcone synthase (CHS) and stilbene synthase (STS) transcripts are differ- entially regulated during noble rot but are uniformly activated in bunch rot, providing another example of competition between CHS and STS pathways in grapevine. Similar competition between these two pathways has been found during development, abiotic stress, or downy mildew infection, which may direct the ﬂow of carbon toward stilbene biosynthesis at the expense of ﬂavonol formation [54]. y p [ ] VIT_18s0001g03470, the V. vinifera ortholog of Arabidopsis Flavonol synthase 1 (FLS1), was repressed in berries subjected to noble rot (Figure 7). FLS1 is the most important ﬂavonol synthase isoform in Arabidopsis with a deﬁning role in ﬂavonol levels com- pared to other isoforms [55]. Similarly, decreased ﬂavonol synthase activity was de- tected in Sémillon berries during noble rot development [7]. Bunch rot, however, did not suppress the transcription of vvFLS1 (VIT_18s0001g03470) in Trincadeira and Marselan berries [2,16]. Interestingly, Arabidopsis orthologs (JAO1 and JAO2) of two noble rot- induced ﬂavonol synthase transcripts (VIT_08s0105g00380 and VIT_13s0067g01020) have been recently shown to inactivate jasmonic acid and down-regulate immunity upon B. cinerea infection [56,57]. Noble rot development in the white grape cultivar Sémillon was associated with an unusual accumulation of anthocyanins [7]. In accordance, a cluster of an- thocyanidin 3-O-glucosyltransferase isoforms encoded on chromosome 3 (VIT_03s0017g02110, VIT_03s0017g02120, VIT_03s0017g02140) exhibited substantial upregulation during no- ble rot in white-skinned Furmint berries. Increased anthocyanin formation is a typical hallmark of ripening red-skinned cultivars, a phenomenon that is normally missing in berries of white-skinned grapevine cultivars [7]. Lastly, transcripts involved in isoﬂavonoid metabolisms, such as isoﬂavone reductase and isoﬂavone hydroxylases, were consistently upregulated due to noble rot (Figure 7). These enzymes have been linked with the synthesis of pterocarpan phytoalexins in legumes [58]. 2.8.4. Further Components of the Grape Berry Antioxidative System of Noble rot) of botrytization. The antioxidative capacity is expressed in Trolox equivalents. Values are means ± SEM, n = 5 for all groups. Different letters indicate statistically significant differences at p ≤ 0.01. Figure 8. The total antioxidative capacity of grapevine berries at four different stages (Healthy, Stages I.–III. of Noble rot) of botrytization. The antioxidative capacity is expressed in Trolox equiva- lents. Values are means ± SEM, n = 5 for all groups. Different letters indicate statistically signiﬁcant differences at p ≤0.01. Figure 8. The total antioxidative capacity of grapevine berries at four different stages (Healthy, Stages I.–III. of Noble rot) of botrytization. The antioxidative capacity is expressed in Trolox equivalents. Values are means ± SEM, n = 5 for all groups. Different letters indicate statistically significant differences at p ≤ 0.01. Figure 8. The total antioxidative capacity of grapevine berries at four different stages (Healthy, Stages I.–III. of Noble rot) of botrytization. The antioxidative capacity is expressed in Trolox equiva- lents. Values are means ± SEM, n = 5 for all groups. Different letters indicate statistically signiﬁcant differences at p ≤0.01. 2.9. Transcription of Redox Signaling-Connected and Redox-Regulated Genes in Noble Rot 2.9.1. NPR1 Interactors NIMIN-1 and NIMIN-2 2.9. Transcription of Redox Signaling-Connected and Redox-Regulated Genes in Noble Rot 2.9.1. NPR1 Interactors NIMIN-1 and NIMIN-2 Non-expression of PR genes 1 (NPR1 or NIM1) is a master redox sensor in plant defense [64]. Although transcription of V. vinifera NPR1/NIM1 isoforms did not seem to be activated in Furmint berries during noble rot, V. vinifera orthologs of NPR1-interacting NIMIN-1 and NIMIN-2 are among the most highly upregulated genes in our samples. Moreover, this robust transcriptional response is noble rot specific because in bunch rot samples NIMIN gene activation was lacking [2,16]. Non-expression of PR genes 1 (NPR1 or NIM1) is a master redox sensor in plant defense [64]. Although transcription of V. vinifera NPR1/NIM1 isoforms did not seem to be activated in Furmint berries during noble rot, V. vinifera orthologs of NPR1-interacting NIMIN-1 and NIMIN-2 are among the most highly upregulated genes in our samples. Moreover, this robust transcriptional response is noble rot speciﬁc because in bunch rot samples NIMIN gene activation was lacking [2,16]. p g g NPR1 protein plays a crucial role in defense responses against phytopathogens by integrating redox and salicylic acid signaling [64]. 2.8.4. Further Components of the Grape Berry Antioxidative System 2.8.4. Further Components of the Grape Berry Antioxidative System Another remarkable redox-associated response of noble rot is the speciﬁc inhibition of some sectors of the plant antioxidative system. Peroxisome-located catalase isoforms (VIT_00s0698g00010 and VIT_04s0044g00020) are repressed, and concomitantly, a set of chloroplastic carotenoid biosynthesis genes (e.g., VIT_05s0062g01110; [59]) is uniformly down-regulated. These antioxidant responses are consistently lacking in berries during bunch rot. In the meantime, noble rot development triggers the gradual decline of total (Trolox equivalent) antioxidant capacity in Furmint berries (Figure 8). Components of the ascorbate-glutathione cycle do not show apparent transcriptional changes. Explor- ing the most reasonable causes that may explain this observed decrease in antioxidant functions compared to bunch rot, the conclusively proved accumulation of anthocyanin phenylpropanoids in noble-rotted berries may be responsible for the downregulation of other antioxidant molecules. Anthocyanins serve as optical ﬁlters that protect chloroplasts from photoinhibition light ﬂuxes, such as in leaves of deciduous trees that turn red in the autumn [60–62]. In addition, anthocyanins are also free radical scavengers themselves that may replace other antioxidant systems in berries during noble rot, resulting in a decrease in (Trolox equivalent) net antioxidant capacity [63]. It should be noted, however, that bunch rot also induces a color change and some increase in the anthocyanin content of berries [16]. 11 of 21 Plants 2022, 11, 864 11 of 21 Figure 7. Transcription of genes encoding components of the phenylpropanoid pathway is g affected by noble rot in Furmint (analyzed in this study). Bunch rot also results in a m transcriptional change observed in Marselan berries [2]. Figure 7. Transcription of genes encoding components of the phenylpropanoid pathway is gre affected by noble rot in Furmint (analyzed in this study). Bunch rot also results in a marked trans tional change observed in Marselan berries [2]. igure 7. Transcription of genes encoding components of the phenylpropanoid pathway is greatly ffected by noble rot in Furmint (analyzed in this study). Bunch rot also results in a marked Figure 7. Transcription of genes encoding components of the phenylpropanoid pathway is greatly affected by noble rot in Furmint (analyzed in this study). Bunch rot also results in a marked transcrip- tional change observed in Marselan berries [2]. Plants 2022, 11, 864 12 of 21 ever, that ontent of 12 of 21 ever, that ontent of Figure 8. The total antioxidative capacity of grapevine berries at four different stages (Healthy, Stages I.–III. 2.8.4. Further Components of the Grape Berry Antioxidative System Transcription of two negative regulators of NPR1, NIMIN-1, and NIMIN-2 (VIT_07s0005g02070, VIT_01s0011g03430), is drastically upregulated during noble rot but not in bunch rot. Since NIMIN proteins are g g NPR1 protein plays a crucial role in defense responses against phytopathogens by integrating redox and salicylic acid signaling [64]. Transcription of two negative regulators of NPR1, NIMIN-1, and NIMIN-2 (VIT_07s0005g02070, VIT_01s0011g03430), is drastically upregulated during noble rot but not in bunch rot. Since NIMIN proteins are suppressors of some salicylic acid-induced responses and systemic acquired resistance [65,66] and salicylic acid signaling typically represses jasmonic acid-mediated signaling pathways [67], activa- tion of NIMIN-1 and NIMIN-2 may therefore contribute to Botrytis resistance during noble rot in comparison with bunch rot. Moreover, TGA9 (TGACG MOTIF-BINDING PROTEIN 9), a redox-sensitive bZIP transcription factor (VIT_06s0080g00360), and its nuclear interactor ROXY1, a CC-type glutaredoxin (VIT_01s0146g00220), are also transcriptionally activated in a noble rot-speciﬁc manner. TGA9 is connected to ROS-mediated Pattern-Triggered Immunity [68], and ROXY1 contributes to hydrogen peroxide accumulation and B. cinerea susceptibility in Arabidopsis [69]. In addition to NIMIN proteins, redox regulator TGA transcription factors also modulate NPR1 activity [70]. 2.9.2. MCP1 Metacaspase Lesion Simulating Disease 1 (LSD1) is an important regulator of ROS and hormonal homeostasis in plants [71]. Metacaspase 1 (MCP1) is a crucial component of programmed cell death develop- ment in plant cells. Arabidopsis MCP1 physically interacts with LSD1 and contributes to superoxide-initiated cell death in juvenile plant tissues, but in aging cells, it counteracts the process of cell death by regulating autophagy and vacuolar lysis [72,73]. Grapevine MCP1 isoforms (encoded by VIT_16s0013g00210 and VIT_16s0013g00220) are transcriptionally activated during noble rot of the berries. Bunch rot, on the other hand, does not lead to the Plants 2022, 11, 864 13 of 21 13 of 21 induction of MCP1 [2,16]. In ripe berries, the second function of MCP1 is more likely to be utilized by the plant cells, and in senescent tissues, MCP1 suppresses B. cinerea-induced tissue necrosis [73]. VvMCP1, a redox-associated plant metacaspase might therefore adjust the spread of programmed cell death in noble rot and orchestrate symptom development caused by the fungus. 2.9.3. Transcripts of Redox-Associated Chloroplastic Proteins Are Suppressed A uniform down-regulation of transcripts encoding redox-associated chloroplastic proteins has been observed in berries affected by noble rot. With a few exceptions, these genes listed in Table 1 correspond with transcripts of redox-sensitive starch metabolizing enzymes and redox-regulated proteins localized in the chloroplast of the model plant Arabidopsis thaliana [74,75]. Decreased mRNA abundance of chloroplastic redox proteins is a characteristic of noble rot, and it is not typical of berries exposed to bunch rot. Table 1. Grapevine genes encoding chloroplast-located redox proteins whose transcription is sup- pressed in Furmint berries as noble rot progresses. A. thaliana AGI Code Corresponding V. 2.9.2. MCP1 Metacaspase vinifera Ensembl ID Function Reference At3g50820 VIT_18s0001g11710 Oxygen-evolving enhancer protein 1 [75] At3g54050 VIT_08s0007g01570 Fructose-1,6-bisphosphatase [75] At1g32060 VIT_02s0109g00080 Phosphoribulokinase [75] At3g55800 VIT_13s0019g03350 Sedoheptulose-1,7-bisphosphatase [75] At2g39730 VIT_06s0004g05180 Ribulose bisphosphate carboxylase [75] At1g09340 VIT_14s0060g00820 Chloroplast stem-loop binding protein [75] At1g07320 VIT_06s0004g06140 50S ribosomal protein [76] At4g20360 VIT_17s0000g09370 Elongation factor TuB [76] At3g29320 VIT_14s0108g01560 Alpha-1,4 glucan phosphorylase L isozyme [74] At2g39930 VIT_11s0078g00310 Isoamylase 1 [74] At1g03310 VIT_07s0104g00370 Isoamylase 2 [74] At4g09020 VIT_18s0001g06520 Isoamylase 3 [74] At5g64860 VIT_07s0031g01540 4-alpha-glucanotransferase [74] At1g11720 VIT_10s0116g01730 Starch synthase 3 [74] At4g18240 VIT_11s0065g00150 Starch synthase 6 [74] At3g20440 VIT_19s0090g00920 1,4-alpha-glucan-branching enzyme 1 [74] At5g03650 VIT_08s0007g03750 1,4-alpha-glucan-branching enzyme 2 [74] At1g44575 VIT_18s0001g02740 Photosystem II 22 kDa protein [77] At1g12900 VIT_14s0068g00680 Glyceraldehyde-3-phosphate dehydrogenase A [77] At1g69830 VIT_14s0068g00420 Alpha-amylase-like 3 [78] At5g04140 VIT_08s0007g05260 Ferredoxin-dependent glutamate synthase [79] At1g68830 VIT_01s0011g03010 Serine/threonine-protein kinase STN7 [79] At1g20020 VIT_04s0023g03510 Ferredoxin-NADP-oxidoreductase 2 [80] At5g38430 VIT_17s0000g03690 Ribulose bisphosphate carboxylase small chain [81] Table 1. Grapevine genes encoding chloroplast-located redox proteins whose transcription is sup- pressed in Furmint berries as noble rot progresses. 2.9.4. Alcohol Dehydrogenase 1 Grapevine orthologs (VIT_04s0044g01110, VIT_04s0044g01120, VIT_04s0044g01130) of Arabidopsis Alcohol Dehydrogenase 1 (AtADH1) are distinctively upregulated during noble rot, which response does not occur in berries under bunch rot development. Alcohol dehydrogenases catalyze the last step of the ethanol fermentation pathway (the reversible conversion of acetaldehyde to ethanol) used by plants to cope with energy deﬁciency during hypoxic stress, and the activity of AtADH1 is controlled by redox cues [82,83]. g yp y y AtADH1 is not only redox-regulated, but it also contributes to the ABA sensitivity and abiotic (drought, salinity) as well as biotic (Pseudomonas syringae pv. tabaci DC3000) Plants 2022, 11, 864 14 of 21 14 of 21 stress resistance of plants. In addition, AtADH1 overexpression results in the accumulation of soluble sugars in leaves [84]. Grapevine orthologs of AtADH1 interactors appearing in the STRING v11 database, such as Pyruvate Decarboxylase-2 (VIT_08s0217g00100) and two aldehyde dehydrogenases (VIT_06s0004g02060, VIT_14s0066g01550), are also differentially expressed in a noble rot speciﬁc manner [85]. VvADH1 isoforms might be partly responsible for sugar accumulation characteristic of berries enduring noble rot [86], and they might be regulated by redox signals and ABA. 3.1. Plant Material and Sampling Healthy and botrytized Vitis vinifera cv. Furmint berry samples were collected in Mád (Tokaj wine region), Hungary in vineyard Betsek, 48◦11′16′′ N 21◦19′03′′ E, on 25 October and 3 November 2016. Collection and handling of the berries were the same at both sampling times. Four different samples were collected: healthy berries, stage I noble rot, stage II noble rot and stage III noble rot berries (Figure 1). Five independent biological replicates were collected at both time points in 50 mL centrifuge tubes for all four types of samples, each tube containing 5 (healthy), or 10 (stage I and II) or 15 (stage III) berries. Each stage of noble rot was represented by a total of ten biological replicates. Samples were frozen immediately in liquid nitrogen, transported in dry ice, and stored at −70 ◦C in an ultra-low temperature laboratory freezer. 2.9.5. WRKY Transcription Factors Transcript levels of two WRKY transcription factors, WRKY40 and WRK46, were distinctively elevated during noble rot. VIT_15s0046g01140, a grapevine ortholog of AtWRKY46 exhibited increasing abundance throughout the development of noble rot compared to bunch rot. Similarly, VIT_04s0008g05760, an AtWRKY40 ortholog, was upreg- ulated in Furmint berry samples subjected to noble rot. Both AtWRKY46 and AtWRKY40 are redox-responsive [29,87]. In Arabidopsis, thaliana WRKY40 interacts with the aforemen- tioned TF ERF109/RRTF1 by binding to the W box sequence of its promoter region [28,88]. ROS-generating stimuli from necrotrophic infections stimulate ERF109 expression, while biotrophic and mutualistic microbes repress ERF109 expression [28]. WRKY46 and WRKY40 are also important regulators of ABA signaling and osmotic stress responses [89–91]. It should be noted that some other WRKY TF genes, such as WRKY7 (VIT_07s0031g00080, VIT_18s0001g10030), WRKY9 (VIT_12s0055g00340), WRKY22 (VIT_15s0046g02190), and WRKY70 (VIT_08s0058g01390, VIT_13s0067g03140) with a less obvious connection to re- dox regulation or ABA signaling have also displayed noble rot-speciﬁc activation in our Furmint samples. 3.2. RNA Extraction Berries were ground to a ﬁne powder in liquid nitrogen by a Retsch laboratory ball mill. Considering the uneven dry weight of the berry samples, varying amounts of frozen plant powder were weighed into new 50 mL centrifuge tubes, 800 mg for healthy and stage I berries, 500 mg for stage II berries and 250 mg for stage III berries. Total RNA extraction was carried out according to Reid et al. [92]. Twenty milliliter extraction buffer [300 mM Tris HCl (pH 8.0), 25 mM EDTA, 2 M NaCl, 2% CTAB, 2% PVPP, 0.05% spermidine trihydrochloride, 2% β-mercaptoethanol (administered immediately before use)] was added to powdered samples which were incubated at 65 ◦C for 10 min and vortexed every 2 min. After that, mixtures were extracted twice with equal volumes of chloroform: isoamyl alcohol (24:1), thoroughly vortexed, and centrifuged at 4800× g for 10 min at 4 ◦C. The supernatant (10 mL) was transferred to a new 50 mL centrifuge tube and combined with 0.1 vol 3M NaOAc (pH 5.2) and 0.6 vol ice-cold isopropanol, mixed and kept for 30 min at −70 ◦C. The nucleic acid pellets were recovered by centrifugation at 4800× g for 30 min at 4 ◦C. The supernatant was removed, and the remaining pellet was washed with 5 mL of 70% Plants 2022, 11, 864 15 of 21 15 of 21 EtOH, then centrifuged at 4800× g for 15 min at 4 ◦C. The liquid was discarded and the precipitate was dried with a lyophilizer. p p y p The nucleic acid pellets were dissolved in 1 mL TE (Tris-EDTA) buffer (pH 7.0) and transferred to new 2 mL microcentrifuge tubes. Selective RNA precipitation was achieved by adding 0.3 vol of 8M LiCl and a subsequent overnight incubation at 4 ◦C. Total grapevine and fungal RNA were harvested by centrifugation at 21,000× g for 30 min at 4 ◦C followed by a washing step with 500 µL 70% ice-cold EtOH and ﬁnally drying by a lyophilizer. Isolated RNA pellets were dissolved in 42 µL nuclease-free ultrapure water, and RNA quality was ﬁrst evaluated using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientiﬁc, Waltham, MA, USA). Then all samples were subjected to DNase treatment using the DNA-free™DNA Removal Kit (Invitrogen, Thermo Fisher Scientiﬁc, Waltham, MA, USA). 3.3. RNA-Seq Library Preparation and Sequencing RNA-Seq libraries were prepared from total RNA using a TruSeq RNA Sample prepa- ration kit (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. Shortly, poly-A RNAs were captured by oligo-dT conjugated magnetic beads then the eluted mRNAs were fragmented at 94 ◦C. First-strand cDNA was synthesized by random priming reverse transcription, and after the second strand synthesis step, double-stranded cDNA was produced. Following repairing ends, A-tailing and adapter ligation steps, adapter- ligated fragments were ampliﬁed in enrichment polymerase chain reaction (PCR), and lastly, a double-stranded mRNA sequencing library was generated. Sequencing runs were performed on Illumina NextSeq500 instrument using single read, 75bp-long sequencing mode and generating 18–20 million sequencing reads for each sample. 3.4. Primary and Secondary Bioinformatics Analyses The quality of fastq ﬁles was evaluated by using FastQC v 0.11.7 software. After that, raw sequence reads were aligned to the reference genome (Vitis vinifera IGGP_12x genome version downloaded from the Ensembl Plants). STAR v2.5.4b bioinformatics tool was used for the mapping, and bam ﬁles were created. StrandNGS software (Agilent Technologies) was used for selecting lists of transcripts, which show differential mRNA abundance between conditions. Bam ﬁles of all individual biological samples were transferred, and raw expression data were normalized using the DESeq2 tool. To identify differentially expressed genes between healthy grape berry samples and various stages of noble rot, an ANOVA test with subsequent Tukey post hoc test and Benjamini-Hochberg FDR for multiple testing correction was used. Gene expression records that passed the following criterion were used: p ≤0.05 and −1 ≥Log2 fold change (FC) ≥1. Supplemental Table S1 contains all the grapevine gene loci discussed in our work, their exact transcript abundance fold change values between healthy grape berry samples and individual noble rot stages, as well as the ANOVA and Tukey p values for each transcript or pairwise comparison. RNA-seq raw data were deposited at ArrayExpress (E-MTAB-11205) and the European Nucleotide Archive (PRJEB48949). 3.2. RNA Extraction Total RNA sample quality was analyzed more precisely on an Agilent BioAnalyzer) using Eukaryotic Total RNA Nano and Pico Kit (Agilent Technologies, Palo Alto, CA, USA) according to the manufacturer’s protocol. Samples with RNA integrity number (RIN) value > 7 were accepted for the library preparation process. 3.8. Dry Weight Assessment of Healthy and Botrytized Berry Samples The dry weight of grape berry samples was determined by keeping 1 g of pulverized grape berry material in a laboratory oven at 105 ◦C for 1 h and weighing the mass of water loss at the end of the process. Three technical replicates were used for each stage of noble rot. 3.5. Validation of RNA-Seq Results RNA-Seq results were veriﬁed by real-time RT-PCR analysis of the expression of 5 grapevines and 5 B. cinerea genes as described before [93]. First-strand cDNA was synthesized with the First Strand cDNA Synthesis Kit (Thermo Fisher Scientiﬁc) and used as a template for real-time PCR analysis in ﬁve-fold dilution. Relative quantiﬁcation analysis was carried out using the comparative 2−∆∆Ct method [94]. To assess the level of gene expression, results were normalized using Ct values from the cDNA ampliﬁcation of the constitutively expressed V. vinifera actin gene VIT_08s0007g06520 [92] and B. cinerea Plants 2022, 11, 864 16 of 21 β-tubulin gene BCIN_01g08040 [95]. Results of RNA-Seq validation, primer sequences and Real-time RT-PCR parameters are shown in Supplemental Tables S2 and S3. β-tubulin gene BCIN_01g08040 [95]. Results of RNA-Seq validation, primer sequences and Real-time RT-PCR parameters are shown in Supplemental Tables S2 and S3. 3.6. Functional Annotation 3.7. UPLC-MS/MS Analysis of Abscisic Acid and Related Metabolites Extraction and chromatographic analyses of abscisic acid (ABA) and its related metabo- lites, namely phaseic acid (PA) and dihydrophaseic acid (DPA), were carried out according to Vrhovsek et al. [97] and Pál et al. [98] with slight modiﬁcations. g Grape samples were homogenized in liquid N2, stored at −80 ◦C until preparation, 0.2 g (F.W.) were weighted into 2 mL safety Eppendorf tubes containing two borosilicate inert glass beads (df = 3 mm). 100 µL water was added to stage III. noble rot samples to compensate for their low water content. Samples were spiked with 20 ng [2H6](+)-cis,trans- abscisic acid (OlChemIm s.r.o. Olomouc, Czech Republic) before extraction to serve as an internal standard. Extraction was carried out with 2 × 1.5 mL of methanol: water (2:1 v/v%) by vortex mixing for 10 s, followed by shaking for 3 min with 1250 rpm in a cryo-cooled rack using a miniG1600 (SPEX SamplePrep.; Metuchen, NJ, USA). After centrifugation (16,500× g; 4 ◦C; 10 min), supernatants were collected and pooled. To remove lipids and carotenoids, 1.2 mL of the supernatant was partitioned by adding 0.6 mL of n-hexane and vortexed for 2 × 10 s. Phases were separated by centrifugation (10,000× g; 4◦; 10 min) and the bottom phase of methanol: water was collected and ﬁltered through 0.22 µm PTFE syringe ﬁlters before analysis. After injecting 2 µL separation was achieved on a Waters (Milford, MA, USA) Acquity I-class UPLC equipped with an HSS T3 column (1.8 µm; 100 mm × 2.1 mm; Waters) at 40 ◦C in a 15 min water: acetonitrile (0.1 v/v% FA) gradient. For detection, a Waters Xevo TQ-XS system equipped with a UniSpray™source was utilized using negative polarity in MRM mode (further details to be found in Supplemental Table S4). All solvents used were UPLC grade and purchased from VWR (Radnor, PA, USA), reference solutions phaseic acid and dihydrophaseic acid were given by the Institute of Experimental Botany, Czech Academy of Science, while abscisic acid was purchased from the Merck-Sigma group (Darmstadt, Germany). 3.6. Functional Annotation Ensembl identiﬁers and transcript sequences of differentially expressed Vitis vinifera genes were used to collect functional information in KEGG, PFAM, GO, and TAIR databases as recently described [96]. Protein-protein interactions were predicted by the STRING v11 database. 4. Conclusions This study gives an insight into two segments of the transcriptome of Furmint berries during natural noble rot development. Exploring results of RNA-Seq analysis, redox- and plant hormone-associated mRNA responses have been highlighted. It has been pointed out that senescence- and stress hormone-connected metabolic pathways are generally activated during noble rot. This is particularly true for ABA, whose early and robust accumulation is probably a crucial molecular cue in the noble rot development of Furmint grape berries. This notion is corroborated by the prompt and uniform upregulation of ABA-metabolic and ABA-responsive transcripts, an extensive gene expression response that does not occur in berries exposed to bunch rot. On the contrary, pathways controlled by juvenile hormones appear either unaffected (cytokinins) or downregulated (auxins) by noble rot. Activation of various GST isoforms is also a consistent transcriptional change in berries affected by noble rot. Since noble rot is accompanied by an increase in the level of anthocyanin compounds, GST proteins may participate in transporting anthocyanins from the cytoplasm to the vacuole of plant cells [44,45]. Induction of the phenylpropanoid pathway seems to be a general B. cinerea interaction-related plant response (it is also typical of grape berries subjected to bunch rot) with some subtle noble rot-speciﬁc features such as the inhibition of chalcone synthase isoforms or Flavonol synthase 1. It is also accentuated that in some sectors of the grape berry antioxidative apparatus (e.g., two catalase isoforms, genes encoding carotenoid biosynthesis transcripts), the total antioxidant capacity and transcripts of an array of redox-associated proteins localized in the chloroplasts are consistently suppressed. These noble rot-speciﬁc inhibitions might be partly attributed to the apparent color change resulting from the accumulation of quinone and anthocyanin compounds [7,86]. Grape berries are plant organs that conduct photosynthesis [99,100], and extended darkness is known to downregulate transcripts of chloroplastic antioxidants [101]. g p p These ﬁndings contribute to the understanding of noble rot, a process vital to the production of botrytized dessert wines in the Tokaj wine region. Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/plants11070864/s1, Table S1: Transcript abundance fold change values for genes discussed in this work; Table S2: RNA-Seq validation: primer sequences and Real-time RT-PCR parameters; Table S3: Real-time RT-PCR veriﬁcation of RNA-Seq results; Table S4: Details of targeted abscisic acid, phaseic acid and dihydrophaseic acid analysis by UPLC-US-MS/MS method. Author Contributions: Conceptualization, M.P., L.K., K.Z.V. and B.B.; Methodology, T.D., J.H.-K. 3.10. Data Analysis 3.10. Data Analysis Experimental data of ABA, DPA, and total antioxidant quantiﬁcations were statistically analyzed by one-way ANOVA and subsequent LSD (Least Signiﬁcant Difference) test for pairwise comparisons. 3.9. Determination of Total Antioxidant Capacity of Grape Berries The total antioxidant capacity of grape berries was determined with an antioxidant assay kit (MAK334, Sigma-Aldrich, St. Louis, MO, USA) according to the manufacturer’s recommendations. During the assay, Cu2+ is reduced by antioxidants to Cu+, which speciﬁcally forms a colored complex with a dye reagent. The color intensity at 570 nm is proportional to the antioxidative capacity of the sample, which is expressed in µM Trolox equivalents. Grape berries were homogenized with liquid nitrogen in mortars, and the tissue suspensions were centrifuged (10,000× g, 10 min, 4 ◦C). Supernatants were diluted with distilled water (1:1), and 80 µL of diluted sample was mixed with 400 µL Reaction Mix. Following 10 min of incubation, the absorbance of solutions was measured at 570 nm. Absorbance values were compared to a calibration curve prepared with 300–1000 µM Trolox solutions. Plants 2022, 11, 864 17 of 21 17 of 21 References 1. Weiberg, A.; Wang, M.; Lin, F.M.; Zhao, H.; Zhang, Z.; Kaloshian, I.; Huang, H.-D.; Jin, H. Fungal small RNAs suppress plant immunity by hijacking host RNA interference pathways. Science 2013, 342, 118–123. y y j g p y 2. Kelloniemi, J.; Trouvelot, S.; Héloir, M.C.; Simon, A.; Dalmais, B.; Frettinger, P.; Cimerman, A.; Fermaud, M.; Roudet, J.; Baulande, S.; et al. Analysis of the molecular dialogue between gray mold (Botrytis cinerea) and grapevine (Vitis vinifera) reveals a clear shift in defense mechanisms during berry ripening. Mol. Plant-Microbe Interact. 2015, 28, 1167–1180. g y p g 3. Ribéreau-Gayon, P.; Dubourdieu, D.; Donèche, B.; Lovaud, A. The Handbook of Enology: The Microbiology of Wine and Viniﬁcations, 2nd ed.; Wiley: Chichester, UK, 2006; Volume 1. y 4. Vannini, A.; Chilosi, G. Botrytis infection: Grey mould and noble rot. In Sweet, Reinforced and Fortiﬁed Wines: Grape Biochemistry, Technology and Viniﬁcation, 1st ed.; Mencarelli, F., Tonutti, P., Eds.; Wiley: Chichester, UK, 2013; pp. 159–170. gy ﬁ y pp 5. Steel, C.C.; Blackman, J.W.; Schmidtke, L.M. Grapevine bunch rots: Impacts on wine compositio procedures for the removal of wine faults. J. Agric. Food Chem. 2013, 61, 5189–5206. an, J.W.; Schmidtke, L.M. Grapevine bunch rots: Impacts on wine composition, quality, and potential emoval of wine faults. J. Agric. Food Chem. 2013, 61, 5189–5206. 6. Dankó, T.; Szelényi, M.; Janda, T.; Molnár, B.P.; Pogány, M. Distinct volatile signatures of bunch rot and noble rot. Physiol. Mol. Plant Pathol. 2021, 114, 101626. 7. Blanco-Ulate, B.; Amrine, K.C.H.; Collins, T.S.; Rivero, R.M.; Vicente, A.R.; Morales-Cruz, A.; Doyle, C.L.; Ye, Z.; Allen, G.; Heymann, H.; et al. Developmental and metabolic plasticity of white-skinned grape berries in response to Botrytis cinerea during noble rot. Plant Physiol. 2015, 169, 2422–2443. y 8. Lovato, A.; Zenoni, S.; Tornielli, G.B.; Colombo, T.; Vandelle, E.; Polverari, A. Speciﬁc molecular inter and Botrytis cinerea are required for noble rot development in grape berries. Postharvest Biol. Techno y q p g p 9. Kuhn, N.; Guan, L.; Dai, Z.W.; Wu, B.H.; Lauvergeat, V.; Gomès, E.; Li, S.H.; Godoy, F.; Arce-Johnson, P.; Delrot, S. Berry ripening: Recently heard through the grapevine. J. Exp. Bot. 2013, 65, 4543–4559. [PubMed] 10. Coelho, J.; Almeida-Trapp, M.; Pimentel, D.; Soares, F.; Reis, P.; Rego, C.; Mithöfer, A.; Fortes, A.M. 4. Conclusions and E.K.-T.; Software, T.D., D.R.S. and B.K.; Validation, J.F. and G.G.; Formal Analysis, T.D. and D.R.S.; Investigation, T.D., E.K.-T., J.H.-K. and K.Á.H.; Resources, M.P. and K.Á.H.; Data Curation, T.D. and B.K.; Writing—Original Draft Preparation, M.P., J.F., G.G. and B.B.; Writing—Review & Editing, M.P., J.F., G.G. and B.B.; Visualization, T.D., M.P. and D.R.S.; Supervision, M.P. and K.Z.V.; Project Administration, M.P., L.K. and K.Z.V.; Funding Acquisition, L.K. and K.Z.V. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by the Széchenyi 2020 programme, the European Regional Develop- ment Fund, and the Hungarian Government (GINOP-2.3.2-15-2016-00061). K.Z.V. is supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences. Data Availability Statement: RNA-seq raw data were deposited at ArrayExpress (E-MTAB-11205) and the European Nucleotide Archive (PRJEB48949). Data Availability Statement: RNA-seq raw data were deposited at ArrayExpress (E-MTAB-11205) and the European Nucleotide Archive (PRJEB48949). Acknowledgments: The authors are grateful to István Szepsy (Szepsy Winery, Mád) for providing the grapevine biological material and sharing his expertise in viticulture and noble rot formation. Acknowledgments: The authors are grateful to István Szepsy (Szepsy Winery, Mád) for providing the grapevine biological material and sharing his expertise in viticulture and noble rot formation. Acknowledgments: The authors are grateful to István Szepsy (Szepsy Winery, Mád) for providing the grapevine biological material and sharing his expertise in viticulture and noble rot formation. 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. Plants 2022, 11, 864 18 of 21 18 of 21 References The study of hormonal metabolism of Trincadeira and Syrah cultivars indicates new roles of salicylic acid, jasmonates, ABA and IAA during grape ripening and upon infection with Botrytis cinerea. Plant Sci. 2019, 283, 266–277. [PubMed] p g p y [ ] 11. Jain, S.; Kumar, A. The pathogenesis related class 10 proteins in plant defense against biotic and abiotic stresses. Adv. Plants Agric. Res. 2015, 2, 305–314. 12. Loreti, E.; Povero, G.; Novi, G.; Solfanelli, C.; Alpi, A.; Perata, P. Gibberellins, jasmonate and abscisic acid modulate the sucrose-induced expression of anthocyanin biosynthetic genes in Arabidopsis. New Phytol. 2008, 179, 1004–1016. [PubMed] 13. Lackman, P.; González-Guzmán, M.; Tilleman, S.; Carqueijeiro, I.; Pérez, A.C.; Moses, T.; Seo, M.; Kanno, Y.; Häkkinen, S.T.; Van Montagu, M.C.E.; et al. Jasmonate signaling involves the abscisic acid receptor PYL4 to regulate metabolic reprogramming in Arabidopsis and tobacco. Proc. Natl. Acad. Sci. USA 2011, 108, 5891–5896. p 14. Hundertmark, M.; Hincha, D.K. LEA (late embryogenesis abundant) proteins and their encoding genes in Arabidopsis thaliana. BMC Genom. 2008, 9, 118. 15. Lim, J.; Lim, C.W.; Lee, S.C. The pepper Late Embryogenesis Abundant protein, CaDIL1, positively regulates drought tolerance and ABA signaling. Front. Plant Sci. 2018, 9, 1301. 16. Agudelo-Romero, P.; Erban, A.; Rego, C.; Carbonell-Bejerano, P.; Nascimento, T.; Sousa, L.; Martínez-Zapater, J.M.; Kopka, J.; Fortes, A.M. Transcriptome and metabolome reprogramming in Vitis vinifera cv. Trincadeira berries upon infection with Botrytis cinerea. J. Exp. Bot. 2015, 66, 1769–1785. p 17. Leonhardt, N.; Kwak, J.M.; Robert, N.; Waner, D.; Leonhardt, G.; Schroeder, J.I. Microarray expressio guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. Pl p 17. Leonhardt, N.; Kwak, J.M.; Robert, N.; Waner, D.; Leonhardt, G.; Schroeder, J.I. Microarray expression analyses of Arabidopsis guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. Plant Cell 2004, 16, 596–615. 18 West G M ; Pascal B D ; Ng L M ; Soon F F ; Melcher K ; Xu H E ; Chalmers M J ; Grifﬁn PR Protein conformation ensembles 17. Leonhardt, N.; Kwak, J.M.; Robert, N.; Waner, D.; Leonhardt, G.; Schroeder, J.I. Microarray expression analyses of Arabidopsis guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. Plant Cell 2004, 16, 596–615. 18. West, G.M.; Pascal, B.D.; Ng, L.M.; Soon, F.F.; Melcher, K.; Xu, H.E.; Chalmers, M.J.; Grifﬁn, P.R. References Protein conformation ensembles monitored by HDX reveal a structural rationale for abscisic acid signaling protein afﬁnities and activities. Structure 2013, 21, 229 235 , ; , J ; , ; , ; , ; , J y p y p guard cells and isolation of a recessive abscisic acid hypersensitive protein phosphatase 2C mutant. Plant Cell 2004, 16, 596–615. 18. West, G.M.; Pascal, B.D.; Ng, L.M.; Soon, F.F.; Melcher, K.; Xu, H.E.; Chalmers, M.J.; Grifﬁn, P.R. Protein conformation ensembles monitored by HDX reveal a structural rationale for abscisic acid signaling protein afﬁnities and activities. Structure 2013, 21, 229–235. 19. Costantini, V.; Bellincontro, A.; De Santis, D.; Botondi, R.; Mencarelli, F. Metabolic changes of Malvasia grapes for wine production during postharvest drying. J. Agric. Food Chem. 2006, 54, 3334–3340. [PubMed] 20. Li, Z.X.; Lan, J.B.; Liu, Y.Q.; Qi, L.W.; Tang, J.M. Investigation of the role of AcTPR2 in kiwifruit and its response to Botrytis cinerea infection. BMC Plant Biol. 2020, 20, 557. Ton, J.; Flors, V.; Mauch-Mani, B. The multifaceted ro 21. Ton, J.; Flors, V.; Mauch-Mani, B. The multifaceted role of ABA in disease resistance. Trends Plant Sci. 2009, 14, 310–317. [PubMed] 22. Audenaert, K.; De Meyer, G.B.; Höfte, M.M. Abscisic acid determines basal susceptibility of tomato to Botrytis cinerea and suppresses salicylic acid-dependent signaling mechanisms. Plant Physiol. 2002, 128, 491–501. 22. Audenaert, K.; De Meyer, G.B.; Höfte, M.M. Abscisic acid determines basal susceptibility of tomato to Botrytis cinerea and suppresses salicylic acid-dependent signaling mechanisms. Plant Physiol. 2002, 128, 491–501. pp y p g g y 23. AbuQamar, S.; Luo, H.; Laluk, K.; Mickelbart, M.V.; Mengiste, T. Crosstalk between biotic and abiotic stress responses in tomato is mediated by the AIM1 transcription factor. Plant J. 2009, 58, 347–360. [PubMed] 24. Thirugnanasambantham, K.; Durairaj, S.; Saravanan, S.; Karikalan, K.; Muralidaran, S.; Islam, V.I.H. Role of ethylene response transcription factor (ERF) and its regulation in response to stress encountered by plants. Plant Mol. Biol. Rep. 2015, 33, 347–357. 25 B hi ldi A At f A Ed i S G d ll N O R d A M H S M Al Att S G Al K d M A Al H j A S 24. Thirugnanasambantham, K.; Durairaj, S.; Saravanan, S.; Karikalan, K.; Muralidaran, S.; Islam, V.I.H. Role of ethylene response transcription factor (ERF) and its regulation in response to stress encountered by plants. Plant Mol. Biol. Rep. 2015, 33, 347–357. y 26. Zhang, G.; Zhao, F.; Chen, L.; Pan, Y.; Sun, L.; Bao, N.; Zhang, T.; Cui, C.X.; Qiu, Z.; Zhang, Y.; et al. Jas signalling promotes plant regeneration. Nat. Plants 2019, 5, 491–497. References Haemoglobin modulates NO emission and hyponasty under hypoxia-related stress in Arabidopsis thaliana. J. Exp. Bot. 2012, 63, 5581–5591. yp y yp p p 36. Weits, D.A.; Giuntoli, B.; Kosmacz, M.; Parlanti, S.; Hubberten, H.M.; Riegler, H.; Hoefgen, R.; Per Licausi, F. Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Na 36. Weits, D.A.; Giuntoli, B.; Kosmacz, M.; Parlanti, S.; Hubberten, H.M.; Riegler, H.; Hoefgen, R.; Perata, P.; Van Dongen, J.T.; Licausi, F. Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nat. Commun. 2014, 5, 3425. 37. Dolferus, R.; Jacobs, M.; Peacock, W.J.; Dennis, E.S. Differential interactions of promoter elements in stress responses of the Arabidopsis Adh gene. Plant Physiol. 1994, 105, 1075–1087. Licausi, F. Plant cysteine oxidases control the oxygen-dependent branch of the N-end-rule pathway. Nat. Commun. 2014, 5, 3425. 37. Dolferus, R.; Jacobs, M.; Peacock, W.J.; Dennis, E.S. Differential interactions of promoter elements in stress responses of the Arabidopsis Adh gene. Plant Physiol. 1994, 105, 1075–1087. p g y 38. Vlot, A.C.; Liu, P.P.; Cameron, R.K.; Park, S.W.; Yang, Y.; Kumar, D.; Zhou, F.; Padukkavidana, T.; Gustafsson, C.; Pichersky, E.; et al. Identiﬁcation of likely orthologs of tobacco salicylic acid-binding protein 2 and their role in systemic acquired resistance in Arabidopsis thaliana. Plant J. 2008, 56, 445–456. [PubMed] ck, J. The multi-protein family of Arabidopsis sulphotransferases and their relatives in other plant species. J 809–1820. [PubMed] 39. Klein, M.; Papenbrock, J. The multi-protein family of Arabidopsis sulphotransferases and their relative Exp. Bot. 2004, 55, 1809–1820. [PubMed] 40. Wagner, U.; Edwards, R.; Dixon, D.P.; Mauch, F. Probing the diversity of the Arabidopsis glutathione S-transferase gene family. Plant Mol. Biol. 2002, 49, 515–532. 41. Gullner, G.; K˝omíves, T.; Király, L.; Schröder, P. Glutathione S-transferase enzymes in plant-pathogen interactions. Front. Plant Sci. 2018, 9, 1836. 42. Kilili, K.G.; Atanassova, N.; Vardanyan, A.; Clatot, N.; Al-Sabarna, K.; Kanellopoulos, P.N.; Makris, A.M.; Kampranis, S.C. Differential roles of tau class glutathione S-transferases in oxidative stress. J. Biol. Chem. 2004, 279, 24540–24551. [PubMed] 43. Martínez-Márquez, A.; Martínez-Esteso, M.J.; Vilella-Antón, M.T.; Sellés-Marchart, S.; Morante-Carriel, J.A.; Hurtado, E.; Palazon, J.; Bru-Martínez, R. A tau class glutathione-S-transferase is involved in trans-resveratrol transport out of grapevine cells. Front. Plant Sci. 2017, 8, 1457. 44. Conn, S.; Curtin, C.; Bézier, A.; Franco, C.; Zhang, W. Puriﬁcation, molecular cloning, and characterization of glutathione S-transferases (GSTs) from pigmented Vitis vinifera L. References 25. Bahieldin, A.; Atef, A.; Edris, S.; Gadalla, N.O.; Ramadan, A.M.; Hassan, S.M.; Al Attas, S.G.; Al-Kordy, M.A.; Al-Hajar, A.S.; Sabir, J.S.; et al. Multifunctional activities of ERF109 as affected by salt stress in Arabidopsis. Sci. Rep. 2018, 8, 6403. , J ; y p p , , 26. Zhang, G.; Zhao, F.; Chen, L.; Pan, Y.; Sun, L.; Bao, N.; Zhang, T.; Cui, C.X.; Qiu, Z.; Zhang, Y.; et al. Jasmonate-mediated wound signalling promotes plant regeneration. Nat. Plants 2019, 5, 491–497. Plants 2022, 11, 864 19 of 21 19 of 21 27. Khandelwal, A.; Elvitigala, T.; Ghosh, B.; Quatrano, R.S. Arabidopsis transcriptome reveals control circuits regulating redox homeostasis and the role of an AP2 transcription factor. Plant Physiol. 2008, 148, 2050–2058. p y 28. Matsuo, M.; Johnson, J.M.; Hieno, A.; Tokizawa, M.; Nomoto, M.; Tada, Y.; Godfrey, R.; Obokata, J.; Sherameti, I.; Yamamoto, Y.Y.; et al. High Redox Responsive Transcription Factor1 levels result in accumulation of reactive oxygen species in Arabidopsis thaliana shoots and roots. Mol. Plant 2015, 8, 1253–1273. p 29. Mehterov, N.; Balazadeh, S.; Hille, J.; Toneva, V.; Mueller-Roeber, B.; Gechev, T. Oxidative stress provokes distinct transcriptional responses in the stress-tolerant atr7 and stress-sensitive loh2 Arabidopsis thaliana mutants as revealed by multi-parallel quantitative real-time PCR analysis of ROS marker and antioxidant genes. Plant Physiol. Biochem. 2012, 59, 20–29. 30. Yin, X.R.; Xie, X.L.; Xia, X.J.; Yu, J.Q.; Ferguson, I.B.; Giovannoni, J.J.; Chen, K.S. Involvement of an ethylene response factor in chlorophyll degradation during citrus fruit degreening. Plant J. 2016, 86, 403–412. 31. Park, H.Y.; Seok, H.Y.; Woo, D.H.; Lee, S.Y.; Tarte, V.N.; Lee, E.H.; Lee, C.H.; Moon, Y.H. AtERF71/H mediates osmotic stress response as well as hypoxia response in Arabidopsis. Biochem. Biophys. Res. Com p yp p p p y 32. Eysholdt-Derzsó, E.; Sauter, M. Root bending is antagonistically affected by hypoxia and ERF-mediated transcription via auxin signaling. Plant Physiol. 2017, 175, 412–423. g g y 33. White, M.D.; Kamps, J.J.; East, S.; Kearney, L.J.T.; Flashman, E. The plant cysteine oxidases from Arabidopsis thaliana are kinetically tailored to act as oxygen sensors. J. Biol. Chem. 2018, 293, 11786–11795. 34. Hebelstrup, K.; Jensen, E.Ø. Expression of NO scavenging hemoglobin is involved in the timing of bolting in Arabidopsis thaliana. Planta 2008, 227, 917–927. 35. Hebelstrup, K.H.; van Zanten, M.; Mandon, J.; Voesenek, L.A.; Harren, F.J.; Cristescu, S.M.; Møller, I.M.; Mur, L.A. References , ; , y g , , [ ] 63. Gould, K.S. Nature’s Swiss army knife: The diverse protective roles of anthocyanins in leaves. J. Biomed. Biotechnol. 2004, 5, 314–320. 64. González-Bosch, C. Priming plant resistance by activation of redox-sensitive genes. Free Radic. Biol. Med. 2018, 122, 171–180. [PubMed] 65. Chern, M.; Canlas, P.E.; Ronald, P.C. Strong suppression of systemic acquired resistance in Arabidopsis by NRR is dependent on its ability to interact with NPR1and its putative repression domain. Mol. Plant 2008, 1, 552–559. ability to interact with NPR1and its putative repression domain. Mol. Plant 2008, 1, 552–559. 66. Zwicker, S.; Mast, S.; Stos, V.; Pﬁtzner, A.J.; Pﬁtzner, U.M. Tobacco NIMIN2 proteins control PR gene induction through transient repression early in systemic acquired resistance Mol Plant Pathol 2007 8 385–400 [PubMed] 66. Zwicker, S.; Mast, S.; Stos, V.; Pﬁtzner, A.J.; Pﬁtzner, U.M. Tobacco NIMIN2 proteins control PR gene repression early in systemic acquired resistance. Mol. Plant Pathol. 2007, 8, 385–400. [PubMed] 67. Glazebrook, J. Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol. 2005, 43, 205–227. 68. Noshi, M.; Mori, D.; Tanabe, N.; Maruta, T.; Shigeoka, S. Arabidopsis clade IV TGA transcription factors, TGA10 and TGA9, are involved in ROS-mediated responses to bacterial PAMP ﬂg22. Plant Sci. 2016, 252, 12–21. [PubMed] 69. Wang, Z.; Xing, S.; Birkenbihl, R.P.; Zachgo, S. Conserved Functions of Arabidopsis and Rice CC-Type Glutaredoxins in Flower Development and Pathogen Response. Mol. Plant 2009, 2, 323–335. 70. Herrera-Vásquez, A.; Salinas, P.; Holuigue, L. Salicylic acid and reactive oxygen species interplay in the transcriptional control of defense genes expression. Front. Plant Sci. 2015, 6, 171. 71. Czarnocka, W.; Karpi´nski, S. Friend or foe? Reactive oxygen species production, scavenging and signa environmental stresses. Free Radic. Biol. Med. 2018, 122, 4–20. pi´nski, S. Friend or foe? Reactive oxygen species production, scavenging and signaling in plant response to ses. Free Radic. Biol. Med. 2018, 122, 4–20. 72. Coll, N.S.; Vercammen, D.; Smidler, A.; Clover, C.; Van Breusegem, F.; Dangl, J.L.; Epple, P. Arabidopsis type I metacaspases control cell death. Science 2010, 330, 1393–1397. [PubMed] [ ] 73. Coll, N.S.; Smidler, A.; Puigvert, M.; Popa, C.; Valls, M.; Dangl, J.L. The plant metacaspase AtMC1 in pathogen-triggered programmed cell death and aging: Functional linkage with autophagy. Cell Death Differ. 2014, 21, 1399–1408. 74. Glaring, M.A.; Skryhan, K.; Kötting, O.; Zeeman, S.C.; Blennow, A. References cell suspension cultures as putative anthocyanin transport proteins. J. Exp. Bot. 2008, 59, 3621–3634. 45. Pérez-Díaz, R.; Madrid-Espinoza, J.; Salinas-Cornejo, J.; González-Villanueva, E.; Ruiz-Lara, S. Differential roles for VviGST1, VviGST3, and VviGST4 in proanthocyanidin and anthocyanin transport in Vitis vinifera. Front. Plant Sci. 2016, 7, 1166. [PubMed] 46. Lallement, P.A.; Meux, E.; Gualberto, J.M.; Prosper, P.; Didierjean, C.; Saul, F.; Haouz, A.; Rouhier, N.; Hecker, A. Structural and enzymatic insights into Lambda glutathione transferases from Populus trichocarpa, monomeric enzymes constituting an early divergent class speciﬁc to terrestrial plants. Biochem. J. 2014, 462, 39–52. [PubMed] g p p 47. Moraga, A.R.; Nohales, P.F.; Pérez, J.A.; Gómez-Gómez, L. Glucosylation of the saffron apocarotenoid crocetin by a glucosyltrans- ferase isolated from Crocus sativus stigmas. Planta 2004, 219, 955–966. [PubMed] 47. Moraga, A.R.; Nohales, P.F.; Pérez, J.A.; Gómez-Gómez, L. Glucosylation of the saffron apo ferase isolated from Crocus sativus stigmas. Planta 2004, 219, 955–966. [PubMed] g 48. Marcheafave, G.G.; Tormena, C.D.; Pauli, E.D.; Rakocevic, M.; Bruns, R.E.; Scarminio, I.S. Experimental mixture design solvent effects on pigment extraction and antioxidant activity from Coffea arabica L. leaves. Microchem. J. 2019, 146, 713–721. ave, G.G.; Tormena, C.D.; Pauli, E.D.; Rakocevic, M.; Bruns, R.E.; Scarminio, I.S. Experimental mixture desi p g y ff J 49. Milani, A.; Basirnejad, M.; Shahbazi, S.; Bolhassani, A. Carotenoids: Biochemistry, pharmacology and treatment. Br. J. Pharmacol. 2017, 174, 1290–1324. 50. Hashemi, M.; Hosseinzadeh, H. A comprehensive review on biological activities and toxicology of crocetin. Food Chem. Toxicol. 2019, 130, 44–60. [PubMed] , E.; Pollastri, S.; Tattini, M. Flavonoids as antioxidants in plants: Location and functional signiﬁcance. Plant . [PubMed] 51. Agati, G.; Azzarello, E.; Pollastri, S.; Tattini, M. Flavonoids as antioxidants in plants: Location and func Sci. 2012, 196, 67–76. [PubMed] 52. Atanasova-Penichon, V.; Barreau, C.; Richard-Forget, F. Antioxidant secondary metabolites in cereals: Potential involvement in resistance to Fusarium and mycotoxin accumulation. Front. Microbiol. 2016, 7, 566. [PubMed] Plants 2022, 11, 864 20 of 21 20 of 21 53. Sharma, A.; Shahzad, B.; Rehman, A.; Bhardwaj, R.; Landi, M.; Zheng, B. Response of phenylpropanoid pathway and the role of polyphenols in plants under abiotic stress. Molecules 2019, 24, 2452. p yp p 54. Vannozzi, A.; Dry, I.B.; Fasoli, M.; Zenoni, S.; Lucchin, M. Genome-wide analysis of the grapevine stilbene synthase multigenic family: Genomic organization and expression proﬁles upon biotic and abiotic stresses. BMC Plant Biol. 2012, 12, 130. References y g p p p 55. Owens, D.K.; Alerding, A.B.; Crosby, K.C.; Bandara, A.B.; Westwood, J.H.; Winkel, B.S. Functional analysis of a predicted ﬂavonol synthase gene family in Arabidopsis. Plant Physiol. 2008, 147, 1046–1061. [PubMed] y g y p y 56. Caarls, L.; Elberse, J.; Awwanah, M.; Ludwig, N.R.; De Vries, M.; Zeilmaker, T.; Van Wees, S.C.; Schuurink, R.C.; Van den Ackerveken, G. Arabidopsis JASMONATE-INDUCED OXYGENASES down-regulate plant immunity by hydrox- ylation and inactivation of the hormone jasmonic acid. Proc. Natl. Acad. Sci. USA 2017, 114, 6388–6393. [PubMed] 56. Caarls, L.; Elberse, J.; Awwanah, M.; Ludwig, N.R.; De Vries, M.; Zeilmaker, T.; Van Wees, S.C.; Schuurink, R.C.; Van den Ackerveken, G. Arabidopsis JASMONATE-INDUCED OXYGENASES down-regulate plant immunity by hydrox- p g p n and inactivation of the hormone jasmonic acid. Proc. Natl. Acad. Sci. USA 2017, 114, 6388–6393. [PubMed ylation and inactivation of the hormone jasmonic acid. Proc. Natl. Acad. Sci. USA 2017, 114, 6388–6393. [PubMed] 57. Smirnova, E.; Marquis, V.; Poirier, L.; Aubert, Y.; Zumsteg, J.; Ménard, R.; Miesch, L.; Heitz, T. Jasmonic acid oxidase 2 hydroxylates jasmonic acid and represses basal defense and resistance responses against Botrytis cinerea infection. Mol. Plant 2017, 10, 1159–1173. 57. Smirnova, E.; Marquis, V.; Poirier, L.; Aubert, Y.; Zumsteg, J.; Ménard, R.; Miesch, L.; Heitz, T. Jasmonic acid oxidase 2 hydroxylates jasmonic acid and represses basal defense and resistance responses against Botrytis cinerea infection. Mol. Plant 2017, 10, 1159–1173. [PubMed] 58. Deavours, B.E.; Dixon, R.A. Metabolic engineering of isoﬂavonoid biosynthesis in alfalfa. Plant Physiol. 2005, 138, 2245–2259. [PubMed] 59. Beltrán, J.; Kloss, B.; Hosler, J.P.; Geng, J.; Liu, A.; Modi, A.; Dawson, J.H.; Sono, M.; Shumskaya, M.; Ampomah-Dwamena, C.; et al. Control of carotenoid biosynthesis through a heme-based cis-trans isomerase. Nat. Chem. Biol. 2015, 11, 598–605. 60. Feild, T.S.; Lee, D.W.; Holbrook, N.M. Why leaves turn red in autumn. The role of anthocyanins in senescing leaves of red-osier dogwood. Plant Physiol. 2001, 127, 566–574. [PubMed] g y 61. Schaberg, P.G.; Van den Berg, A.K.; Murakami, P.F.; Shane, J.B.; Donnelly, J.R. Factors inﬂuencing red expression in autumn foliage of sugar maple trees. Tree Physiol. 2003, 23, 325–333. [PubMed] 62. Neill, S.O.; Gould, K.S. Anthocyanins in leaves: Light attenuators or antioxidants? Funct. Plant Biol. 2003, 30, 865–873. [PubMed] 63. Gould, K.S. Nature’s Swiss army knife: The diverse protective roles of anthocyanins in leaves. J. Biomed. Biotechnol. 2004, 5, 314–320. p 80. Hall, M.; Mata-Cabana, A.; Åkerlund, H.E.; Florencio, F.J.; Schröder, W.P.; Lindahl, M.; Kieselbach, T. Thioredoxin targets of the plant chloroplast lumen and their implications for plastid function. Proteomics 2010, 10, 987–1001. References Comprehensive survey of redox sensitive starch metabolising enzymes in Arabidopsis thaliana. Plant Physiol. Biochem. 2012, 58, 89–97. [PubMed] 75. Ströher, E.; Dietz, K.J. The dynamic thiol–disulphide redox proteome of the Arabidopsis thaliana chloroplast as revealed by differential electrophoretic mobility. Physiol. Plant. 2008, 133, 566–583. 76. Lindahl, M.; Kieselbach, T. Disulphide proteomes and interactions with thioredoxin on the track towards understanding redox regulation in chloroplasts and cyanobacteria. J. Proteom. 2009, 72, 416–438. 77. Muthuramalingam, M.; Seidel, T.; Laxa, M.; De Miranda, S.M.; Gärtner, F.; Ströher, E.; Kandlbinder, A.; Dietz, K.J. Multiple redox and non-redox interactions deﬁne 2-Cys peroxiredoxin as a regulatory hub in the chloroplast. Mol. Plant 2009, 2, 1273–1288. [PubMed] 78. Seung, D.; Thalmann, M.; Sparla, F.; Abou Hachem, M.; Lee, S.K.; Issakidis-Bourguet, E.; Svensson, B.; Zeeman, S.C.; Santelia, D. Arabidopsis thaliana AMY3 is a unique redox-regulated chloroplastic α-amylase. J. Biol. Chem. 2013, 288, 33620–33633. [PubMed] 79. Muthuramalingam, M.; Matros, A.; Scheibe, R.; Mock, H.P.; Dietz, K.J. The hydrogen peroxide-sensitive proteome of the chloroplast in vitro and in vivo. Front. Plant Sci. 2013, 4, 54. [PubMed] 78. Seung, D.; Thalmann, M.; Sparla, F.; Abou Hachem, M.; Lee, S.K.; Issakidis-Bourguet, E.; Svensson, B.; Zeeman, S.C.; Santelia, D. Arabidopsis thaliana AMY3 is a unique redox-regulated chloroplastic α-amylase. J. Biol. Chem. 2013, 288, 33620–33633. [PubMed] 79 M h li M M A S h ib R M k H P Di K J Th h d id i i f h 80. Hall, M.; Mata-Cabana, A.; Åkerlund, H.E.; Florencio, F.J.; Schröder, W.P.; Lindahl, M.; Kieselbach, T. Thioredoxin targets of the plant chloroplast lumen and their implications for plastid function. Proteomics 2010, 10, 987–1001. Plants 2022, 11, 864 21 of 21 21 of 21 81. Marchand, C.; Le Maréchal, P.; Meyer, Y.; Decottignies, P. Comparative proteomic approaches for the isolation of proteins interacting with thioredoxin. Proteomics 2006, 6, 6528–6537. [PubMed] g 82. Dumont, S.; Bykova, N.V.; Khaou, A.; Besserour, Y.; Dorval, M.; Rivoal, J. Arabidopsis thaliana alcohol dehydrogenase is differently affected by several redox modiﬁcations. PLoS ONE 2018, 13, e0204530. y 83. Wang, H.; Wang, S.; Lu, Y.; Alvarez, S.; Hicks, L.M.; Ge, X.; Xia, Y. Proteomic analysis of early-responsive redox-sensitive proteins in Arabidopsis. J. Proteome Res. 2012, 11, 412–424. [PubMed] 84. Shi, H.; Liu, W.; Yao, Y.; Wei, Y.; Chan, Z. Alcohol dehydrogenase 1 (ADH1) confers both abiotic and biotic stress resistance in Arabidopsis. Plant Sci. 2017, 262, 24–31. 85. References Szklarczyk, D.; Gable, A.L.; Lyon, D.; Junge, A.; Wyder, S.; Huerta-Cepas, J.; Simonovic, M.; Doncheva, N.T.; Morris, J.H.; Bork, P.; et al. STRING v11: Protein–protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 2019, 47, D607–D613. [PubMed] 86. Magyar, I. Botrytized wines. In Advances in Food and Nutrition Research; Jackson, R.S., Ed.; Academic Press: Burlington, NJ, USA, 2011; pp. 147–206. pp 87. Scarpeci, T.E.; Zanor, M.I.; Carrillo, N.; Mueller-Roeber, B.; Valle, E.M. Generation of superoxide anion in chloroplasts of Arabidopsis thaliana during active photosynthesis: A focus on rapidly induced genes. Plant Mol. Biol. 2008, 66, 361–378. [PubMed] 88. Pandey, S.P.; Roccaro, M.; Schön, M.; Logemann, E.; Somssich, I.E. Transcriptional reprogramming regulated by WRKY18 and WRKY40 facilitates powdery mildew infection of Arabidopsis Plant J 2010 64 912–923 [PubMed] pp 87. Scarpeci, T.E.; Zanor, M.I.; Carrillo, N.; Mueller-Roeber, B.; Valle, E.M. Generation of superoxide anion in chloroplasts of Arabidopsis thaliana during active photosynthesis: A focus on rapidly induced genes. Plant Mol. Biol. 2008, 66, 361–378. [PubMed] 88 P d S P R M S hö M L E S i h I E T i ti l i l t d b WRKY18 d p g p y p y g 88. Pandey, S.P.; Roccaro, M.; Schön, M.; Logemann, E.; Somssich, I.E. Transcriptional reprogramming WRKY40 facilitates powdery mildew infection of Arabidopsis. Plant J. 2010, 64, 912–923. [PubMed] p y p J [ ] 89. Ding, Z.J.; Yan, J.Y.; Xu, X.Y.; Yu, D.Q.; Li, G.X.; Zhang, S.Q.; Zheng, S.J. Transcription factor WRKY 46 regulates osmotic stress responses and stomatal movement independently in Arabidopsis. Plant J. 2014, 79, 13–27. [PubMed] 90. Ding, Z.J.; Yan, J.Y.; Li, C.X.; Li, G.X.; Wu, Y.R.; Zheng, S.J. Transcription factor WRKY 46 modulates the development of Arabidopsis lateral roots in osmotic/salt stress conditions via regulation of ABA signaling and auxin homeostasis. Plant J. 2015, 84, 56–69. [PubMed] [ ] 91. Chen, H.; Lai, Z.; Shi, J.; Xiao, Y.; Chen, Z.; Xu, X. Roles of Arabidopsis WRKY18, WRKY40 and WRKY60 transcription factors in plant responses to abscisic acid and abiotic stress. BMC Plant Biol. 2010, 10, 281. p p 92. Reid, K.E.; Olsson, N.; Schlosser, J.; Peng, F.; Lund, S.T. An optimized grapevine RNA isolation procedure and statistical determination of reference genes for real-time RT-PCR during berry development. BMC Plant Biol. 2006, 6, 27. 93. References Kámán-Tóth, E.; Dankó, T.; Gullner, G.; Bozsó, Z.; Palkovics, L.; Pogány, M. Contribution of cell wall peroxidase-and NADPH oxidase-derived reactive oxygen species to Alternaria brassicicola-induced oxidative burst in Arabidopsis. Mol. Plant Pathol. 2019, 20, 485–499. [PubMed] 94. Livak, K.J.; Schmittgen, T.D. Analysis of relative gene expression data using real-time quantitative PCR and the 2−∆∆CT method. Methods 2001, 25, 402–408. 95. Ren, H.; Wu, X.; Lyu, Y.; Zhou, H.; Xie, X.; Zhang, X.; Yang, H. Selection of reliable reference genes for gene expression studies in Botrytis cinerea. J. Microbiol. Methods 2017, 142, 71–75. 96. Kalapos, B.; Juhász, C.; Balogh, E.; Kocsy, G.; Tóbiás, I.; Gullner, G. Transcriptome proﬁling of pepper leaves by RNA-Seq during an incompatible and a compatible pepper-tobamovirus interaction. Sci. Rep. 2021, 11, 20680. 97. Vrhovsek, U.; Masuero, D.; Gasperotti, M.; Franceschi, P.; Caputi, L.; Viola, R.; Mattivi, F. A versatile targeted metabolomics method for the rapid quantiﬁcation of multiple classes of phenolics in fruits and beverages. J. Agric. Food Chem. 2012, 60, 8831–8840. [PubMed] 98. Pál, M.; Ivanovska, B.; Oláh, T.; Tajti, J.; Hamow, K.Á.; Szalai, G.; Khalil, R.; Vanková, R.; Dobrev, P.; Misheva, S.P.; et al. Role of polyamines in plant growth regulation of Rht wheat mutants. Plant Physiol. Biochem. 2019, 137, 189–202. [PubMed] 99 Pandey R M ; Farmahan H L Changes in the rate of photosynthesis and respiration in leaves and berries of Vitis vinifera 98. Pál, M.; Ivanovska, B.; Oláh, T.; Tajti, J.; Hamow, K.Á.; Szalai, G.; Khalil, R.; Vanková, R.; Dobrev, P.; Misheva, S.P.; et al. Role of polyamines in plant growth regulation of Rht wheat mutants. Plant Physiol. Biochem. 2019, 137, 189–202. [PubMed] 99. Pandey, R.M.; Farmahan, H.L. Changes in the rate of photosynthesis and respiration in leaves and berries of Vitis vinifera grapevines at various stages of berry development. Vitis 1977, 16, 106–111. Á p y p g g y 99. Pandey, R.M.; Farmahan, H.L. Changes in the rate of photosynthesis and respiration in leaves and berries of Vitis vinifera grapevines at various stages of berry development. Vitis 1977, 16, 106–111. Á g p g y p 100. Martínez-Esteso, M.J.; Vilella-Antón, M.T.; Pedreño, M.Á.; Valero, M.L.; Bru-Martínez, R. iTRAQ-based p insights into the central metabolism changes driving grape berry development and ripening. BMC Pla 100. Martínez-Esteso, M.J.; Vilella-Antón, M.T.; Pedreño, M.Á.; Valero, M.L.; Bru-Martínez, R. iTRAQ-based protein proﬁling provides insights into the central metabolism changes driving grape berry development and ripening. References BMC Plant Biol. 2013, 13, 167. 101. Rosenwasser, S.; Rot, I.; Sollner, E.; Meyer, A.J.; Smith, Y.; Leviatan, N.; Fluhr, R.; Friedman, H. Organelles contribute differentially to reactive oxygen species-related events during extended darkness. Plant Physiol. 2011, 156, 185–201. [PubMed] insights into the central metabolism changes driving grape berry development and ripening. BMC Plant Biol. 2013, 13, 167. 101. Rosenwasser, S.; Rot, I.; Sollner, E.; Meyer, A.J.; Smith, Y.; Leviatan, N.; Fluhr, R.; Friedman, H. Organelles contribute differentially to reactive oxygen species-related events during extended darkness. Plant Physiol. 2011, 156, 185–201. [PubMed]
https://openalex.org/W301255261	https://hal.inria.fr/hal-01387261/file/978-3-662-44733-8_38_Chapter.pdf	English	null	Slack Resource as Evolutionary Determinant of International Manufacturing Joint Venture’s Growth Performance	Lecture notes in computer science	2,014	cc-by	3,843	To cite this version: Stig B. Taps, Thomas Ditlev Brunø, Kjeld Nielsen, Kaj A. Joergensen. Slack Resource as Evolutionary Determinant of International Manufacturing Joint Venture’s Growth Performance. IFIP International Conference on Advances in Production Management Systems (APMS), Sep 2014, Ajaccio, France. pp.304-312, 10.1007/978-3-662-44733-8_38. hal-01387261 Slack Resource as Evolutionary Determinant of International Manufacturing Joint Venture’s Growth Performance Stig B. Taps, Thomas Ditlev Brunø, Kjeld Nielsen, Kaj A. Joergensen To cite this version: Stig B. Taps, Thomas Ditlev Brunø, Kjeld Nielsen, Kaj A. Joergensen. Slack Resource as Evolutionary Determinant of International Manufacturing Joint Venture’s Growth Performance. IFIP International Conference on Advances in Production Management Systems (APMS), Sep 2014, Ajaccio, France. pp.304-312, 10.1007/978-3-662-44733-8_38. hal-01387261 Distributed under a Creative Commons Attribution 4.0 International License Slack Resource as Evolutionary Determinant of International Manufacturing Joint Venture's Growth Performance. Stig B. Taps, Thomas D. Brunoe, Kjeld Nielsen and Kaj A. Joergensen Department of Mechanical and Manufacturing Engineering, Aalborg University, Denmark taps@m-tech.aau.dk Abstract. At present time in a world of change our understanding are constantly challenged with new knowledge that radically makes demand on prevailing theories and concept the scientific world for long have believed was unchangeable. This article makes a back flash on innovation performance with some suggestion on how International Manufacturing Joint Ventures can enhance their learning capacity. Manufacturing in a global world is a complicated management challenge because the supply chain is located in different national settings. Networkning and knowledge absorption is a key to get international competitiveness. This paper claims a framework for building absorption capacity in international manufacturing firms. Keywords: Growth performance. Innovation conceptualization. Slack re- sources. Absorptive capacity HAL Id: hal-01387261 https://inria.hal.science/hal-01387261v1 Submitted on 25 Oct 2016 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 1 Introduction International Manufacturing Joint Ventures (IMJV) is an organizational entity, con- structed by two or more firms of different nationality [11], pooling a portion of their resources within a common legal organization [12]. These resources comprise firm- specific knowledge, and the knowledge base created in the entity is affected by each partner’s sharing of knowledge with the other (Buckley, 1996). Productive resources of human technologies and mechanical technologies [5] are bundled from different sources, i.e., from partner transfer, by acquisition, and by engagement. Physical facili- ties are acquired, and technological equipment arrives, and is installed according to the focus. Managers, technical experts and labor are engaged, bundled together in a firm without a history by the simple fact that the IMJV is a new firm construct, a new 'child', to stand alone, and seeking to its own performance s through efficiency of combining those diverse bundling of resources. The IMJV is a synthetic organization [25], simultaneously establishing a structure and carry on operation. People have fragmented knowledge on parts of the focus, and the knowledge technology embed- ded in people is diffused and distinct, and acquired in working experiences from dif- ferent organizational settings, maybe too, from other geographical locations outside the IMJV. Taking actions and making decisions within this newly structured 'child' are not supported by common understandings of actors or by common social and technological experiences. g p How can such a synthetic construct growth? Managers' must transform the IMJV from a synthetic organization through a pro- cess implying entrenched behaviors to stabilize organizational continuity whether as deliberate plans to establish patterns of behavior or as emergent patterns by which such patterns get established [16]. Accordingly the transformation process can be one of designing formal planning systems to imply entrenched organizational behavior through standardization of work, skills or outputs; and by focusing on collective so- cialization and individual cognition to establish an emergent pattern of coherence in behavior. Inter-correlation drive different parts in organization to coherence among their parts by means of manager's “tuning” of coordination mechanisms e.g., their choice among mechanisms to entrench behaviors in desired outcome and synergistic complementation between mechanisms. Further, this dynamic capability approach acknowledge that resource endowments are “sticky”, firms are to some degree stuck with what they have and may have to live with what they lack [24]. 1 Introduction Features as bundling of resources, structural form of IMJV , appointment of managers to various departmental responsibility, and strategic direc- tion are commended contractual agreements among partners' founding conditions and therefore sticky for managerial actions within IMJV. Such imprinted stickiness makes demand on the ability of an IMJV to grow and innovate. Conditions could be barriers to transfer best practice inside the firm e.g., difficulty to transfer knowledge among subunits, and be hindering to integration and operational streamlining of practices, structures and systems. The paper is organized in the following way. First, grow performance is elabo- rated from an innovative perspective and it is argued that building organizational slack resources to handle uncertainty influence firm's innovative performance. Second, a constructive approach for research in management capabilities to ex- ploit resources for innovative purpose is argued. 2 Innovative performance Innovation has been conceptualized in many ways and studied from several perspec- tives. For instance, distinctions are made between studies of adoption and diffusion of innovations and at different levels of analysis: individual, organizations, industry and nations. 2.1 Economists Economists conceptualize innovation at a high level of aggregation or abstraction, seeing it as one of the factors that cause increased productivity and economic growth at the industry or even national level [7]. According to Urabe [26]: Innovation consists of the generation of a new idea and its implementation into new product, process or service, leading to the dynamic growth of the national economy and the increase of employment as well as to creation of pure profit for the innovative business enterprise. Firm is innovative only when 'newness' is transformed to economic benefit to soci- ety. Firm’s innovativeness and rent generation are the outcome of the analytical fit between structure-conduct-performance of firms [2]. But when there are rapid chang- es in technology and market forces and defensible feedback effects from competitors such a monopoly rent generating strategy is difficult to defend. The growing uncer- tainty in today’s turbulent competitive environment together with customers changing and unpredictable demand after products and services, accelerating pace of technolog- ical change affecting newness in manufacturing processes. For example, Pine claims [20] that our understanding of the market and the marketing concept is a misunder- standing because the market only exists as an individualist. Understanding customers idiosyncratic needs by development of a solution space makes increasing demand on robust process design [17]. 2.2 Technologists By technology researchers innovation is defined as a continuous process during which the innovation changes form as it is used and assimilated [21]. The processes of gen- erating new technology and improving upon existing technology concern technolo- gists and they study both product and process innovations from this perspective. Some technologists study the nature and dynamics of technological change within the con- text of the industry and attempt to understand the relationship between technological changes at the industry level and firm-level adaptations to these changes [1]. Other technologists researcher are concerned with factors that influence the generation of technological innovation at the level of organizational subunits such as the R&D or the manufacturing departments. They focus on factors that improve technical performances and criteria that influence the choice and use of technological innovations in various organizational subunits. Thus, both generation factors and adoption phases of the innovation process within subunits is under study. For example, Salvador [22] proposes a process model ap- proach based on three fundamental organizational capabilities: Solution Space devel- opment, Robust Process Design, and Choice Navigation that constitutes radical inno- vation to crack the code of mass customization. Wheelwright & Clark [28] argue that problem s in new product and new process development arises because of mismatch between anticipated benefits in resource utilization, organizational renewal and mar- ket position with strong defensible competitive barriers. They find causes for these mismatches from organization's analytical capability to alignment of external reality, and from lack of communication among functions within an organization and from locking into product and processes without innovative uniqueness and from managers fail to plan sufficiently. 2.3 Sociologists Also sociologists research in innovation at the focus on the organizational features that are compatible with the adoption of innovation within the organizations. One definition of innovation within this school is that innovation is adoption of an inter- nally generated or purchased device, system, policy, program, process, product, or service that is new to the adopting organization [6]. This school studies factors (as technologists) but factors as the importance of dis- tinct contextual, structural and individual features that explain variance in innovation adoption at the level of organization. Hage [10]claim that complexity of the division of labor is critical for organizational innovation because the division would lead to much greater adaptiveness or flexibility of the intellectual or problem-salving capaci- ties or learning capacities of the organization, what Cohen & Levintahl [4] refer to as absorptive capacity. From the above discussion innovation is conceptualized in different ways. Innova- tion is seen as a product; a new idea; method or device and as a process of introducing something new, newness either as technology or as organization. Organization can either be generator or adopter on innovation. Organization as generator of innovation is linked to the organization's problem solving and decision making capability to re- move ambiguities between successive stages of the innovation process: idea genera- tion, project definition, problem-solving, design and development, and commerciali- zation of new product s and processes. 3 The level and type of analysis The level of analysis is conceptualized differently across four levels: industry, organi- zation, subunits and the innovation itself. At the industry level researcher are con- cerned with factors that distinguish innovation development patterns among industries such as technological appropriateness to the stages of the industry life-cycle. Research at the organizational level seeks to identify contextual, structural and behavioral char- acteristics that differentiate innovative from non-innovative organizations. They try to find group of organizational characteristics which relative more efficiently determine the outcome of organizational innovation such as its adaptability to the environment, capacity to allocate resources to innovative programs and overall performances. At the subunit level researcher try to identify and analyze factors that either facilitate or hinder innovations within a division or a department. The most widely studied organi- zational subunits are R&D departments. The type of innovation under investigation also differs. Process innovations are defined as tools, devices, and knowledge in throughput technology that mediate between inputs and outputs and are new to an industry, organization or subunit. Product innovations, in contrast, are outputs or services that are developed to the benefit of customers. Type of innovation is also conceptualized as either radical or incremental. Rad- ical innovations changes fundamentally existing practices within an organization or an industry into new practices on how to make activities around in transforming competences to new standards. Radical innovations give rise to new business pos- sibilities, new strategies and new structures under increased environmental uncer- tainty. Incremental innovation improves the old product or process within the existing structure and strategy and mainly reinforces the existing capabilities of the organization's practices. Type of innovation can be viewed as either technical or social. This distinction reflects a more general distinction between technical innovation and administra- tive innovation. Technical innovation includes products, processes and technolo- gies used to produce product s or render services related to the technical core of an organization. Administrative innovation is related to management processes in designing organizational structure, administrative processes and human resources. Differences in conceptualization of innovation among economists, technologists and sociologists are summarized in table 1. Each group is only addressing a limited aspect of innovation complexity whenever the position is focused and broad defini- tion, and the product and process view s and the radical or incremental perspective. 3 The level and type of analysis At the introduction we argued that dynamic capabilities are the key role in adop- tion, integrating and configuring internal and external organizational skills, resources and functional competences and are important determinants for IMJV's growth per- formance. Therefore in the next section the focus is on the organizational level of analysis. The insights from other levels of analysis are gained to broaden the follow- ing argumentation about slack resource availability in organization as important de- terminants for IMJV's innovation. 4 Slack resources, absorptive capacity to innovation In the following section we shall shortly argue a complete theoretical framework we propose to be. 4 Slack resources, absorptive capacity to innovation The slack serves as a buffer between organization and environmental discontinuities and provides the organization with the internal capability to invest for innovation [8]. Evans[9] has labeled innovative capability as strategic flexibility keeping the organi- zation both agile and versatile. Bugelman [1] argue that the rate which mutant ideas were pursued seemed to depend on the amount of slack resources available at the operational level. To keep it simple, we define slack as the pool of resources in an Table 1. Differences in conceptualization of innovation among economists, technologists and sociologist (Based on [1]) Table 1. Differences in conceptualization of innovation among economists, technologists and sociologist (Based on [1]) organization that is in excess of the minimum necessary to produce a level of organi- zational output [18]. The excess of absorptive capacity an organization on a given time has over the capacity to run the daily operation is the resource available to inno- vative performance. A firm's absorptive capacity is defined as the ability prior related knowledge con- fers to recognize the value of new information, assimilate it, and apply it to commer- cial ends [4]. These abilities, collectively, constitute what they call a firm's absorptive capacity. They view absorptive capacity, at the firm level construct, as an ability the firm develops over time by accumulating a base of knowledge. The firm absorptive capacity is generated in a variety of ways: by product of a firm's R&D activities; by production experience, reorganize or automate particular manufacturing processes; and by investment in absorptive capacity directly, when sending personal for tech- nical training. How manager design the structural features of the institutional, managerial, and technical level [19]of a firm, and how they coordinate interrelated activities to a co- herent whole, have great influence on a firm's absorptive capacity. This design pro- cess is complicated by the resource bundling, in so far as, an IMJV is a new “child”, with people without experience in working together, and with prior knowledge ac- quired in variety of different organizational contexts. Knowledge may be nominally acquired, but not well utilized, because the individual do not possess the appropriate contextual knowledge necessary to make new knowledge full y intelligible [14]. Therefore, firm’s absorptive capacity is not simply the sum of the absorptive capaci- ties of its employees, but is dependent on the firm's capacity to transfer knowledge across and within subunits. 5 Theoretical framework. Bundling of complex, heterogeneous resources, and capabilities of organizations that are usually dissimilar in culture, size, industry, an often country of origin present a managerial challenge. Because of partners' bundling of complementary resources, management within IMJVs typically lacks administrative tools. The organization is operated by people, who do not have a common tradition, and the organization does not form an entity. Lack of planning systems or blueprints force management to es- tablish new structures and simultaneously carry on operations. Uncertainty is high on, how to manage an IMJV to a coherent whole entity. Because of the bundling of re- sources, complexity and differentiation increases, as well as in the risk of conflicts, opportunistic behavior, and competing goals between subunits. As a result manage- ment is confronted by the increasingly crucial need to monitor, coordinate, and inte- grate the activities of the IMJV [3], [15]. A structuration process [27] develops knowledge technologies e.g., the meaning and concepts to use mechanical technolo- gies and human technologies in production [5], by which uncertainties are trans- formed to relatively certainties of how to act. Structuration suggest, then, that systems are built from rules and interactions; that resources, like actions, are tools people use to enact organizations; and that structures are the medium and outcome of the interac- tion [27]. Manager are bounded rational [23] in their acting, because the continuous, recipro- cal interaction among variables, have a continuum of outcomes, that should all alter- natives be examined no action would occur. Manager act accordingly to what they can see and take advantage of in their belief on what is satisfyingly. They are bounded by, what they know, and by what they value, and they are sensitive to norms of what is appropriate behavior [13]. IMJV, as a synthetic organization, is without organiza- tional rationality, because of lack of rules, routines, and procedures, and "knowing that", and "knowing how" [13], is dispersed among organizational members, without common interpretation of the world. It is left to manager 's bounded rationality to implement a structuration process, satisfying their belief on the relation between the imprented targets from partner, knowledge technologies encoded in the context , and choice of coordination tuning, to facilitate an emerging learning process with in the IMJV. The theoretical framework for the research is outlined in Figure 1. 5 Theoretical framework. The first category of determinants in the framework comprises two exogenous de- terminants of IMJV forms: technology and organization structure. Technology is linked to organization structures through its requirements for procedures to control work and requirement for information, coordination and control intensity. How man- agers integrate mechanical technologies with the human technologies is dependent on their choices between mechanism: formalization, to make task simpler by routiniza- tion, and building information linkages; centralization, to distribute responsibility and decisions authority; socialization, the degree of motivation of workers and reduction of distractions. The organizational absorptive capacity is thus an outcome of man- agement’s cognitive skills to combine coordination mechanisms in a structuration process and building communication linkages supporting exploration of the new pos- sibilities. 6 Conclusion This conceptual approach is descriptive. The intended empirical research is case study to explore, in natural settings, how this approach might contribute to deeper under- standing of the phenomenon. But much more work must be done, especially in the scientific understanding and methodology proposal. Fig. 1. Theoretical framework Fig. 1. Theoretical framework Fig. 1. Theoretical framework References 1. Burgelman, R.: Design for Corporate Entrepreneurship in Established Firms. Calif. Manage. Rev., 26 (1984) 154 2. Chandler, A. D. MIT Press, Cambridge (1962) 3. Child, J.: Organizations: A guide to problem and practise. Harper & Row, New York (1977) 3. Child, J.: Organizations: A guide to problem and 4. Cohen, W. M., & Levinthal, D. A.: Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quaterly, 35 (1990) 128 4. Cohen, W. M., & Levinthal, D. A.: Absorptive Capacity: A Ne Innovation. Administrative Science Quaterly, 35 (1990) 128 5. Collin, P. D., Hage, J., Hull, F.: A framework for analyzing technical systems in complex organ- izations. In: Anonymous Research in the Sociology of organizations. JAI Press, Greenwich (1986) 6. Damanpour, F.: Organizational Innovation: A Meta-Analysis of Effects of Determinants and Moderators. Academy of Management Journal, 35 (1991) 555 7. Damanpour, F., & Gopalakrishnan, S.: Theories of Organizational Structure and Innovation Adoption: The Role of Environmental Change. J. Eng. Technol. Manage., 15 (1998) 1-24 p g g g ( 8. Eppkin, D. J.: Planning for Strategic Flexibility. Long Range Plann., 11 (1978) 9 9. Evans, J. S.: Strategic Flexibility for High Technology Manoeuvres: A Conceptual Framework. Journal of management studies, 28 (1991) 69-89 10. Hage, J. T.: Organizational Innovation and Organizational Change . Annual Review of Soci- ology, 25 (1999) 597 11. Inkpen, P., & Beamish, P.: Knowledge, Bargaining Powder, and the Instability of International Joint Ventures. Academy of Management Review, 22 (1997) 177 Joint Ventures. Academy of Management Review, 22 (19 12. Kogut, B.: Joint Ventures: Theoretical and Empirical Perspectives. Strategic Manage. J., p (1988) 319 13. Kogut, B., & Zander, U.: What Firms do? Coordination, Identity and Learning. Organization Science, 7 (1996) 502 14. Lindsay, P. H., & Norman, D. A.: {Human Information Processing}. (1977) 15. Mintzberg, H.: The structuring of organizations, a synthesis of the research. Prentice-Hall, New York (1979) 16. Mintzberg, H., Anestrand, B., Lampel, J.: The strategy safari. The Free Press, New York (2000) 17. Nielsen, K., Brunø, T. D., Jørgensen, K. A.: A Framework Study on Assessment of Mass Customization Capbilities. Proceedings of MCP-CE 2012, Novi Sad, Serbia (2012) 18. Nohria, N., & Gulati, R.: What is the Optimum Amount of Organizational Slack?: A Study of the Relationship between Slack and Innovation in Multinational Firms . European Manage- ment Journal, 15 (1997) 603 19. Parson, T.: Structure and process in mordern societies. References Free Press, Glencoe, III (1960) 19. Parson, T.: Structure and process in mordern societies. Free Press, Glencoe, III (1960) 20. Pine, B. J.: Mass customization: The new frontier in business competition. Harvard Business 19. Parson, T.: Structure and process in mordern societies. Free Press, Glencoe, III (1960) 20. Pine, B. J.: Mass customization: The new frontier in business competition. Harvard Business School Press, Boston, Mass. (1993) 20. Pine, B. J.: Mass customization: The new frontier in business competition. Harvard Business School Press, Boston, Mass. (1993) 21. Sahal, D.: Original Research ArticleTechnological Forecasting and Social Change. Technolo- gy, productivity, and industry structure, 1 (1983) 1 21. Sahal, D.: Original Research ArticleTechnological F 21. Sahal, D.: Original Research ArticleTechnological Forecasting and Social Change gy, productivity, and industry structure, 1 (1983) 1 gy, productivity, and industry structure, 1 (1983) 1 gy p y y ( ) 22. Salvador, F., De Holan, P. M., Piller, F.: Cracking the Code of Mass Customization. MIT Sloan Management Review, 50 (2009) 71-78 gy p y y 22. Salvador, F., De Holan, P. M., Piller, F.: Cracki Sloan Management Review, 50 (2009) 71-78 22. Salvador, F., De Holan, P. M., Piller, F.: Cracking Sloan Management Review, 50 (2009) 71-78 g ( ) 23. Simon, H.: Economic, Bounded Rationality and the Cognitive Revolution. In: Egidi, M., Mar- ris, R. and Viale, R. (eds.). Edward Publisher, Hants (1992) 24. Teece, D. J., Pisano, G., Shuen, A.: Dynamic Capabilities and Strategic Management. Strategic Manage. J., 18 (1997) 509-533 25. Thomson, J. D.: Organization in aaction. McGraw-Hill, New York (1967 26. Urabe, K.: Innovation and the Japanese Mangement System. (1988) 3 27. Weick, K. E.: Technology as Equivoque: Sensemaking in New Technologies. In: Godmand, P.S. and Sproull, L.S. (eds.) Technology and Organizations 28. Wheelwright, S., & Clark, K.: Competing through Development Capability in a Manufactur- ing-Based Organization . Bus. Horiz., 35 (1992) 29 28. Wheelwright, S., & Clark, K.: Competing through Developmen ing-Based Organization . Bus. Horiz., 35 (1992) 29
https://openalex.org/W2110458076	https://digital.library.adelaide.edu.au/dspace/bitstream/2440/93697/2/hdl_93697.pdf	English	null	The prognostic value of estrogen receptor beta and proline-, glutamic acid- and leucine-rich protein 1 (PELP1) expression in ovarian cancer	BMC cancer	2,013	cc-by	5,451	http://hdl.handle.net/2440/93697 PUBLISHED VERSION Stefanie Aust, Peter Horak, Dietmar Pils, Sophie Pils, Christoph Grimm, Reinhard Horvat, Dan Tong, Bernd Schmid, Paul Speiser, Alexander Reinthaller, and Stephan Polterauer The prognostic value of estrogen receptor beta and proline-, glutamic acid- and leucine-rich protein 1 (PELP1) expression in ovarian cancer BMC Cancer, 2013; 13(1):115-1-115-7 © 2013 Aust 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. Originally published at: http://doi.org/10.1186/1471-2407-13-115 PERMISSIONS http://creativecommons.org/licenses/by/2.0/ Abstract Background: Proline-, glutamic acid-, and leucine-rich protein 1 (PELP1), a coregulator of the estrogen receptors (ERs) alpha and beta, is a potential proto-oncogene in hormone dependent gynecological malignancies. To better understand the role of PELP1 in epithelial ovarian cancer (EOC), the protein expression and prognostic significance of PELP1 was evaluated together with ERalpha and ERbeta in EOC tissues. Methods: The expression of PELP1, ERalpha, and ERbeta was characterized in tumor tissues of 63 EOC patients. The prognostic value was calculated performing log-rank tests and multivariate Cox-Regression analysis. In a second step, validation analysis in an independent set of 86 serous EOC patients was performed. Results: Nuclear PELP1 expression was present in 76.2% of the samples. Prevalence of PELP1 expression in mucinous tumors was significantly lower (37.5%) compared to serous (85.7%) and endometrioid tumors (86.7%). A significant association between PELP1 expression and nuclear ERbeta staining was found (p=0.01). Positive PELP1 expression was associated with better disease-free survival (DFS) (p=0.004) and overall survival (OS) (p=0.04). The combined expression of ERbeta+/PELP1+ revealed an independent association with better DFS (HR 0.3 [0.1-0.7], p=0.004) and OS (HR 0.3 [0.1-0.7], p=0.005). In the validation set, the combined expression of ERbeta+/PELP1+ was not associated with DFS (HR 0.7 [0.4-1.3], p=0.3) and OS (HR 0.7 [0.3-1.4], p=0.3). Conclusion: Positive immunohistochemical staining for the ER coregulator PELP1, alone and in combination with ERbeta, might be of prognostic relevance in EOC. Keywords: PELP1, Estrogen receptor alpha, Estrogen receptor beta, Immunohistochemistry, Prognosis, Ovarian cancer Keywords: PELP1, Estrogen receptor alpha, Estrogen receptor beta, Immunohistochemistry, Progno cancer In ovarian cancer cells, estrogen is biologically active by binding to the nuclear receptors (NRs) ERalpha and ERbeta. NR signaling leads to a translocation of the ER to the nucleus, followed by a binding to the target gene pro- moters and consequently to a specific gene transcription [4]. Inhibition or enhancement of transcription is achieved with the help of ligand-regulated coactivators and core- pressors, stabilizing and enhancing or repressing tran- scription [5]. Deregulation of such NR-coactivators - or coregulators - in ovarian cancer cells is likely to interact with proliferation and survival of cancer cells. The increas- ing interest in coregulators as mediators of NR function has revealed proline-, glutamic acid-, and leucine-rich pro- tein 1 (PELP) 1, also named “modulator of nongenomic RESEARCH ARTICLE Open Access http://hdl.handle.net/2440/93697 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Aust et al. BMC Cancer 2013, 13:115 © 2013 Aust 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. The prognostic value of estrogen receptor beta and proline-, glutamic acid- and leucine-rich protein 1 (PELP1) expression in ovarian cancer Stefanie Aust1, Peter Horak2, Dietmar Pils1, Sophie Pils1, Christoph Grimm1, Reinhard Horvat3, Dan Tong1, Bernd Schmid4, Paul Speiser1, Alexander Reinthaller1 and Stephan Polterauer1* * Correspondence: stephan.polterauer@meduniwien.ac.at 1Department of Gynaecology and Gynaecological Oncology, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria Full list of author information is available at the end of the article Background Despite increasing knowledge in the etiology and treat- ment of epithelial ovarian cancer (EOC), ovarian cancer still accounts for more death women than any other gynecological malignancy [1]. As ovarian cancer is an endocrine-related cancer, the role of estrogen in the complex network of ovarian cancer pathogenesis, as well as the role of estrogen receptors (ERs) as prognostic markers, are being intensively discussed [2,3]. * Correspondence: stephan.polterauer@meduniwien.ac.at 1Department of Gynaecology and Gynaecological Oncology, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria Full list of author information is available at the end of the article Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Page 2 of 7 FIGO guidelines was performed, including hysterectomy, bilateralsalpingo-oophorectomy,pelvicand/orparaaortic lymphadenectomy, appendectomy, omentectomy and cytoreductive procedure in order to resect all gross tumor. All patients with tumor stages FIGO Ic to III and all pa- tients with clear cell carcinoma received a platinum-based chemotherapy. Patients wishing to preserve fertility and tumor stage FIGO Ia were treated with conservative sur- gery (unilateral salpingo-oophorectomy) and full surgical staging including washings, omentectomy, appendectomy, node biopsies and a thorough abdominal exploration with biopsies of all suspicious areas. Residual tumor load was defined as negative, if macroscopically absent. Post- therapeutically all patients were followed up four times an- nually, including pelvic examination, abdominal ultrasound examination, and serum tumor marker evaluation. Overall survival was the time interval between diagnosis and death. Overall observation time was the time interval between diagnosis and last contact, defined as death from the dis- ease or last follow-up. Recurrent disease was defined as at least a twofold increase in the nadir serum CA-125 level after first-line chemotherapy or radiological diagnosis. Pa- tients without recurrence, cancer progression or death were censored at the time of last follow-up. Experienced gyne- cological oncologists and an experienced pathologist per- formed the clinical and histopathological evaluation and the evaluation of response to first-line treatment. actions of estrogen receptor” (MNAR), as mediator of not only genomic but also non-genomic actions of ERs and as potential proto-oncogene in hormonal cancers [6]. actions of estrogen receptor” (MNAR), as mediator of not only genomic but also non-genomic actions of ERs and as potential proto-oncogene in hormonal cancers [6]. PELP1 interacts with various NRs, including ERs, androgen receptors, glucocorticoid receptors and pro- gesterone receptors [7]. In cancer cells, PELP1 seems to effect ER-mediated gene expression, subsequently influencing cell proliferation and differentiation [8]. Background In endometrial cancer, PELP1 functionally interacts with both, ERalpha and ERbeta, and enhances their tran- scriptional response [9]. Furthermore, PELP1 has been associated with increased cell motility and invasion in cancer cells [10]. Elevated PELP1 expression has also been associated with poor outcome in ER positive/lu- minal-like breast cancer tissue [11]. In ovarian cancer cell line models and in nude mouse models elevated PELP1 expression leads to increased cell migration, metastasis and tumor progression [12]. In hu- man ovarian cancer tissue, only one study has focused on the role of PELP1 protein expression [13]. PELP1 has been described to be overexpressed in 60% of ovarian cancers and to be deregulated in several subtypes of ovarian tumors [13]. It has been proposed, that during ovarian cancer progression, modification of PELP1 ex- pression might occur. According to these findings, the prognostic relevance of PELP1 in ovarian cancer needs to be further elucidated. Additionally, the prog- nostic importance of ERalpha and ERbeta in ovarian cancer is still discussed controversially, despite their importance in breast and endometrial cancer [14-18]. Ovarian tissue microarray and Immunohistochemistry Paraffin embedded tissue-blocks were processed using stan- dardized procedures. Tissue microarrays (TMAs) was as- sembled by taking three core needle ‘biopsies’ from defined tumor regions in the preexisting paraffin-embedded tissue blocks, using techniques and an apparatus developed by Beecher Instruments Inc., Micro-Array Technology (Sun Prairie, WI, USA). Separate TMAs were constructed for the test and validation set. Immunohistochemistry (IHC) pro- cedures were performed at room temperature. Samples were deparaffinized, rehydrated and treated with 3% H2O2 for 10 minutes to quench endogenous peroxidase. For ERbeta antigen heat retrieval was performed using heat- induced epitope retrieval in DakoCytomation Target Re- trieval Solution (No. S1700, DAKO, Denmark), and for ERalpha and PELP1 heat-induced epitope retrieval was performed using citrate buffer (Citra-BioGenex No. HK 087-5K). The sections were incubated at 4°C overnight with primary antibodies (ERalpha, 1:50, clone 1D5, mouse IgG1, Dako, Denmark; ERbeta1, 1:20, clone PPG5/10, mouse IgG2a, Dako, Denmark; PELP1, 1:500, polyclonal rabbit, No. IHC-00013, Bethyl Laboratories, USA). As positive controls, FFPE sections of ER positive human breast adenocarcinoma were used. Negative control mouse and rabbit isotypes were used as negative controls. Slides were The major aim of this study was to further define the role of PELP1 in human ovarian cancer and to evaluate the prognostic role of PELP1, ERalpha and ERbeta in EOC patients receiving platinum/taxane- based chemotherapy. Distribution of PELP1 expression Of the 63 cancer tissues, 48 (76.2%) of patients were found to express nuclear staining of PELP1. Out of the 48 patients classified as PELP1 positive, a moderate PELP1 expression was found in 24 (38.1%) tissues and a strong expression was observed in 23 (36.5%) of the samples. Representative 1 Endometrioid: n= 16; Mucinous: n= 9; Undifferentiated carcinoma: n= 7; Clear cell carcinoma: n= 3; Data analysis and Statistics Table 1 Patients’ characteristics Screening set (n = 63) Validation set (n = 86) Characteristics n (%) n (%) Histology Serous 28 (44.4) 86 (100.0) Non-serous 35 (55.6) 1 - FIGO I 19 (30.2) 6 (7.0) II 6 (9.5) 4 (4.7) III 32 (50.8) 61 (70.9) IV 6 (9.5) 15 (17.4) Grade Grade 1 10 (15.9) 6 (7.0) Grade 2 20 (31.7) 19 (22.1) Grade 3 33 (52.4) 61 (70.9) Residual tumor no 35 (55.6) 37 (43.0) > 0 cm 28 (44.4) 49 (57.0) ERα positive 13 (20.6) 27 (31.4) negative 49 (77.8) 55 (63.9) missing 1 (1.5) 4 (4.7) ERbeta positive 45 (71.4) 14 (16.3) negative 15 (23.8) 68 (79.0) missing 3 (4.7) 4 (4.7) PELP1 positive 48 (76.2) 21 (24.4) negative 12 (19.0) 61 (70.9) missing 3 (4.7) 4 (4.7) 1 Endometrioid: n= 16; Mucinous: n= 9; Undifferentiated carcinoma: n= 7; Clear cell carcinoma: n= 3; Table 1 Patients’ characteristics Screening set (n = 63) Validation set (n = 86) Table 1 Patients’ characteristics Study population Paraffinembeddedovariantumortissuefromprimarysur- gery was used from 63 patients with EOC FIGO stage I-IV, receiving cytoreductive surgery at the Department of General Gynaecology and Gynaecological Oncology, Medical University of Vienna, Austria, between 1996 and 2001.Inasecondstep,avalidationsetwasusedcomprising 86 EOC patients of only serous histology, receiving cytoreductive surgery at the Medical University of Vienna, Austria, between 2004 and 2010. An informed consent according to the criteria of the Medical University of Vienna was obtained from all patients. Approval for this study was obtained by the institutional review board of the Medical University of Vienna (IRB-no.:266/2010). Patientsweretreatedaccordingtostandardsofthepresent institution with upfront surgery and adjuvant platinum– based chemotherapy. Surgical staging according to Page 3 of 7 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 incubated for 25 minutes with a biotinylated secondary antibody (Link, No. K0673, Dako, Denmark), followed by incubation with Streptavidin Peroxidase-HRP (25 Minutes, No. K 0673, Dako, Denmark). The slides were stained with diamino-benzidine (DAB Chromogen 1:50 in DAB Sub- strate Buffer, K0673, Dako, Denmark) for 2 minutes. For counterstaining the slides were dipped into hematoxylin for 25 seconds. Intra-nuclear distribution of PELP1 was deter- mined by immunofluorescence staining. The fluorescence labeled secondary antibody, goat anti-rabbit (1:5000; Invitrogen, AlexaFluorW 488 fragment of goat anti- rabbit IgG (H+L)) was used besides DAPI for nuclear counterstaining. observation period, 36 patients died (57.1%) and 36 patients (57.1%) experienced tumor recurrence. 50 patients (79.4%) received carboplatin-paclitaxel based standard chemother- apy. Out of these 50 patients, 46 women (73.0%) showed initial response to chemotherapy and a total of 34 women died (68.0%). The characteristics of the validation set are likewise presented in Table 1. In this validation set the mean observation time was 51.3 months and a total of 45 (52.3%) patients died. Data analysis and Statistics The intensity patterns and nuclear positivity of staining were analyzed by two independent co-workers, including a gynecological pathologist (RH), applying a semi-quanti- tative scale of ImmunoReactive Score (IRS) calculated by multiplication of the number of positively labeled cells (4 percentage groups) with the intensity of the staining reac- tion (3 grades) [19]. The intensity of reaction was scored as negative (intensity IRS 0–2, no reaction, and IRS 3–4 showing a very weak reaction of staining with <10% posi- tive cells) or positive (IRS 6–8 with a moderate reaction and 10 to 50% positive cells; IRS 9–12 with a strong reac- tion and 51% to 80% positive cells). Statistical analyses were performed using SPSS software version 19 (IBM Corporation, Armonk, New York, United States). P-values <0.05 were considered statistically significant. Associations between ERalpha-expression, ERbeta-expression, PELP1- expression, and clinicopathological parameters were calculated using Pearson’s χ2 or Fisher’s exact tests as ap- propriate. Impact of the potential new prognostic factors on disease-free survival (DFS) and overall survival (OS) was determined by univariate and multiple Cox proportional- Hazards model analyses, whereby in the test set, only the patients receiving chemotherapy according to standardized protocols were included in these analyses (n=50). In the validation set, only patients receiving chemotherapy were included. Survival analyses were performed for all three pa- rameters and the combination pattern ERbeta+/PELP1+ versus ERbeta+/Pelp1-, ERbeta-/PELP1+ and ERbeta- /Pelp1- tumors as well as for the combination pattern ERalpha+/PELP1+ versus ERalpha+/Pelp1-, ERalpha- /PELP1+ and ERalpha-/Pelp1- tumors. Study population The characteristics of the patients included in the study show a typical heterogeneous ovarian cancer population and are depicted in Table 1. Mean age of the EOC patients at time of cytoreductive surgery was 58.3 years (±13.9years). The mean observation period was 57.8 months. Within the Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Page 4 of 7 immunohistochemical examples of PELP1 positive and negative stainings in ovarian cancer tissue are shown in Figure 1A-D. Using immunofluorescence, intra-nuclear distribution of PELP1 was analyzed. As depicted in Figure 2, an intensive PELP1 staining within the nucleoli of the tumor cells can be observed. Further examination of histological subtypes revealed expression of PELP1 in all included subtypes of ovarian cancer. All undifferenti- ated carcinomas were PELP1 positive, as was the majority of serous and endometrioid tumors (85.7% and 86.7%, re- spectively), whereas the majority of mucinous tumors (62.5%) showed no nuclear PELP1 expression (p=0.02, Fisher’s exact). No significant difference of PELP1 expres- sion regarding age (<55 years vs > 55 years), FIGO stage, grade or residual tumor load after cytoreductive surgery could be observed. Figure 2 Intra-nuclear distribution of PELP1 in a serous EOC sample. An intense PELP1 staining (green) can be observed within the nucleolar compartments (white arrows). DAPI (blue) was used for nuclear counterstaining (Pictures were taken with the confocal microscope LSM700). Coexpression of PELP1 and ERs In a next step, we analyzed the coexpression of PELP1 to- gether with ERalpha and ERbeta. A significant association between PELP1 expression and nuclear ERbeta staining was found (p=0.01, Fisher’s exact), whereas no significant association between PELP1 and nuclear ERalpha staining was observed (p=0.3, Fisher’s exact). Expression of PELP1 and both ERs was observed in 10 patients (15.8%). To provide more profound and histotype specific sur- vival data we have additionally performed a validation analysis in a validation set of 86 EOC with only serous histology. In univariate and multivariate analysis, ERbeta+/PELP1+ expression had no significant influ- ence on survival in this validation set of only serous histology. Survival data is presented in Table 2B. Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Page 5 of 7 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 (18.8%) and undifferentiated carcinomas (28.6%). No ERalpha expression was seen in the clear cell carcinomas and the mucinous ovarian carcinomas. Regarding age (≤55 years vs > 55 years), FIGO-stage, grade or residual tumor load after debulking surgery, no significant differences were found among ERalpha and/or ERbeta expressing tumors (data not shown). PELP1 and ERbeta turned out to be the most relevant prognostic factor in univariate survival analysis (DFS: p=0.03, OS: p=0.02). In a univariate analysis, the clinico- pathological parameters FIGO-stage (p < 0.001 and p < 0.001) and residual tumor (p < 0.001 and p < 0.001) were associated with DFS and OS, respectively. In a next step, we added the coexpression of PELP1 and ERbeta and all clinicopathological parameters in a Cox proportional-hazards regression model. Multiple Cox regression analysis revealed that ERbeta+/PELP1+ tumors had the strongest independent impact on sur- vival, with a significantly longer DFS (HR 0.3 [0.1-0.7], p = 0.004) and OS (HR 0.3 [0.1-0.7], p = 0.005). Survival analyses Table 2A shows the association between PELP1, ER- alpha, ERbeta and survival. In a univariate analysis, ovar- ian cancer patients with PELP1 expressing tumor tissue had a better OS and DFS (p = 0.04, p = 0.004; respect- ively) compared to patients without PELP1 expression. ERbeta and ERalpha had no significant univariate influ- ence on survival. Interestingly, the coexpression of Distribution of ERalpha and ERbeta expression Nuclear staining of ERalpha and ERbeta was positive in 13 (20.6%) and in 45 (71.4%) ovarian cancer patients, respect- ively. Examination of ERalpha and ERbeta expression among the different histological tumor types revealed no significant differences (p=0.4 and p=0.2, respectively; Fish- er’s exact). ERbeta expression could be observed in all sub- types. A total of 85.2% serous, 44.4% mucinous, 73.3% endometrioid, 71.4% undifferentiated and both clear cell carcinomas (100%) were ERbeta expression positive. A higher percentage of ERalpha expressing tumors was ob- served among serous (29.6%) compared to endometrioid Figure 2 Intra-nuclear distribution of PELP1 in a serous EOC sample. An intense PELP1 staining (green) can be observed within the nucleolar compartments (white arrows). DAPI (blue) was used for nuclear counterstaining (Pictures were taken with the confocal microscope LSM700). Figure 1 Representative immunohistochemical examples of PELP1 staining. In (A) PELP1 positive low differentiated epithelial ovarian cancer, (B) PELP1 positive high differentiated epithelial ovarian cancer, (C) mucinous ovarian cancer with nearly complete absence of nuclear PELP1 staining and (D) PELP1 negative low differentiated epithelial ovarian cancer is shown. Pictures were taken using TissueFAXS (TissueGnostics; Vienna, Austria; magnification x200). Figure 1 Representative immunohistochemical examples of PELP1 staining. In (A) PELP1 positive low differentiated epithelial ovarian cancer, (B) PELP1 positive high differentiated epithelial ovarian cancer, (C) mucinous ovarian cancer with nearly complete absence of nuclear PELP1 staining and (D) PELP1 negative low differentiated epithelial ovarian cancer is shown. Pictures were taken using TissueFAXS (TissueGnostics Vienna, Austria; magnification x200). Discussion We know that the patient number of our test set is a limitation for survival analysis, but in view of the low patient number, the positive impact we could observe for PELP1 in univariate survival analysis is even more interesting. As we have observed a correlation between PELP1 and ERbeta expression, we add- itionally analyzed the coexpression of ERbeta and PELP1 regarding survival. Besides the clinicopathological factors FIGO stage and residual tumor load after cytoreductive sur- gery, this coexpression pattern (ERbeta+/PELP1+) turned out to be the most relevant prognostic factor in univariate and multivariate survival analysis, revealing a significantly longer DFS (HR 0.3 [0.1-0.7], p = 0.004) and OS (HR 0.3 [0.1-0.7], p = 0.005). and ERbeta in 63 human EOC tissues. Nuclear PELP1 expression was present in 76.2% and was found in all histological subtypes. To our knowledge, only one study has previously reported on PELP1 expression in human EOC tissue [13], whereby no differences could be observed between the four major types of EOC. Vadlamudi et al. speculated that PELP1modulates rDNA transcription and accelerates cell cycle progression. Using immunofluores- cence staining of MCF7 and HeLa cells, they discovered that PELP1 localizes in the nucleolar compartments [6]. Thus we decided to perform immunofluorescence to in- vestigate intranuclear expression of PELP1 in EOC. Our results show, that in serous EOC, an intense nucleolar ex- pression of PELP1 can be observed. Further studies are re- quired to determine PELP1 expression during cell-cycle progression in EOC. In our study-population, the majority of serous and endometrioid tumor tissues were PELP1 positive, whereas we observed a significantly reduced number of PELP1 ex- pressing mucinous tumors (37.5%). Additionally, none of the mucinous tumors expressed ERalpha. Primary mucin- ous tumors are comparatively rare and they were cautiously classified in this study [20]. Molecular changes in mucinous tumors have not been studied to the same extend as in ser- ous and endometrioid tumors [21], highlighting the import- ance to further characterize this histological subtype. Matching our results, a previous study has reported mucin- ous EOC to be ERalpha negative [22]. Additionally, a sig- nificantly higher ER expression was observed in serous (43% positive) versus mucinous (4% positive) EOCs in a large study, including over 700 ovarian cancer patients [14]. Our results show, that mucinous tumors remained ERbeta positive, which is also in accordance with the findings by Lindgren et al. [22]. Discussion In this study, coexpression of PELP1 and ERbeta was asso- ciated with a better prognosis in patients with EOC. We primarily investigated the expression of PELP1, ERalpha Table 2 Survival analysis of ovarian cancer patients including coexpression of PELP1/ERbeta in the test set (A) and validation set (B) Disease-free survival Overall survival Univariate1 Multivariate2 Univariate1 Multivariate2 P-Value P-Value HR (95%CI) P-Value P-Value HR (95%CI) Patients’ age 0.9 0.09 2.1 (0.9-4.9) 0.9 0.4 1.4 (0.6-3.0) Histological type (serous vs. non-serous) 0.6 0.6 0.8 (0.4-1.8) 0.4 0.9 0.9 (0.5-2.2) FIGO stage (I vs. II vs. III vs. IV) <0.001 <0.001 4.4 (1.9-10.1) <0.001 0.007 3.2 (1.4-7.6) Residual tumor(no vs. yes) <0.001 0.9 0.9 (0.4-2.6) <0.001 0.4 1.6 (0.6-4.3) ERalpha (negative vs. positive) 0.8 - - 0.9 - - ERbeta (negative vs. positive) 0.3 - - 0.2 - - PELP1 (negative vs. positive) 0.004 - - 0.04 - - PELP1/ERbeta (ERbeta+/PELP1+ vs. others) 0.03 0.004 0.3 (0.1-0.7) 0.02 0.005 0.3 (0.1-0.7) Disease-free survival Overall survival Univariate1 Multivariate2 Univariate1 Multivariate2 P-Value P-Value HR (95%CI) P-Value P-Value HR (95%CI) Patients’ age 0.09 0.4 1.3 (0.7-2.2) 0.005 0.04 2.4 (1.1-5.3) FIGO stage (I vs. II vs. III vs. IV) 0.001 0.02 1.8 (1.1-3.0) 0.001 0.005 2.5 (1.3-4.7) Residual tumor (no vs. yes) 0.003 0.1 1.6 (0.9-1.8) 0.02 0.3 1.4 (0.7-2.7) ERalpha (negative vs. positive) 0.2 - - 0.6 - - ERbeta (negative vs. positive) 0.7 - - 0.2 - - PELP1 (negative vs. positive) 0.8 - - 0.9 - - PELP1/ERbeta (ERbeta+/PELP1+ vs. others) 0.6 0.3 0.7 (0.4-1.3) 0.9 0.3 0.7 (0.3-1.4) 1Log rank test; 2multivariate Cox-regression analysis, HR=Hazard Ratio, 95%CI= 95% Confidence Interval. nalysis of ovarian cancer patients including coexpression of PELP1/ERbeta in the test set (A) and Table 2 Survival analysis of ovarian cancer patients including coexpression of PELP1/ERbeta in the test set (A) and validation set (B) Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Page 6 of 7 survival [17,24] have likewise been published. To determine the association between ER and survival, coregulators interacting with these NRs and often deregulated in hormone-interacting tumors [25] should likewise be taken into account. Our analysis revealed that women with posi- tive PELP1 expression had a better OS and DFS compared to patients without PELP1 expression. Discussion Due to these positive findings we validated the survival data in a validation set. The validation was performed in an independent more homogenous set of 86 EOC patients with only serous histology. Unfortunately we could not re- produce our findings regarding the protective effect of the coexpression of ERbeta and PELP1. Both results are of rele- vance as little is known about the role of PELP1 in EOC. PELP1 is being discussed as a potentially targetable proto- oncogene in ER positive breast cancer [26], associated with rapid tumor growth in xenograft models and high grade as well as node-positivity in human breast cancer [27]. PELP1 seems to be of importance in breast cancer, due to an in- volvement in hormone therapy response and resistance [25]. In EOC, hormone therapy has been studied in a lim- ited number of trials, including mainly patients with recur- rent or refractory ovarian cancer. To understand not only the prognostic, but also the therapeutic role of ERs in EOC, further studies on coregulators and the mechanisms through which hormones interact with EOC need to be designed. Our findings reflect relatively low frequencies of ERalpha expression versus ERbeta expression, with 20.6% versus 71.4%, respectively. This is in accordance with a previous immunohistochemical evaluation in EOC, describing a higher ERbeta than ERalpha expression [15]. Interacting proteins have been described for PELP1 and both nuclear hormone receptors [8]. In our study, a similar expression pattern was observed between ERbeta and PELP1, as proven by a significant correlation. This may indicate a functional interaction especially between ERbeta and PELP1 in EOC. In ER transactivation assays, PELP1 seems to potentiate ERbeta mediated transcriptional activity in endometrial cells [9]. To determine the impact of PELP1 on ER transcriptional activity, additional in vitro studies with ovarian cancer cells need to be set up. The findings of our test set remain of importance as Grivas et al. have described PELP1 overexpression in epithelial colorectal cancer cells to be associated with in- creased OS. Comparable to our results, a correlation be- tween ERbeta and PELP1 was observed, and PELP1 was a positive prognostic marker in the subset of ERbeta positive carcinomas [28]. References 23. Halon A, Materna V, Drag-Zalesinska M, Nowak-Markwitz E, Gansukh T, Donizy P, Spaczynski M, Zabel M, Dietel M, Lage H, et al: Estrogen receptor alpha expression in ovarian cancer predicts longer overall survival. Pathol Oncol Res 2011, 17(3):511–518. 1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ: CA Cancer J Clin 2007, 57(1):43–66. 2. Gallo D, Ferlini C, Scambia G: The epithelial-mesenchymal transition and the estrogen-signaling in ovarian cancer. Curr Drug Targets 2010, 11(4):474–481. 24. Tangjitgamol S, Manusirivithaya S, Khunnarong J, Jesadapatarakul S, Tanwanich S: Expressions of estrogen and progesterone receptors in epithelial ovarian cancer: a clinicopathologic study. Int J Gynecol Cancer 2009, 19(4):620–627. 3. Park SH, Cheung LW, Wong AS, Leung PC: Estrogen regulates Snail and Slug in the down-regulation of E-cadherin and induces metastatic potential of ovarian cancer cells through estrogen receptor alpha. Mol Endocrinol 2008, 22(9):2085–2098. 25. Chakravarty D, Tekmal RR, Vadlamudi RK: PELP1: A novel therapeutic target for hormonal cancers. IUBMB Life 2010, 62(3):162–169. 4. Horwitz KB, Jackson TA, Bain DL, Richer JK, Takimoto GS, Tung L: Nuclear receptor coactivators and corepressors. Mol Endocrinol 1996, 10(10):1167–1177. 26. Cortez V, Mann M, Tekmal S, Suzuki T, Miyata N, Rodriguez-Aguayo C, Lopez-Berestein G, Sood AK, Vadlamudi RK: Targeting the PELP1-KDM1 axis as a potential therapeutic strategy for breast cancer. Breast Cancer Res 2012, 14(4):R108. 5. O'Malley BW: Coregulators: from whence came these "master genes". Mol Endocrinol 2007, 21(5):1009–1013. 6. Gonugunta VK, Nair BC, Rajhans R, Sareddy GR, Nair SS, Vadlamudi RK: Regulation of rDNA transcription by proto-oncogene PELP1. PLoS One 2011, 6(6):e21095. 6. Gonugunta VK, Nair BC, Rajhans R, Sareddy GR, Nair SS, Vadlamudi RK: Regulation of rDNA transcription by proto-oncogene PELP1. PLoS One 2011, 6(6):e21095. 27. Rajhans R, Nair S, Holden AH, Kumar R, Tekmal RR, Vadlamudi RK: Oncogenic potential of the nuclear receptor coregulator proline-, glutamic acid-, leucine-rich protein 1/modulator of the nongenomic actions of the estrogen receptor. Cancer Res 2007, 67(11):5505–5512. 7. Vadlamudi RK, Kumar R: Functional and biological properties of the nuclear receptor coregulator PELP1/MNAR. Nucl Recept Signal 2007, 5:e004. 7. Vadlamudi RK, Kumar R: Functional and biological properties of the nuclear receptor coregulator PELP1/MNAR. Nucl Recept Signal 2007, 5:e004. 28. Grivas PD, Tzelepi V, Sotiropoulou-Bonikou G, Kefalopoulou Z, Papavassiliou AG, Kalofonos H: Expression of ERalpha, ERbeta and co-regulator PELP1/ MNAR in colorectal cancer: prognostic significance and clinicopathologic correlations. Cell Oncol 2009, 31(3):235–247. 8. Competing interests 13. Dimple C, Nair SS, Rajhans R, Pitcheswara PR, Liu J, Balasenthil S, Le XF, Burow ME, Auersperg N, Tekmal RR, et al: Role of PELP1/MNAR signaling in ovarian tumorigenesis. Cancer Res 2008, 68(12):4902–4909. Competing interests The authors have no competing interests to declare. The authors have no competing interests to declare. 14. Hogdall EV, Christensen L, Hogdall CK, Blaakaer J, Gayther S, Jacobs IJ, Christensen IJ, Kjaer SK: Prognostic value of estrogen receptor and progesterone receptor tumor expression in Danish ovarian cancer patients: from the 'MALOVA' ovarian cancer study. Oncol Rep 2007, 18(5):1051–1059. Received: 24 July 2012 Accepted: 6 March 2013 Published: 14 March 2013 21. Cho KR: Shih Ie M: Ovarian cancer. Annu Rev Pathol 2009, 4:287–313. 22. Lindgren PR, Cajander S, Backstrom T, Gustafsson JA, Makela S, Olofsson JI: Estrogen and progesterone receptors in ovarian epithelial tumors. Mol Cell Endocrinol 2004, 221(1–2):97–104. Authors’ contribution SA carried out the immunohistochemical analyses and has been involved in the draft of the manuscript. PH has been involved in construction of the validation TMA and data generation. DP carried out the statistical analyses and has been involved in data interpretation. SP has been involved in clinical data generation. StP has been involved in drafting of the manuscript and the clinical data generation. RH carried out the pathological examination of the immunohistochemical stainings. DT carried out the statistical analyses. BS has been involved in the construction of the test TMA and the data generation. PS has been involved in the clinical data generation. AR has been involved in the data interpretation and participated in the design of the study. CG participated in the design of the study and helped to draft the manuscript. All authors read and approved the final manuscript. 15. Burges A, Bruning A, Dannenmann C, Blankenstein T, Jeschke U, Shabani N, Friese K, Mylonas I: Prognostic significance of estrogen receptor alpha and beta expression in human serous carcinomas of the ovary. Arch Gynecol Obstet 2010, 281(3):511–517. 16. Arias-Pulido H, Smith HO, Joste NE, Bocklage T, Qualls CR, Chavez A, Prossnitz ER, Verschraegen CF: Estrogen and progesterone receptor status and outcome in epithelial ovarian cancers and low malignant potential tumors. Gynecol Oncol 2009, 114(3):480–485. 17. Lee P, Rosen DG, Zhu C, Silva EG, Liu J: Expression of progesterone receptor is a favorable prognostic marker in ovarian cancer. Gynecol Oncol 2005, 96(3):671–677. Author details 1D t t f 1Department of Gynaecology and Gynaecological Oncology, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria. 2Department of Internal Medicine, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria. 3Department of Pathology, Medical University of Vienna, 1090, Vienna, Austria. 4Department of Gynaecology and Gynaecological Oncology, Hietzing Hospital Vienna, 1130, 1Department of Gynaecology and Gynaecological Oncology, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria. 18. Munstedt K, Steen J, Knauf AG, Buch T, von Georgi R, Franke FE: Steroid hormone receptors and long term survival in invasive ovarian cancer. Cancer 2000, 89(8):1783–1791. 2Department of Internal Medicine, Comprehensive Cancer Center, Medical University of Vienna, 1090, Vienna, Austria. 3Department of Pathology, Medical University of Vienna 1090 Vienna Austria 4Department of 19. Remmele W: Stegner HE: [Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue]. Pathologe 1987, 8(3):138–140. Gynaecology and Gynaecological Oncology, Hietzing Hospital Vienna, 1130, Vienna, Austria. 20. Hart WR: Mucinous tumors of the ovary: a review. Int J Gynecol Pathol 2005, 24(1):4–25. Received: 24 July 2012 Accepted: 6 March 2013 Published: 14 March 2013 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Aust et al. BMC Cancer 2013, 13:115 http://www.biomedcentral.com/1471-2407/13/115 Conclusion The prognostic impact of ER expression on patients’ sur- vival has been discussed in literature over years but still re- mains controversial. ER expression was described as a favorable marker in a variety of studies [14,15,23]. However, reports on ER as not significantly influencing patients’ Potential prognostic markers for women diagnosed with ovarian cancer are urgently needed. In conclusion, our results show that PELP1 expression, alone and in com- bination with ERbeta, may be an interesting research tar- get in this cancer entity. Page 7 of 7 References Wong CW, McNally C, Nickbarg E, Komm BS, Cheskis BJ: Estrogen receptor- interacting protein that modulates its nongenomic activity-crosstalk with Src/Erk phosphorylation cascade. Proc Natl Acad Sci U S A 2002, 99(23):14783–14788. doi:10.1186/1471-2407-13-115 Cite this article as: Aust et al.: The prognostic value of estrogen receptor beta and proline-, glutamic acid- and leucine-rich protein 1 (PELP1) expression in ovarian cancer. BMC Cancer 2013 13:115. 9. Vadlamudi RK, Balasenthil S, Broaddus RR, Gustafsson JA, Kumar R: Deregulation of estrogen receptor coactivator proline-, glutamic acid-, and leucine-rich protein-1/modulator of nongenomic activity of estrogen receptor in human endometrial tumors. J Clin Endocrinol Metab 2004, 89(12):6130–6138. 10. Chakravarty D, Nair SS, Santhamma B, Nair BC, Wang L, Bandyopadhyay A, Agyin JK, Brann D, Sun LZ, Yeh IT, et al: Extranuclear functions of ER impact invasive migration and metastasis by breast cancer cells. Cancer Res 2010, 70(10):4092–4101. 11. Habashy HO, Powe DG, Rakha EA, Ball G, Macmillan RD, Green AR, Ellis IO: The prognostic significance of PELP1 expression in invasive breast cancer with emphasis on the ER-positive luminal-like subtype. Breast Cancer Res Treat 2010, 120(3):603–612. 12. Chakravarty D, Roy SS, Babu CR, Dandamudi R, Curiel TJ, Vivas-Mejia P, Lopez-Berestein G, Sood AK, Vadlamudi RK: Therapeutic targeting of PELP1 prevents ovarian cancer growth and metastasis. Clin Cancer Res 2011, 17(8):2250–2259.
https://openalex.org/W2166894028	https://repositorioaberto.uab.pt/bitstream/10400.2/9772/1/Formative%20assessment%20of%20teachers%20in%20the%20context%20of%20an%20online%20learning%20environment.pdf	English	null	Formative assessment of teachers in the context of an online learning environment	Teachers and teaching	2,013	cc-by-sa	7,592	Teachers and Teaching: theory and practice To cite this article: Luís Tinoca & Isolina Oliveira (2013) Formative assessment of teachers in the context of an online learning environment, Teachers and Teaching: theory and practice, 19:2, 214-227, DOI: 10.1080/13540602.2013.741836 To link to this article: http://dx.doi.org/10.1080/13540602.2013.741836 This article was downloaded by: [79.169.70.250] On: 30 October 2014, At: 04:00 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK This article was downloaded by: [79.169.70.250] On: 30 October 2014, At: 04:00 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK g Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions Teachers and Teaching: theory and practice, 2013 Vol. 19, No. 2, 214–227, http://dx.doi.org/10.1080/13540602.2013.741836 Teachers and Teaching: theory and practice, 2013 Vol. 19, No. 2, 214–227, http://dx.doi.org/10.1080/13540602.2013.741836 Corresponding author. Email: ltinoca@ie.ul.pt Luís Tinocaa and Isolina Oliveirab aInstitute of Education, University of Lisbon, Alameda da Universidade, Lisbon, Portugal; bLE@D, Universidade Aberta, Lisbon, Portugal (Received 6 June 2012; ﬁnal version received 17 October 2012) aInstitute of Education, University of Lisbon, Alameda da Universidade, Lisbon, Portugal; bLE@D, Universidade Aberta, Lisbon, Portugal (Received 6 June 2012; ﬁnal version received 17 October 2012) aInstitute of Education, University of Lisbon, Alameda da Universidade, Lisbon, Portugal; bLE@D, Universidade Aberta, Lisbon, Portugal (Received 6 June 2012; ﬁnal version received 17 October 2012) (Received 6 June 2012; ﬁnal version received 17 October 2012) This case study looked at how an assessment strategy designed for an online learning environment can support teachers’ professional development. More spe- ciﬁcally, we intended to evaluate how a particular online assessment design can help the participating in-service teachers to recognize the added value of forma- tive assessment, and promote their own use of formative assessment in their pro- fessional classroom practices. The presented assessment design consists of a combination of different, non-standard assessment methods in an online environ- ment. We analysed data from 494 questionnaires, the participants’ critical reﬂec- tions about their learning and the participants’ produced artefacts. The ﬁndings illustrate the participants’ recognition of the formative character of the proposed assessment design, reﬂected not only by the high scores reported in seven of the eight themes explored, but also from the qualitative analysis of the participants’ reﬂections and artefacts. Suggestions are offered for the improvement of the pro- posed assessment design, so as to better promote the formative character of assessment. In particular, implications for the development of formative assess- ment in online professional development are discussed taking into account its potential to promote the participants’ self-regulatory learning processes. Keywords: in-service teacher education; professional development; formative assessment; assessment design 2012 Taylor & Francis Professional development and formative assessment The teaching career is framed as a continuous personal and professional development process, including not only the development of the teachers’ knowledge and compe- tences, but also of their personal self, with its associate beliefs, idiosyncrasies and life stories and the context where their activity takes place (Fullan & Hargreaves, 1992). In fact, teachers go through a professional development process that includes not only what they learn in formal settings, but also the knowledge that emerges from their own experiences and practices. This process includes a continuous dia- logue between theory and practice supported by critical reﬂection (Ponte, 1998). The teacher is recognized as a reﬂexive practitioner with previous experiences that build new knowledge from the reﬂection of his own practice. Professional development is, therefore, a complex process where a variety of facts take part such as the working contexts, training opportunities and individual characteristics (Fullan & Hargreaves, 1992; Lieberman, 1994; Ponte, 1998). To become a professional implies the expan- sion, development and reconstruction of our own knowledge and practices, demand- ing that the individual acts as a participant in his own learning process. Therefore, when we aim to promote teachers’ professional development, it is important to understand the process promoting teachers’ professional growth and the conditions that facilitate it. One of the main goals of professional development must be to help the teachers to revisit theories and practices (Tillema & Smith, 2009), always taking into account that their professional development is the corner- stone to assure the quality of learning of their students. In Portugal, the current legislation (Ministério da Educação, 2001) deﬁnes four dimensions for the teachers’ competence proﬁle: professional, social and ethical, development of teaching and learning, participating in school and relations with community and lifelong professional development. In particular, according to this last dimension, teachers should incorporate professional development into their pro- fessional practice as a cornerstone, building it from their identiﬁed needs, and through the problematized analysis of their pedagogical practice, and the reﬂection of the development of their profession with the resource to research about their practice, with the cooperation of other professionals. Teachers are required to be creative and reﬂexive practitioners with the ability to develop knowledge independently and solve complex problems. This represents a new paradigm for teaching where it is necessary to promote learning environ- ments capable of sustaining deep learning anchored in real contexts. Introduction With the development of Information and Communication Technologies and the emergence of web 2.0, teaching has increasingly resourced to technological tools for content production and distribution, as well as computer-mediated communica- tion. The e-learning explosion and the emergence of new digitally supported learn- ing environments emphasize the necessity of rethinking current teaching and learning perspectives. If properly used, these new technologies can promote the transformation of teaching practices towards a student-centred perspective, and so add value to the learning process. This new learning culture, discussed by Garrison and Anderson (2003), McCon- nell, (2006), Pereira et al. (2009), and Anderson and Dron (2011), where technolog- ical mediation is the cornerstone, is characterized as being fundamentally collaborative and emphasizing the importance of multiple perspectives, attributing 215 Teachers and Teaching: theory and practice to the learner a central role in his own learning process, and accentuating the formative nature of assessment. In this article, we introduce an assessment design for an online teacher profes- sional development context. This assessment design is framed by a conceptual framework for formative digital assessment (Pereira et al., 2009; Pereira, Oliveira, & Tinoca, 2010). Moreover, we intended to evaluate the participating teachers’ per- ception of the proposed assessment design and its perceived impact on their profes- sional development. Professional development and formative assessment Taking into account these new settings, assessment must also go beyond the measurement of knowledge reproduction which emphasizes the necessity for a new assessment design. Furthermore, this paradigm shift is illustrated by Elwood and Klenowski (2002), when they distinguish ‘assessment of learning’ as assessment focused on 216 L. Tinoca and I. Oliveira measurement and scaling, from ‘assessment for learning’ as assessment meant for the students, through feedback, to understand their own learning processes and the goals that they intend to achieve. measurement and scaling, from ‘assessment for learning’ as assessment meant for the students, through feedback, to understand their own learning processes and the goals that they intend to achieve. This new assessment culture is characterized by: • Seamless integration of assessment with the teaching progress (Birenbaum, 1996) • Student participation in the development of his own assessment in dialogue with the teacher (McConnell, 2006) • Assessing both products and processes (Linn, Baker, & Dunbar, 1991) • Non-standardized assessment methods associated with teaching practices (Dierick & Dochy, 2001) • Using a variety of assessment methods similar to real/professional life con- texts (Herrington & Herrington, 1998; Resnick, 1987) ( g g , ; , ) • Emphasis on student reﬂection about their learning (McConnell, 2006) Downloaded by [79.169.70.250] at 04:00 30 October 2014 • Valuing a qualitative description of performance in detriment of a quantitative classiﬁcation (Birenbaum, 1996) As Sainsbury and Walker (2007) emphasize, it is necessary to take an approach which incorporates collaboration into a wider range of assessments, and which pro- vides useful timely feedback, and thus has the potential to harness the motivating force of assessment into the effective promotion of learning during the assessment process itself. Nicol and Macfarlane (2006, p. 199) consider feedback as the infor- mation student’s receive about how their present state relates to proposed goals and criteria. Students create internal feedback when they are capable of monitoring their development of the learning tasks, and assess their own progress towards the pro- posed goals. For these authors, this concept includes not only the cognitive process, but is also regulated by motivational beliefs. As these authors emphasize, ‘formative assessment and feedback should be used to empower students as self-regulated learners’. According to Hattie and Timperley (2007, p. 8) ‘feedback is conceptualized as infor- mation provided by an agent (e.g. teacher, peer, book, self, experience) regarding aspects or one’s performance or understandings’. Assessment design in an online environment In teacher professional development in an online environment, the assessment design should aim to promote the development and use of metacognitive skills. As Vovides, Sanchez-Alonso, Mitropoulou, and Nickmans (2007, p. 64) emphasize, ‘e-learning environments can have potential added learning beneﬁts and can improve students’ and educators’ self-regulation skills, in particular their metacogni- tive skills’. Our assessment design was developed taking into account the quality of assessment in an online environment, and emphasizing the four dimensions pro- posed by Pereira et al. (2010): authenticity, consistency, transparency and practica- bility. Authenticity is conceptualized as the need to warrant that online assessment tasks are complex, related to real-life contexts and recognized as signiﬁcant by stu- dents, teachers and employers; consistency stresses the importance of aligning the competences being assessed with the assessment strategies being used and the assessment criteria, as well as the need to use a variety of indicators; transparency is related with participant engagement in online tasks through the democratization and visibility of the assessment strategies being used. What is more, assessment in an online environment refers to practices that may provide several beneﬁts, such as: immediate feedback, greater variety and authenticity in assessment designs, improved learner engagement (for example through interactive formative assess- ments with adaptive feedback) and increased opportunities for learners to act on feedback (for example by encouraging reﬂection in online forums). ( p y g g ) Within this context, our assessment design consists of a combination of non- standard assessment methods, including alternative forms of assessment: the e-folio and the p-folio. The e-folio ‘is a short digital document elaborated by the student and published online to be visualized by the teacher, and should clearly demonstrate that the student acquired or developed a given competence’ (Pereira, Mendes, Morgado, Amante, & Bidarra, 2007, p. 19). The e-folios are complemented by a p-folio that takes place in a face-to-face setting. Moreover, these new assessment strategies are framed by a pedagogical model for online learning (Pereira et al., 2007) supported by a curricular unit plan (CUP) and a formative activities plan (FAP). The CUP is a document which is presented at the beginning of the course/ curricular unit and guides the learning process. It is accompanied by an online forum where the participants can question and debate its contents. Professional development and formative assessment Without feedback, participants lack the opportunity to frame their learning and development. Feedback can also be insuf- ﬁcient to enhance further learning, such as when only a numerical classiﬁcation is provided, or when received too long after the assessment task. These authors present four levels of feedback: (i) task level, feedback related with the task interpretation or with the attained results; (ii) process level, focused on the processes necessary for the task development, such as giving cues that lead to better strategies; (iii) self-regu- lation level, related to self-monitoring, directing, and regulating of actions; and (iv) self-level, centred on personal evaluations and their effect on the learner. Also, in online contexts, we can ﬁnd all these levels of feedback; such as when during an online forum the proposed activities are discussed and clariﬁed, or when, at the end of an assessment task, the lecturer provides the participant with feedback about his or her performance and suggestions for improvement. Furthermore, the cognitive gains fuelled by feedback may enable reﬂection and an expansion of the knowledge to other situations, as well as bigger metacognitive awareness while promoting self- assessment and self-awareness (Herrington & Herrington, 1998). 217 Teachers and Teaching: theory and practice Assessment design in an online environment This document clariﬁes aspects about the course, such as the contents, the objectives and the com- petences to develop, the formative activities plan, the assessment methods and crite- ria as well as the teacher’s expectations regarding the in-service teachers’ participation. Furthermore, by creating the conditions for student’s self-regulatory processes through self-monitoring and self-evaluation strategies, we are promoting the development of their own self-efﬁcacy (Schunk & Pajares, 2001; Zimmerman, 2000). The aim of the FAP is to potentiate the formative processes of this model by guiding the participants’ exploration of the available resources, prompting and sup- porting their interaction in the online forums. The FAP requires that the participants complete a set of tasks proposed by the lecturer, such as a research assignment or developing an artefact, and consecutively share their experiences in the online for- ums with their colleagues. In this way, promoting critical conversations about their experiences where students ‘co-construct their professional selves through interac- tions with informed and involved partners/mentors invested in their development’ 218 L. Tinoca and I. Oliveira (Carter, 2005, p. 485). Consequently, the FAP helps participants to identify strengths and weaknesses in their learning and to reﬂect upon these aspects while interacting with their peers and with the lecturer. Therefore, the FAP is a corner- stone to the development of the in-service teachers’ critical metacognitive skills such as planning, self-regulation and self-evaluation, to manage the learning process and engage in online inquiry (Vovides, Sanchez-Alonso, Mitropoulou, & Nickmans, 2007). Figure 1 illustrates the general learning path under the proposed model. This learning path is framed by the presented CUP and supported by several moments of interaction and activities throughout the semester. This model is characterized by cyclical modes of interaction where the participants engage in online forums, com- plete the proposed FAP, and culminate in a formal assessment activity – the e-folio and the p-folio. These cyclical modes of interaction should be complemented by individualized and reﬂective learning. Even though each of the available forums is scheduled to take place during certain periods/weeks, they are always available online and can be revisited by the participants as often as necessary, and at any moment throughout the semester. The e-folios include instructions detailing the steps to be taken in order to com- plete the proposed task, the activity’s assessment criteria and instructions related to formal aspects (such as online submission procedures). Assessment design in an online environment The e-folios can last from a few days to a full month depending on the nature of the required task. Pereira et al. (2009) presented a summary of the types of tasks usually required in the e-folios (see Table 1). It is important to notice that some of the e-folios required more than one task (for example an essay and a practical exercise). As mentioned before, the e-folios are complemented by a p-folio that takes place in a face-to-face setting. The p-folio was introduced as a crucial tool for the certiﬁcation of the assessment process as a face-to-face moment where the partici- pants are required to be present at an institutional regional centre, taking the form Figure 1. An illustration of the proposed learning path. Figure 1. An illustration of the proposed learning path. Teachers and Teaching: theory and practice Teachers and Teaching: theory and practice 219 Table 1. E-folios tasks and deﬁnitions. Task Deﬁnition Essay writing composition, reﬂection, analysis Practical exercise application, demonstration Construction of an artefact blog, pictures, poster, video, slideshow Critical comment stating an opinion Project designing a planned action Report of ﬁeldwork, of experimental work Reading review reading guide summarizing the main ideas in a text/excerpt Table 1. E-folios tasks and deﬁnitions. of a written examination, presentation of a project or a report. Both the e-folios and p-folios applied in combination ‘provide evidence of the extent to which the student achieved the competencies expected in a given curricular unit’ (Pereira et al., 2007, p. 20). Moreover, students are required to achieve at least 50% of the total score on the sum of the e-folios in order to be allowed to go on and complete the p-folio at the end of the semester. According to the presented model, the p-folio must repre- sent 60% of the ﬁnal grade. Downloaded by [79.169.70.250] at 04:00 30 October 2014 With this study, we aim to illustrate how an assessment design for online learn- ing, composed by a combination of different non-standard assessment methods, can be used to support in-service teachers’ professional development. The proposed research questions are: how useful do teachers ﬁnd the proposed assessment design for online learning for their professional development? In particular, what is the for- mative potential of the FAP and the online forums? And how do the e-folio tool and the associated feedback contribute to the teachers’ professional development? The context During the last decade, due the broadening offer in Portuguese K-12 education, a large amount of teachers have entered the educational system, from a variety of sci- entiﬁc areas, without formal pedagogical teacher training. Universidade Aberta, being a public distance teaching university, developed an online professional devel- opment programme for teachers with a duration of two semesters, and composed of a set of seven courses: Education and Society, Conﬂict Management, Practice and ICT, Pedagogical Assessment Models, Didactical Principles, Ethics and Education and the Practicum Seminar, for a total of 40 ECTS. The present study was devel- oped in two of these courses: ‘Practice and ICT’ and ‘Pedagogical Assessment Models’; chosen because of their particular emphasis on assessment practices and on the use of ICT in education. Instruments For the present study, we used three main data-collection strategies: a large-scale anonymous online questionnaire to be completed at the end of the semester; the participating teachers’ critical reﬂections about the assessment strategies used; and their productions throughout the semester. Participants This programme was developed in an online learning environment during the 2010/ 2011 academic year and enroled 630 in-service teachers organized in 13 virtual classes. Each course was chaired by a lecturer responsible for the entire instructional design and assessment, and supported by a team of 5–7 online tutors. The in-service teachers had been developing their teaching practice for, at least, ﬁve L. Tinoca and I. Oliveira 220 years, but some of them had been teaching for as long as 10 or 12 years, and were teaching at the 5–12th grade levels. The participating teachers represented a variety of scientiﬁc areas such as Music, Science, Mathematics, Portuguese, English, Geog- raphy, History and Computer Science. Procedure The two authors worked collaboratively as the lecturers in charge of the two chosen courses. Given the framework established by the institutional pedagogical model, the proposed assessment design was developed trying to capitalize on the potential of the available resources. Namely, the use of the online forums and of the forma- tive activities plan, as well as the different types of feedback associated with each activity. The e-folios and p-folios were regarded as complementary, and required the participating teachers the development of different types of tasks. In particular, the e-folios required the teachers to construct classroom artefacts and critically comment on their application. The required artefacts included the construction of alternative assessment tools and rubrics such as blogs, e-portfolios and wikis. The p-folio took the form of a written examination where the participating teachers were required to reﬂect about their learning process throughout the course and its relationship with their professional practice. At the end of the semester, the partici- pants were requested to ﬁll out an evaluative questionnaire, and to write a critical reﬂection about their experiences throughout the courses. Downloaded by [79.169.70.250] at 04:00 30 October 2014 Artefacts The participants’ productions throughout the semester, including the interactions in the online forums, the e-folios and the p-folios, were also collected in order to illus- trate their reﬂective and metacognitive skills and accomplishment of the proposed competences. Deﬁnition Downloaded by [79.169.70.250] at 04:00 30 October 2014 observed between the two chosen courses. All themes included three to four items. Examples of items are: ‘After the feedback from the ﬁrst e-folio I improved my study plan’ (feedback); ‘the participants interaction in the online forums was very productive’ (online forums interactions); and ‘the proposed assessment activities where adequate to the competences that we needed to develop in this course’ (transparency). observed between the two chosen courses. All themes included three to four items. Examples of items are: ‘After the feedback from the ﬁrst e-folio I improved my study plan’ (feedback); ‘the participants interaction in the online forums was very productive’ (online forums interactions); and ‘the proposed assessment activities where adequate to the competences that we needed to develop in this course’ (transparency). Critical reﬂections Also at the end of the semester, the participants were requested to write a critical reﬂection about the most and least positive aspects of the course, taking into partic- ular account the proposed assessment strategies, as well as any suggestions for improving the course. The course evaluation questionnaire The online questionnaire was adapted from Pereira (2009) taking into account the proposed assessment design, including ﬁve sections focused on the proposed con- tents, the pedagogical methods, the learning materials and the assessment process. The survey included 40 questions with a four-level likert scale, ranging from totally disagree, to totally agree. The questionnaire was developed with eight main themes: contents, digital learning resources, formative activities plan, online forums interac- tions, regulatory processes, transparency, e-folios and feedback. These themes were developed taking into account not only the pedagogical strategies inherent to this pedagogical model (Pereira et al., 2007), but also the presented framework for digi- tal assessment (Pereira et al., 2010). In order to validate the developed themes, for this new population, the Cron- bach’s alpha index was calculated for each theme. Table 2 presents a description of the proposed themes and the associated Cronbach’s alpha indexes. The internal con- sistency of all themes was high (Cronbach’s alpha larger than .75) illustrating the reliability of the proposed themes. The total number of responses to the survey was of 265 and 229, respectively, for both studied courses, giving us a total N = 494. No signiﬁcant differences were Teachers and Teaching: theory and practice 221 Table 2. Dimensions of analysis. Themes Deﬁnition Cronbach’s alpha Contents The contents studied in each course .810 Digital learning resources All online learning resources (books, articles, videos, blogs, wikis, etc.) .757 Formative activities Set of tasks presented by the instructor including guidelines towards possible answers .828 Online forums interactions Discussions on the online forums .805 Regulatory processes Analysis and reﬂection aimed at the improvement of one’s own learning and practice .891 Transparency Adequacy and justice of the assessment practices .894 e-folios A short digital document elaborated by the student and published online .844 Feedback Information given to the participants about the e-folio to perform it more effectively and to improve their learning and practice .862 Analysis This research is based on qualitative case study methods as described by Yin (2003) and Stake (1995). This case study is supported by a mixed methods approach (Cohen, Manion, & Morrison, 2000; Mercer, Littleton, & Wegerif, 2009) aimed at producing a description of the proposed assessment design. This account was based on several triangulated sources of data, including written artefacts of the participants’ work, an online survey of the in-service teachers’ perceptions and experiences and transcripts from the online discussion forums. 222 L. Tinoca and I. Oliveira The course evaluation questionnaire was analysed as follows. For each theme, the mean score and standard deviation were calculated. In order to facilitate interpretation, the results were converted into a percentage scale. With regard to the scores, we made a distinction between low (0–33), medium (34–66) and high (67–100) scores (Baartman, Gulikers, Dijkstra, & Blankert, 2011). Moreover, to get an impression of which categories got high and low scores, the overall mean score for all categories was calculated. Since the normal distribution and homogeneity of variance requirements were veriﬁed, one-sample Student’s t-tests were used to compare the scores on the eight themes to this overall mean score. The analysis of both the participants’ online postings and their critical comments involved: (a) reading and questioning the data taking into account the posed research questions; (b) identiﬁcation of the main concepts in accordance with deﬁned themes; and (c) subsequent reﬁnement of the analysis. Coding was con- ducted by both researchers, and any differences were further discussed allowing intercoder reliability to reach 93%. The e-folios and p-folio were graded according to an analytical rubric developed by the course lecturers. Formative activities and online discussion forums The formative activities and the online discussion forums represent the bulk of the work done by the participants throughout the semester. Here, we can notice two clearly different reactions from the participants. Concerning the formative activities plan, the participants recognize their value, as can be seen when Carla says ‘in my opinion the formative activities plan contributed for me to be able to bring together practice and theory’ or when Pedro commented ‘in our profession practice is so important, that it really helped to have the theory being accompanied by the forma- tive activities’. However, as far as the online discussion forums were concerned, the reactions were clearly less positive. Teresa recognizes this saying that ‘the lack of participation in the forums was clearly the point that I want to highlight as less positive’. It should be noted that in the used pedagogical model, the participation in these online forums is voluntary and not assessed. Nevertheless, and in spite of this less positive result, those participants who did interact in the forums report a posi- tive opinion, such as Carlos when he says ‘I felt the forums were important and allowed me to clarify and explore the questions I had throughout the semester’ or Vanda when she comments ‘the tutors always accompanied our work (in the for- ums) and clariﬁed all questions when necessary’. Contents and digital learning resources Concerning the themes related to the proposed contents and digital resources explored in these two courses, it can be observed that the teachers have clearly positive reactions to the proposed contents and to the variety of available resources. This can be further emphasized from their reactions on the critical commentary, such as when Ana says ‘the proposed contents and the way they were tackled was of tremendous importance’; or when Susana commented that ‘it was very positive to have such a wide range of resources available, and also to be able to go even further and research new material beyond the suggested resources’. However, even though most teachers recognized the relevance of the proposed contents for their practice, such as when Carlos stated ‘the contents studied here were of great per- sonal and professional interest, giving me the opportunity to implement new assess- ment and teaching methods’, there were also some, like Patricia, who felt that the contents were too theoretical saying ‘I ﬁnd this course too theoretical, given its rele- vance for classroom practice’. Downloaded by [79.169.70.250] at 04:00 30 October 2014 Results Taking into account that the used virtual pedagogical model requires participants to achieve a minimum of 50% in their set of e-folios, in order to be allowed to con- tinue their assessment path towards the p-folio, we should start by stressing that 98% of the participants were allowed to continue towards the p-folio. Moreover, the overall mean score in the e-folios was of 80%, illustrating the participants’ high level of achievement of the proposed competences, in particular, the construction of complex classroom artefacts and the development of a coherent discourse sustained in their professional practice. Table 3 shows the mean scores and standard deviations for each of the proposed themes. From this analysis, two categories stood out as signiﬁcantly different from the overall mean score (M = 73): the e-folios (t(485) = 120.35, p < .01) and the online forums interactions (t(493) = 102.35, p < .01). Three different levels clearly emerge from the presented data. Firstly, the e-folio was clearly the theme with the highest score, presenting a mean of 83%, illustrating the considerable value attributed to it by the participating teachers. Secondly, there were a set of six themes with high scores (more than 67%) but clearly indistinguishable between them including: contents, digital learning resources, formative activities, feedback, Table 3. Mean score and standard deviation for the 8 established dimensions. Dimensions M (%) SD (%) Themes (4 items) 74 14 Digital learning resources (4 items) 74 13 Formative activities (3 items) 73 16 Online forums interactions (4 items) 64⁄ 14 e-folios (4 items) 83⁄ 15 Feedback (3 items) 71 18 Transparency (3 items) 72 15 Regulatory processes (4 items) 73 14 Note: ⁄p < 01 Table 3. Mean score and standard deviation for the 8 established dimensions. Teachers and Teaching: theory and practice 223 transparency and regulatory processes. And ﬁnally, there was one theme – online forums interactions – with only a medium score (64%). It should also be noted that the standard deviation for the feedback theme was slightly larger than average. Possible explanations for this are presented later. Next, we present these eight themes in four groups of two each and illustrate their interpretation with the participating teachers’ critical reﬂections about the proposed assessment design. e-folio tools and feedback Clearly, the most positive result from this study was the participants’ very positive reaction towards the e-folio tool (83%), which also contributed to the very good L. Tinoca and I. Oliveira 224 attained results. First of all, earlier research has shown that, even though the imposed pedagogical model advocates for the use of this tool, a variety of interpre- tations may exist with regard to what can be requested from the students (Pereira et al., 2009). However, in this case, the researchers/lecturers adopted a similar peda- gogical approach and structure for the proposed e-folios in the two studied courses. The formal structure was similar and included four sections (even though not always in the same order): (1) introduction and presentation of the required tasks in a clear and objective tone; (2) description of the competences being assessed; (3) presentation of the assessment and scoring criteria; and (4) formal speciﬁcations about the electronic submission of the ﬁnal product. Also, both lecturers designed e-folios where the participants were given one week to analyse or develop an arte- fact relevant for the course and their professional practice (such as an assessment tool or a blog). The positive reactions towards the e-folios are also clearly illustrated by the par- ticipants’ comments such as when Rita said that ‘the e-folios, despite the time they took, were very important, because they allowed me to tackle very important con- tents for my practice in a very stimulating way’ or when Carolina stated that ‘the completion of the e-folios was a very effective way to motivate us to research and further our knowledge’. However, the feedback not only particularly associated with the e-folios, but also throughout the other activities during the semester, even though having a posi- tive reaction from the participants (71%), was not at the same level as the e-folios themselves. This is illustrated by the participants’ critical comments, such as when Rafael recognizes that ‘the feedback to our e-folios represented guidance for our study and contributed to my self-assessment’. Nonetheless, it should also be pointed out that this was the theme with the highest standard deviation. e-folio tools and feedback This may be due to the large number of participants per class, and the very good results attained in the e-folios meant that the lecturers opted for only providing short reinforcing feedback to those participants who had done very well with their tasks, that ended up repre- senting a signiﬁcant portion of the group. Discussion The results of this study support the proposed assessment design’s capacity to enhance in-service teachers’ professional development, contributing to the emer- gence of critical reﬂection centred on their professional practice. This is partially promoted, as Carter (2005) argues, by the opportunities given to the teachers to share and discuss, in the online forums, their tacit knowledge with their colleagues, coming from a variety of geographical locations throughout the country (both rural and urban) and with a very diverse set of professional experiences. As Fullan and Hargreaves (1992) and Ponte (1998) stress, teachers’ professional development is a complex individual and collective process, dependent on professional contexts and training opportunities. In this study, the teachers’ attention was further focused on the learning promoted by the assessment task (e-folio) and accompanying feedback by the lecturer, even more than on the online debates. The e-folio was the tool towards which the participants showed the most positive reactions, particularly rec- ognizing its impact on their formative assessment and regulatory processes (Nicol & Macfarlane, 2006). The lecturers develop e-folios where the in-service teachers were asked to complete open-ended, unstructured, tasks related to their professional life, adapted to their own classroom practice, and requiring them to analyse and reﬂect about its classroom impact, promoting the development of their own self-efﬁ- cacy (Zimmerman, 2000). In such a way, the in-service teachers were granted with the opportunity of incorporating training as a central piece of their professional development, focused on their perceived needs, and on the analysis and reﬂection on their own practices (Ministério da Educação, 2001). The participants’ comments, and positive reactions towards the e-folio tool, strongly suggest its recognition as an authentic activity, aligned with their professional practice (Herrington & Herrigton, 1998). Moreover, the clariﬁcation of the assessment criteria allowed the in-service teachers to interpret and answer the assessment task (e-folio) taking into account their appropriation of knowledge. The lecturer’s work was aligned with the proposed criteria, and the feedback was provided not only as a ﬁnal score but also as written commentary, targeted at the self-level feedback centred on the partici- pants’ productions (Hattie & Timperley, 2007). On the other hand, the feedback provided during the forum interactions that is more aligned with the task-level and the process-level feedbacks, has not been explored in this study in its fullest poten- tial. Transparency and regulatory processes According to our perspective, transparency contributes to the formative nature of assessment by clearly making available to the participants all information regarding how and when the assessment tasks are going to take place, and about the assess- ment and scoring criteria. In this case, this was attained through the information provided in the PUC, on the e-folio instructions and, when requested, on the accompanying forums. Also, we were particularly interested in evaluating our assessment design’s contribution to the participants’ regulatory processes and asso- ciated formative nature of the proposed assessment tasks. The achieved results in these themes were quite satisfactory (72 and 73%) representing the participants’ rec- ognition of these features in their courses. This can be further illustrated by the par- ticipants’ critical comments such as when Duarte says: I had a clear idea (since the beginning) of what the assessment criteria were and about how my classroom practice could improve towards as much transparency as possible … it’s curious that something so evident is often the exception instead of the rule. 225 Teachers and Teaching: theory and practice Discussion This is clearly a scenario where the potential for assessment as communication between the in-service teachers and lecturer can be further explored. As Tillema and Smith (2009) suggest, teachers often ﬁnd it hard to expose their reasoning and, in this context, this may also be due to the participants’ reported lack of time, and with their difﬁculty to assume a more participatory role than what they are used to. Even so, as McConnell (2006) points out, the participants who are unable or unwill- ing to fully participate in these online forums, and adopt a more ‘lurking’ attitude, can still beneﬁt from the visibility of their colleagues online discussions. Furthermore, we consider that the different types of feedback and the interac- tions in the online forums are crucial in the proposed assessment design, as reported Furthermore, we consider that the different types of feedback and the interac- tions in the online forums are crucial in the proposed assessment design, as reported by the in-service teachers. However, these characteristics represent an increased workload for the lecturers, responsible for online classrooms with up to 60 partici- pants, and coordinating a total of 13 classes. This problem was also discussed by Nicol and Macfarlane (2006) and can be one of the reasons for the increased vari- ance of the participating teachers’ perceptions regarding the provided feedback. 226 L. Tinoca and I. Oliveira Implications One of the main contributions this study intends to make is promoting the quality of assessment in general and its capability to promote the formative character of assessment, particularly in the context of online professional development. Here, the construction of the assessment strategy was designed under the framework of the pedagogical model being used, but taking into careful consideration the authenticity and transparency of the proposed strategies and their formative impact on the teach- ers’ practice. However, there are other dimensions contributing to the quality of assessment that should be further explored and introduced into the development of online assessment strategies. Pereira et al. (2010) proposed a new conceptual frame- work for assessment in an online environment based on four dimensions: authentic- ity, consistency, transparency and practicability; and further developed into 15 criteria. In order to enhance the quality of this assessment design, these dimensions and criteria should be incorporated in the development of the proposed assessment strategies. Downloaded by [79.169.70.250] at 04:00 30 October 2014 Furthermore, considering the proposed assessment design, two main implications should be emphasized. On the one hand, the quality of the online forum discussions should be reviewed. In online distance education, the online forums are a pivotal tool to mediate the participants’ interaction processes (McConnell, 2006). In the present study, it was clear that this was a tool that did not beneﬁt all participants equally. Also, the size of the virtual classroom should be reviewed, as a smaller classroom may promote greater commitment and enhance the participants’ interac- tion. On the other hand, the potential of the e-folio tool must be emphasized. If designed in accordance with the suggested criteria, it has a clear potential to pro- mote the formative character of assessment, serve as a motivator for the participants learning and as a valuable source of feedback for their regulatory processes. Anderson, T., & Dron, J. (2011). Three generations of distance education pedagogy. The International Review of Research in Open and distance Learning. Special Issue – Con- nectivism: Design and Delivery of Social Networked Learning. http://www.irrodl.org Baartman L K J Gulikers J Dijkstra A & Blankert H (2011) Self evaluation of assess References Anderson, T., & Dron, J. (2011). Three generations of distance education pedagogy. The International Review of Research in Open and distance Learning. Special Issue – Con- nectivism: Design and Delivery of Social Networked Learning. http://www.irrodl.org Baartman, L.K.J., Gulikers, J., Dijkstra, A., & Blankert, H. (2011). Self-evaluation of assess- ment quality in higher vocational education: Assessment quality, points for improvement and students’ involvement. Proceedings of the EARLI 2011 Conference, Exeter, UK. Birenbaum, M. (1996). Assessment 2000: Towards a pluralistic approach to assessment. In M. Birenbaum & F.J.R.C. Dochy (Eds.), Alternatives in assessment of achievement, learning processes and prior knowledge (pp. 3–29). Boston, MA: Kluwer Academic. Carter, D. (2005). Distributed practicum supervision in a managed learning environment (MLE). Teachers and Teaching: Theory and Practice, 11(5), 481–497. ( ) g y ( ) Cohen, L., Manion, L., & Morrison, K. (2000). Research Methods in Education. London: Routledge Falmer. g Dierick, S., & Dochy, F.J.R.C. (2001). New lines in edumetrics: New forms of assessment lead to new assessment criteria. Studies in Educational Evaluation, 27, 307–329. Elwood, J., & Klenowski, V. (2002). Creating communities of shared practice: The chal- lenges of assessment use in learning and teaching. Assessment & Evaluation in Higher Education, 27(3), 243–256. ( ), Fullan, M., & Hargreaves, A. (1992). Teacher education and educational change. In M. Fullan & A. Hargreaves (Eds.), Teacher Education and Educational Change (pp. 1–9). London: Falmer. arrison, R., & Anderson, T. (2003). E-Learning in the 21st century: A framework for research and practice. London: Routledge/Falmer. Teachers and Teaching: theory and practice 227 Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. ( ) Herrington, J., & Herrington, A. (1998). Authentic assessment and multimedia: How univer- sity students respond to a model of authentic assessment. Higher Education Research & Development, 17(3), 305–322. p ( ) Lieberman, A. (1994). Teacher development: Commitment and challenge. In P. Grimmett & J. Neufeld (Eds.), Teacher development and the struggle for authenticity: Professional growth and restructuring in the context of change (pp. 15–30). New York, NY: Teachers College Press. g Linn, R.L., Baker, E.L., & Dunbar, S.B. (1991). Complex, performance-based assessme Expectations and validation criteria. Educational Researcher, 20(8), 15–21. McConnell, D. (2006). E-learning groups and communities. Berkshire: Open University Press. Mercer, N., Littleton, K., & Wegerif, R. (2009). Methods for studying the processes of interaction and collaborative activity in computer-based educational activities. In K. Kumpulainen, C.E. References Hmelo-Silver, & M. César (Eds.), Investigating Classroom Interaction (pp. 27–42). Rotterdam: Sense. Ministério da Educação (Ministry of Education) (2001). DL 240/2001 Perﬁl geral de dese- mpenho do educador de infância e dos professores dos ensinos básico e secundário [General performance proﬁle for kindergarten, elementary and secondary school teach- ers]. Diário da República, 201(I-A), 5570–5572. ] p ( ), Nicol, D.J., & Macfarlane-Dick, D. (2006). Formative assessment and self-regulated learn- ing: A model and seven principles of good feedback practice. Higher Education, 31(2), 199–218. Pereira, A. (2009). Interacção entre estudantes online: estratégias adequadas a grandes Números [Online student interactions: Strategies for dealing with large groups]. Proceed- ings of the XIII Encontro Iberoamericano de Educação Superior a Distância. Lisboa, Por- tugal 16/18 September 2009. Fundação Calouste Gulbenkian. g p Pereira, A., Mendes, A.Q., Morgado, L., Amante, L., & Bidarra, A. (2007). Universidade aberta’s pedagogical model for distance education: A university for the future. Lisboa: Universidade Aberta. Pereira, A., Oliveira, I., & Tinoca, L. (2010). A cultura de avaliação: que dimensões? [The assessment culture: Which dimensions?] Proceedings of the TICeduca 2011, University of Lisbon. Pereira, A., Oliveira, I., Tinoca, L., Amante, L., Relvas, M.J., Pinto, M.C., & Moreira, D. (2009). Evaluating continuous assessment quality in competence-based education online: the case of the e-folio. European Journal of Open, Distance and E-learning: EDEN. http://www.eurodl.org/ Ponte, J.P. (1998). Da formação ao desenvolvimento proﬁssional [From training to profes- sional development]. In APM (Ed.), Actas do ProfMat 98 (pp. 27–44). Lisboa: APM. i k ( ) i i h l d d l h ( ) p ] ( ) f (pp ) Resnick, L.B. (1987). Learning in school and out. Educational Researcher, 16(9), 13–20. S i b & lk A (200 ) A hi l f l i di Sainsbury, E.J., & Walker, R.A. (2007). Assessment as a vehicle for learning: Extending col- laboration into testing. Assessment & Evaluation in Higher Education, 33(2), 103–117. Schunk, D.H., & Pajares, F. (2001). The development of academic self-efﬁcacy. In A. Wig- ﬁeld & J. Eccles (Eds.), Development of Achievement Motivation. San Diego, CA: Academic Press. Stake, R.E. (1995). The art of case study research. London: Sage. Tillema, H., & Smith, K. (2009). Assessment orientation to formative assessment of learning to teach teachers and teaching: theory and practice. Teachers and teaching: Theory and practice, 15(3), 389–402. p ( ), Vovides, Y., Sanchez-Alonso, S., Mitropoulou, V., & Nickmans, G. (2007). References The use of e-learning course management systems to support learning strategies and to improve self-regulated learning. Educational Research Review, 2, 64–74. g g , Yin, R.K. (2003). Case study research design and methods (3rd ed.). Thousand Oaks, CA: Sage. g Zimmerman, B.J. (2000). Self-efﬁcacy: An essential motive to learn. Contemporary Educational Psychology, 25, 82–91.
https://openalex.org/W4384108471	https://pubs.rsc.org/en/content/articlepdf/2023/ob/d3ob90104a	English	null	Contents list	Organic & biomolecular chemistry	2,023	cc-by	3,583	An international journal of synthetic, physical and biomolecular organic chemistry View Article On An international journal of synthetic, physical and biomolecular organic chemistry rsc.li/obc View Article Online View Journal \| View Issue The Royal Society of Chemistry is the world's leading chemistry community. Through our high impact journals and publications we connect the world with the chemical sciences and invest the profits back into the chemistry community. ISSN 1477-0520 CODEN OBCRAK 21(27) 5483–5660 (2023) Cover See Sudhakar Gaikwad, Susnata Pramanik et al., pp. 5492–5505. Carbohydrate recognition using metal–ligand assemblies Raﬁq Ahamed, Jayashree Venkatesh, Rakshantha Srithar, Sudhakar Gaikwad* and Susnata Pramanik* This journal is © The Royal Society of Chemistry 2023. Apart from fair dealing for the purposes of research or private study for non-commercial purposes, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988 and the Copyright and Related Rights Regulation 2003, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the Publishers or in the case of reprographic reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. US copyright law is applicable to users in the USA. Organozincs for versatile synthesis of non-symmetric azoarenes Editorial Staff Executive Editor Rebecca Garton Deputy Editor Jack Washington Development Editor Daniel Robertshaw Editorial Production Manager Sarah Whitehouse Publishing Editors Nicola Burton, Tom Cozens, Katie Fernandez, Ryan Kean, Roxane Owen, Alex Rowles Editorial Assistant Amy Cook Publishing Assistant Andrea Whiteside Publisher Sam Keltie For queries about submitted papers, please contact Sarah Whitehouse, Editorial Production Manager in the first instance. E-mail: obc@rsc.org For pre-submission queries please contact Rebecca Garton, Executive Editor. Email: obc-rsc@rsc.org Organic & Biomolecular Chemistry (electronic: ISSN 1477- 0539) is published 48 times a year by the Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, UK CB4 0WF. All orders, with cheques made payable to the Royal Society of Chemistry, should be sent to the Royal Society of Chemistry Order Department, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, CB4 0WF, UK Tel +44 (0)1223 432398; E-mail orders@rsc.org 2023 Annual (electronic) subscription price: £5164; US$9267. Customers in Canada will be subject to a surcharge to cover GST. Customers in the EU subscribing to the electronic version only will be charged VAT. If you take an institutional subscription to any RSC journal you are entitled to free, site-wide web access to that journal. You can arrange access via Internet Protocol (IP) address at www.rsc.org/ip Customers should make payments by cheque in sterling payable on a UK clearing bank or in US dollars payable on a US clearing bank. Whilst this material has been produced with all due care, the Royal Society of Chemistry cannot be held responsible or liable for its accuracy and completeness, nor for any consequences arising from any errors or the use of the information contained in this publication. The publication of advertisements does not constitute any endorsement by the Royal Society of Chemistry or Authors of any products advertised. The views and opinions advanced by contributors do not necessarily Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Advisory Board Ramachary, University of Hyderabad, India 2023 Annual (electronic) subscription price: £5164; US$9267. Customers in Canada will be subject to a surcharge to cover GST. Customers in the EU subscribing to the electronic version only will be charged VAT. Ratmir Derda, University of Alberta, Canada Antonio Echavarren, Institute of Chemical Research of Catalonia, Spain Jian Zhou, East China Normal University, China Ben Feringa, University of Groningen, The Netherlands If you take an institutional subscription to any RSC journal you are entitled to free, site-wide web access to that journal. You can arrange access via Internet Protocol (IP) address at www.rsc.org/ip Paolo Scrimin, University of Padova, Italy Oliver Seitz, Humboldt University of Berlin, Paolo Scrimin, University of Padova, Italy Oliver Seitz, Humboldt University of Berlin, Germany Amar Flood, Indiana University Bloomington, USA Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03 This article is licensed under a Creative Commons Attribution 3.0 Un Christian Hackenberger, Leibniz-Institut für Molekulare Pharmakologie and Humboldt Universität zu Berlin, Germanyf Katrina Jolliffe, University of Sydney, Australia Motomu Kanai, University of Tokyo, Japan Katrina Jolliffe, University of Sydney, Australia Motomu Kanai, University of Tokyo, Japan Judy I-Chia Wu, University of Houston, USA Advisory Board Carmen Galan, University of Bristol, UK Jason Harper, University of New South Wales, Australia Elizabeth Krenske, University of Queensland, Australia Mahesh Lakshman, The City College of New York, USA Shih-Yuan Liu, Boston College, USA Geraldine Masson, Institut de Chimie des Substances Naturelles (CNRS), France Elizabeth New, University of Sydney, Australia Dhevalapally B. Organozincs for versatile synthesis of non-symmetric azoarenes Organozincs for versatile synthesis of non-symmetric azoarenes Zhili Duan, Shoucheng Dong and Jie Li* Org. Biomol. Chem., 2023, 21, 5485–5491 \| 5485 This journal is © The Royal Society of Chemistry 2023 View Article Online Editorial Staff Executive Editor Rebecca Garton Deputy Editor Jack Washington Development Editor Daniel Robertshaw Editorial Production Manager Sarah Whitehouse Publishing Editors Nicola Burton, Tom Cozens, Katie Fernandez, Ryan Kean, Roxane Owen, Alex Rowles Editorial Assistant Amy Cook Publishing Assistant Andrea Whiteside Publisher Sam Keltie For queries about submitted papers, please contact Sarah Whitehouse, Editorial Production Manager in the first instance. E-mail: obc@rsc.org For pre-submission queries please contact Rebecca Garton, Executive Editor. Email: obc-rsc@rsc.org Organic & Biomolecular Chemistry (electronic: ISSN 1477- 0539) is published 48 times a year by the Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, UK CB4 0WF. All orders, with cheques made payable to the Royal Society of Chemistry, should be sent to the Royal Society of Chemistry Order Department, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, CB4 0WF, UK Tel +44 (0)1223 432398; E-mail orders@rsc.org 2023 Annual (electronic) subscription price: £5164; US$9267. Customers in Canada will be subject to a surcharge to cover GST. Customers in the EU subscribing to the electronic version only will be charged VAT. If you take an institutional subscription to any RSC journal you are entitled to free, site-wide web access to that journal. You can arrange access via Internet Protocol (IP) address at www.rsc.org/ip Customers should make payments by cheque in sterling payable on a UK clearing bank or in US dollars payable on a US clearing bank. Whilst this material has been produced with all due care, the Royal Society of Chemistry cannot be held responsible or liable for its accuracy and completeness, nor for any consequences arising from any errors or the use of the information contained in this publication. The publication of advertisements does not constitute any endorsement by the Royal Society of Chemistry or Authors of any products advertised. The views This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Rapid publication of high quality organic chemistry research rsc.li/obc Organic & Biomolecular Chemistry is a weekly journal for the publication of highly significant original research and reviews in all areas of organic chemistry, including organic synthesis, physical organic chemistry, and organic aspects of supramolecular chemistry and chemical biology. Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Advisory Board Ramachary, University of Hyderabad, India Paolo Scrimin University of Padova Italy Kyo Han Ahn, Pohang University of Science and Technology, Korea Igor Alabugin, Florida State University, USA Gonçalo Bernardes, University of Cambridge UK Shunsuke Chiba, Nanyang Technological University, Singapore Andre Cobb, Kings College London, UK Steven Cobb, Durham University, UK Ratmir Derda, University of Alberta, Canada Antonio Echavarren, Institute of Chemical Research of Catalonia, Spain Ben Feringa, University of Groningen, The Jay Siegel, University of Zürich, Switzerland Corey Stephenson, University of Michigan, USA Dean Tantillo, University of California Davis, USA Mark Taylor, University of Toronto, Canada Georgios Vassilikogiannakis, University of Crete, Greece Helma Wennemers, ETH Zürich, Switzerland Peter Wipf, University of Pittsburgh, USA Shuli You, Shanghai Institute of Organic Chemistry, China Jian Zhou, East China Normal University, China Jay Siegel, University of Zürich, Switzerland Corey Stephenson, University of Michigan, USA Dean Tantillo, University of California Davis, USA Mark Taylor, University of Toronto, Canada Georgios Vassilikogiannakis, University of Crete, Greece Helma Wennemers, ETH Zürich, Switzerland Peter Wipf, University of Pittsburgh, USA Shuli You, Shanghai Institute of Organic Chemistry, China Jian Zhou, East China Normal University, China Jay Siegel, University of Zürich, Switzerland Corey Stephenson, University of Michigan, USA Jay Siegel, University of Zürich, Switzerland Corey Stephenson, University of Michigan, USA Jay Siegel, University of Zürich, Switzerland Corey Stephenson, University of Michigan, USA Dean Tantillo, University of California Davis, USA Mark Taylor, University of Toronto, Canada Georgios Vassilikogiannakis, University of Crete, Greece Helma Wennemers, ETH Zürich, Switzerland Peter Wipf, University of Pittsburgh, USA Shuli You, Shanghai Institute of Organic Chemistry, China Jian Zhou, East China Normal University, China Kyo Han Ahn, Pohang University of Science and Technology, Korea Kyo Han Ahn, Pohang University of Science and Technology, Korea All orders, with cheques made payable to the Royal Society of Chemistry, should be sent to the Royal Society of Chemistry Order Department, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, CB4 0WF, UK Igor Alabugin, Florida State University, USA Gonçalo Bernardes, University of Cambridge, UK Elizabeth Krenske, University of Queensland, Australia Shunsuke Chiba, Nanyang Technological University, Singapore Mahesh Lakshman, The City College of New York, USA Tel +44 (0)1223 432398; E-mail orders@rsc.org Andre Cobb, Kings College London, UK Steven Cobb Durham University UK Andre Cobb, Kings College London, UK Steven Cobb, Durham University, UK Shih-Yuan Liu, Boston College, USA Geraldine Masson, Institut de Chimie des Substances Naturelles (CNRS), France Elizabeth New, University of Sydney, Australia Dhevalapally B. Chair Lei Liu, Tsinghua University, China Xiaohua Liu, Sichuan University, China Santanu Mukherjee, Indian Institute of Science, Bangalore, India Scott Silverman, University of Illinois at Urbana-Champaign, USA Cristina Trujillo, University of Manchester, UK Members Ivan Huc, Ludwig-Maximilian University of Munich, Germany S.S.V Ramasastry, Indian Institute of Science Education and Research Mohali, India Corinna Schindler, University of Michigan, USA Judy I-Chia Wu, University of Houston, USA Anthony Davis, University of Bristol, UK Editorial Board Chair Anthony Davis, University of Bristol, UK Associate Editors Christian Hackenberger, Leibniz-Institut für Molekulare Pharmakologie and Humboldt Universität zu Berlin, Germany Katrina Jolliffe, University of Sydney, Australia Motomu Kanai, University of Tokyo, Japan 5511 5511 Visible light-induced synthesis of polysubstituted oxazoles from diazo compounds Jinrui Bai, Dan Qi, Zhuoheng Song, Bin Li, Lin Guo, Chao Yang* and Wujiong Xia* Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 A This article is licensed under a Creative Commons Attribution 3.0 Unporte 5516 Synthesis of non-anomeric C-glycosyl pyrazolidinone derivatives via visible-light photoredox catalysis Renan. O. Gonçalves, Pedro H. R. Oliveira, Iva S. de Jesus, Natalí P. Debia, Diogo S. Lüdtke* and Márcio W. Paixão* 5521 I2/TBHP-mediated oxidative cascade cyclization of vinyl azide and benzylamine to construct 2,5-di- substituted oxazoles Mohanreddy Pothireddy, Rana Chatterjee, Vijaya Babu Penke and Rambabu Dandela* 5527 Chemoenzymatic synthesis of (+)-isoagatholactone, (+)-spongian-16-one, and 3-deoxychavalone A via biocatalytic polyene cyclization Wen Xiao, Su-Jing Wang, Ming-Zhu Yu, Xue-Jie Zhang and Zheng Xiang* Information for Authors Information for Authors Customers should make payments by cheque in sterling payable on a UK clearing bank or in US dollars payable on a US clearing bank. Full details on how to submit material for publication in Organic & Biomolecular Chemistry are given in the Instructions for Authors (available from http://www.rsc.org/authors). S b i i h ld b d i th j l’ h li/ b This journal is © The Royal Society of Chemistry 2023. Apart from fair dealing for the purposes of research or private study for non-commercial purposes, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988 and the Copyright and Related Rights Regulation 2003, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the Publishers or in the case of reprographic reproduction in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. US copyright law is applicable to users in the USA. Whilst this material has been produced with all due care, the Royal Society of Chemistry cannot be held responsible or liable for its accuracy and completeness, nor for any consequences arising from any errors or the use of the information contained in this publication. The publication of advertisements does not constitute any endorsement by the Royal Society of Chemistry or Authors of any products advertised. The views and opinions advanced by contributors do not necessarily reflect those of the Royal Society of Chemistry which shall not be liable for any resulting loss or damage arising as a result of reliance upon this material. The Royal Society of Chemistry is a charity, registered in England and Wales, Number 207890, and a company incorporated in England by Royal Charter (Registered No. RC000524), registered office: Burlington House Piccadilly London W1J 0BA UK Authors may reproduce/republish portions of their published contribution without seeking permission from the Royal Society of Chemistry, provided that any such republication is accompanied by an acknowledgement in the form: (Original Citation)–Reproduced by permission of the Royal Society of Chemistry. Registered charity number: 207890 Telephone: +44 (0) 207 4378 6556. For marketing opportunities relating to this journal, contact marketing@rsc.org View Article Online COMMUNICATIONS 5516 Renan. O. Gonçalves, Pedro H. R. Oliveira, Iva S. de Jesus, Natalí P. Debia, Diogo S. Lüdtke* and Márcio W. Paixão* Visible light-induced synthesis of polysubstituted oxazoles from diazo compounds Jinrui Bai, Dan Qi, Zhuoheng Song, Bin Li, Lin Guo, Chao Yang* and Wujiong Xia* This journal is © The Royal Society of Chemistry 2023 5521 5527 Chemoenzymatic synthesis of (+)-isoagatholactone, (+)-spongian-16-one, and 3-deoxychavalone A via biocatalytic polyene cyclization Chemoenzymatic synthesis of (+)-isoagatholactone, (+)-spongian-16-one, and 3-deoxychavalone A via biocatalytic polyene cyclization Wen Xiao, Su-Jing Wang, Ming-Zhu Yu, Xue-Jie Zhang and Zheng Xiang* Org. Biomol. Chem., 2023, 21, 5485–5491 \| 5487 This journal is © The Royal Society of Chemistry 2023 View Article Online COMMUNICATIONS 5532 Brønsted acid-promoted ring-opening and annulation of thioamides and 2H-azirines to synthesize 2,4,5-trisubstituted thiazoles Meng Wang, Jingyi Ma, Hesong Wang, Fangpeng Hu, Bo Sun, Taiyan Tan, Minglang Li and Guosheng Huang* 5537 Cinchona-alkaloid-derived NN ligands for ruthenium catalyzed asymmetric hydrogenation of ketones Hao Sun, Pinli Dai, Jie Tian, Qian Xu, Qian Chen, Linlin Li, Xin Meng, Lin Zhang* and Chun Li* 5542 A reusable polymer anchored pyridine mediated formal [4 + 1] annulation reaction for the diastereoselective synthesis of 2,3-dihydrobenzofurans Akanksha Kumari, Anshul Jain, Khyati Shukla, Ranjan Patra and Nirmal K. Rana* 5547 Catalyst-free electrochemical sulfonylation of amines with sulfonyl hydrazide in aqueous medium Wei Chen,* Haojian Xu, Run Wu, Yang Chen, Pingbing Yu and Yanxi Jin Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5532 Brønsted acid-promoted ring-opening and annulation of thioamides and 2H-azirines to synthesize 2,4,5-trisubstituted thiazoles Brønsted acid-promoted ring-opening and annulation of thioamides and 2H-azirines to synthesize 2,4,5-trisubstituted thiazoles Meng Wang, Jingyi Ma, Hesong Wang, Fangpeng Hu, Bo Sun, Taiyan Tan, Minglang Li and Guosheng Huang* Meng Wang, Jingyi Ma, Hesong Wang, Fangpeng Hu, Bo Sun, Taiyan Tan, Minglang Li and Guosheng Huang* cess Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5537 Cinchona-alkalo ruthenium catal ketones Hao Sun, Pinli Da Linlin Li, Xin Meng 5537 Cinchona-alkaloid-derived NN ligands for ruthenium catalyzed asymmetric hydrogenation of ketones Hao Sun, Pinli Dai, Jie Tian, Qian Xu, Qian Chen, Linlin Li, Xin Meng, Lin Zhang* and Chun Li* under a Creative Commons Attribution 3 5537 Cinchona-alkaloid-derived NN ligands for ruthenium catalyzed asymmetric hydrogenation of ketones Hao Sun, Pinli Dai, Jie Tian, Qian Xu, Qian Chen, Linlin Li, Xin Meng, Lin Zhang* and Chun Li* This article is 5547 Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Peerawat Saejong, Juan J. Rojas, Camille Denis, Andrew J. P. White, Anne Sophie Voisin-Chiret, Chulho Choi and James A. Bull* Catalyst-free electrochemical sulfonylation of amines with sulfonyl hydrazide in aqueous medium Catalyst-free electrochemical sulfonylation of amines with sulfonyl hydrazide in aqueous medium Catalyst-free electrochemical sulfonylation of amines with sulfonyl hydrazide in aqueous medium Wei Chen,* Haojian Xu, Run Wu, Yang Chen, Pingbing Yu and Yanxi Jin Wei Chen,* Haojian Xu, Run Wu, Yang Chen, Pingbing Yu and Yanxi Jin 5488 \| Org. Biomol. Chem., 2023, 21, 5485–5491 View Article Online PAPERS This journal is © The Royal Society of Chemistry 2023 Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5553 Synthesis of oxetane and azetidine ethers as ester isosteres by Brønsted acid catalysed alkylation of alcohols with 3-aryl-oxetanols and 3-aryl- azetidinols Peerawat Saejong, Juan J. Rojas, Camille Denis, Andrew J. P. White, Anne Sophie Voisin-Chiret, Chulho Choi and James A. Bull* 5560 A triphenylamine-based ﬂuorescent probe with phenylboronic acid for highly selective detection of Hg2+ and CH3Hg+ in groundwater Lin Li, Hao Ouyang, Zhiqing Long, Qinpeng Zhang, Yunqi Jiang, Meng Cai, Shenglun Xiong, Sangshan Peng, Guangyu Xu* and Qing He* 5567 Ru(II) catalyzed chelation assisted C(sp2)–H bond functionalization along with concomitant (4 + 2) annulation Anindita Sarkar, Moumita Saha and Asish R. Das* 5587 Pd-catalyzed homo-coupling of benzofurans: one- pot synthesis of diverse 3,3’-bisbenzofurans Kishan Gugulothu, Ramanna Jatoth, Ramakrishna Gudipati, Nampally Rajitha, Rambabu Dandela and K. Shiva Kumar* p Synthesis of oxetane and azetidine ethers as ester isosteres by Brønsted acid catalysed alkylation of alcohols with 3-aryl-oxetanols and 3-aryl- azetidinols This journal is © The Royal Society of Chemistry 2023 5560 e-based ﬂuorescent probe with cid for highly selective detection Hg+ in groundwater g, Zhiqing Long, Qinpeng Zhang, Cai, Shenglun Xiong, uangyu Xu* and Qing He* A triphenylamine-based ﬂuorescent probe with phenylboronic acid for highly selective detection of Hg2+ and CH3Hg+ in groundwater Lin Li, Hao Ouyang, Zhiqing Long, Qinpeng Zhang, Yunqi Jiang, Meng Cai, Shenglun Xiong, Sangshan Peng, Guangyu Xu* and Qing He* 5587 Pd-catalyzed homo-coupling of benzofurans: one- pot synthesis of diverse 3,3’-bisbenzofurans Org. Biomol. Chem., 2023, 21, 5485–5491 \| 5489 This journal is © The Royal Society of Chemistry 2023 View Article Online PAPERS 5592 Metal-free synthesis of N-sulfonyl imines from benzyl alcohol derivatives and iminoiodinanes via mechanochemistry Souvik Guha, Ritwik Bhattacharya, Jesni Jacob, Mahesh Ravva and Subhabrata Sen* 5601 A Pd–Sn heterobimetallic catalyst for carbonylative Suzuki, Sonogashira and aminocarbonylation reactions using chloroform as a CO surrogate Anuradha Mohanty, Mukesh Kumar Nayak and Sujit Roy* 5609 Electrochemical oxidation of 3-substituted indoles Juan J. Arteaga Giraldo, Ashley C. Lindsay, Rachel Chae-Young Seo, Paul A. Kilmartin and Jonathan Sperry* 5616 Thiol-promoted intermolecular cyclization to synthesize 1,2,4-oxadiazoles including tioxazafen under transition metal-free conditions Congcong Yan, Min Zhang, Jiaxin Li, Jinli Zhang* and Yangjie Wu Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Metal-free synthesis of N-sulfonyl imines from benzyl alcohol derivatives and iminoiodinanes via mechanochemistry Metal-free synthesis of N-sulfonyl imines from benzyl alcohol derivatives and iminoiodinanes via mechanochemistry Souvik Guha, Ritwik Bhattacharya, Jesni Jacob, Mahesh Ravva and Subhabrata Sen* Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5609 Electrochemical oxidation of 3-substituted indoles Juan J. Arteaga Giraldo, Ashley C. Lindsay, Rachel Chae-Young Seo, Paul A. Kilmartin and Jonathan Sperry* Electrochemical oxidation of 3-substituted indoles Juan J. Arteaga Giraldo, Ashley C. Lindsay, Rachel Chae-Young Seo, Paul A. Kilmartin and Jonathan Sperry* Th 5616 Thiol-promoted intermolecular cyclization to synthesize 1,2,4-oxadiazoles including tioxazafen under transition metal-free conditions Thiol-promoted intermolecular cyclization to synthesize 1,2,4-oxadiazoles including tioxazafen under transition metal-free conditions Congcong Yan, Min Zhang, Jiaxin Li, Jinli Zhang* and Yangjie Wu Congcong Yan, Min Zhang, Jiaxin Li, Jinli Zhang* and Yangjie Wu 5490 \| Org. Biomol. Chem., 2023, 21, 5485–5491 View Article Online PAPERS 5622 Flavin based supramolecular gel displaying multi- stimuli triggered sol–gel transition M. S. S. This journal is © The Royal Society of Chemistry 2023 5560 Vinod Mouli and Ashutosh Kumar Mishra* 5629 Towards a qualitative understanding of carbonyl reactivity of α-substituted ethyl 5-oxohomoadamantyl-4-carboxylates Ilya M. Tkachenko,* Vadim A. Shiryaev and Yuri N. Klimochkin 5643 Alkyltin ﬂuorides as alkylating reagent in aminoalkylation of maleimides Jingjing Wang, Lin Zhao, Yafei Zhang, Peisen Zheng, Peng Zou* and Ge Wu* This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. Flavin based supramolecular gel displaying multi- stimuli triggered sol–gel transition M. S. S. Vinod Mouli and Ashutosh Kumar Mishra* Open Access Article. Published on 12 July 2023. Downloaded on 10/24/2024 5:03:40 AM. This article is licensed under a Creative Commons Attribution 3.0 Unported Licence. 5648 Photo-induced crosslinking uncovers an antiparallel strand orientation in heterodimeric (EIAALEK)3/(KIAALKE)3 and (EIAALEK)3/(RIAALRE)3 coiled-coil systems D. Aerssens, L. Miret-Casals,* D. Gomez, D. Sousa-Neves, Y. Levy, M. De Vleesschouwer, A. Manicardi* and A. Madder* Org. Biomol. Chem., 2023, 21, 5485–5491 \| 5491 This journal is © The Royal Society of Chemistry 2023
https://openalex.org/W4308063481	https://cris.unibo.it/bitstream/11585/935153/1/perini%20et%20al.%2c%202023%20ITAN.pdf	English	null	Candidate biomarkers from the integration of methylation and gene expression in discordant autistic sibling pairs	Research Square (Research Square)	2,022	cc-by	11,949	INTRODUCTION A i normal brain functions [22–26]. The interaction between ASD- associated genes and DNAm occurs through changes in the chromatin state driven by DNAm alterations that ultimately affect the expression of neurodevelopmental genes. As an example, Nguyen et al. provided strong evidence for epigenetic regulation via differential DNAm of the activity of two genes (BCL-2 and RORA) in the autistic brain, validating their methylation status of BCL-2 and RORA in lymphoblastoid cells of discordant mono- zygotic ASD twins and unaffected siblings [27]. Addressing the relationship between DNAm and expression of disease-associated genes is crucial for a full understanding of the etiology of heterogeneous diseases, but for ASD this relationship has been examined for a limited number of risk genes [28–31]. INTRODUCTION Autism spectrum disorders (ASD) are diagnosed in children at a young age if they exhibit difﬁculties in social interaction and communication and have restricted interests and repetitive behaviors [1, 2]. Affected children are ~1–2% of the population, and the prevalence is at least three times higher in boys than in girls [3]. Genetic studies have been instrumental in the discovery of multiple types of risk variants and the gain of new insights into the biology of ASD [4–7]. It has been acknowledged in several studies that ASD liability likely depends on additive effects of common and rare variants [8–11], and recent sequencing efforts have led to the identiﬁcation of over 100 putative ASD risk genes, the majority of which are neuronally expressed [7]. Consistently, post-mortem investigations on brain samples from ASD subjects have revealed gene expression signatures associated with multi- ple synaptic functions [5, 12–14], in addition to a dysregulation of immune function genes, which has also been shown by investigations conducted in blood [15–19]. However, like for other complex disorders, genetic variants (including structural and chromosomal variations) do not fully explain the heritability of the disease, suggesting that the risk for the disease is not exclusively driven by genetic variants [20, 21]. INTRODUCTION A i DNA methylation (DNAm) represents one of the potential epigenetic mechanisms that may contribute to the risk of ASD due to interactions with genetic elements during the development of Received: 29 October 2022 Revised: 18 March 2023 Accepted: 21 March 1Department of Cellular, Computational and Integrative Biology (CIBIO), University of Tr Research - University of Trento Centre for Computational and Systems Biology (COSBI), Rov ✉email: enrico.domenici@unitn.it Autism spectrum disorders (ASD) are diagnosed in children at a young age if they exhibit difﬁculties in social interaction and communication and have restricted interests and repetitive behaviors [1, 2]. Affected children are ~1–2% of the population, and the prevalence is at least three times higher in boys than in girls [3]. Genetic studies have been instrumental in the discovery of multiple types of risk variants and the gain of new insights into the biology of ASD [4–7]. It has been acknowledged in several studies that ASD liability likely depends on additive effects of common and rare variants [8–11], and recent sequencing efforts have led to the identiﬁcation of over 100 putative ASD risk genes, the majority of which are neuronally expressed [7]. Consistently, post-mortem investigations on brain samples from ASD subjects have revealed gene expression signatures associated with multi- ple synaptic functions [5, 12–14], in addition to a dysregulation of immune function genes, which has also been shown by investigations conducted in blood [15–19]. However, like for other complex disorders, genetic variants (including structural and chromosomal variations) do not fully explain the heritability of the disease, suggesting that the risk for the disease is not exclusively driven by genetic variants [20, 21]. Genes may be up-or down-regulated as a response to modiﬁcations of DNAm that are linked to environmental variables and DNA sequence variations. Environmental variables such as advanced paternal age and chemical exposures have been shown to have strong effects on DNAm proﬁle, and at the same time to increase the risk for ASD diagnosis [28, 32, 33]. Genetic variants may also inﬂuence DNAm levels through allele-speciﬁc methyla- tion or methylation quantitative trait loci (mQTL), as it has been shown by large-scale investigations conducted in blood [34–37] and in brain [38, 39]. Translational Psychiatry Translational Psychiatry www.nature.com/tp 1Department of Cellular, Computational and Integrative Biology (CIBIO), University of Trento, Trento (TN), Italy. 2EURAC Research, Bolzano, Italy. 3Fondazione The Microsoft Research - University of Trento Centre for Computational and Systems Biology (COSBI), Rovereto (TN), Italy. A list of authors and their afﬁliations appears at the end of the paper. ✉email: enrico domenici@unitn it ARTICLE OPEN Candidate biomarkers from the integration of methylation and gene expression in discordant autistic sibling pairs Samuel Perini1, Michele Filosi1,2, The Italian Autism Network and Enrico Domenici 1,3✉ © The Author(s) 2023 While the genetics of autism spectrum disorders (ASD) has been intensively studied, resulting in the identiﬁcation of over 100 putative risk genes, the epigenetics of ASD has received less attention, and results have been inconsistent across studies. We aimed to investigate the contribution of DNA methylation (DNAm) to the risk of ASD and identify candidate biomarkers arising from the interaction of epigenetic mechanisms with genotype, gene expression, and cellular proportions. We performed DNAm differential analysis using whole blood samples from 75 discordant sibling pairs of the Italian Autism Network collection and estimated their cellular composition. We studied the correlation between DNAm and gene expression accounting for the potential effects of different genotypes on DNAm. We showed that the proportion of NK cells was signiﬁcantly reduced in ASD siblings suggesting an imbalance in their immune system. We identiﬁed differentially methylated regions (DMRs) involved in neurogenesis and synaptic organization. Among candidate loci for ASD, we detected a DMR mapping to CLEC11A (neighboring SHANK1) where DNAm and gene expression were signiﬁcantly and negatively correlated, independently from genotype effects. As reported in previous studies, we conﬁrmed the involvement of immune functions in the pathophysiology of ASD. Notwithstanding the complexity of the disorder, suitable biomarkers such as CLEC11A and its neighbor SHANK1 can be discovered using integrative analyses even with peripheral tissues. Translational Psychiatry (2023) 13:109 ; https://doi.org/10.1038/s41398-023-02407-4 Translational Psychiatry (2023) 13:109 ; https://doi.org/10.1038/s41398-023-02407-4 Received: 29 October 2022 Revised: 18 March 2023 Accepted: 21 March 2023 EPIC array data processing y p g The analysis of the EPIC array data was developed from the pipeline described in the ChAMP R package [58]. Raw IDAT ﬁles were imported and checked for quality using the minﬁR package [59]. The quality control steps included default ﬁlters for probes with a detection P-value > 0.01 in at least one sample, for probes with a bead count < 3 in at least 5% of samples, for probes that were not found in CpG islands, for SNP-related probes according to the general recommendations [60], for cross-reactive probes [61, 62], for probes that aligned to multiple locations and ﬁnally, for probes located in chromosome X and Y. For each probe that passed the quality ﬁlters (i.e., CpG site), we calculated the Beta-value for each sample by taking the ratio between the number of methylated cytosines and the total number of cytosines at the speciﬁc CpG site. We then run BMIQ intra- array normalization [63], which is an effective method for adjusting Beta- values for bias introduced by the Inﬁnium type 2 probe design [64]. We visualized the similarity of samples based on the normalized Beta-values of the 1,000 most variable probes using a multidimensional scaling plot, and we removed outlier families from downstream analyses if the metadata were incorrectly reported. Here, we examined methylation data from whole blood samples of ASD discordant siblings that belonged to the Italian Autism Network (ITAN) cohort. We aimed to detect regions of the genome showing discordant levels of DNAm which correlated with the expression of nearby genes. As this association between diagnosis, DNAm, and gene expression is likely to be biased by demographic factors [50], genetic variants [34], biological and non-biological variables [46, 51, 52], we estimated the difference in DNAm between ASD and non-ASD siblings, and also the correlation between DNAm and gene expression conditionally on these confounding factors. Finally, we investigated biological pathways that were predicted to be signiﬁcantly affected by the identiﬁed DNAm alterations. The contribution of batch effects to DNAm variation was determined using singular value decomposition (SVD [65]), and we used ComBat [66] to correct the Beta-values for technical features that showed a signiﬁcant association (P-value < 0.05) with the ﬁrst 20 principal components of DNAm variation. These batch effects were slide, plate, and array number, and we ensured that the biological variable of interest (ASD vs. non-ASD) was not confounded with these technical variables [67–69]. INTRODUCTION A i Of note, mQTLs have been found to be highly correlated between independent brain and blood samples [40], opening the possibility to learn about epigenetic mechanisms in y y g DNA methylation (DNAm) represents one of the potential epigenetic mechanisms that may contribute to the risk of ASD due to interactions with genetic elements during the development of S. Perini et al. 2 the brain by studying peripheral tissues, and identifying biomar- kers that reﬂect complex epigenetic interactions contributing to ASD risk, which would be otherwise overlooked. the brain by studying peripheral tissues, and identifying biomar- kers that reﬂect complex epigenetic interactions contributing to ASD risk, which would be otherwise overlooked. LightningTM MagPrep protocol (Zymo Research, California, USA). This protocol consisted of adding the ready-to-use Lightning Conversion Reagent directly to the DNA coupled to a magnetic bead-based clean-up method. Secondly, the methylation proﬁle of the converted DNA was determined across the human genome (hg19) using the Inﬁnium HumanMethylationEPIC (“EPIC”) array by Illumina [55]. The DNA methyla- tion data that support the ﬁndings of this study are available from the ITAN Foundation (see https://www.fondazioneitan.org/en for details) upon submission of an ofﬁcial request. Data are released only for research purposes, upon assessment of a project proposal by the ITAN Scientiﬁc Committee. LightningTM MagPrep protocol (Zymo Research, California, USA). This protocol consisted of adding the ready-to-use Lightning Conversion Reagent directly to the DNA coupled to a magnetic bead-based clean-up method. Secondly, the methylation proﬁle of the converted DNA was determined across the human genome (hg19) using the Inﬁnium HumanMethylationEPIC (“EPIC”) array by Illumina [55]. The DNA methyla- tion data that support the ﬁndings of this study are available from the ITAN Foundation (see https://www.fondazioneitan.org/en for details) upon submission of an ofﬁcial request. Data are released only for research purposes, upon assessment of a project proposal by the ITAN Scientiﬁc Committee. , Diagnosis of ASD still relies on the fulﬁllment of descriptive criteria, even though there is a strong motivation to develop effective biomarkers that can be used for early diagnosis. Some of the main candidate biomarkers for ASD have been identiﬁed through differential DNAm analysis. The receptor of the oxytocin hormone is one well-studied example. Its genomic region has been found differentially methylated in several studies with ASD and non-ASD controls, but the reported effects of DNAm on gene expression and protein synthesis have not been consistent across studies [41–43]. DNAm proﬁling DNA was extracted from peripheral blood samples using the Puregene Blood Kit (Gentra Systems, Minneapolis, MN, US), a modiﬁed salting-out precipitation method, following the manufacturer’s instructions. Each DNA sample was then checked for quality and quantity using NanoDrop ND 1000 spectrophotometer (Thermo Scientiﬁc, Wilmington, DE, US). Follow- ing sample randomization, DNA methylation quantiﬁcation was performed at Life & Brain GmbH, Bonn, in two main steps. Firstly, DNA samples were submitted to bisulﬁte conversion using the EZ-96 DNA Methylation- Subjects j For the ITAN collection, thirteen centers were involved after approval by the Verona Hospital Ethical Review Board (study protocol AUT-SFK001, CE1419) and by the Ethical Review Committees. More than 800 individuals across 256 families were recruited and diagnosed with ASD with their consent or the parents’ consent [53]. The diagnosis was performed by experienced child psychiatrists that followed the DSM IV [54] and used standard tools: Broader Phenotype Autism Symptom Scale (BPASS), Autism Diagnostic Interview-Revised (ADI-R), Autism Diagnostic Observation Schedule (ADOS), and Krug Asperger Disorder Index screening. Of the ITAN collection, we selected 76 families having two siblings discordant for ASD diagnosis, and the affected child between 4 and 18 years old. The total number of subjects was 152 (Supplementary Table 1). Peripheral cell-type composition We calculated cell-type composition for each sample from the Beta-values using the RefbaseEWAS method implemented in ChAMP [72]. We ﬁrst assessed whether there was a signiﬁcant difference in cell-type composi- tion between ASD and non-ASD subjects using Dirichlet regression in R. Then, we extracted the variation in cell-type composition that was not partly explained by the diagnosis (i.e. the residuals of the Dirichlet regression), and we used these residuals of cell-type composition as covariates of a linear model to examine changes in DNAm that were not driven by cell-type heterogeneity (see below). EPIC array data processing We also applied control probe adjustment [70] to further reduce the effects of technical biases on the methylation signal. This speciﬁc step involved the use of signal intensities for the EPIC array control probes which were representative of the efﬁciency of bisulﬁte conversion, and other aspects related to the chemistry of methylation quantiﬁcation. We performed a principal component analysis in R [71] of these control probe intensities to deal with the high correlation between them, and then we determined whether there were signiﬁcant batch effects of these principal components using SVD. INTRODUCTION A i Differential DNA methylation analyses have been conducted in post-mortem tissues from ASD subjects and neurotypical controls, resulting in the identiﬁcation of signatures of DNAm alterations and multiple biological pathways involved [44–46]. For biomarker discovery, brain tissues are less suitable than peripheral ones as they are much less accessible, and they also tend to be affected by post-mortem degradation. However, it must be taken into account that epigenetic signatures tend to be tissue-speciﬁc: interindividual variation in whole blood is not a strong predictor of interindividual variation in the brain, although DNA methylation in whole blood signiﬁcantly co-varies with that in the brain at some genomic loci, including loci relevant for neurodevelopmental disorders [34, 47]. Therefore, whilst on the one hand, peripheral tissues might represent a valid alternative to brain tissues for detecting and developing effective biomarkers for ASD, on the other hand, they contain only a small amount of information that is shared with brain samples [48] and detecting signiﬁcant signals may require very large sample sizes [49]. DNA array genotyping and transcriptome analysis DNA samples were genotyped by the Autism Sequencing Consortium [56] using Illumina GSA v2 arrays (~658,000 markers). Genotype data were called using the genotyping command line interface of Illumina array analysis platform. Transcriptome data were generated by Poly-A RNA sequencing on Illumina RNASeq Platform as previously described [17]. Exome sequence data were also available as a result of a collaboration with the Autism Sequencing Consortium [57]. Differential methylation We used the Benjamini-Hochberg corrected P-values for detecting probe-gene pairs that may be potentially associated with ASD diagnosis. model was ﬁtted to the distribution of Beta-values across samples, with age, sex, family, and diagnosis as covariates. By including family in the linear model, we also expected to capture most of the variation explained by ethnicity because, for most samples, ethnicity was the same within families and different between families. A moderated gene-wise variance was then computed based on the empirical Bayes method [74] to test the null hypothesis that a difference in ﬁtted values between ASD and non- ASD subjects was equal to zero. We called a CpG site a DMP if such a difference had a Benjamini-Hochberg (BH) corrected P-value < 0.05. To address the interactions between neighboring CpG sites and the net effect that these sites have on associated gene(s), we determined the methylation status of clusters of probes by running the DMRcate algorithm [75] built in ChAMP. This algorithm identiﬁed differentially methylated regions (DMRs) between ASD and non-ASD subjects by applying a Gaussian smoothing to the moderated statistics of the limma output using windows of 1000 bp (default setting). DMRcate does not require a priori annotations of the genome for calling DMRs, and it does not combine genomically nearby CpG sites based on the direction of DNAm. Because of these two features of the algorithm, we were able ﬁrstly to study genomic regions that either had (e.g., promoter) or did not have (e.g., intergenic) explicit gene associations and, secondly, to analyze regulatory regions that showed hypermethylated and hypomethylated probes in ASD cases compared to non-ASD siblings (or vice versa). We used a quantile-quantile (Q–Q) plot between observed DMP/DMR P-values and expected P values under a uniform distribution of range [0,1] to check for spurious estimated P-values. g The correlation between DNAm and gene expression may also depend on genetic components. To understand potential genetic effects on DNAm and gene expression, we used the genotype data of our subjects and integrated them with published data of independent DNAm quantitative trait locus (mQTL) [34] and of cis-expression quantitative trait loci (eQTL) [87]. First, we searched for cis-eQTL of the gene which was signiﬁcantly associated with the differentially methylated distal probe. Then, we checked if such a genetic variant was also a mQTL. Differential methylation If a given SNP was a mQTL and corresponded to our distal probe, we estimated its effect on DNAm using a beta regression model [88] where the independent variables were the diagnosis and the genotype of the subjects multiplied by the Beta-value of the mQTL. Each linear model included two different top-ranked mQTL which were estimated to be in linkage equilibrium using LDlink [89]. To assess the accuracy of the identiﬁed DMRs to classify ASD subjects, we ﬁrst built a classiﬁer model based on a Random Forest (RF) algorithm using the R package randomForest [76] and fed it with the median DNAm values of the DMRs. The model was trained and tested on 70% of the samples using a 5-fold cross-validation that was repeated ten times (R package caret [77]). The remaining 30% of the samples were used for validating the model classiﬁer performance to predict ASD diagnosis. To avoid leakage of information from the test set into the training set, we then run differential methylation analysis on 70% of the samples, and then tested the ability of the identiﬁed DMRs to correctly classify the remaining 30% of the samples in terms of ASD diagnosis. The classiﬁer was based on a Random Forest (RF) algorithm using the R package randomForest, and fed with the median DNAm values of the top 50 DMRs. A 5-cross-validation procedure was repeated ten times as above. Differential methylation After ﬁltering probes and samples for quality and correcting Beta-values for batch effects and cell-type composition, the next step of the DNAm pipeline was to identify differentially methylated probes (DMPs) using the limma R package [73, 74] implemented in ChAMP. For each probe, a linear Translational Psychiatry (2023) 13:109 S. Perini et al. 3 in supervised mode. From the list of DMPs generated using limma (see above), distal EPIC probes (at least 2 Kb far from transcriptional start sites) were kept for the integration with gene expression data if the DNAm difference at these probes was \|0.03\| between ASD and non-ASD subjects. Each differentially methylated distal probe was then paired with the closest ten upstream genes and the closest ten downstream genes. For each probe-gene pair, we computed the inverse correlation between the DNAm of the probe, and the expression of the gene and further analyzed the difference in gene expression between ASD and non-ASD subjects using the Mann-Whitney U test. We used the Benjamini-Hochberg corrected P-values for detecting probe-gene pairs that may be potentially associated with ASD diagnosis. model was ﬁtted to the distribution of Beta-values across samples, with age, sex, family, and diagnosis as covariates. By including family in the linear model, we also expected to capture most of the variation explained by ethnicity because, for most samples, ethnicity was the same within families and different between families. A moderated gene-wise variance was then computed based on the empirical Bayes method [74] to test the null hypothesis that a difference in ﬁtted values between ASD and non- ASD subjects was equal to zero. We called a CpG site a DMP if such a difference had a Benjamini-Hochberg (BH) corrected P-value < 0.05. in supervised mode. From the list of DMPs generated using limma (see above), distal EPIC probes (at least 2 Kb far from transcriptional start sites) were kept for the integration with gene expression data if the DNAm difference at these probes was \|0.03\| between ASD and non-ASD subjects. Each differentially methylated distal probe was then paired with the closest ten upstream genes and the closest ten downstream genes. For each probe-gene pair, we computed the inverse correlation between the DNAm of the probe, and the expression of the gene and further analyzed the difference in gene expression between ASD and non-ASD subjects using the Mann-Whitney U test. Difference in cell-type composition We estimated per-sample cell-type proportions from DNAm data and analyzed these compositional data using Dirichlet regression to determine whether there was a difference between ASD and non-ASD subjects. We found no signiﬁcant difference in estimated cell-type composition, except for NK cells where the proportion was signiﬁcantly lower in ASD subjects compared to non-ASD subjects (ASD = 0.01, non-ASD = 0.04, P-value < 0.001) (Fig. 1; Correlation between methylation in blood and in reference brain datasets We obtained data from the Image-CpG database [90] to assess to which extent the methylation level within the identiﬁed DMRs could reﬂect brain methylation. Blood-brain correlations parameters for EPIC Illumina EPIC array were available as summary statistics. As described in Roberson-Nay et al. [91] we computed the median rho correlation, together with the minimum and maximum rho, for all CpGs included in each DMR. Gene set enrichment Both DMP and DMR datasets were analyzed for enriched gene sets to ﬁnd biological functions signiﬁcantly affected by differential DNAm between ASD and non-ASD subjects. We performed gene set enrichment analysis (GSEA) using bioinformatic tools that were developed explicitly for either DMPs or DMRs and that allowed us to correct gene sets P-values for the different numbers of CpGs between genes (i.e., probe number bias [78]). For DMPs, GSEA was performed using functional class scoring, which is an approach implemented in the methylGSA R package [79]. Compared to other GSEA methods such as over-representation analysis, functional class scoring allowed us to rank all CpGs and not only DMPs by their P-values and, thus, to use the entire list of CpG-associated genes for each gene set to determine which biological functions were signiﬁcantly inﬂuenced by differential DNAm between ASD and non-ASD subjects. Gene sets and the corresponding biological functions were extracted from widely used databases (Gene Ontology, KEGG, and Reactome) and also from specialized databases which were composed of genes implicated in psychiatric disorders (Gandal et al. [80]. and SFARI Gene database [81]), genes highly expressed in the human brain (The Human Protein Atlas [82]) and ﬁnally, genes involved in synapse functions (SynGO [83]). Fig. 1 Estimated cell-type proportions in ASD and non-ASD subjects. Statistical signiﬁcance was calculated using Dirichlet regression and only the proportions of NK cells were signiﬁcantly different between ASD (green) and non-ASD (orange) subjects (* meaning a P-value < 0.001). The error bars represent the standard deviation. For DMRs, we used an empirical Bayes GSEA method implemented in ChAMP [84] that has been developed to overcome the probe number bias and to avoid using only CpGs below an arbitrary signiﬁcance threshold [85]. All CpGs of each DMR were mapped to genes, and these genes were ranked by their overall level of differential DNAm. This list of ranked genes was then examined for enriched biological pathways which were obtained from widely used databases (Gene Ontology, KEGG, and Reactome). Fig. 1 Estimated cell-type proportions in ASD and non-ASD subjects. Statistical signiﬁcance was calculated using Dirichlet regression and only the proportions of NK cells were signiﬁcantly different between ASD (green) and non-ASD (orange) subjects (* meaning a P-value < 0.001). The error bars represent the standard deviation. Identiﬁcation of DMPs and DMRs We examined EPIC methylation arrays of 76 discordant autistic sibling pairs to understand the role of epigenetics in ASD. We identiﬁed two outlier families using a multidimensional scaling plot based on DNAm Beta-values (Supplementary Fig. 2), but we discarded only one outlier family in which the sampling was not performed on the same date for both siblings; in all the other families, each pair was sampled on the same date. After removing this outlier, the methylation proﬁle of the 75 families was overall homogeneous, and largely shared between discordant siblings of the same family (Supplementary Fig. 3). Notably, one subject is carrier of a 15q11.2-13.1 duplication and three ASD subjects carry a de novo protein truncating mutation in one of the 72 associated genes identiﬁed by Fu et al. [57]. However, since all of them clustered with the corresponding control sibling in the multi- variate analysis (Supplementary Figs. 2, 3), we retained them in the analysis. Family was, in fact, one of the factors that resulted to be signiﬁcantly associated with the ﬁrst 20 principal components of the SVD analysis. The other factors that explained a signiﬁcant variation in Beta-values were either involved with the array processing (e.g., slide and control probes intensities) or with the heterogeneity of cell types (e.g., granulocytes and B cells), and no signiﬁcant variation was explained by the diagnosis (Supplemen- tary Fig. 4). However, after ﬁtting a linear model to each probe’s Beta-values to correct for family, age, sex, batch effects, control probes intensities, and cell-type composition, we were able to capture the portion of DNAm variation that was signiﬁcantly associated with the diagnosis (Supplementary Fig. 4). Fig. 2 Statistical signiﬁcance of genome-wide differential methy- lation between ASD and non-ASD subjects. Top-ranked DMP (red point) of each top 2% DMR and its corresponding gene (text box). To assess the accuracy of the identiﬁed DMRs to classify ASD subjects, we then built classiﬁer models based on a Random Forest algorithm as described in Methods. The ﬁrst model was fed with the median DNAm values of the 418 DMRs identiﬁed by differential methylation on the full cohort. Despite the small differences in methylation between ASD and non-ASD siblings (see the mean values of DNAm for each DMR in ASD and non-ASD subjects and their distribution in Supplementary File 4) and the potential p-value inﬂation, the overall model performance was 0.758 in terms of AUC. Identiﬁcation of DMPs and DMRs Since the performance of the ﬁrst model can be skewed away from the true performance due to leakage of information from the test set into the training set, we performed a second model based on DMRs differential methylation analysis on 70% of the samples, and then tested the ability of the identiﬁed top 50 DMRs to correctly classify the remaining 30% of the samples in terms of ASD diagnosis, as described in Methods. In this case, the model performance was 0.72 (see Supplementary Fig 8). Among the newly identiﬁed DMRs, 42 had exactly the same genomic position of the DMR identiﬁed with the full cohort analysis. As expected, their mean difference between ASD and control subjects correlated with their feature importance in the RF-based classiﬁcation (see Supplementary Fig 9). By restricting the classiﬁer to these 42 DMRs (and feeding the classiﬁer with their median DNAm values), the overall performance raised up to 0.75. For the differential methylation analysis between ASD and non- ASD siblings, we used the adjusted Beta-values and we identiﬁed 37,643 DMPs across the genome with a BH corrected P-value < 0.05 and a range of Beta-values difference between −0.05 and 0.05 (Supplementary File 1; for the Q–Q plot, see Supplementary Fig. 5). The data on single DMPs may not accurately reﬂect the interactions between neighboring CpG sites and the net effect that these sites have on associated gene(s). To overcome these limitations, we ran the DMRcate algorithm [75] and identiﬁed 418 DMRs with Stouffer’s P-value < 0.05 that were located genome-wide, as expected from the distribution of DMPs (Supplementary File 2). Of these DMPs and DMRs, the top-ranked sites were determined to potentially affect genes that have been reported to be involved in neurodevelop- mental diseases such as TBX1 [92], SHANK2 [93], and TTC23 [94] (see Fig. 2 and Supplementary Fig 6 which reports the location of the DMRs relative to the genes discussed here). Notably, out of the 418 signiﬁcant DMRs, only 26 has a statistically signiﬁcant difference when comparing males vs males (Stouffer’s P-value < 0.05, see Supplementary File 2), and none of the neurodevelop- mental related genes here discussed displays a signiﬁcantly different methylation in males vs females. To further address the potential confounding effect of sex on diagnosis-related ﬁndings, we additionally conducted an analysis stratiﬁed by sex (i.e. a separate analysis for males and females), see Supplementary File 3. Integration with gene expression and genotype data Integration with gene expression and genotype data To examine the potential effects that differential DNAm may have on transcription activity and gene regulation, we integrated the DNAm proﬁle with gene expression data from the same set of samples [17] using the ELMER R package (version 2.8.3) [86]. This type of analysis was performed Translational Psychiatry (2023) 13:109 Fig. 2 Statistical signiﬁcance of genome-wide differential methy- lation between ASD and non-ASD subjects. Top-ranked DMP (red point) of each top 2% DMR and its corresponding gene (text box). S. Perini et al. 4 Supplementary Table 2). These differences in cell-type composi- tion were consistent with the results of a previous study on ASD in which the same subjects were analyzed using gene expression data [17]. Even though we performed a DNAm-based cell-type deconvolution method, our estimated proportions were corre- lated (Pearson’s r > 0.25) with the proportions that were calculated in the previous study where it was used an RNA-based deconvolution algorithm (Supplementary Fig. 1a, b). Supplementary Table 2). These differences in cell-type composi- tion were consistent with the results of a previous study on ASD in which the same subjects were analyzed using gene expression data [17]. Even though we performed a DNAm-based cell-type deconvolution method, our estimated proportions were corre- lated (Pearson’s r > 0.25) with the proportions that were calculated in the previous study where it was used an RNA-based deconvolution algorithm (Supplementary Fig. 1a, b). Identiﬁcation of DMPs and DMRs Consistently with the smaller sample size of females compared to males, we detected fewer DMRs reaching statistical signiﬁcance and larger p-values in females (641 DMRs in females and 7570 DMRs in males; 170 DMRs were shared between females and males). However, we found a strong correlation (Pearson’s correlation 0.96, P-value < 0.001) between male and female coefﬁcients estimated by sex-stratiﬁed DMP analysis, see Supplementary Fig 7. Enriched biological processes in ASD g p We performed GSEA to understand which biological processes may have been altered by differential methylation between ASD and non-ASD subjects. We extracted biological processes from a total of seven databases (Gene Ontology, KEGG, Reactome, Gandal et al. [80], SFARI Gene [81], The Human Protein Atlas [82], and SynGO [83]), and identiﬁed 457 and 2,305 processes that were signiﬁcantly represented (FDR < 0.05) by the list of genes associated with DMPs and DMRs, respectively (Supplementary File 5; Supplementary File 6). Of the top 1% signiﬁcantly enriched processes with the highest Normalized Enrichment Score, two processes involved genes associated with synapse disassembly and structure, three processes were associated with neurogenesis, and one process involved learning genes. Finally, we found a signiﬁcant enrichment for SFARI ASD risk genes (Fig. 3). To further corroborate the evidence of the enrichment of neurodevelop- mental processes, we built a 2 × 2 contingency table with the number of index DMPs and non-DMPs (at gene level) that are associated with the top ranked neurodevelopmental processes (displayed in blue in Fig. 3), and the number of DMPs and non- Translational Psychiatry (2023) 13:109 S. Perini et al. Fig. 4 Gene expression and DNAm of CLEC11A. Gene expression (y-axis) was normalized and adjusted for batch effects and demographic parameters [17]. Median DNAm (i-axis) was calculated for the DMPs that were signiﬁcantly associated with CLEC11A. Mean (large and black circles) and 95% conﬁdence intervals (black bars) of CLEC11A expression and DNAm are shown for ASD (green; DNAm = 0.514 (0.511–0.517); DGE = −0.151 (−0.307–0.005)) and non-ASD (orange; DNAm = 0.519 (0.516–0.522); DGE = 0.061 (−0.091–0.212)) siblings. 5 1.26 1.31 1.28 1.28 1.26 1.28 1.34 1.28 1.29 1.28 1.26 1.3 1.29 1.31 1.26 1.32 1.31 1.27 1.29 1.28 1.29 1.28 1.26 1.26 1.33 1.45 ASD Risk Genes Gastric Acid Secretion Melanogenesis Phosphatidylinositol Signaling System Adherens Junction Organization Cell−Cell Adherens Junction Cortical Actin Cytoskeleton Cortical Cytoskeleton Inner Ear Morphogenesis Learning Regulation Of Ras Protein Signal Transduction Response To Axon Injury Sodium Ion Transmembrane Transporter Activity Specific Granule Membrane Nerve Development Neuron Projection Regeneration O−Glycan Processing T Cell Differentiation In Thymus Inositol Phosphate Metabolism Cardiac Muscle Cell Contraction Multi−Organism Behavior Solute:sodium Symporter Activity Translation Factor Activity Non−Small Cell Lung Cancer Synapse Structure Synapse Disassembly SFARI GO KEGG SynGO Organization 0.01 0.02 0.03 0.04 FDR Fig. 3 Top 1% enriched gene sets. Enriched biological processes in ASD Biological functions (y-axis) were extracted from four databases (x-axis) and they were ranked by FDR (red gradient) and Normalized Enrichment Score (values inside the tiles). Of the top 1% functions, seven may have an important contribution to the etiology of ASD (highlighted in blue). Fig. 4 Gene expression and DNAm of CLEC11A. Gene expression (y-axis) was normalized and adjusted for batch effects and demographic parameters [17]. Median DNAm (i-axis) was calculated for the DMPs that were signiﬁcantly associated with CLEC11A. Mean (large and black circles) and 95% conﬁdence intervals (black bars) of CLEC11A expression and DNAm are shown for ASD (green; DNAm = 0.514 (0.511–0.517); DGE = −0.151 (−0.307–0.005)) and non-ASD (orange; DNAm = 0.519 (0.516–0.522); DGE = 0.061 (−0.091–0.212)) siblings. Fig. 3 Top 1% enriched gene sets. Biological functions (y-axis) were extracted from four databases (x-axis) and they were ranked by FDR (red gradient) and Normalized Enrichment Score (values inside the tiles). Of the top 1% functions, seven may have an important contribution to the etiology of ASD (highlighted in blue). Table 1. Probe-gene pair association using ELMER. Table 1. Probe-gene pair association using ELMER. Hypermethylated Gene Distance FDR cg17729891 C19orf48 193401 0.01 cg20315590 TSEN15 −1959694 0.01 cg17729891 CLEC11A 119026 0.04 cg10841563 GNAQ −2633643 0.01 cg10841563 KIF27 3171593 0.01 cg10841563 HNRNPK 3302978 0.01 cg05890377 MOB1A 21940 0.03 Hypermethylated (left) and hypomethylated (right) distal probes in ASD subjects and putative target genes. impact on DNAm of CLEC11A and MOB1A. We investigated the effects of the SNPs genotype on the estimated differential DNAm between ASD and non-ASD subjects, but only for MOB1A the genotype contribution was strong enough to remove the differential DNAm signal (Supplementary Table 4). For CLEC11A, we found that differential DNAm was not lost when taking into account the presence of a mQTL. This strong signal for CLEC11A was supported by both the DMR analysis in which the genomic region involving CLEC11A was characterized as DMR (Fig. 2; Supplementary File 2), and the estimated correlation between the median DNAm across the four DMPs at 200–1500 bases upstream of CLEC11A transcriptional start site and CLEC11A expression (Pearson’s r = −0.39, P-value < 0.001; Fig. 4). Although the difference in DNAm of CLEC11A between ASD and non-ASD siblings was signiﬁcant, gene expression was only marginally different between the two groups (Fig. 4). Hypermethylated (left) and hypomethylated (right) distal probes in ASD subjects and putative target genes. Enriched biological processes in ASD DMPs associated with other processes (see Supplementary Table 3. We then assessed whether DMPs were overrepresented for neurodevelopment-related processes using a hypergeometric distribution (one side Fisher’s exact test). The result conﬁrms that neurodevelopment-related processes are indeed enriched within the DMPs (odds ratio = 2.96, P-value < 0.001). Finally, we exploited the dataset reported by Braun et al. [90] to investigate the degree of correlation between blood and brain for some of the identiﬁed DMRs, i.e. TTC23, TBX1, SHANK2 and CLEC11A, by computing the Spearman correlation for all CpGs included in each DMR. All four DMRs showed a trend for positive correlation between blood and brain methylation, with TTC23 displaying a consistent trend across all CpGs (see Supplementary Tab 5), suggesting the potential of using blood for detecting changes which may be relevant in the brain. Correlation between DNAm and gene expression g p Differential methylation between different biological conditions may contribute to phenotypic differences through changes in the underlying gene expression. Here, we investigated whether ASD and non-ASD siblings showed genes or genomic regions that differed in the joint pattern of DNAm and gene expression using ELMER analysis in supervised mode. We identiﬁed 346,224 probes on EPIC arrays that were at least 2 kb away from a transcriptional start site. Six of these distal probes were hypermethylated and ten were hypomethylated in ASD subjects with FDR < 0.05 and Beta- values difference > \|0.03 \| (Supplementary Fig. 10). We then identiﬁed putative target genes for these differentially methylated distal probes and found a signiﬁcant association for three of the hypermethylated probe-gene pairs and four of the hypomethy- lated probe-gene pairs (Table 1). Notably, we identiﬁed signiﬁcant associations for CLEC11A and MOB1A, genes known to be associated with immune and mental disorders [95, 96]. Translational Psychiatry (2023) 13:109 DISCUSSION Id tif i b We ﬁrst calculated the correlation between gene expression and DNAm, identifying a signiﬁcant and negative correlation between DNAm and expres- sion at the DMR that was mapped to CLEC11A. We then combined genotype data from the same samples and methylation eQTL data and showed that the correlation was not driven by genotype effects of candidate SNPs located in the same region. In our study, we analyzed the methylation proﬁle of a subset of the ITAN collection, the same subset previously examined at the transcription level [17]. We estimated cell-type proportions using DNAm array data of whole blood mixtures and showed a reduced fraction of NK cells in ASD siblings. A similar reduction in NK cells was also found using transcriptome data of the same cohort [17], consistent with previous investigations reporting either a decrease of NK cells [103] or a reduction of their activity [104–107]. NK cells act against viral infections, representing an important defensive mechanism of the innate immune system [108], and their decreased level or activity would suggest a suboptimal function of the immune system in ASD. In ASD children, immune pathways are typically dysregulated, and the expression of genes that are involved in the immune response has been found to be correlated with changes of DNAm in their regulatory regions [109]. For example, Nardone et al. [46]. found that hypomethylated sites were associated with the overexpression of transcription factors regulating the development of the microglia. It has also been reported that the immune system of mothers of ASD cases responded slowly to infections due to DNAm alterations at regulatory regions of immune genes [110]. Such DNAm alterations may be inherited, potentially contributing to the risk of ASD in children of immunocompromised mothers. More recently, a genome-wide methylation analysis in the blood of ASD subject vs neurotypical controls has led to the identiﬁcation of a putative epigenetically distinct subset of children driven by altered blood cell composition [111], further supporting the role of the immune systems in the pathophysiology of ASD. Nevertheless, given the known dynamic nature of the immune system and possible variations in cell proportions and subsets reﬂecting hormonal changes or environmental inﬂuences [112], the reduction in NK cells observed might well be reﬂecting a transient state, and would need to be conﬁrmed in longitudinal studies. DISCUSSION Id tif i b g In other studies, the region of CLEC11A has been identiﬁed as a candidate locus for the etiology of neurodevelopmental disorders due to its position nearby SHANK1, a gene strongly associated with ASD [96]. SHANK family genes encode postsynaptic proteins that are needed for functional electrochemical communication at the level of synapses, and mutations in any of the three genes (SHANK1, SHANK2, and SHANK3) have been reported to disrupt neuronal activity [96, 114]. Different SHANK mutations have been found in ASD individuals, including a deletion covering CLEC11A and SHANK1 in males with ASD [96, 115, 116]. In addition, differential DNAm between ASD cases and controls has been determined at the regulatory regions of SHANK1 and SHANK3 [117, 118]. We could not detect DNAm changes that were speciﬁcally associated with SHANK1 nor SHANK3 because these biomarkers are absent in the blood, but we did ﬁnd a DMR associated with the SHANK2 gene which is instead also expressed in the blood [119], showing a trend for correlation with the methylation level assessed in brain datasets. y Compared to studies based on brain tissues, our analysis using whole blood samples was limited in its power to capture neurodevelopmental genes and pathways, as we could only examine those that were not exclusively regulated in the brain. Still, we found differentially methylated sites in the blood that were associated with genes involved in neurodevelopment and ASD. We are aware that for some of these candidate genes differential DNAm may reﬂect a correlation between ASD diagnosis and some other variables that were not measured in our study, and may be transient, thus further investigations would be required to validate our ﬁndings. Nevertheless, our results provide further support for the possibility of using peripheral tissues to identify candidate epigenetic ASD biomarkers through integrative analyses across genomic data, as in the case of the CLEC11A-SHANK1 region. When validated, the methylation markers might become a valuable asset to support early diagnosis, particularly for non-syndromic or “idiopathic autism”, and assist the identiﬁcation of epigenetically distinct subtypes of autism [111, 120], thus facilitating stratiﬁed therapies. Autism is a polygenic neurodevelopmental disorder and, besides immune genes, there are other genes that have been associated with ASD [5, 6, 8, 11, 19]. REFERENCES 1. Lai MC, Lombardo MV, Baron-Cohen S. Autism. Lancet. 2014;383:896–910. 2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 5th edn. American Psychiatric Association; 2013. 2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 5th edn. American Psychiatric Association; 2013. 3. Zeidan J, Fombonne E, Scorah J, Ibrahim A, Durkin MS, Saxena S, et al. Global prevalence of autism: a systematic review update. Autism Res. 2022;15:778–90. 3. Zeidan J, Fombonne E, Scorah J, Ibrahim A, Durkin MS, Saxena S, et al. Global prevalence of autism: a systematic review update. Autism Res. 2022;15:778–90. 4. Vorstman JAS, Parr JR, Moreno-De-Luca D, Anney RJL, Nurnberger JI Jr, Hall- mayer JF. Autism genetics: opportunities and challenges for clinical translation. Nat Rev Genet. 2017;18:362–76. 4. Vorstman JAS, Parr JR, Moreno-De-Luca D, Anney RJL, Nurnberger JI Jr, Hall- mayer JF. Autism genetics: opportunities and challenges for clinical translation. Nat Rev Genet. 2017;18:362–76. By conducting gene set enrichment analysis (GSEA), we showed that identiﬁed DMPs were associated with biological processes related to synapse functions and neurogenesis. Although neurode- velopmental pathways are preferably studied using brain tissues, our results support the potential for identifying epigenetic biomarkers of neurodevelopment even in peripheral tissues [47, 48]. 5. de la Torre-Ubieta L, Won H, Stein JL, Geschwind DH. Advancing the understanding of autism disease mechanisms through genetics. Nat Med. 2016;22:345–61. 5. de la Torre-Ubieta L, Won H, Stein JL, Geschwind DH. Advancing the understanding of autism disease mechanisms through genetics. Nat Med. 2016;22:345–61. 6. Robinson EB, St, Pourcain B, Anttila V, Kosmicki JA, Bulik-Sullivan B, Grove J, et al. Genetic risk for autism spectrum disorders and neuropsychiatric variation in the general population. Nat Genet. 2016;48:552–5. 6. Robinson EB, St, Pourcain B, Anttila V, Kosmicki JA, Bulik-Sullivan B, Grove J, et al. Genetic risk for autism spectrum disorders and neuropsychiatric variation in the general population. Nat Genet. 2016;48:552–5. 7. Satterstrom FK, Kosmicki JA, Wang J, Breen MS, De Rubeis S, An JY, et al. Large- scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell. 2020;180:568.e523. Differences in cellular composition, DNAm, and function of neurodevelopmental pathways are expected to be associated with differences in gene expression between affected subjects and controls. Ultimately, the association between multiple mechanisms may strongly depend on genetic variants that, for example, are 8. Weiner DJ, Wigdor EM, Ripke S, Walters RK, Kosmicki JA, Grove J, et al. DISCUSSION Id tif i b By comparing the genome- wide methylation proﬁle of ASD subjects versus their healthy sibling controls, we identiﬁed hundreds of genes that were differentially methylated after accounting for demographic data, batch effects, and cell-type composition. In particular, 37,643 genomic sites (DMPs) and 418 regions (DMRs) were differentially methylated between ASD and non-ASD siblings. We detected DNAm alterations in TBX1, SHANK2, and TTC23, genes that have been previously shown to be affected by epigenetic processes in ASD and other neurodevelopmental diseases [92–94]. As expected, genes that have a signiﬁcant contribution to the risk of ASD, such as TBX1, SHANK2, and TTC23, are regulated through DNAm, suggesting that the etiology of ASD can be better understood by also looking at DNAm alterations. DISCUSSION Id tif i b Identifying robust biomarkers for ASD has been a challenging task. This group of neurodevelopmental disorders is polygenic, phenotypically heterogeneous, and inﬂuenced by both genetic and environmental factors [97–100]. As there is no single common genetic variant showing a strong predictive value for ASD diagnosis [101, 102], candidate biomarkers have been searched for outside the protein-coding sequences of genes such as in enhancers and intergenic regions. In these regulatory regions, biomarkers can be found by detecting changes in DNA methyla- tion (DNAm) that may potentially affect gene functions [22–26]. However, although the contribution of DNAm to the risk of ASD has been acknowledged, questions remain about the potential of this type of epigenetic biomarkers to help in the development of diagnostic and therapeutic approaches to ASD. Correlated DNAm and gene expression changes may be driven by genetic variants inﬂuencing DNAm at the nearby target gene. We identiﬁed two SNPs as candidate genetic variants for such an Translational Psychiatry (2023) 13:109 S. Perini et al. 6 Here, we analyzed the methylation proﬁle of whole blood samples from 75 sibling pairs with a discordant ASD diagnosis. We aimed to analyze DNAm in subjects with ASD and detect dysregulated genomic regions that may be involved in the etiology of the disorder. In addition, we investigated possible differences in cell-type proportions between ASD and non-ASD siblings that might have been involved in the pathophysiology of the disorder. responsible for differential DNAm and the regulation of gene activity [113]. We integrated the methylation Beta-values of DMPs of our samples with the corresponding gene expression and genotype data to understand the potential relationship between genetic and non-genetic components in ASD. We ﬁrst calculated the correlation between gene expression and DNAm, identifying a signiﬁcant and negative correlation between DNAm and expres- sion at the DMR that was mapped to CLEC11A. We then combined genotype data from the same samples and methylation eQTL data and showed that the correlation was not driven by genotype effects of candidate SNPs located in the same region. responsible for differential DNAm and the regulation of gene activity [113]. We integrated the methylation Beta-values of DMPs of our samples with the corresponding gene expression and genotype data to understand the potential relationship between genetic and non-genetic components in ASD. 3. Zeidan J, Fombonne E, Scorah J, Ibrahim A, Durkin MS, Saxena S, et al. Global prevalence of autism: a systematic review update. Autism Res. 2022;15:778–90. REFERENCES Polygenic transmission disequilibrium conﬁrms that common and rare variation act addi- tively to create risk for autism spectrum disorders. Nat Genet. 2017;49:978–85. Translational Psychiatry (2023) 13:109 S. Perini et al. 7 9. Oetjens MT, Kelly MA, Sturm AC, Martin CL, Ledbetter DH. Quantifying the polygenic contribution to variable expressivity in eleven rare genetic disorders. Nat Commun. 2019;10:4897. 35. Kerkel K, Spadola A, Yuan E, Kosek J, Jiang L, Hod E, et al. Genomic surveys by methylation-sensitive SNP analysis identify sequence-dependent allele-speciﬁc DNA methylation. Nat Genet. 2008;40:904–8. 10. Niemi MEK, Martin HC, Rice DL, Gallone G, Gordon S, Kelemen M, et al. Common genetic variants contribute to risk of rare severe neurodevelopmental disorders. Nature. 2018;562:268–71. 36. van Dongen J, Nivard MG, Willemsen G, Hottenga J-J, Helmer Q, Dolan CV, et al. Genetic and environmental inﬂuences interact with age and sex in shaping the human methylome. Nat Commun. 2016;7:11115. 11. Klei L, McClain LL, Mahjani B, Panayidou K, De Rubeis S, Grahnat A-CS, et al. How rare and common risk variation jointly affect liability for autism spectrum dis- order. Mol Autism. 2021;12:66. 37. Hannon E, Knox O, Sugden K, Burrage J, Wong CCY, Belsky DW, et al. Char- acterizing genetic and environmental inﬂuences on variable DNA methylation using monozygotic and dizygotic twins. PLoS Genet. 2018;14:e1007544. 12. Ebert DH, Greenberg ME. Activity-dependent neuronal signalling and autism spectrum disorder. Nature. 2013;493:327–37. 38. Hannon E, Spiers H, Viana J, Pidsley R, Burrage J, Murphy TM, et al. Methylation QTLs in the developing brain and their enrichment in schizophrenia risk loci. Nat Neurosci. 2016;19:48–54. 13. De Rubeis S, He X, Goldberg AP, Poultney CS, Samocha K, Ercument Cicek A, et al. Synaptic, transcriptional and chromatin genes disrupted in autism. Nature. 2014;515:209–15. 39. Ng B, White CC, Klein H-U, Sieberts SK, McCabe C, Patrick E, et al. An xQTL map integrates the genetic architecture of the human brain’s transcriptome and epigenome. Nat Neurosci. 2017;20:1418–26. 14. Voineagu I, Wang X, Johnston P, Lowe JK, Tian Y, Horvath S, et al. Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature. 2011;474:380–4. 40. Qi T, Wu Y, Zeng J, Zhang F, Xue A, Jiang L, et al. Identifying gene targets for brain-related traits using transcriptomic and methylomic data from blood. Nat Commun. 2018;9:2282. 15. Pramparo T, Pierce K, Lombardo MV, Carter Barnes C, Marinero S, Ahrens- Barbeau C, et al. REFERENCES Prediction of autism by translation and immune/inﬂammation coexpressed genes in toddlers from pediatric community practices. JAMA Psy- chiatry. 2015;72:386–94. 41. Siu MT, Goodman SJ, Yellan I, Butcher DT, Jangjoo M, Grafodatskaya D, et al. DNA methylation of the oxytocin receptor across neurodevelopmental dis- orders. J Autism Dev Disord. 2021;51:3610–23. 42. Gregory SG, Connelly JJ, Towers AJ, Johnson J, Biscocho D, Markunas CA, et al. Genomic and epigenetic evidence for oxytocin receptor deﬁciency in autism. BMC Med. 2009;7:62. 16. Gupta S, Ellis SE, Ashar FN, Moes A, Bader JS, Zhan J, et al. Transcriptome analysis reveals dysregulation of innate immune response genes and neuronal activity-dependent genes in autism. Nat Commun. 2014;5:5748. 43. Maud C, Ryan J, McIntosh JE, Olsson CA. The role of oxytocin receptor gene (OXTR) DNA methylation (DNAm) in human social and emotional functioning: a systematic narrative review. BMC Psychiatry. 2018;18:154. 17. Filosi M, Kam-Thong T, Essioux L, Muglia P, Trabetti E, Spooren W, et al. Tran- scriptome signatures from discordant sibling pairs reveal changes in peripheral blood immune cell composition in Autism Spectrum Disorder. Transl Psychiatry. 2020;10:106. 44. Wong CCY, Smith RG, Hannon E, Ramaswami G, Parikshak NN, Assary E, et al. Genome-wide DNA methylation proﬁling identiﬁes convergent molecular sig- natures associated with idiopathic and syndromic autism in post-mortem human brain tissue. Hum Mol Genet. 2019;28:2201–11. 18. Tylee DS, Hess JL, Quinn TP, Barve R, Huang H, Zhang-James Y, et al. Blood transcriptomic comparison of individuals with and without autism spectrum disorder: a combined-samples mega-analysis. Am J Med Genet B. 2017;174:181–201. 45. Ladd-Acosta C, Hansen KD, Briem E, Fallin MD, Kaufmann WE, Feinberg AP. Common DNA methylation alterations in multiple brain regions in autism. Mol Psychiatry. 2014;19:862–71. 19. Grove J, Ripke S, Als TD, Mattheisen M, Walters RK, Won H, et al. Identiﬁcation of common genetic risk variants for autism spectrum disorder. Nat Genet. 2019;51:431–44. 46. Nardone S, Sams DS, Reuveni E, Getselter D, Oron O, Karpuj M, et al. DNA methylation analysis of the autistic brain reveals multiple dysregulated biolo- gical pathways. Transl Psychiatry. 2014;4:e433. 20. Chaste P, Leboyer M. Autism risk factors: genes, environment, and gene- environment interactions. Dialogues Clin Neurosci. 2012;14:281–92. 21. Rylaarsdam L, Guemez-Gamboa A. Genetic causes and modiﬁers of autism spectrum disorder. Front Cell Neurosci. 2019;13:385. 47. Davies MN, Volta M, Pidsley R, Lunnon K, Dixit A, Lovestone S, et al. REFERENCES Functional annotation of the human brain methylome identiﬁes tissue-speciﬁc epigenetic variation across brain and blood. Genome Biol. 2012;13:R43. 22. Vijayakumar NT, Judy MV. Autism spectrum disorders: Integration of the gen- ome, transcriptome and the environment. J Neurol Sci. 2016;364:167–76. 48. Hannon E, Lunnon K, Schalkwyk L, Mill J. Interindividual methylomic variation across blood, cortex, and cerebellum: implications for epigenetic studies of neurological and neuropsychiatric phenotypes. Epigenetics. 2015;10:1024–32. 23. Masini E, Loi E, Vega-Benedetti AF, Carta M, Doneddu G, Fadda R, et al. An overview of the main genetic, epigenetic and environmental factors involved in autism spectrum disorder focusing on synaptic activity. Int J Mol Sci. 2020;21:8290. 49. Andrews SV, Sheppard B, Windham GC, Schieve LA, Schendel DE, Croen LA, et al. Case-control meta-analysis of blood DNA methylation and autism spec- trum disorder. Mol Autism. 2018;9:40. 24. Abdolmaleky HM, Zhou JR, Thiagalingam S. An update on the epigenetics of psychotic diseases and autism. Epigenomics. 2015;7:427–49. 50. Ansel A, Rosenzweig JP, Zisman PD, Melamed M, Gesundheit B. Variation in gene expression in autism spectrum disorders: an extensive review of tran- scriptomic studies. Front Neurosci. 2017;10:601. 25. Abdolmaleky HM, Zhou JR, Thiagalingam S. Cataloging recent advances in epigenetic alterations in major mental disorders and autism. Epigenomics. 2021;13:1231–45. 51. Lee SC, Quinn TP, Lai J, Kong SW, Hertz-Picciotto I, Glatt SJ, et al. Solving for X: Evidence for sex-speciﬁc autism biomarkers across multiple transcriptomic studies. Am J Med Genet Part B. 2019;180:377–89. 26. Coppedè F. The diagnostic potential of the epigenome in autism spectrum disorders. Epigenomics. 2021;13:1587–90. 52. Werling DM, Parikshak NN, Geschwind DH. Gene expression in human brain implicates sexually dimorphic pathways in autism spectrum disorders. Nat Commun. 2016;7:10717. 27. Nguyen A, Rauch TA, Pfeifer GP, Hu VW. Global methylation proﬁling of lym- phoblastoid cell lines reveals epigenetic contributions to autism spectrum dis- orders and a novel autism candidate gene, RORA, whose protein product is reduced in autistic brain. FASEB J. 2010;24:3036–51. 53. Muglia P, Filosi M, Da Ros L, Kam-Thong T, Nardocci F, Trabetti E, et al. The Italian autism network (ITAN): a resource for molecular genetics and biomarker investigations. BMC Psychiatry. 2018;18:369. 28. Cheroni C, Caporale N, Testa G. Autism spectrum disorder at the crossroad between genes and environment: contributions, convergences, and interactions in ASD developmental pathophysiology. Mol Autism. 2020;11:69. 54. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edn. American Psychiatric Association; 1994. 29. REFERENCES McCartney DL, Walker RM, Morris SW, McIntosh AM, Porteous DJ, Evans KL. Identiﬁcation of polymorphic and off-target probe binding sites on the Illumina Inﬁnium MethylationEPIC BeadChip. Genom Data. 2016;9:22–24. g g g p 88. Cribari-Neto F, Zeileis A. Beta regression in R. J Stat Softw. 2010;34:1–24. 62. Pidsley R, Zotenko E, Peters TJ, Lawrence MG, Risbridger GP, Molloy P, et al. Critical evaluation of the Illumina MethylationEPIC BeadChip microarray for whole-genome DNA methylation proﬁling. Genome Biol. 2016;17:208. 89. Machiela MJ, Chanock SJ. LDlink: a web-based application for exploring population-speciﬁc haplotype structure and linking correlated alleles of possible functional variants. Bioinformatics. 2015;31:3555–7. 63. Teschendorff AE, Marabita F, Lechner M, Bartlett T, Tegner J, Gomez-Cabrero D, et al. A beta-mixture quantile normalization method for correcting probe design bias in Illumina Inﬁnium 450 k DNA methylation data. Bioinformatics. 2013;29:189–96. 90. Braun PR, Han S, Hing B, Nagahama Y, Gaul LN, Heinzman JT, et al. Genome- wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019;9:47. tissues within individuals. Transl Psychiatry. 2019;9:47. 91. Roberson-Nay R, Lapato DM, Wolen AR, Lancaster EE, Webb BT, Verhulst B, et al. An epigenome-wide association study of early-onset major depression in monozygotic twins. Transl Psychiatry. 2020;10:301. 64. Marabita F, Almgren M, Lindholm ME, Ruhrmann S, Fagerström-Billai F, Jagodic M, et al. An evaluation of analysis pipelines for DNA methylation proﬁling using the Illumina HumanMethylation450 BeadChip platform. Epigenetics. 2013;8:333–46. 92. Paylor R, Glaser B, Mupo A, Ataliotis P, Spencer C, Sobotka A, et al. Tbx1 hap- loinsufﬁciency is linked to behavioral disorders in mice and humans: Implica- tions for 22q11 deletion syndrome. Proc Natl Acad Sci USA. 2006;103:7729–34. 65. Teschendorff AE, Zhuang J, Widschwendter M. Independent surrogate variable analysis to deconvolve confounding factors in large-scale microarray proﬁling studies. Bioinformatics. 2011;27:1496–505. 93. Berkel S, Marshall CR, Weiss B, Howe J, Roeth R, Moog U, et al. Mutations in the SHANK2 synaptic scaffolding gene in autism spectrum disorder and mental retardation. Nat Genet. 2010;42:489–91. 66. Johnson WE, Li C, Rabinovic A. Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics. 2007;8:118–27. 94. Maccani JZJ, Koestler DC, Lester B, Houseman EA, Armstrong DA, Kelsey KT, et al. Placental DNA methylation related to both infant toenail mercury and adverse neurobehavioral outcomes. Environ Health Perspect. 2015;123:723–9. 67. Buhule OD, Minster RL, Hawley NL, Medvedovic M, Sun G, Viali S, et al. REFERENCES Ginsberg MR, Rubin RA, Falcone T, Ting AH, Natowicz MR. Brain transcriptional and epigenetic associations with autism. PLoS ONE. 2012;7:e44736. disorders, 4th edn. American Psychiatric Association; 1994. 55. Moran S, Arribas C, Esteller M. Validation of a DNA methylation microarray for 850,000 CpG sites of the human genome enriched in enhancer sequences. Epigenomics. 2016;8:389–99. 30. Iakoucheva LM, Muotri AR, Sebat J. Getting to the cores of autism. Cell. 2019;178:1287–98. 56. Buxbaum JD, Daly MJ, Devlin B, Lehner T, Roeder K, State MW. The autism sequencing consortium: large-scale, high-throughput sequencing in autism spectrum disorders. Neuron. 2012;76:1052–6. 31. Stenz L, Rahban R, Prados J, Nef S, Paoloni-Giacobino A. Genetic resistance to DEHP-induced transgenerational endocrine disruption. PLoS ONE. 2019;14:e0208371. 57. Fu JM, Satterstrom FK, Peng M, Brand H, Collins RL, Dong S, et al. Rare coding variation provides insight into the genetic architecture and phenotypic context of autism. Nat Genet. 2022;54:1320–31. 32. Kubota T, Mochizuki K. Epigenetic effect of environmental factors on autism spectrum disorders. Int J Environ Res Public Health. 2016;13:504. 33. Emberti Gialloreti L, Mazzone L, Benvenuto A, Fasano A, Alcon AG, Kraneveld A, et al. Risk and protective environmental factors associated with autism spec- trum disorder: evidence-based principles and recommendations. J Clin Med. 2019;8:217. 58. Morris TJ, Butcher LM, Feber A, Teschendorff AE, Chakravarthy AR, Wojdacz TK, et al. ChAMP: 450k chip analysis methylation pipeline. Bioinformatics. 2014;30:428–30. 34. Min JL, Hemani G, Hannon E, Dekkers KF, Castillo-Fernandez J, Luijk R, et al. Genomic and phenotypic insights from an atlas of genetic effects on DNA methylation. Nat Genet. 2021;53:1311–21. 59. Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minﬁ: a ﬂexible and comprehensive Bioconductor package for the analysis of Inﬁnium DNA methylation microarrays. Bioinformatics. 2014;30:1363–9. Translational Psychiatry (2023) 13:109 S. Perini et al. 8 60. Zhou W, Laird PW, Shen H. Comprehensive characterization, annotation and innovative use of Inﬁnium DNA methylation BeadChip probes. Nucleic Acids Res. 2017;45:e22. 86. Silva TC, Coetzee SG, Gull N, Yao L, Hazelett DJ, Noushmehr H, et al. ELMER v.2: an R/Bioconductor package to reconstruct gene regulatory networks from DNA methylation and transcriptome proﬁles. Bioinformatics. 2018;35:1974–7. y 87. Võsa U, Claringbould A, Westra H-J, Bonder MJ, Deelen P, Zeng B, et al. Large- scale cis- and trans-eQTL analyses identify thousands of genetic loci and poly- genic scores that regulate blood gene expression. Nat Genet. 2021;53:1300–10. 61. REFERENCES Stratiﬁed randomization controls better for batch effects in 450K methylation analysis: a cautionary tale. Front Genet. 2014;5:354. 68. Nygaard V, Rødland EA, Hovig E. Methods that remove batch effects while retaining group differences may lead to exaggerated conﬁdence in downstream analyses. Biostatistics. 2016;17:29–39. 95. Wang M, Guo J, Zhang L, Kuek V, Xu J, Zou J. Molecular structure, expression, and functional role of Clec11a in skeletal biology and cancers. J Cell Physiol. 2020;235:6357–65. 69. Price EM, Robinson WP. Adjusting for batch effects in DNA methylation microarray data, a lesson learned. Front Genet. 2018;9:83. 96. Sato D, Lionel AC, Leblond CS, Prasad A, Pinto D, Walker S, et al. SHANK1 Deletions in Males with Autism Spectrum Disorder. Am J Hum Genet. 2012;90:879–87. 70. Lehne B, Drong AW, Loh M, Zhang W, Scott WR, Tan S-T, et al. A coherent approach for analysis of the Illumina HumanMethylation450 BeadChip improves data quality and performance in epigenome-wide association studies. Genome Biol. 2015;16:37. 97. Hallmayer J, Cleveland S, Torres A, Phillips J, Cohen B, Torigoe T, et al. Genetic heritability and shared environmental factors among twin pairs with autism. Arch Gen Psychiatry. 2011;68:1095–102. 71. R Foundation for Statistical Computing. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing, Vienna, Austria; 2020. 98. Kalkbrenner AE, Schmidt RJ, Penlesky AC. Environmental chemical exposures and autism spectrum disorders: a review of the epidemiological evidence. Curr Probl Pediatr Adolesc Health Care. 2014;44:277–318. 99. Krumm N, O’Roak BJ, Shendure J, Eichler EEA. de novo convergence of autism genetics and molecular neuroscience. Trends Neurosci. 2014;37:95–105. 72. Houseman EA, Accomando WP, Koestler DC, Christensen BC, Marsit CJ, Nelson HH, et al. DNA methylation arrays as surrogate measures of cell mixture dis- tribution. BMC Bioinform. 2012;13:86. 100. Vogel Ciernia A, LaSalle J. The landscape of DNA methylation amid a perfect storm of autism aetiologies. Nat Rev Neurosci. 2016;17:411–23. 73. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W. et al. limma powers dif- ferential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. 101. Anney R, Klei L, Pinto D, Almeida J, Bacchelli E, Baird G, et al. Individual common variants exert weak effects on the risk for autism spectrum disorders. Hum Mol Genet. 2012;21:4781–92. variants exert weak effects on the risk for autism spectrum disorders. Hum Mol Genet. 2012;21:4781–92. 74. Smyth GK. REFERENCES Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol. 2004;3:3. 102. Bourgeron T. From the genetic architecture to synaptic plasticity in autism spectrum disorder. Nat Rev Neurosci. 2015;16:551–63. 75. Peters TJ, Buckley MJ, Statham AL, Pidsley R, Samaras KV, Lord R, et al. De novo identiﬁcation of differentially methylated regions in the human genome. Epi- genetics Chromatin. 2015;8:6. 103. Siniscalco D, Mijatovic T, Bosmans E, Cirillo A, Kruzliak P, Lombardi VC, et al. Decreased numbers of CD57+CD3- cells identify potential innate immune dif- ferences in patients with autism spectrum disorder. Vivo. 2016;30:83–89. 76. Liaw A, Wiener M. Classiﬁcation and regression by randomForest. R News. 2002;2:18–22. 104. Enstrom AM, Lit L, Onore CE, Gregg JP, Hansen RL, Pessah IN, et al. Altered gene expression and function of peripheral blood natural killer cells in children with autism. Brain Behav Immun. 2009;23:124–33. 77. Kuhn M. caret: classiﬁcation and regression training. R package version. 2022;6:0–93. 105. Ashwood P, Corbett BA, Kantor A, Schulman H, Van de Water J, Amaral DG. In search of cellular immunophenotypes in the blood of children with autism. PLoS ONE. 2011;6:e19299. 78. Geeleher P, Hartnett L, Egan LJ, Golden A, Raja Ali RA, Seoighe C. Gene-set analysis is severely biased when applied to genome-wide methylation data. Bioinformatics. 2013;29:1851–7. 79. Ren X, Kuan PF. methylGSA: a Bioconductor package and Shiny app for DNA methylation data length bias adjustment in gene set testing. Bioinformatics. 2018;35:1958–9. 106. Vojdani A, Mumper E, Granpeesheh D, Mielke L, Traver D, Bock K, et al. Low natural killer cell cytotoxic activity in autism: the role of glutathione, IL-2 and IL- 15. J Neuroimmunol. 2008;205:148–54. 107. Warren RP, Foster A, Margaretten NC. Reduced natural killer cell activity in autism. J Am Acad Child Adolesc Psychiatry. 1987;26:333–5. 80. Gandal MJ, Haney JR, Parikshak NN, Leppa V, Ramaswami G, Hartl C, et al. Shared molecular neuropathology across major psychiatric disorders parallels polygenic overlap. Science. 2018;359:693–7. 108. Jiao Y, Huntington ND, Belz GT, Seillet C. Type 1 innate lymphoid cell biology: lessons learnt from natural killer cells. Front Immunol. 2016;7:426. 108. Jiao Y, Huntington ND, Belz GT, Seillet C. Type 1 innate lymp 81. SFARI gene database [Internet]. 2021. Available from: https://gene.sfari.org/ database/human-gene/. lessons learnt from natural killer cells. Front Immunol. 2016;7:426 109. Nardone S, Elliott E. 4Unit of Child Neurology and Psychiatry, ASST Spedali Civili of Brescia, Brescia, Italy. 5Department of Psychology and Cognitive Sciences, University of Trento, Rovereto, Italy. 6Child Neurology and Psychiatry Unit, Tor Vergata University of Rome, Rome, Italy. 7Oasi Research Institute - IRCCS, Troina, Italy. 8Department of Translational Medical Sciences, Federico II University, Naples, Italy. 9Faculty of Law, University of Trento, Trento, Italy. 10Child Psychopathology Unit, Scientiﬁc Institute, IRCCS Eugenio Medea, Bosisio Parini, LC, Italy. 11Child Neuropsychiatry, Department of Surgical Sciences, Dentistry, Gynecology and Pediatrics, University of Verona, Verona, Italy. 12Fondazione Smith Kline, Verona, Italy. 13Child Neuropsychiatry Complex Operative Unit, University of Bari “Aldo Moro”, Bari, Italy. 14Department of Basic Medical Sciences, Neuroscience and Sense Organs, University of Bari Aldo Moro, Bari, Italy. 15Center for Autism Spectrum Disorders, Child Psychiatry Unit, Provincial Health Service of Catania (ASP CT), Catania, Italy. 16Child and Adolescent Psychiatry Unit, Department of Biomedical Sciences, University of Cagliari, Cagliari, Italy. 17Department of Clinical and Experimental Sciences, University of Brescia, Brescia, Italy. 18Children and Adolescents Neuropsychiatry Unit, Woman and Child’s Health Department, Padova University Hospital, Padova, Italy. 19Center for Autism Spectrum Disorder - Child and Adolescent Neuropsychiatry Unit, Rimini - Romagna Health Department, Rimini, Italy. 20Child and Adolescent Neuropsychiatry Unit, Infermi Hospital, Rimini, Italy. 21Department of Neurosciences, Biomedicine and Movement Sciences, Section of Biology and Genetics, University of Verona, Verona, Italy. 22IRCCS Stella Maris Foundation, Pisa, Italy. 23GRIN Therapeutics Inc. New York, New York, USA. 24Department of Mental and Physical Health and Preventive Medicine, University of Campania, Naples, Italy. 25Child Neurology and Psychiatry Unit, IRCCS ISNB, S. Orsola-Malpighi Hospital, Department of Medical and Surgical Sciences, University of Bologna, Bologna, Italy. 26Centro Ricerche Cliniche di Verona, Integrated University Hospital Verona, Verona, Italy. 27Child and Adolescent Neuropsychiatry Unit, Maternal-Child Integrated Care Department, Integrated University Hospital Verona, Verona, Italy. Reprints and permission information is available at http://www.nature.com/ reprints Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 120. Hu VW, Hong Y, Xu M, Shu HT. Altered DNA methylation in a severe subtype of idiopathic autism: Evidence for sex differences in affected metabolic pathways. Autism. 2021;25:887–910. 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/. REFERENCES The interaction between the immune system and epige- netics in the etiology of autism spectrum disorders. Front Neurosci. 2016;10:329. 82. Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, et al. Tissue-based map of the human proteome. Science. 2015;347:1260419. 110. Chen S-W, Zhong X-S, Jiang L-N, Zheng X-Y, Xiong Y-Q, Ma S-J, et al. Maternal autoimmune diseases and the risk of autism spectrum disorders in offspring: a 110. Chen S-W, Zhong X-S, Jiang L-N, Zheng X-Y, Xiong Y-Q, Ma S-J, et al. Maternal autoimmune diseases and the risk of autism spectrum disorders in offspring: a systematic review and meta-analysis. Behav Brain Res. 2016;296:61–69. 83. Koopmans F, van Nierop P, Andres-Alonso M, Byrnes A, Cijsouw T, Coba MP, et al. SynGO: an evidence-based, expert-curated knowledge base for the synapse. Neuron. 2019;103:217.e214. 111. Jangjoo M, Goodman SJ, Choufani S, Trost B, Scherer SW, Kelley E, et al. An epigenetically distinct subset of children with autism spectrum disorder resulting from differences in blood cell composition. Front Neurol. 2021;12:612817. 84. Dong D, Tian Y, Zheng SC, Teschendorff AE. ebGSEA: an improved Gene Set Enrichment Analysis method for Epigenome-Wide-Association Studies. Bioin- formatics. 2019;35:3514–6. 112. Gyurova IE, Schlums H, Sucharew H, Ambroggio L, Ochayon DE, Win HT, et al. Dynamic changes in natural killer cell subset frequencies in the absence of cytomegalovirus infection. Front Immunol. 2019;10:2728. 85. Goeman JJ, van de Geer SA, de Kort F, van Houwelingen HC. A global test for groups of genes: testing association with a clinical outcome. Bioinformatics. 2004;20:93–99. Translational Psychiatry (2023) 13:109 Translational Psychiatry (2023) 13:109 S. Perini et al. 9 AUTHOR CONTRIBUTIONS 113. Ziller MJ, Gu H, Müller F, Donaghey J, Tsai LT, Kohlbacher O, et al. Charting a dynamic DNA methylation landscape of the human genome. Nature. 2013;500:477–81. ED conceptualized and designed the study and contributed to the interpretation of the data. SP performed the bioinformatic analysis and prepared the manuscript including ﬁgures. MF contributed to the design of the study and the analysis of the data. All authors reviewed and edited the manuscript. The Italian Autism Network (Fondazione ITAN). 114. Sala C, Vicidomini C, Bigi I, Mossa A, Verpelli C. Shank synaptic scaffold proteins: keys to understanding the pathogenesis of autism and other synaptic disorders. J Neurochem. 2015;135:849–58. 115. Leblond CS, Nava C, Polge A, Gauthier J, Huguet G, Lumbroso S, et al. Meta- analysis of SHANK mutations in autism spectrum disorders: a gradient of severity in cognitive impairments. PLoS Genet. 2014;10:e1004580. ADDITIONAL INFORMATION Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41398-023-02407-4. 118. Zhu L, Wang X, Li X-L, Towers A, Cao X, Wang P, et al. Epigenetic dysregulation of SHANK3 in brain tissues from individuals with autism spectrum disorders. Hum Mol Genet. 2013;23:1563–78. Correspondence and requests for materials should be addressed to Enrico Domenici. 119. Lim S, Naisbitt S, Yoon J, Hwang JI, Suh PG, Sheng M, et al. Characterization of the Shank family of synaptic proteins. Multiple genes, alternative splicing, and differential expression in brain and development. J Biol Chem. 1999;274:29510–8. Reprints and permission information is available at http://www.nature.com/ reprints COMPETING INTERESTS 116. May HJ, Jeong J, Revah-Politi A, Cohen JS, Chassevent A, Baptista J, et al. Truncating variants in the SHANK1 gene are associated with a spectrum of neurodevelopmental disorders. Genet Med. 2021;23:1912–21. The authors declare no competing interests. 117. Bahado-Singh RO, Vishweswaraiah S, Aydas B, Mishra NK, Yilmaz A, Guda C, et al. Artiﬁcial intelligence analysis of newborn leucocyte epigenomic markers for the prediction of autism. Brain Res. 2019;1724:146457. ACKNOWLEDGEMENTS The authors would like to thank all family members who provided their time and samples that made this investigation possible. We would like to acknowledge the Scientiﬁc Committee of the Italian Autism Network - Fondazione ITAN) for continuous support, and Mike Talkowski and the Autism Sequencing Consortium for genotyping the DNA samples and for exome sequence data. This work was supported by a grant from the Fondazione Italiana Autismo (FIA-2018/53). Michele Filosi was supported by the Strategic Project TRAIN - Trentino Autism Initiative. This article is dedicated to the memory of our friend and colleague Alessandro Zuddas, whose contribution to the development of the Italian Autism Network is deeply acknowledged. © The Author(s) 2023 THE ITALIAN AUTISM NETWORK Giovanni Allibrio4, Ilaria Basadonne5, Arianna Benvenuto6, Seraﬁno Buono7, Carmela Bravaccio8, Carlo Casonato9, Elisa Ceppi10, Paolo Curatolo6, Bernardo Dalla Bernardina11, Lucio Da Ros12, Francesca Darra11, Anna Eusebi4, Alessandra Gabellone13, Andrea De Giacomo14, Grazia Distefano15, Enrico Domenici 1,3✉, Federica Donno16, Maurizio Elia7, Elisa Fazzi4,17, Michela Gatta18, Stefania Giusto15, Serenella Grittani19, Evamaria Lanzarini20, Giovanni Malerba21, Elisa Mani10, Barbara Manzi6, Lucia Margari13, Lucia Marzulli13, Gabriele Masi22, Paola Mattei4, Luigi Mazzone6, Massimo Molteni10, Pierandrea Muglia23, Sebastiano Musumeci7, Antonio Narzisi22, Antonio Pascotto24, Cinzia Pari19, Antonia Parmeggiani25, Maria Giuseppina Petruzzelli13, Alessia Raffagnato18, Emiliangelo Ratti26, Maria Paola Rossaro18, Maria Pia Riccio8, Paolo Rizzini12, Renato Scifo15, Martina Siracusano6, Raffaella Tancredi22, Alessandra Tiberti4, Elisabetta Trabetti21, Annalisa Traverso18, Paola Venuti5, Leonardo Zoccante27 and Alessandro Zuddas16 Translational Psychiatry (2023) 13:109
https://openalex.org/W3196340693	https://discovery.ucl.ac.uk/id/eprint/10065830/1/Schofield_10-1108_JD-07-2018-0114.pdf	English	null	Transforming scholarship in the archives through handwritten text recognition: Transkribus as a case study	CrimRxiv	2,019	cc-by	14,660	The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0022-0418.htm The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0022-0418.htm The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0022-0418.htm 954 Received 18 July 2018 Revised 20 December 2018 Accepted 27 December 2018 Guenter Muehlberger, Louise Seaward, Melissa Terras, Sofia Ares Oliveira, Vicente Bosch, Maximilian Bryan, Sebastian Colutto, Hervé Déjean, Markus Diem, Stefan Fiel, Basilis Gatos, Albert Greinoecker, Tobias Grüning, Guenter Hackl, Vili Haukkovaara, Gerhard Heyer, Lauri Hirvonen, Tobias Hodel, Matti Jokinen, Philip Kahle, Mario Kallio, Frederic Kaplan, Florian Kleber, Roger Labahn, Eva Maria Lang, Sören Laube, Gundram Leifert, Georgios Louloudis, Rory McNicholl, Jean-Luc Meunier, Johannes Michael, Elena Mühlbauer, Nathanael Philipp, Ioannis Pratikakis, Joan Puigcerver Pérez, Hannelore Putz, George Retsinas, Verónica Romero, Robert Sablatnig, Joan Andreu Sánchez, Philip Schofield, Giorgos Sfikas, Christian Sieber, Nikolaos Stamatopoulos, Tobias Strauß, Tamara Terbul, Alejandro Héctor Toselli, Berthold Ulreich, Mauricio Villegas, Enrique Vidal, Johanna Walcher, Max Weidemann, Herbert Wurster and Konstantinos Zagoris (Information about the authors can be found at the end of this article.) Received 18 July 2018 Revised 20 December 2018 Accepted 27 December 2018 Transforming scholarship in the archives through handwritten text recognition Transkribus as a case study JD 75,5 JD 75,5 954 Journal of Documentation Vol. 75 No. 5, 2019 pp. 954-976 Emerald Publishing Limited 0022-0418 DOI 10.1108/JD-07-2018-0114 Abstract Purpose – An overview of the current use of handwritten text recognition (HTR) on archival manuscript material, as provided by the EU H2020 funded Transkribus platform. It explains HTR, demonstrates Transkribus, gives examples of use cases, highlights the affect HTR may have on scholarship, and evidences this turning point of the advanced use of digitised heritage content. The paper aims to discuss these issues. © Guenter Muehlberger, Louise Seaward, Melissa Terras, Sofia Ares Oliveira, Vicente Bosch, Maximilian Bryan, Sebastian Colutto, Hervé Déjean, Markus Diem, Stefan Fiel, Basilis Gatos, Albert Greinoecker, Tobias Grüning, Guenter Hackl, Vili Haukkovaara, Gerhard Heyer, Lauri Hirvonen, Tobias Hodel, Matti Jokinen, Philip Kahle, Mario Kallio, Frederic Kaplan, Florian Kleber, Roger Labahn, Eva Maria Lang, Sören Laube, Gundram Leifert, Georgios Louloudis, Rory McNicholl, Jean-Luc Meunier, Johannes Michael, Elena Mühlbauer, Nathanael Philipp, Ioannis Pratikakis, Joan Puigcerver Pérez, Hannelore Putz, George Retsinas, Verónica Romero, Robert Sablatnig, Joan Andreu Sánchez, Philip Schofield, Giorgos Sfikas, Christian Sieber, Nikolaos Stamatopoulos, Tobias Strauß, Tamara Terbul, Alejandro Héctor Toselli, Berthold Ulreich, Mauricio Villegas, Enrique Vidal, Johanna Walcher, Max Weidemann, Herbert Wurster and Konstantinos Zagoris. Published by Emerald Publishing Limited. This article is published under the Creative Commons Attribution (CC BY 4.0) licence. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial & non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this licence may be seen at http://creativecommons. org/licences/by/4.0/legalcode Journal of Documentation Vol. 75 No. 5, 2019 pp. 954-976 Emerald Publishing Limited 0022-0418 DOI 10.1108/JD-07-2018-0114 Transforming scholarship in the archives Transforming scholarship in the archives Design/methodology/approach – This paper adopts a case study approach, using the development an delivery of the one openly available HTR platform for manuscript material. i di k ib h d d h i bl h l h b l d i Findings – Transkribus has demonstrated that HTR is now a useable technology that can be employed in conjunction with mass digitisation to generate accurate transcripts of archival material. Use cases are demonstrated, and a cooperative model is suggested as a way to ensure sustainability and scaling of the platform. However, funding and resourcing issues are identified. p , g g Research limitations/implications – The paper presents results from projects: further user studies coul be undertaken involving interviews, surveys, etc. Abstract 955 g y Practical implications – Only HTR provided via Transkribus is covered: however, this is the only publicly available platform for HTR on individual collections of historical documents at time of writing and it represents the current state-of-the-art in this field. Social implications – The increased access to information contained within historical texts has th potential to be transformational for both institutions and individuals. p Originality/value – This is the first published overview of how HTR is used by a wide archival studies community, reporting and showcasing current application of handwriting technology in the cultural heritage sector. Keywords User studies, Library, Archives, Transcription, Neural networks, Digital humanities, Digital library infrastructure, Handwritten text recognition, HTR, Transcribing Introduction Archives are increasingly investing in the digitisation of their manuscript collections but until recently the textual content of the resulting digital images has only been available to those who have the time to study and transcribe individual passages. The use of computers to process and search images of historical papers using handwritten text recognition (HTR) has the potential to transform access to our written past for the use of researchers, institutions and the general public. This paper reports on the Recognition and Enrichment of Archival Documents (READ) European Union Horizon 2020 project which is developing advanced text recognition technology on the basis of artificial neural networks and resulting in a publicly available infrastructure: the Transkribus platform. Users of Transkribus (whether institutional or individual) are able to extract data from handwritten and printed texts via HTR, while simultaneously contributing to the improvement of the same technology thanks to machine learning principles. The automated recognition of a wide variety of historical texts has significant implications for the accessibility of the written records of global cultural heritage. Thi h T k ib l f d f i h d l This paper uses the Transkribus platform as a case study, focusing on the development, application and impact of HTR technology. It demonstrates that HTR has the capacity to make a significant contribution to the archival mission by making it easier for anyone to read, transcribe, process and mine historical documents. It shows that the technology fits neatly into the archival workflow, making direct use of growing repositories of digitised images of historical texts. By providing examples of institutions and researchers who are generating new resources with Transkribus, the paper shows how HTR can extend the existing research infrastructure of the archives, libraries and humanities domain. Looking to the future, this paper argues that this form of machine learning has the potential to change the nature and scope of historical research. Finally, it suggests that a cooperative approach from the archives, library and humanities community is the best way to support and sustain the benefits of the technology offered through Transkribus. Handwritten text recognition – an overview HTR i i h i h i HTR is an active research area in the computational sciences, dating back to the mid- twentieth century (Dimond, 1957). HTR was originally closely aligned to the development of optical character recognition (OCR) technology, where scanned images of printed text are converted into machine-encoded text, generally by comparing individual characters with existing templates (Govindan and Shivaprasad, 1990; Schantz, 1982; Ul-Hasan et al., 2016). HTR developed into a research area in its own right due to the variability of different hands, and the computational complexity of the task (Bertolami and Bunke, 2008; Kichuk, 2015; JD 75,5 JD 75,5 Leedham, 1994; Sudholt and Fink, 2016). Statistical advances in the 1980s, and advanced pattern recognition combined with artificial intelligence in the 1990s were followed by the development of deep neural network approaches in the 2000s and 2010s[1]. This, combined with the availability of increased computer processing power, has resulted in improvements in the recognition of handwritten historical documents, as is regularly evidenced at scientific competitions in the two major conferences in this area: the International Conference of Document Analysis and Recognition and the International Conference on Frontiers in Handwriting Recognition. Researchers originally developed this technology with handwritten materials in mind and it is widely known in the computer science field under the initials HTR. However, the technology can equally be applied to early printed texts that are too complex to be processed adequately with OCR techniques. 956 Most prior application of HTR has been in the financial and commercial sectors (e.g. for postal address interpretation (Pal et al., 2012), bank-cheque processing (Dimauro et al., 1997), signature verification (Hafemann et al., 2015) and biometric writer identification (Morera et al., 2018)). However, recent successes in HTR coincide with the availability of affordable, high-quality digital imaging technologies, related online systems for hosting images and subsequent programs of mass digitisation which are being carried out by most major libraries and archives worldwide to increase access to their collections (Borowiecki and Navarrete, 2016; Ogilvie, 2016; Terras, 2010). Unfortunately, it has long been the problem that: […] there are growing numbers of scanned manuscripts that current OCR and handwriting recognition techniques cannot transcribe, because the systems are not trained for the scripts in which these manuscripts are written. Documents in this category range from illuminated medieval manuscripts to handwritten letters to early printed works. Without transcriptions, these documents remain unsearchable. (Edwards, 2007, p. The Transkribus platform Various projects have undertaken work on the OCR of early printed materials and experimented with the recognition of handwritten manuscripts (Bulacu et al., 2009; Edwards, 2007; Firmani et al., 2018; Fischer et al., 2009; Springmann and Lüdeling, 2017; Terras, 2006; Weber et al., 2018; Van der Zant et al., 2009). However, there has not yet been sufficient interdisciplinary work applying deep neural network models to manuscript material and more importantly, there was previously no user-friendly platform to make this technology accessible. With Transkribus, historical manuscripts of all dates, languages and formats can be read, transcribed and searched by means of automated recognition. The Transkribus research infrastructure aims to provide a complete and reliable workflow for this process. Users work with Transkribus to create “ground truth”[8] data that is suitable for machine learning. From submitted images and transcripts, the HTR engines[9] learn to decipher (historical) handwritten or printed text from digital images and can then automatically generate transcripts of similar material. Transkribus was conceived and launched in 2015 with funding from the European Commission’s Seventh Framework Programme (FP7), as part of the tranScriptorium project (2013–2015)[10]. Whilst tranScriptorium largely focused on computer science research, the implementation and development of Transkribus was placed at the centre of a successor project called READ (2016–2019). Funded once more by the European Commission, this time under the Horizon 2020 scheme, READ aims to maintain, develop and promote a functioning online research infrastructure where new technologies can feed innovation in archival research. Dr Günter Mühlberger of the University of Innsbruck coordinates the READ project, working with collaborators from thirteen other European universities and research institutions[11] representing ten teams of computer scientists and developers who contribute to and construct components of the Transkribus infrastructure[12], but also teams engaged in humanities scholarship and institutions with archival material. Transkribus services are currently freely available online, and directed towards four intended user groups: archivists, humanities scholars, computer scientists and members of the public, all of whom are interested in the study and exploitation of historical documents. The interests of these user groups overlap and each make a vital contribution to the Transkribus infrastructure. Memory institutions, humanities scholars and the public can provide digitised images and transcripts as ground truth for HTR training, whilst computer scientists deliver the necessary research to sustain this technology. Handwritten text recognition – an overview HTR i i h i h i 1) […] there are growing numbers of scanned manuscripts that current OCR and handwriting recognition techniques cannot transcribe, because the systems are not trained for the scripts in which these manuscripts are written. Documents in this category range from illuminated medieval manuscripts to handwritten letters to early printed works. Without transcriptions, these documents remain unsearchable. (Edwards, 2007, p. 1) Mass digitisation of historical material, in combination with traditional archival catalogues and finding aids, is already broadening access to document collections. Automated transcription and searching of digitised texts goes further, expanding the existing possibilities of historical enquiry for scholars, institutions, commercial providers and other users. Successful development of HTR will improve and increase access to collections, allowing users to quickly and efficiently pinpoint particular topics, words, people, places and events in documents, but also changing the understanding of context, and multiplying research possibilities. The generation of machine-readable textual transcripts will provide the basis for advanced semantic, linguistic, and geo-spatial computational analysis of historical primary source material (see Gregory et al., 2015; Meroño-Peñuela et al., 2015; Weisser, 2016 for possibilities). The research questions which can then be asked of historic manuscripts change: the way institutions can deliver and present archival material will be similarly transformed (Estill and Levy, 2016). Commercial digitisation providers are moving into this space, undertaking digitisation on behalf of under-funded institutions and licensing back access to the resulting resources. As of early 2018, Adam Matthew Digital describes itself as “currently the only publisher to utilise artificial intelligence to offer Handwritten Text Recognition (HTR) for its handwritten manuscript collections” (Adam Matthew Digital, 2018). At the time of writing, it offers the same software as Transkribus, allowing HTR-based searching across several of its themed digitised archive collections and starting to provide HTR as a part of a collection management service via its Quartex platform[2]. However, this commercial exercise restricts HTR to contributing organisations and means that researchers and other individuals are unable to engage with the development and application of the technology. Machine learning is not a panacea and critical appraisal of its training process and its underlying data are essential if this technology is to be integrated into archival practice and scholarly research in a meaningful way. Handwritten text recognition – an overview HTR i i h i h i Transforming scholarship in the archives Transforming scholarship in the archives It is within this framework, and with open aims, that the large-scale READ[3] research initiative has provided Transkribus[4] as the platform to deliver HTR technology to institutions and individual users. Although the READ project has published numerous research papers on the computational aspects of HTR[5], as well as data sets[6] and other project deliverables[7], this is the first publication from the project to cover the research programme from the perspective of the active user community. In considering examples of projects working with Transkribus, it indicates that the combination of HTR and digitised content has potential to extend existing methods of scholarship in significant new directions. 957 Figure 1. Transkribus stakeholders, representing what they can contribute, but also what they can gain, from interacting with the HTR technological platform supplied by READ The Transkribus platform Each user group can also derive tangible benefits from the initiative: archives can deliver searchable digitised collections for their users, humanities scholars can conduct research efficiently and members of the public can study their family history or contribute by transcribing or correcting transcripts of historical documents. Computer scientists can also request to reuse a wealth of data, in the form of images and transcripts of historical material, for their HTR research. This growing user network is central to the success of Transkribus: machine learning means that HTR becomes stronger with every document processed in the platform (Figure 1). Since 2015, the Digitisation and Electronic Archiving (DEA) group at the University of Innsbruck have been responsible for maintaining and developing the Transkribus Graphical User Interface (GUI): currently freely available via a downloadable, Java-based programme (for Windows, MacOS and Linux) that requires local installation[13]. Access to the platform is granted to users who have registered for an account at the Transkribus website. The Transkribus GUI stabilised with the release of version 1.0 at the beginning of 2017 (Kahle et al., 2017): its most important features are fully integrated and functional, although new developments continue to improve the platform. In addition, all services and uploaded documents are also available in machine-readable form via a REST interface[14]. Most of the Transkribus software is available open source via GitHub[15], with two exceptions. The HTR module in Transkribus is licensed for use within the READ project by the Computational Intelligence Laboratory (CITlab) at the University of Rostock (one of the READ project members) and its collaborator, the German technology company PLANET[16]. An open-source neural network toolkit for HTR, known as Laia, has been created by the Pattern Recognition and Human Language Technology (PRHLT) research centre at the Polytechnic University of Valencia (another of the READ project members) [17]. It will be integrated into the Transkribus GUI in 2019. The server component of Transkribus is also restricted at present in order to avoid the establishment of a competing service platform. The Transkribus platform ARCHIVES Deliver Documents Work with documents Get results Contribute (Crowd-Sourcing) Advanced search Technology Data Enriched documents STORAGE TOOLS EXPERT INTERFACES WEB UI TRANS- KRIBUS COMPUTER SCIENCE AND TECHNOLOGY PROVIDERS HUMANITIES SCHOLARS PUBLIC VOLUNTEERS/CROWD ARCHIVES Deliver Documents Work with documents Get results Contribute (Crowd-Sourcing) Advanced search Technology Data Enriched documents STORAGE TOOLS EXPERT INTERFACES WEB UI TRANS- KRIBUS COMPUTER SCIENCE AND TECHNOLOGY PROVIDERS HUMANITIES SCHOLARS PUBLIC VOLUNTEERS/CROWD Figure 1. Transkribus stakeholders, representing what they can contribute, but also what they can gain, from interacting with the HTR technological platform supplied by READ 958 JD 75,5 JD 75,5 JD 75 5 JD Enriched documents Deliver Documents 958 Work with documents Data Get results Technology Advanced search Contribute (Crowd-Sourcing) User Interface (GUI): currently freely available via a downloadable, Java-based programme (for Windows, MacOS and Linux) that requires local installation[13]. Access to the platform is granted to users who have registered for an account at the Transkribus website. The Transkribus GUI stabilised with the release of version 1.0 at the beginning of 2017 (Kahle et al., 2017): its most important features are fully integrated and functional, although new developments continue to improve the platform. In addition, all services and uploaded documents are also available in machine-readable form via a REST interface[14]. Most of the Transkribus software is available open source via GitHub[15], with two exceptions. The HTR module in Transkribus is licensed for use within the READ project by the Computational Intelligence Laboratory (CITlab) at the University of Rostock (one of the READ project members) and its collaborator, the German technology company PLANET[16]. An open-source neural network toolkit for HTR, known as Laia, has been created by the Pattern Recognition and Human Language Technology (PRHLT) research centre at the Polytechnic University of Valencia (another of the READ project members) [17]. It will be integrated into the Transkribus GUI in 2019. The server component of Transkribus is also restricted at present in order to avoid the establishment of a competing service platform. Transkribus workflow Th l d i H The latest advances in HTR research, based on deep neural networks, have been implemented in the Transkribus GUI. Neural networks can be trained to recognise a particular style of writing by examining and processing digitised images and transcriptions of documents. The result of the training process is what is known as an HTR “model”, a computational system tailored to automatically transcribe a set of historical material. HTR technology is language-independent: the neural network training process for any type of alphabet, from any date, is the same. This means there is potential to train up models for any script, from any period (Leifert et al., 2016). The technology follows a line-oriented approach where the image of a baseline (a horizontal line running underneath a line of text in a digitised image) and the corresponding correctly transcribed text represent the input for the learning algorithms of neural networks (Romero et al., 2015). Transforming scholarship in the archives Transforming scholarship in the archives Digitised images and their transcripts are the main prerequisite for working with HTR. They must be pre-processed in the Transkribus GUI in order to become ground truth data that can be used to train an HTR model to transcribe a specific collection of historical material, either that written by one writer or a set of similar types of writing. There are three main stages to creating ground truth in the Transkribus GUI. The first is uploading digitised images to the platform. The second is using layout analysis tools to segment the digitised images into lines. The third is accurately transcribing the text of each of the lines in a digitised image. 959 g g Transkribus accepts a range of image formats, and has sufficient server space to process large collections. When a user uploads images to Transkribus, these images remain private to their user account and are not made publicly available. A collection owner can allow other Transkribus users to view or work with their documents if they wish. Training data for HTR should be representative of the different parts of an archival collection, reflecting an appropriate variety of layouts, vocabulary and writing styles. Users can therefore select specific pages to become ground truth or simply choose pages from regular intervals within a collection (e.g. every tenth page). Transkribus workflow Th l d i H A ground truth data set of 15,000 transcribed words (or around 75 pages) is generally sufficient for training an HTR engine to recognise text written in one hand. A model can be trained to recognise printed text with just 5,000 transcribed words (or around 25 pages). According to the principle of machine learning, the more words of ground truth that a user submits, the more accurate the results are likely to be. Indeed, if a collection contains documents written in several hands or languages, it is recommended that users create ground truth for a higher number of transcribed words. To give an example, one of the strongest HTR models has been trained by the Bentham Project at University College London, one of the members of the READ project[18]. This model was trained on over 50,000 words from papers written by the English philosopher Jeremy Bentham (1748–1832) and his secretaries. In the best cases, it generates an output where around 95 per cent of characters on similar pages from the Bentham collection are transcribed correctly by the programme. This model is publicly available to all Transkribus users under the title “English Writing M1”. The figures which appear later in this paper relate to this model. Once images reside on the Transkribus server, they are ready for layout analysis or segmentation. Recent technological breakthroughs have enhanced the accuracy of this crucial process, making it easier for machines to identify text on archival documents which have more complex layouts (Diem et al., 2017; Grüning et al., 2017; Leifert et al., 2016). The Transkribus GUI contains both automatic and manual segmentation tools that allow users to mark their images with three segmentation elements: text regions around each block of text, line regions around each line of text and baselines running along the bottom of each line of text[19]. Transkribus users can commence automated segmentation on a batch of pages and the tool works in minutes to find the lines in images where words are set out relatively neatly on a page. The results of automated segmentation can sometimes be less precise when documents have a more complicated structure, such as a tabular form. In such cases, a combination of automated and manual segmentation will work to divide a page into lines. HTR engines rely most heavily on information they derive from the baselines running along the bottom of each line of text. Transforming scholarship in the archives Figure 2. Screenshot showing an image of a manuscript loaded into the Transkribus GUI after automated baseline detection Transkribus workflow Th l d i H Users can therefore check the positioning of automatically generated baselines and correct them if necessary (Figure 2). Transcription is the third and final part of creating ground truth. After segmentation, the Transkribus GUI displays a text editor field divided into lines which are connected to the lines drawn on the image. Users need to produce a consistent transcript of each line of the text in the image, replicating any spelling mistakes, unusual symbols or abbreviations. The neural networks can also learn from normalised transcriptions, where abbreviations have been expanded (Thöle, 2017). Users have the option to Sources: Jeremy Bentham, “of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections JD 75,5 JD 75,5 960 Figure 2. Screenshot showing an image of a manuscript loaded into the Transkribus GUI after automated baseline detection Sources: Jeremy Bentham, “of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections transcribe their documents in the Transkribus Web interface[20], a streamlined version of Transkribus that makes transcription simpler and quicker for larger teams or volunteers. The Transkribus GUI has a suite of tagging tools for those users who wish to create rich transcripts that could form part of a digital edition. At the current time, there is no benefit in marking up transcripts that are being prepared as ground truth. HTR engines are programmed to ignore tags and instead focus on recognising text. However, developments in Named Entity Recognition technology should permit the recognition of tagged content in the near future (Figures 3 and 4). In summary, users upload images to the Transkribus GUI, segment each page into lines and then transcribe each page with a high level of consistency. With these three simple steps, ground truth creation is complete. Users who have existing transcriptions of their documents also have the option to truncate the process of creating training data, thanks to a Text2Image matching tool[21]. Once images and text file transcriptions have been uploaded to Transkribus, the Text2Image algorithm seeks to match the lines in the images to the lines of the transcribed text. Only lines that have been matched with a certain predefined confidence value will be included in the training data. Figure 4. Screenshot showing a line from an image of a manuscript transcribed manually by a human transcriber in the Transkribus Web interface Transkribus workflow Th l d i H Image courtesy of UCL Library Services, Special Collections 961 Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Figure 3. Screenshot showing an image of a manuscript which has been segmented automatically and then manually transcribed by a human transcriber in the Transkribus GUI Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections their ground truth pages. At this stage, users can also send files containing relevant dictionaries or vocabulary lists which can improve the accuracy of the recognition. The process of model generation is complex: the learning effect of the HTR is achieved by adapting its respective hypotheses to the existing training data in an iterative process and thus independently finding those rules which provide the best output (the correct text) with a given input (the picture of the line), but for the user in Transkribus this complexity is resolved to a few parameters (Grüning et al., 2016; Sánchez et al., 2014, 2017; Strauß et al., 2016; Weidemann et al., 2017). It takes between several hours and several days to train a model, depending on the size of the training data and the load on the computing infrastructure. The actual result of the training process is a model which is capable of recognising handwritten or printed documents which are similar to the ground truth. However, the output is not the transcription of the page itself, but rather a confidence matrix showing the likelihood of the appearance of each character in the alphabet at a given spot in the image of a line. With this confidence matrix further actions are possible, such as their ground truth pages. Figure 3. Screenshot showing an image of a manuscript which has been segmented automatically and then manually transcribed by a human transcriber in the Transkribus GUI Transkribus workflow Th l d i H The Text2Image matching tool therefore represents a simple and cost-efficient entry point into ground truth production and HTR for those who have collated existing transcriptions. g p Users can request access to train their own HTR models or e-mail the Transkribus team at the University of Innsbruck to request that a model be trained to recognise the text from Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Figure 3. Screenshot showing an image of a manuscript which has been segmented automatically and then manually transcribed by a human transcriber in the Transkribus GUI Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Figure 4. Screenshot showing a line from an image of a manuscript transcribed manually by a human transcriber in the Transkribus Web interface 961 Transforming scholarship in the archives Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Figure 3 Screenshot showin an image of manuscript which ha been segmente automatically an then manuall transcribed by human transcriber i the Transkribus GU 961 Transforming scholarship in the archive 961 Transforming scholarship in the archives Note: The blue line in the image represents a baseline and its corresponding text is shown in blue in the text editor below Sources: Jeremy Bentham, “Of (Disposition)”, 1778, Box xxvii, Fol. 58, Bentham Papers, UCL Special Collections. Transkribus workflow Th l d i H At this stage, users can also send files containing relevant dictionaries or vocabulary lists which can improve the accuracy of the recognition. The process of model generation is complex: the learning effect of the HTR is achieved by adapting its respective hypotheses to the existing training data in an iterative process and thus independently finding those rules which provide the best output (the correct text) with a given input (the picture of the line), but for the user in Transkribus this complexity is resolved to a few parameters (Grüning et al., 2016; Sánchez et al., 2014, 2017; Strauß et al., 2016; Weidemann et al., 2017). It takes between several hours and several days to train a model, depending on the size of the training data and the load on the computing infrastructure. The actual result of the training process is a model which is capable of recognising handwritten or printed documents which are similar to the ground truth. However, the output is not the transcription of the page itself, but rather a confidence matrix showing the likelihood of the appearance of each character in the alphabet at a given spot in the image of a line. With this confidence matrix further actions are possible, such as JD 75,5 JD 75,5 decoding the confidences into transcribed text, taking them as an input for keyword searching or in the future, using them to correct an automated transcript (Figure 5). Once training is complete, users can access their model in the Transkribus GUI and generate an automated transcript of a page from their ground truth set. Any pages from the same collection that were not used as training data must be uploaded to the Transkribus GUI and then segmented into lines before they too can be automatically transcribed with HTR. In the current set-up 30 pages (with an average of 40 lines on each page) can be automatically transcribed in just over 28 min. It would therefore take 32 days and 18 hours to automatically recognise 50,000 pages. New GPU servers are due to be installed at the University of Innsbruck, which will consequently improve these processing rates. 962 The Transkribus GUI displays standardised information about each HTR model in a particular collection, including its name, the documents on which it was trained and its accuracy level. Figure 5. Workflow for HTR in Transkribus Transkribus workflow Th l d i H Workflow for HTR in Transkribus Transforming scholarship in the archives Transforming scholarship in the archives Transforming scholarship in the archives confidence values assigned to characters as part of the HTR process and recovers all possible matches for a given word (this is known as a “Query by String” approach). The results will return what the engine deems to be the best matches, as well other possible matches for that word based on alternative readings of each character on the page. This means that Keyword Spotting technology can find words in a collection, even if those words have been transcribed incorrectly by HTR. Moreover, it can recognise and retrieve results for words where there are historical or personal variations in spelling. Thus, this form of searching can produce useable results with HTR models that have higher error rates, up to 30 per cent CER (Giotis et al., 2017; Puigcerver et al., 2015, 2017; Retsinas et al., 2016; Strauß et al., 2016; Toselli et al., 2017). The platform displays the results of a Keyword Spotting query as a list of transcribed words, thumbnail images of the portion of the digitised pages on which those words appear and a confidence rating for each word. In a future version of the Transkribus GUI, users will benefit from further research which facilitates search queries relating to partial words and graphical symbols (known as the “Query by Example” approach) (Zagoris et al., 2017). Users will also be able to export their Keyword Spotting results as a data matrix for examining the contents of a document collection. A validation tool is being developed which will help users to easily eliminate incorrect results for their search term and create a controlled index of occurrences of that word (Figure 7). 963 Once HTR has been completed on any given set of documents, it is up to the user to work with the resulting transcriptions in any way they feel appropriate. They can be included in digital editions, subjected to further computational analysis using semantic or linguistic techniques, or (in the case of large scale collections provided by institutions) ingested into Notes: The automated transcription of this page has a Character Error Rate (CER) of around 34 per cent, which is useless for transcription. Nevertheless, the word “magnitude” is identified correctly via Keyword Spotting in Transkribus Sources: Jeremy Bentham, “Annuity Notes”, 1800, Box ii, Fol. Transkribus workflow Th l d i H Users are supplied with a learning curve which indicates the number of words used in the training and the best values achieved in generating the model. The platform determines the overall accuracy of the HTR model using a measurement of Character Error Rate (CER), which refers to the average percentage of characters transcribed incorrectly by the programme[22]. During the training process, a small selection of pages from the ground truth is set aside as a test set and is not used to train the HTR. This means that Transkribus can provide CERs relating to the automated transcription of previously processed pages, as well as unknown pages from the same data set. The platform also has a comparison function that enables users to compute and generate a visualisation of the accuracy of the computer-generated transcription of any page from the ground truth. In the best cases, HTR can produce automated transcripts of handwritten material with a CER of below 5 per cent (meaning that 95 per cent of the characters are correct). Outputs from models trained on printed material can be even better, reaching CERs of 1-2 per cent. The use of dictionaries will, in many cases, improve the HTR results but the accuracy of neural networks on a purely visual level is high. The experience of Transkribus users indicates that transcripts with these accuracy rates can be proofread and corrected relatively quickly, with less effort than would be required to transcribe each page from scratch (Alvermann and Blüggel, 2017) (Figure 6). If an HTR model is less accurate, with a CER of more than 10 per cent, experiments suggest that automated transcriptions become less useful as a research resource in themselves because correcting myriad errors is more time consuming than manual transcription. However, it does not follow that less accurate results are ultimately useless. Indeed, HTR output can still be a solid foundation for searching and indexing vast collections of digitised documents. The Transkribus GUI provides access to a sophisticated searching technology known as Keyword Spotting[23]. This tool searches through the Importing documents Segmentation Create training data Keyword Spotting Apply existing model Model training Recognised text Figure 5. Figure 7. Screenshot showing an image of a line of a text in a manuscript transcribed automatically in the Transkribus GUI using the “English Writing M1” model and dictionary Figure 6. Screenshot showing an image of a manuscript transcribed automatically in the Transkribus GUI using the “English Writing M1” model and dictionary Transforming scholarship in the archives The Transkribus user community The digitised archives and repositories that READ project members have provided are the primary test cases for the development of Transkribus. The Bentham Project at University College London has trained a succession of models with the aim of improving the automated recognition of Jeremy Bentham’s handwriting[24]. Following a collaboration with the PRHLT research team at the Polytechnic University of Valencia, there is now an online platform for the Keyword Spotting of the near entirety of Bentham’s papers (around 90,000 digitised images) [25]. The Bentham Project is also considering how to integrate HTR technology into the workflow of Transcribe Bentham, its scholarly crowdsourcing initiative that asks members of the public to transcribe Bentham’s writings (Causer and Terras, 2014). HTR could provide volunteers with automated transcripts of simple pages to check and correct or help them to decipher complex passages by providing suggested readings of each word on a page (Seaward, 2016). Passau Diocesan Archives[26] are utilising Transkribus to transcribe and search their large collections of sacramental registers (Wurster et al., 2017). Experiments with a set of 1,200 images of death registers written in nineteenth-century German (around 400,000 words written by 40 different scribes) have shown that a CER between 17 and 19 per cent can be achieved. Passau Diocesan Archives are also working to improve the automated Layout Analysis of tabular data by sorting tables from their collection into different categories that can be used as training templates. Improved table recognition and the possibility of exporting tabular data will have significant implications for the field, since many archival documents are laid out in tables and forms (Clinchant et al., 2018). The National Archives of Finland[27] would like to enhance the usability of their vast collections of governmental records, many of which are digitised but not transcribed. They have worked with students and volunteers to produce training data for three collections: nineteenth-century court records written in Swedish, estate inventories of the Finnish nobility also written in nineteenth-century Swedish and diaries from the Second World War written in Finnish (Kallio, 2017). The best results came from the court records, where 75,000 words of ground truth produced a model capable of transcribing pages with a CER of around 12 per cent. The multiplicity of writers in the other two collections meant that the results were somewhat weaker. Transkribus workflow Th l d i H 29, Bentham Papers, UCL Special Collections. Image courtesy of UCL Special Collections Notes: The automated transcription of this page has a Character Error Rate (CER) of around 9 per cent Sources: Jeremy Bentham, “How to measure Pain and Pleasure”, 1775, Box xxvii, Fol. 36a, Bentham Papers, UCL Special Collections. Image courtesy of UCL Library Services, Special Collections Notes: The automated transcription of this page has a Character Error Rate (CER) of around 9 per cent Notes: The automated transcription of this page has a Character Error Rate (CER) of around 34 per cent, which is useless for transcription. Nevertheless, the word “magnitude” is identified correctly via Keyword Spotting in Transkribus JD 75,5 JD 75,5 content management systems to be used as a finding aid to locate the content of collections. There is therefore much potential to support the archival and manuscript studies community via the reliable transcription and searching of handwritten and printed texts. The Transkribus user community A total of 144,000 words of training data for the Second World War diaries led to an output with a CER of around 17 per cent, whilst 99,000 words of the estate inventories trained a model that transcribed pages with a CER of around 24 per cent. The National Archives of Finland will continue to engage students and volunteers in the creation of further training data in the hope of improving these accuracy rates. The State Archives of Zurich[28] had a head start in exploring the potential of HTR technology because they are in possession of nearly 200,000 pages of transcribed text relating to one of the main series of their archival collections from the nineteenth century (Hodel, 2017)[29]. They have experimented with the Text2Image matching tool to pair 100,000 pages of these existing transcripts with corresponding digitised images, laying the groundwork for future training of HTR on a large scale. Training has already been undertaken on part of this data set, which comprises German language documents written between 1848 and 1853, totalling around 2,750,000 words. The output reaches a CER of around 6 per cent when the model is applied to documents written in the same hand. A model has also been trained on a smaller subset of these documents (around 570,000 words from the years 1803–1882) and the results have a CER of around 18 per cent. The accuracy of this model is sufficient to also recognise other texts written in nineteenth-century German and will be used as a basis for Keyword Spotting. From 2015, when the Transkribus GUI became available, scholars, archives, volunteers and computer scientists were encouraged to download and benefit from it At the time of 964 From 2015, when the Transkribus GUI became available, scholars, archives, volunteers and computer scientists were encouraged to download and benefit from it. At the time of Transforming scholarship in the archives Transforming scholarship in the archives writing there are now more than 17,000 registered individual users of the Transkribus platform, and more than 80 institutions and projects, including many libraries and archives, have signed a Memorandum of Understanding with READ[30]. Every month Transkribus users are generating about 130 HTR models trained on collections of different scripts, dates and languages: from medieval Hebrew to twentieth-century German Kurrent. The Transkribus user community Their experiences already demonstrate that the outputs of automated transcription and searching expedite connections with historical material and have the potential to reshape research practice. A detailed study of the Transkribus user community, which will explore the impact of HTR in specific archives and research projects, is forthcoming. The current paper presents an overview of success stories shared at the first Transkribus User Conference at the Technical University of Vienna, Austria, on 2–3 November 2017. 965 The library of the University of Greifswald has been working productively with Transkribus since early 2016, building up a significant set of ground truth based on a collection of minutes from the central administrative body of their University produced by three hands during the late eighteenth and early nineteenth century (Alvermann and Blüggel, 2017). The team have used this expanding data set to train a succession of models with increasingly accurate results. In the latest experiment, 410,000 words of ground truth have created a model that can produce transcripts of parts of the collection with a CER of just 5 per cent. With such a low error rate, even experienced archivists can correct the automated text faster than typing manually. A version of this model, trained on 250,000 words is freely available to all registered Transkribus users under the title of “Konzilsprotokolle v1”, so they can experiment with similar documents and see the potential of HTR. The team at the University of Greifswald have also shown how HTR technology can feed into the archival workflow by using Goobi software to provide online access to the searchable text of 20 volumes of the collection via their library catalogue[31]. The Georgian Papers Programme[32], led by William & Mary University Libraries has a similar desire to use HTR to open up a collection of historical material. The initiative will provide an online platform of transcriptions of papers from the British Royal Collection Trust, some generated by HTR and others by scholars, students and volunteers. First experiments with some 365,000 words of ground truth from a cache of essays by the English King George III (1738–1820) led to an HTR model capable of producing transcripts with a CER of around 16 per cent (Cornell, 2017). The Transkribus user community They have collaborated with the PRHLT centre at the Polytechnic University of Valencia to create an online resource for the Keyword Spotting of a selection of manuscripts relating to Spanish Golden Age theatre[36]. The Bavarian Academy of Sciences and Humanities, the University of Augsburg and the Berlin-Brandenburg Academy of Sciences and Humanities are collaborating on a long-term project to create an annotated digital edition of medieval German translations of the Gospels. They use Transkribus as a transcription tool to manually produce rich and exportable transcripts with XML tags that will form part of this digital edition (Vetter, 2017). 966 p g p g ( , ) Other users are establishing that HTR technology can be applied fruitfully to early printed text. As part of the OCR-D[37] project, designed to improve the automated recognition of texts printed between the sixteenth and nineteenth centuries, the Berlin-Brandenburg Academy of Sciences and Humanities are compiling a large ground truth data set of different printed sources (Boenig and Würzner, 2017). Dario Kampkaspar and colleagues at the Austrian Centre for Digital Humanities (part of the Austrian Academy of Sciences) are already in the process of training a model for a digital edition of the printed text of the eighteenth-century Wienerisches Diarium newspaper[38]. The team use a mixture of OCR (using the ABBY FineReader tool available in the Transkribus GUI) and HTR to produce transcripts, correct these transcripts and then use these corrections to retrain their HTR model in the hope of improving the accuracy of the recognition (Kampkaspar, 2017). Karen Thöle’s work at the University of Göttingen shows that Transkribus can also cope with more challenging printed texts, in this case an incunable written in late Medieval Latin. With a ground truth set of around 35,000 words, Thöle has produced a model that is able to both recognise the text with a CER of around 5 per cent and also acknowledge and expand frequently used abbreviations (Thöle, 2017). These diverse examples illustrate how Transkribus users are recognising material of different dates, languages and styles. Automated transcripts allow for an unprecedented scale of access to digitised historical material, providing a basis for scholarly editing and research work. The Transkribus user community The Georgian Papers Programme is now working to better these results by improving the consistency of their ground truth and combining their original model with other models already trained to recognise eighteenth- and nineteenth-century English writing. At the time of writing, HTR tends to be strongest for Western scripts because it can draw upon a larger reserve of training data for common languages like English, French or Latin. However, Transkribus users are also starting to generate good results on texts written in non-Western languages. The University of Belgrade Library are working with Transkribus with a view to allowing users of their archive to access transcribed and searchable text. They have used Transkribus to train an HTR model to recognise Cyrillic handwriting from the twentieth century. A training set of some 7,000 words has generated a result where the CER is as low as around 2 per cent on material that the programme has seen before (Jerkov and Sofronijevic, 2017). With more words of training data, the recognition of previously unseen material should become stronger. Transkribus users have also benefited from integrating the platform into the workflow of existing research projects. The Barlach 2020 project at the University of Rostock is working on a digital edition of letters written by the German sculptor and writer Ernst Barlach (1870–1938). They have trained an HTR model with some 42,000 words of Barlach’s writing, integrating an earlier edition of Barlach’s letters into the training process as a dictionary (Lemke and Onasch, 2017). The resulting model transcribes pages with a CER of around 9 per cent and the team are now using these automated transcripts as a starting point for JD 75,5 scholarly editing. The Centre for Manuscript Genetics at the University of Antwerp[33] is working on a digital edition representing the genesis of works by the Irish writer Samuel Beckett (1906–1989)[34]. The team have trained models which can recognise Beckett’s writings in both English and French with CERs of around 12 and 18 per cent, respectively. The project team is interested in using these transcripts to analyse the multiple drafts, layers and noise in Beckett’s personal notes (Dillen, 2017). The PROLOPE research group at the Autonomous University of Barcelona[35] are working a digital edition of plays by the Spanish playwright Félix Lope de Vega (1562–1635) (Gázquez, 2017). The Transkribus user community Moreover, it must be acknowledged that all of the statistics presented in this paper are likely to improve significantly following a major update of the technology known as HTR+, which has been developed by the CITlab team at the University of Rostock. HTR+ draws on the Tensorflow[39] software library developed by Google, which means that deep neural networks can be constructed more efficiently than ever before. Experiments on three handwritten data sets suggest that the training of HTR+ is up to ten times faster than previous versions of HTR (Michael et al., 2018). Most importantly, this technology can improve the CER of automated transcriptions by between 5 and 10 per cent. HTR+ is now available in the Transkribus GUI upon request, and all existing HTR models will be retrained with this technology to allow all users to benefit from this latest advance in machine learning. Improving Transkribus The expanding network of Transkribus users has advantages for the usability and efficacy of HTR technology. User feedback and bug-reporting feeds directly into development work on the platform. As a system of machine learning, Transkribus also becomes stronger as more and more data are processed (Carbonell et al., 2013). All documents uploaded to the Transkribus GUI remain private and are not publicly shared. In the background however, the neural networks are learning from every piece of ground truth submitted in the system and Transforming scholarship in the archives Transforming scholarship in the archives consequently becoming better at recognising different hands, scripts and languages. The more Transkribus users there are, the stronger the HTR will be. In the computational science field, the automated recognition of manuscripts on the basis of a sufficiently large set of ground truth is now viewed as a scientifically solved problem. The next goals for computer scientists are to optimise methods so that they require less training data to achieve comparable results, and build generic models that can work on similar fonts and hands. In the future “out-of-the-box” models could make it easier for even more users to engage with and benefit from HTR, particularly those members of the public who are interested in studying historical documents. Legally, the sharing of the models is unproblematic, because HTR training does not violate any copyright or moral rights: ground truth images and transcripts are used for training but do not actually become part of the resulting neural network model. 967 This network effect, which is made possible by the sharing of data, will play a decisive role in the expansion of the platform in the coming years. The growth of a research community will be facilitated in two ways. First, users can already exchange models among themselves or between different collections and this will be made easier in the future. Second, as has been suggested above, the Transkribus team will train global models, which will unite different sets of training data and thus cover a wide variety of document types and writing styles. It makes more sense to adapt existing models, benefiting from the training data that is already in the system rather than training every new model from scratch. y g y In addition, there are future developments that need to occur in the technology behind Transkribus. Improving Transkribus g p g p y p With the Transkribus Learn[40] platform users can practice reading historical handwriting and the DocScan[41] mobile app and ScanTent[42] device enable users to take high-quality images of documents using a mobile phone. Future work includes allowing users to make meaningful contributions to the indexing of historical holdings by helping to validate the search results delivered by Keyword Spotting, flagging false positives and creating an index of controlled search words. Moreover, it may be possible to develop automated “search agents”, who browse the ever-expanding range of digitised files for specific keywords and, if there are particularly interesting occurrences and accumulations, inform the user accordingly. Central to this is, of course, the user community. The Transkribus team will continue to engage with users to ensure the development of an infrastructure that supports their approaches. A future study of the activities of the user community will also highlight how this new suite of tools is changing humanities research practice. Improving Transkribus There remain problems with the recognition of documents with a layout that is tabular or otherwise complex. Research will continue into the recognition of structural elements such as marginalia, headlines, addresses, dates, salutations and signatures. Computational analysis of writing styles is making writer identification possible, with the potential to attribute authorship to previously obscure documents. With the improvement of the system, the expansion of training data, and the increasing accuracy of the models, comes new opportunities. The READ project team have already constructed a number of prototype tools as part of the wider Transkribus infrastructure that are designed to expedite digitisation, the teaching of palaeography skills and the involvement of the public in historical research. With the Transkribus Learn[40] platform users can practice reading historical handwriting and the DocScan[41] mobile app and ScanTent[42] device enable users to take high-quality images of documents using a mobile phone. Future work includes allowing users to make meaningful contributions to the indexing of historical holdings by helping to validate the search results delivered by Keyword Spotting, flagging false positives and creating an index of controlled search words. Moreover, it may be possible to develop automated “search agents”, who browse the ever-expanding range of digitised files for specific keywords and, if there are particularly interesting occurrences and accumulations, inform the user accordingly. Central to this is, of course, the user community. The Transkribus team will continue to engage with users to ensure the development of an infrastructure that supports their approaches. A future study of the activities of the user community will also highlight how this new suite of tools is changing humanities research practice. In addition, there are future developments that need to occur in the technology behind Transkribus. There remain problems with the recognition of documents with a layout that is tabular or otherwise complex. Research will continue into the recognition of structural elements such as marginalia, headlines, addresses, dates, salutations and signatures. Computational analysis of writing styles is making writer identification possible, with the potential to attribute authorship to previously obscure documents. With the improvement of the system, the expansion of training data, and the increasing accuracy of the models, comes new opportunities. The READ project team have already constructed a number of prototype tools as part of the wider Transkribus infrastructure that are designed to expedite digitisation, the teaching of palaeography skills and the involvement of the public in historical research. Discussion A j b A major benefit to Transkribus is the cooperative manner of working, where all workflow steps – such as loading the documents into the platform, transcribing the texts, training the models and applying them to new documents – are carried out by the user group independently and under its own responsibility. The job of the Transkribus team is to ensure the availability of the platform, explain the various features, provide general support and grow the user community, while ever-improving the underlying HTR technology. However, this also raises the question of the sustainability of such a platform. Considerable resources have already been channelled into the development of the Transkribus infrastructure. JD 75,5 The high number of users and the fact that cooperation agreements have already been established with memory institutions and research groups from all over the world, show that the technology of text recognition meets with great interest and is generally perceived as a central element of the future indexing of historical documents. As detailed above, research projects from across Europe are already using the Transkribus GUI as a productive tool. The more users, the better the recognition of handwritten and printed text of all kinds: the platform must therefore be scalable, as well as sustainable. 968 p , Legal and business models for the continued operation of the platform are currently being developed to prepare for the end of the EU-funded phase of the project in mid-2019. From this point onwards, Transkribus services will be provided as part of a European Cooperative Society (SCE) based at the University of Innsbruck. This is a legal entity founded with the objective of fulfilling the needs of its members, where profit is shared between members and used to improve services[43]. The working title for the initiative is READ-COOP. This legal basis is intended to promote cooperation between archives, libraries, universities and the general public. At the time of writing, a freemium service model is planned, with a mixture of free and paid-for services. The availability of documents, tools and data for the members of the network will continue to be a central element of the platform, promoting open research which allows confidence in the results generated from the system. Preparations for the implementation of READ-COOP were presented at the second Transkribus User Conference in November 2018 (Dellinger, 2018). Regular business operations will begin on 1 July 2019. Discussion A j b To become viable in the long term, the platform needs to continue to support its research community, while generating enough resources to cover staffing and infrastructure costs. The alternative, of course, is that commercial digitisation providers will act as gatekeepers to HTR technologies that they can afford to underpin: restricting access to particular collections, and subscribing users (and seldom making their computational methods transparent). These concerns come at a time when HTR is ready to bring potential change to the wider archival environment. Training data currently available in Transkribus can already be used to create models that provide the basis for Keyword Spotting to search substantial parts of the archive stock in English and German. For other languages, such as Dutch or Finnish, there is also sufficient training data now available to achieve useful results at least for parts of the document stock. In a few years’ time, it can be predicted that sufficient training data will be available to make the majority of digitised archival holdings in Europe searchable with this technology. It is therefore imperative that HTR remains accessible to libraries, archives and individuals who would benefit from it, to allow vastly improved access to our written cultural heritage. Transkribus users will always be able to access and export their ground truth data. These data allow users to analyse the assumptions upon which their results were built and consider possible limitations and biases of automated transcriptions. Open discussion of algorithmic provenance and dependencies is important to develop trust in the reliability of research resources generated with artificial intelligence (Dayhoff and DeLeo, 2001; Samek et al., 2017). Forthcoming new metrics for assessing the accuracy of HTR in the Transkribus GUI will also help users to gain a more assured understanding of the strengths and possible constraints of the technology. There is no doubt that the approaches of historians and genealogists will be heavily affected as this technology becomes embedded into available research methods. A future study of Transkribus users will be needed to examine the ramifications of this technology, establishing how it is challenging and extending the scope of historical analysis and how these new approaches can be best conceptualised, taught and supported. Conclusion Thi h This paper has provided the first published overview of research undertaken in the tranScriptorium and READ EU-funded projects, which has resulted in the establishment of Transforming scholarship in the archives Transforming scholarship in the archives the Transkribus platform for the automated recognition of historical documents. For over three years, the platform has been providing free access to HTR technology that can be applied to banks of images of digitised manuscripts, allowing useful transcripts of the material to be generated and improving the underlying technology for current and future users via machine learning. Such a project is only possible with an interdisciplinary collaboration of computer scientists, developers, humanities scholars, archivists and librarians. The resulting infrastructure has the potential to change the reach and scope of research questions that depend on handwritten primary historical sources and this paper has supplied evidence of a range of research projects that are already successfully engaging with the Transkribus GUI to transcribe and study archival documents. There are benefits to be realised if HTR can be integrated into the digitisation cycle of manuscript material: using the results from this technology as a finding aid across mass-digitised content will vastly improve access to images of historical digitised manuscripts for all. Questions of sustainability now become pressing, just as the technology has become operational: raising issues of ownership, access and collegiate working across the heritage sector. There is no doubt that improving access to historical texts in this way will change the use and analysis of archival material: it is now time to create and implement sustainable support structures that will allow this technology to be available to as wide a research community as possible. 969 Acknowledgements This research was funded as part of the Recognition and Enrichment of Archival Documents (READ) project. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 674943. This research was previously funded as part of the tranScriptorium project. This project received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under Grant Agreement No. 600707. The authors would like to extend thanks to all users of the Transkribus platform. Notes 1. Neural networks are computational systems of hardware and software that are loosely modelled on the biological networks found in animal brains. They learn and improve their performance by training on a series of examples. Deep neural networks are a class of machine learning algorithms that use multiple layers of processing to learn and analyse their task. 1. Neural networks are computational systems of hardware and software that are loosely modelled on the biological networks found in animal brains. They learn and improve their performance by training on a series of examples. Deep neural networks are a class of machine learning algorithms that use multiple layers of processing to learn and analyse their task. 2. www.quartexcollections.com/ (accessed 13 December 2018). 2. www.quartexcollections.com/ (accessed 13 December 2018). 3. Recognition and Enrichment of Archival Documents (READ), Project ID 674943, H2020-EINFRA- 2015-1 European Commission (2016), https://read.transkribus.eu/ (accessed 13 December 2018). 3. Recognition and Enrichment of Archival Documents (READ), Project ID 674943, H2020-EINFRA- 2015-1 European Commission (2016), https://read.transkribus.eu/ (accessed 13 December 2018). 4. https://transkribus.eu/ (accessed 13 December 2018). 5. Computational science research generated by the READ project includes that on layout analysis (Diem et al., 2017; Grüning et al., 2017), keyword spotting (Puigcerver et al., 2017; Toselli et al., 2017) and word spotting (Zagoris et al., 2017). For other research see https://read.transkribus.eu/ research-publications/ (accessed 13 December 2018). 5. Computational science research generated by the READ project includes that on layout analysis (Diem et al., 2017; Grüning et al., 2017), keyword spotting (Puigcerver et al., 2017; Toselli et al., 2017) and word spotting (Zagoris et al., 2017). For other research see https://read.transkribus.eu/ research-publications/ (accessed 13 December 2018). 6. https://zenodo.org/communities/scriptnet/?page=1&size=20 (accessed 13 December 2018). 6. https://zenodo.org/communities/scriptnet/?page=1&size=20 (accessed 13 December 2018). 7. https://read.transkribus.eu/deliverables/ (accessed 13 December 2018). 7. https://read.transkribus.eu/deliverables/ (accessed 13 December 2018). 8. “Ground truth” is a term commonly used in machine learning to refer to accurate, objective information provided by empirical, direct processes, rather than that inferred from sources via the statistical calculation of uncertainty. In the case of Transkribus, “ground-truth” information is gathered by training the system with enough data (usually around 15,000 words or 75 pages) relating an individual script, which can then be used to create a model that can be applied successfully to large volumes of the same script. JD 75,5 9. Notes The current implementation of the HTR engine comes from the CITlab group of the University of Rostock (see Leifert et al., 2016). 10. TranScriptorium, Project ID 600707, FP7-ICT-2011-9, European Commission (2015), http:// transcriptorium.eu/ (accessed 13 December 2018). 11. Institutions contributing to READ include the University of Innsbruck (coordinator/Austria), the Polytechnic University of Valencia (Spain), University College London (UK), National Center for Scientific Research “Demokritos” (Greece), Democritus University of Thrace (Greece), CoSector – University of London (UK), Technical University of Vienna (Austria), University of Rostock (Germany), University of Leipzig (Germany), NAVER LABS Europe (France), École Polytechnique Féderale de Lausanne (Switzerland), National Archives of Finland (Finland), State Archives of Zurich (Switzerland), Passau Diocesan Archives (Germany), and more recently the University of Edinburgh (UK). 970 12. Essential components of the technology come from the CITlab group of the University of Rostock, and the PRHLT group of the Polytechnic University of Valencia, with additional contributions from the wider consortium. 12. Essential components of the technology come from the CITlab group of the University of Rostock, and the PRHLT group of the Polytechnic University of Valencia, with additional contributions from the wider consortium. 13. Downloadable from http://transkribus.eu/ (accessed 13 December 2018). 14. https://transkribus.eu/wiki/index.php/REST_Interface (accessed 13 December 2018). 15. https://github.com/Transkribus/ (accessed 13 December 2018). 16. http://planet.de/ (accessed 13 December 2018). 17. https://github.com/jpuigcerver/Laia (accessed 13 December 2018). 18. For UCL’s contribution to READ see, www.ucl.ac.uk/bentham-project and https://read. transkribus.eu/network/university-college-london/ (accessed 13 December 2018). 18. For UCL’s contribution to READ see, www.ucl.ac.uk/bentham-project and https://read. transkribus.eu/network/university-college-london/ (accessed 13 December 2018). 19. Transkribus includes layout analysis tools by researchers at the National Center for Scientific Research “Demokritos”, the CITlab group at the University of Rostock and the Computer Vision Lab at the Technical University of Vienna. The CITlab tool currently generates the best results on complex historical documents. 19. Transkribus includes layout analysis tools by researchers at the National Center for Scientific Research “Demokritos”, the CITlab group at the University of Rostock and the Computer Vision Lab at the Technical University of Vienna. The CITlab tool currently generates the best results on complex historical documents. 20. http://transkribus.eu/r/read/ (accessed 13 December 2018). 21. The CITlab group of the University of Rostock developed the Text2Image matching tool. For more information see https://transkribus.eu/wiki/images/6/6f/HowToUseExistingTranscriptions. pdf (accessed 13 December 2018). 21. The CITlab group of the University of Rostock developed the Text2Image matching tool. For more information see https://transkribus.eu/wiki/images/6/6f/HowToUseExistingTranscriptions. pdf (accessed 13 December 2018). 22. Notes https://learn.transkribus.eu (accessed 13 December 2018). 40. https://learn.transkribus.eu (accessed 13 December 2018). 41. https://play.google.com/store/apps/details?id=at.ac.tuwien.caa.docscan (accessed 13 December 2018 42. https://scantent.cvl.tuwien.ac.at/en/ (accessed 13 December 2018). 43. https://ec.europa.eu/growth/sectors/social-economy/cooperatives/european-cooperative-society en (accessed 13 December 2018). Notes The Transkribus GUI also allows users to compute accuracy according to a Word Error Rate (WER). Additional metrics for measuring the accuracy of automated transcriptions will be made available in the platform over the coming months. The CERs mentioned in this paper are rounded off to the nearest whole number. 22. The Transkribus GUI also allows users to compute accuracy according to a Word Error Rate (WER). Additional metrics for measuring the accuracy of automated transcriptions will be made available in the platform over the coming months. The CERs mentioned in this paper are rounded off to the nearest whole number. 23. Various approaches to Keyword Spotting are being researched by groups at the University of Rostock, the Polytechnic University of Valencia, Democritus University of Thrace and the National Center for Scientific Research “Demokritos”. The Transkribus GUI currently provides access to a “Query by String” method by the CITlab group at the University of Rostock. 23. Various approaches to Keyword Spotting are being researched by groups at the University of Rostock, the Polytechnic University of Valencia, Democritus University of Thrace and the National Center for Scientific Research “Demokritos”. The Transkribus GUI currently provides access to a “Query by String” method by the CITlab group at the University of Rostock. 24. http://blogs.ucl.ac.uk/transcribe-bentham/2018/11/28/project-update-automated-recognition- bentham-handwriting/ (accessed 13 December 2018). 25. http://prhlt-carabela.prhlt.upv.es/bentham/ (accessed 13 December 2018). 26. www.bistum-passau.de/bistum/archiv and https://read.transkribus.eu/network/passau-diocesan- archives/ for their contribution to READ (accessed 13 December 2018). 27. www.arkisto.fi/en/frontpage and https://read.transkribus.eu/network/national-archives-finland/ for their contribution to READ (accessed 13 December 2018). 28. https://staatsarchiv.zh.ch/internet/justiz_inneres/sta/en/home.html and https://read.transkribus. eu/network/zurich-state-archives/ for their contribution to READ (accessed 13 December 2018). 29. www.archives-quickaccess.ch/search/stazh/krp (accessed 13 December 2018). Transforming 29. www.archives quickaccess.ch/search/stazh/krp (accessed 13 December 2018). 30. https://read.transkribus.eu/network/ (accessed 13 December 2018). 31 www digitale bibliothek mv de/viewer/image/PPNUAG 0 1 St 666/1/LOG 0003/ (accessed Transforming scholarship in the archives 30. https://read.transkribus.eu/network/ (accessed 13 December 2018). scholarship in th hi 31. www.digitale-bibliothek-mv.de/viewer/image/PPNUAG_0_1_St_666/1/LOG_0003/ (accessed 13 December 2018). the archives 32. http://georgianpapersprogramme.com/ (accessed 13 December 2018). 971 33. www.uantwerpen.be/en/research-groups/centre-for-manuscript-genetics/ (accessed 13 December 2018). 971 33. www.uantwerpen.be/en/research-groups/centre-for-manuscript-genetics/ (accessed 13 December 2018 34. www.beckettarchive.org/ (accessed 13 December 2018). 34. www.beckettarchive.org/ (accessed 13 December 2018). 35. http://prolope.uab.cat/ (accessed 13 December 2018). 36. http://prhlt-carabela.prhlt.upv.es/tso/ (accessed 13 December 2018). 36. http://prhlt-carabela.prhlt.upv.es/tso/ (accessed 13 December 2018). 37. http://ocr-d.de/eng (accessed 13 December 2018). 37. http://ocr-d.de/eng (accessed 13 December 2018). 38. www.oeaw.ac.at/en/acdh/projects/wiennerisches-diarium-digital/ (accessed 13 December 2018). 39. www.tensorflow.org/ (accessed 13 December 2018). 39. www.tensorflow.org/ (accessed 13 December 2018). 40. References Adam Matthew Digital (2018), “Handwritten text recognition: artificial intelligence transforms discoverability of handwritten manuscripts”, available at: www.amdigital.co.uk/products/ handwritten-text-recognition (accessed 13 December 2018). Alvermann, D. and Blüggel, B. (2017), “Transkribus at Greifswald. Idea, practice, results, perspective”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/2017/07/ Alvermann_Bluegel_Greifswald.pdf (accessed 13 December 2018). Bertolami, R. and Bunke, H. (2008), “Hidden Markov model-based ensemble methods for offline handwritten text line recognition”, Pattern Recognition, Vol. 41 No. 11, pp. 3452-3460. Boenig, M. and Würzner, K.-M. (2017), “Compilation of a large ground-truth data set using transkribus”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/2017/07/Boenig_ Wuerzner_Groundtruth-1.pdf (accessed 13 December 2018). Borowiecki, K.J. and Navarrete, T. (2016), “Digitization of heritage collections as indicator of innovation”, Economics of Innovation and New Technology, Vol. 26 No. 3, pp. 227-246. Bulacu, M., Brink, A., van der Zant, T. and Schomaker, L. (2009), “Recognition of handwritten numerical fields in a large single-writer historical collection”, 10th International Conference on Document Analysis and Recognition, IEEE, pp. 808-812. Carbonell, J.G., Michalski, R.S. and Mitchell, T.M. (2013), “An overview of machine learning”, in Michalski, R.S., Carbonell, J.G. and Mitchell, T.M. (Eds), Machine Learning: An Artificial Intelligence Approach, Springer-Verlag, Berlin and Heidelberg, pp. 3-23. Causer, T. and Terras, M. (2014), “ ‘Many hands make light work. Many hands together make merry work’: transcribe Bentham and Crowdsourcing manuscript collections”, in Ridge, M. (Ed.), Crowdsourcing Our Cultural Heritage, Ashgate, Farnham, pp. 57-88. Clinchant, S., Déjean, H., Meunier, J., Lang, E.M. and Kleber, F. (2018), “Comparing machine learning approaches for table recognition in historical register books”, 13th IAPR International Workshop on Document Analysis Systems (DAS), IEEE, pp.133-138. JD 75,5 JD 75 5 Cornell, D. (2017), “Georgian papers programme”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/ wp-content/uploads/2017/07/Cornell_Georgian_Papers.pdf (accessed 13 December 2018). Dayhoff, J.E. and DeLeo, J.M. (2001), “Artificial neural networks: opening the black box”, Cancer, Vol. 91 No. 91, pp. 1615-1635. Dellinger, M. (2018), “The READ-COOP: working together for the future of digital cultural heritage”, paper presented at Transkribus User Conference 2018, 8–9 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/2018/11/ DELLINGER-SCE.pdf (accessed 13 December 2018). 972 Diem, M., Kleber, F., Fiel, S., Grüning, T. and Gatos, B. References (2017), “cBAD: ICDAR2017 competition on baseline detection”, 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), IEEE, pp. 1355-1360. Dillen, W. (2017), “Transkribus in practice”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp- content/uploads/2017/07/Dillen_Beckett_edition.pdf (accessed 13 December 2018). Dimauro, G., Impedovo, S., Pirlo, G. and Salzo, A. (1997), “Automatic bankcheck processing: a new engineered system”, International Journal of Pattern Recognition and Artificial Intelligence, Vol. 11 No. 4, pp. 467-504. Dimond, T.L. (1957), “Devices for reading handwritten characters”, papers and discussions presented at the December 9–13, Eastern Joint Computer Conference: Computers with Deadlines to Meet, ACM, Washington, DC, pp. 232-37. Edwards, J.A. (2007), “Easily adaptable handwriting recognition in historical manuscripts”, PhD Dissertation, University of California, Berkeley, CA, available at: www2.eecs.berkeley.edu/Pubs/ TechRpts/2007/EECS-2007-76.pdf (accessed 13 December 2018). Estill, L. and Levy, M. (2016), “Chapter 12 evaluating digital remediations of women’s manuscripts”, Digital Studies/Le champ numérique, Vol. 6, available at: http://doi.org/10.16995/dscn.12 (accessed 13 December 2018). European Commission (2015), “tranScriptorium”, Community Research and Development Information Service, available at: https://cordis.europa.eu/project/rcn/106843_en.html (accessed 13 December 2018). European Commission (2016), “Recognition and enrichment of archival documents”, Community Research and Development Information Service, available at: https://cordis.europa.eu/project/ rcn/198756_en.html (accessed 13 December 2018). Firmani, D., Maiorino, M., Merialdo, P. and Nieddu, E. (2018), “Towards knowledge discovery from the Vatican secret archives. In Codice ratio – episode 1: machine transcription of the manuscripts”, available at: https://arxiv.org/abs/1803.03200 (accessed 13 December 2018). Fischer, A., Wüthrich, M., Liwicki, M., Frinken, V., Bunke, H., Viehhauser, G. and Stolz, M. (2009), “Automatic transcription of handwritten medieval documents”, 15th International Conference on Virtual Systems and Multimedia, IEEE, pp. 137-142. Gázquez, R.V. (2017), “La edición del teatro de Lope”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read. transkribus.eu/wp-content/uploads/2017/07/Valdes_PROLOPE.pdf (accessed 13 December 2018). Giotis, A.P., Sfikas, G., Gatos, B. and Nikou, C. (2017), “A survey of document image word spotting techniques”, Pattern Recognition, Vol. 68, August, pp. 310-332. Govindan, V.K. and Shivaprasad, A.P. (1990), “Character recognition – a review”, Pattern Recognition, Vol. 23 No. 7, pp. 671-683. Gregory, I., Donaldson, C., Murrieta-Flores, P. and Rayson, P. (2015), “Geoparsing, GIS and textual analysis: current developments in spatial humanities research”, International Journal of Humanities and Arts Computing, Vol. 9 No. 1, pp. 1-14. Grüning, T., Leifert, G., Strauß, T. and Labahn, R. Cornell, D. (2017), Georgian papers programme , paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/ wp-content/uploads/2017/07/Cornell_Georgian_Papers.pdf (accessed 13 December 2018). References (2016), “Deliverable 7.7, HTR engine based on neural networks P1”, Deliverable submitted to the European Commission, available at: http://read. transkribus.eu/wp-content/uploads/2017/01/READ_D7.7_HTRbasedonNN.pdf (accessed 13 December 2018). Transforming scholarship in the archives Grüning, T., Leifert, G., Strauß, T. and Labahn, R. (2017), “A Robust and Binarization-free approach for text line detection in historical documents”, 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), IEEE, pp. 236-241. Hafemann, L.G., Sabourin, R. and Oliveira, L.S. (2015), “Offline handwritten signature verification – literature review”, available at: https://arxiv.org/abs/1507.07909 (accessed 13 December 2018). Hodel, T. (2017), “Deliverable 8.4, large scale demonstrators. Evaluation and bootstrapping”, Deliverable submitted to the European Commission, available at: https://read.transkribus.eu/ wp-content/uploads/2017/12/READ_8.5-StAZH-v1.pdf (accessed 13 December 2018). 973 Jerkov, A. and Sofronijevic, A. (2017), “University of Belgrade library, Transkribus user experience”, paper presented at Transkribus User Conference 2017, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/2017/07/Jerkov_ Uni_Belgrade_library.pdf (accessed 13 December 2018). Kahle, P., Colutto, S., Hackl, G. and Mühlberger, G. (2017), “Deliverable 4.2, READ platform and service maintenance”, Deliverable submitted to the European Commission, available at: https://read. transkribus.eu/wp-content/uploads/2017/12/D4.2.pdf (accessed 13 December 2018). Kallio, M. (2017), “Deliverable 8.8, layout analysis and crowdsourcing”, Deliverable submitted to the European Commission, available at: https://read.transkribus.eu/wp-content/uploads/2017/12/ Deliverable_8.8.pdf (accessed 13 December 2018). Kampkaspar, D. (2017), “Digital edition of the ‘Wienerisches Diarium’ ”, paper presented at Transkribus User Conference, 2–3 November, Technical University of Vienna, Vienna, available at: https:// read.transkribus.eu/wp-content/uploads/2017/07/Kampkaspar_Wienerisches_Diarium.pdf (accessed 13 December 2018). Kichuk, D. (2015), “Loose, falling characters and sentences: the persistence of the OCR problem in digital repository e-books”, Libraries and the Academy, Vol. 15 No. 1, pp. 59-91. Leedham, C.G. (1994), “Historical perspectives of handwriting recognition systems”, IEE Colloquium on Handwriting and Pen-Based Input, IEEE, pp. 1-3. Leifert, G., Strauß, T., Grüning, T. and Labahn, R. (2016), “CITlab ARGUS for historical handwritten documents”, available at: http://arxiv.org/abs/1605.08412 (accessed 13 December 2018). Lemke, K. and Onasch, P. (2017), “ ‘I do not find hot punch as nice as writing letters.’ Transcribing Ernst Barlach’s letters (1870–1938) with Transkribus”, paper presented at Transkribus User Conference, 2–3 November, Technical University of Vienna, Vienna, available at: https://read. transkribus.eu/wp-content/uploads/2017/07/Onasch_Lemke_Barlach.pdf (accessed 13 December 2018). Meroño-Peñuela, A., Ashkpour, A., Van Erp, M., Mandemakers, K., Breure, L., Scharnhorst, A., Schlobach, S. and Van Harmelen, F. (2015), “Semantic technologies for historical research: a survey”, Semantic Web, Vol. 6 No. 6, pp. 539-564. Michael, J., Weidemann, M. and Labahn, R. References (2018), “Deliverable 7.9, HTR engine based on neural networks P3”, Deliverable submitted to the European Commission, available at: https://read. transkribus.eu/wp-content/uploads/2018/12/Del_D7_9.pdf (accessed 13 December 2018). Morera, Á., Sánchez, Á., Vélez, J.F. and Moreno, A.B. (2018), “Gender and handedness prediction from offline handwriting using convolutional neural networks”, Complexity, Vol. 2018, available at: https://doi.org/10.1155/2018/3891624 (accessed 13 December 2018). Ogilvie, B. (2016), “Scientific archives in the age of digitization”, Isis, Vol. 107 No. 1, pp. 77-85. Pal, U., Roy, R.K. and Kimura, F. (2012), “Multi-lingual city name recognition for Indian postal automation”, Proceedings of the 2012 International Conference on Frontiers in Handwriting Recognition, IEEE, pp. 169-173. Puigcerver, J., Toselli, A.H. and Vidal, E. (2015), “ICDAR2015 competition on keyword spotting for handwritten documents”, 13th International Conference on Document Analysis and Recognition (ICDAR), IEEE, pp. 1176-1180. Puigcerver, J., Toselli, A.H. and Vidal, E. (2017), “Querying out-of-vocabulary words in lexicon-based keyword spotting”, Neural Computing and Applications, Vol. 28 No. 9, pp. 2373-2382. JD 75,5 JD 75,5 Retsinas, G., Louloudis, G., Stamatopoulos, N. and Gatos, B. (2016), “Keyword spotting in handwritten documents using projections of oriented gradients”, 12th IAPR Workshop on Document Analysis Systems (DAS), IEEE, pp. 411-16. Romero, V., Sánchez, J.A., Bosch, V., Depuydt, K. and De Does, J. (2015), “Influence of text line segmentation in handwritten text recognition”, 13th International Conference on Document Analysis and Recognition (ICDAR), pp. 536-540. Samek, W., Wiegand, T. and Müller, K.-R. (2017), “Explainable artificial intelligence: understanding, visualizing and interpreting deep learning models”, available at: https://arxiv.org/abs/1708.082 96 (accessed 13 December 2018). 974 Sánchez, J.A., Romero, V., Toselli, A.H. and Vidal, E. (2014), “ICFHR2014 competition on handwritten text recognition on transcriptorium datasets (HTRtS)”, 14th International Conference on Frontiers in Handwriting Recognition (ICFHR), IEEE, pp. 785-790. Sánchez, J.A., Romero, V., Toselli, A.H., Villegas, M. and Vidal, E. (2017), “ICDAR2017 competition on handwritten text recognition on the READ dataset”, 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), IEEE, pp. 1383-1388. Schantz, H.F. (1982), The History of OCR (Optical Character Recognition), Recognition Technologies Users Association, Manchester Center, VT. Seaward, L. (2016), “Deliverable 4.10, transcribe Bentham”, Deliverable submitted to the European Commission, available at: http://read.transkribus.eu/wp-content/uploads/2017/01/READ_D4. 10_TranscribeBentham.pdf (accessed 13 December 2018). Springmann, U. and Lüdeling, A. (2017), “OCR of historical printings with an application to building diachronic corpora: a case study using the RIDGES herbal corpus”, Digital Humanities Quarterly, Vol. 11 No. 2, available at: http://digitalhumanities.org/dhq/vol/11/2/000288/000288. html (accessed 13 December 2018). References Strauß, T., Grüning, T., Leifert, G. and Labahn, R. (2016), “CITlab ARGUS for keyword search in historical handwritten documents – description of CITlab’s system for the ImageCLEF 2016 handwritten scanned document retrieval task”, CLEF Working Notes, pp. 399-412. Sudholt, S. and Fink, G.A. (2016), “PHOCNet: a deep convolutional neural network for word spotting in handwritten documents”, 15th International Conference on Frontiers in Handwriting Recognition (ICFHR), IEEE, pp. 277-282. Terras, M. (2006), Image to Interpretation: An Intelligent System to Aid Historians in the Reading of the Vindolanda Texts, Oxford University Press, Oxford. Terras, M. (2010), “The rise of digitisation: an overview”, in Rukowski, R. (Ed.), Digitisation Perspectives, Sense Publishers, pp. 3-20. Thöle, K. (2017), “Transcribing a highly abbreviated incunable (and some more manuscript sources)”, paper presented at Transkribus User Conference, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/2017/07/Thoele_ Incunable.pdf (accessed 13 December 2018). Toselli, A.H., Romero, V. and Vidal, E. (2017), “Word graphs size impact on the performance of handwriting document applications”, Neural Computing and Applications, Vol. 28 No. 9, pp. 2477-2487. Ul-Hasan, A., Bukhari, S.S. and Dengel, A. (2016), “OCRoRACT: a sequence learning OCR system trained on isolated characters”, 12th IAPR Workshop on Document Analysis Systems (DAS), IEEE, pp. 174-79. Van der Zant, T., Schomaker, L., Zinger, S. and Van Schie, H. (2009), “Where are the search engines for handwritten documents?”, Interdisciplinary Science Reviews, Vol. 34 Nos 2–3, pp. 224-235. Vetter, A. (2017), “The Austrian Bible translator – the word of God in German. Annotated critical hybrid edition”, paper presented at Transkribus User Conference, 2–3 November, Technical University of Vienna, Vienna, available at: https://read.transkribus.eu/wp-content/uploads/20 17/07/Vetter_Austrian_Bible.pdf (accessed 13 December 2018). Weber, A., Ameryan, M., Wolstencroft, K., Stork, L., Heerlien, M. and Schomaker, L. (2018), “Towards a digital infrastructure for illustrated handwritten archives”, in Ioannides, M. (Ed.), Digital Cultural Heritage, Springer International, pp. 155-166. Transforming scholarship in the archives Weidemann, M., Michael, J., Grüning, T. and Labahn, R. (2017), “Deliverable 7.8, HTR engine based on neural networks P2”, Deliverable submitted to the European Commission, available at: https://read.transkribus.eu/wp-content/uploads/2017/12/Del_D7_8.pdf (accessed 13 December 2018). Weisser, M. (2016), Practical Corpus Linguistics: An Introduction to Corpus-Based Language Analysis, Wiley-Blackwell. Wurster, H.W., Putz, H., Lang, E.M., Fronhöfer, W., Fronhöfer, A. and Mühlbauer, E. (2017), “Deliverable 8.11, large scale demonstrators. Keyword spotting in registry books P2”, Deliverable submitted to the European Commission, available at: https://read.transkribus.eu/ wp-content/uploads/2017/12/READ_D8_11_LSD_Passau.pdf (accessed 13 December 2018). 975 Zagoris, K., Pratikakis, I. References and Gatos, B. (2017), “Unsupervised word spotting in historical handwritten document images using document-oriented local features”, IEEE Transactions on Image Processing, Vol. 26 No. 8, pp. 4032-4041. Further reading Mühlberger, G. (2015), “Die automatisierte Volltexterkennung historischer Handschriften als gemeinsame Aufgabe von Archiven, Geistes- und Computerwissenschaftlern. Das Modell einer zentralen Transkriptionsplattform als virtuelle Forschungsumgebung”, in Becker, I.C. and Oertel, S. (Eds), Digitalisierung im Archiv. Neue Wege der Bereitstellung des Archivguts. Beiträge des 18. Archivwissenschaftlichen Kolloquiums am 26. und 27. November 2013, Archivschule Marburg, Marburg, pp. 87-116. Transforming scholarship in the archives Author Affiliations Guenter Muehlberger, University of Innsbruck, Innsbruck, Austria Guenter Muehlberger, University of Innsbruck, Innsbruck, Austria Guenter Muehlberger, University of Innsbruck, Innsbruck, Austria Louise Seaward, Bentham Project, Faculty of Laws, University College London, London, UK Melissa Terras, University of Edinburgh, Edinburgh, UK Melissa Terras, University of Edinburgh, Edinburgh, UK Melissa Terras, University of Edinburgh, Edinburgh, UK Sofia Ares Oliveira, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland Sofia Ares Oliveira, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland Vicente Bosch, Universitat Politecnica de Valencia, Valencia, Spain Frederic Kaplan, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland Florian Kleber, Visual Computing & Human-Centered Technology, Wien, Austria Roger Labahn, Universitat Rostock, Rostock, Germany Eva Maria Lang, Passau Diocesan Archives, Passau, Germany Sören Laube, Universitat Leipzig, Leipzig, Germany Gundram Leifert, Universitat Rostock, Rostock, Germany Georgios Louloudis, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Rory McNicholl, University of London, UK Jean-Luc Meunier, Naver Labs Europe, Meylan, France Johannes Michael, Universitat Rostock, Rostock, Germany Elena Mühlbauer, Passau Diocesan Archives, Passau, Germany Nathanael Philipp, Universitat Leipzig, Leipzig, Germany Ioannis Pratikakis, Democritus University of Thrace, Xanthi, Greece Joan Puigcerver Pérez, Universitat Politecnica de Valencia, Valencia, Spain Hannelore Putz, Passau Diocesan Archives, Passau, Germany George Retsinas, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Verónica Romero, Universitat Politecnica de Valencia, Valencia, Spain Robert Sablatnig, Visual Computing & Human-Centered Technology, Wien, Austria Joan Andreu Sánchez, Universitat Politecnica de Valencia, Valencia, Spain Philip Schofield, Bentham Project, Faculty of Laws, University College London, London, UK Giorgos Sfikas, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Christian Sieber, State Archives of Zurich, Zurich, Switzerland Nikolaos Stamatopoulos, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Tobias Strauß, Universitat Rostock, Rostock, Germany Tamara Terbul, University of Innsbruck, Innsbruck, Austria Alejandro Héctor Toselli, Universitat Politecnica de Valencia, Valencia, Spain Berthold Ulreich, University of Innsbruck, Innsbruck, Austria Mauricio Villegas, Universitat Politecnica de Valencia, Valencia, Spain Enrique Vidal, Universitat Politecnica de Valencia, Valencia, Spain Johanna Walcher, University of Innsbruck, Innsbruck, Austria Max Weidemann, Universitat Rostock, Rostock, Germany Herbert Wurster, Passau Diocesan Archives, Passau, Germany, and Konstantinos Zagoris, Democritus University of Thrace, Xanthi, Greece Corresponding author Melissa Terras can be contacted at: M.Terras@ed.ac.uk For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight com Frederic Kaplan, Ecole Polytechnique Federale de Lausanne, Lausanne, Switzerland Florian Kleber, Visual Computing & Human-Centered Technology, Wien, Austria Roger Labahn, Universitat Rostock, Rostock, Germany Eva Maria Lang, Passau Diocesan Archives, Passau, Germany Sören Laube, Universitat Leipzig, Leipzig, Germany Gundram Leifert, Universitat Rostock, Rostock, Germany Georgios Louloudis, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Rory McNicholl, University of London, UK Jean-Luc Meunier, Naver Labs Europe, Meylan, France Johannes Michael, Universitat Rostock, Rostock, Germany Elena Mühlbauer, Passau Diocesan Archives, Passau, Germany Nathanael Philipp, Universitat Leipzig, Leipzig, Germany Ioannis Pratikakis, Democritus University of Thrace, Xanthi, Greece Joan Puigcerver Pérez, Universitat Politecnica de Valencia, Valencia, Spain Hannelore Putz, Passau Diocesan Archives, Passau, Germany George Retsinas, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Verónica Romero, Universitat Politecnica de Valencia, Valencia, Spain Robert Sablatnig, Visual Computing & Human-Centered Technology, Wien, Austria Joan Andreu Sánchez, Universitat Politecnica de Valencia, Valencia, Spain Philip Schofield, Bentham Project, Faculty of Laws, University College London, London, UK Giorgos Sfikas, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Christian Sieber, State Archives of Zurich, Zurich, Switzerland Nikolaos Stamatopoulos, National Center for Scientific Research “DEMOKRITOS”, Agia Paraskevi, Greece Tobias Strauß, Universitat Rostock, Rostock, Germany Tamara Terbul, University of Innsbruck, Innsbruck, Austria Alejandro Héctor Toselli, Universitat Politecnica de Valencia, Valencia, Spain Berthold Ulreich, University of Innsbruck, Innsbruck, Austria Mauricio Villegas, Universitat Politecnica de Valencia, Valencia, Spain Enrique Vidal, Universitat Politecnica de Valencia, Valencia, Spain Johanna Walcher, University of Innsbruck, Innsbruck, Austria Max Weidemann, Universitat Rostock, Rostock, Germany Herbert Wurster, Passau Diocesan Archives, Passau, Germany, and Konstantinos Zagoris, Democritus University of Thrace, Xanthi, Greece Corresponding author Melissa Terras can be contacted at: M.Terras@ed.ac.uk For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm O f f h d il i i @ ldi i h JD 75,5 976 For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com Corresponding author li b p g Melissa Terras can be contacted at: M.Terras@ed.ac.uk Melissa Terras can be contacted at: M.Terras@ed.ac.uk For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com g pp g g p Or contact us for further details: permissions@emeraldinsight.com
https://openalex.org/W3185563284	https://www.journals.vu.lt/open-series/article/download/24522/23772	English	null	Louis Hjelmslev and the Baltic countries	Vilnius University open series	2,021	cc-by	2,320	381 Copyright © 2021 Bohumil Vykypěl. Published by Vilnius University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. https://doi.org/10.15388/SBOL.2021.21 Louis Hjelmslev and the Baltic countries https://doi.org/10.15388/SBOL.2021.21 Bohumil Vykypěl Czech Language Institute of the CAS, Brno, Czech Republic Abstract. This paper deals with texts by Louis Hjelmslev on Baltic topics, in particular an article from 1946 in which he defended the right of the Baltic countries to independence. Keywords: Louis Hjelmslev, Baltic countries, structuralism, World War II, Soviet Union Soviet Union In 2005, Axel Holvoet published my book on the phonology of Lithuanian and Latvian viewed through Hjelmslev’s glossematics (Vykypěl 2005), which is perhaps the best-looking one of my books. I wish to contribute to his festschrift with a small paper on a topic which is even more thrilling than the phonology of the Baltic languages, namely Hjelmslev’s relationship to the Baltic coun tries. In 2005, Axel Holvoet published my book on the phonology of Lithuanian and Latvian viewed through Hjelmslev’s glossematics (Vykypěl 2005), which is perhaps the best-looking one of my books. I wish to contribute to his festschrift with a small paper on a topic which is even more thrilling than the phonology of the Baltic languages, namely Hjelmslev’s relationship to the Baltic coun tries. The external history of Hjelmslev and matters of the Baltic has been described on several occasions (Palionis 1978; Gregersen 1991, I, 174–177; Rasmussen 1992, 3–4, 14–15; Sabaliauskas 2008). In the summer of 1921, Hjelmslev visited Lithuania as it behooved a student of comparative linguistics. He depicted what he had seen there in two journal articles with the title Impressions from Lithuania (Hjelmslev 1922). These articles are, indeed, more impressions than a deeper analysis; only the attempt at characterizing the nature of Lithuanian nationalism is perhaps note worthy, this being, according to Hjelmslev, an opportunistic reality-nation alism. As Gregersen (1991, I, 175) remarked, Hjelmslev did not see in his 1922-impressions the differences between the individual groups of Lithua 381 nians, treating the nation as an ideal whole. Similarly, in a later booklet on the Lithuanian people and their language (Hjelmslev 1930) he wrote in a romantic vein about the antiquity and beauty of Lithuanian or about the dainos as the most typical expression of Lithuanian psyche, belonging to a nation that had, in spite of oppression, managed to maintain its own language and culture. 1 See Gregersen (1991, I, 181–182) on Hjelmslev’s stay in Czechoslovakia. https://doi.org/10.15388/SBOL.2021.21 Louis Hjelmslev and the Baltic countries Hjelmslev’s texts on Lithuania seem to be a typical product of a member of the Western European upper middle class of his time: their middle-class character is manifested by the mentioned “holistic” treatment of the concept of nation; the Western European perspective, in turn, prevented the author, apparently, from seeing more “heroic” or active aspects of Lithuanian history (such as the Grand Duchy of Lithuania, a regional great power of its time, the foundation of Vilnius University in 1579, the second oldest university in the Baltic region, or some of the remarkable works of Lithuanian literature which are not a few). Besides these published texts, Hjelmslev also gave several public lectures on Lithuania as well as Latvia and Estonia: 1928 broadcast lecture on the occasion of the Lithuanian national day; 1930 lecture on Estonia, Latvia and Lithuania; 1931 lecture on the Latvian language and folk poetry in the Danish- Latvian Association; 1933 lecture on Lithuania in the Danish-Lithuanian Soci ety (see Registrant, 5, 13). Hjelmslev also had a practical knowledge of Lithuanian as is attested by his letters written in Lithuanian and published by Palionis (1978). Three of them concern the visit to Lithuania in the summer of 1921: Hjelmslev described briefly what he had visited travelling around the Lithuanian countryside, and after returning home, he expressed gratitude for the hospitality he had met with; in addition, we learn that Hjelmslev (1922) was first not accepted for publication in the newspaper Politiken, and the author was then promised that the text would be published in the journal Tilskueren. More interesting are two other letters: they were written in 1924 when Hjelmslev was studying in Czechoslovakia,1 and their topic is a lecturer position at the University of Kaunas for which Hjelmslev applied unsuccessfully. It was a different time: a Dane who was awarded a generous Czechoslovak scholarship was writing about his application for an academic position in Lithuania; it was the time before the communist destruction of the eastern half of Europe. 382 The largest text devoted to a Baltic topic was Hjelmslev’s dissertation in which he dealt with some intricate problems of Baltic phonology (Hjelmslev 1932); he returned to these questions in a later text (Hjelmslev 1956). Baltic phonology also provided material on which Hjelmslev demonstrated an early version of his glossematic theory (Hjelmslev 1936–1937). 2 The polemic was continued by other authors (see Gregersen 1991, I, 177, fn. *). https://doi.org/10.15388/SBOL.2021.21 Louis Hjelmslev and the Baltic countries These contributions by Hjelmslev to Baltistics have been discussed and analyzed elsewhere (see – besides the special works on Baltic phonology – Sabaliauskas 1979, 225–226; Gregersen 1991, I, 208–218, II, 32–36; Rasmussen 1992, 476). In my view, however, his most important “Baltistic” paper is a newspaper article from 1946 (Hjelmslev 1946). This article was a polemic against an article with the title Three small anti-demo cratic states which had appeared in one of the previous issues of the newspaper. Hjelmslev refuted the claim that the Baltic countries had been fascist dictatorships and that the exiles from these countries were fascists. He defended the right of the Baltic countries to independence and rejected the absurd contention that the Baltic nations had asked for accession to the Soviet Union.2 Having said above that Hjelmslev’s early texts on Lithuania are in essence a product of his time, I have to say that his article from 1946 seems, in contrast, not to be typical of the period. Certainly, more detailed research would be needed (of which I am not aware) on the question of how the occupation of the Baltic countries after World War II was reflected in the world. But I am afraid that reactions of the type manifested in Hjelmslev’s article were in the minority. For instance, one finds neither in the texts nor in the extensive correspondence of the Czech linguist and Balticist Václav Machek a single mention of the fate of the Baltic nations (see Janyšková et al. 2011; Boček–Malčík 2011–2016), which is, however, rather understandable in a country in which the NKVD arrested in 1945 not a few of its citizens of Russian nationality. More diffi cult to understand is, in turn, the poor Archbishop of Canterbury, who called the Baltic partisans fascists whose deportation was justified (see Lowe 2013, 352); recently, he has even found a sad continuator in Jean-Claude Juncker who refused to watch in the European Parliament The Soviet Story by Edvīns Šnore in order not to support such an “anti-Soviet movie”. What is remarkable is that Hjelmslev’s article also mirrors, as I believe, general features of his thinking which is empirical, contextual and inspiring. 383 First is the empirical foundation: the starting point should always be the expe rience data, not aprioristic constructions, in this case Soviet propaganda. 3 See Cigana (2014) on the notion of participation in Hjelmslev’s theory. https://doi.org/10.15388/SBOL.2021.21 Louis Hjelmslev and the Baltic countries Next is the structural contextuality or the relative nature of values and the participa tive character of the inner arrangement of each system as opposed to thinking in absolute terms:3 before World War II the Baltic countries indeed were, as Hjelmslev writes, dictatorships, but not fascist ones and the context in which they existed and in which their leaders had to act was intricate and not that in which contemporary Western democracies existed. In connection with this, what is inspiring – in particular for a historian of the Czech countries – is the comparison sketched by Hjelmslev of the pre-war Baltic countries and the so-called third Czechoslovak Republic (1945–1948): “Men de tre omtalte Personer [viz. the Baltic leaders] var ikke fascistiske Diktatorer, lige saa lidt som Benes er kommunistisk Diktator i Tjekoslovakiet. [But the three mentioned persons were not fascist dictators, just as little as Beneš is a communist dictator in Czechoslovakia.]” (Hjelmslev 1946, 7). The third Czechoslovak Republic, with its “socializing democracy” as even Edvard Beneš put it, viz. democracy allowing only those political subjects which agree with a basic socialist programme, may be considered as a left version of the traditionalistic or conservative dictatorships in the Baltic coun tries in the 1930s. We can lastly pose the question of Hjelmslev’s motivation for his defence of the Balts. He surely had an emotional relationship with Lithuania as Gregersen (1991, I, 177) remarked. However, as noticed more accurately by Blatt (1965), the reason why Hjelmslev “was immune to Eastern propaganda” was not only his “familiarity with Baltic conditions”, but also his “common sense” (gode forstand). Indeed, I think that it is even possible to find eloquent links with his linguistic theory.l Readers of Hjelmslev have perhaps felt a certain utopian flavour in his linguistic thinking (pointed out recently by Cigana in press), based on a belief in the power of theory: in this respect Hjelmslev is a son of his time believing in science and its force. What is, however, striking is that he did not connect this belief with sympathy for totalitarian regimes, as is witnessed also by his text from 1946; thus he stood in contrast to many others professing such a utopian 384 scientism.4 The reason is perhaps that Hjelmslev differentiated between univer sal or realized (which is necessary) vs. 5 See especially: “les rapports paradigmatiques ne traduisent pas la normativité qu’on est sou vent amené à rattacher à la langue comme système de contraints prescriptives. Ceux-ci sont au-delà de la dimension prescriptive, car ils représentent les conditions ou le champ de possi bilités à disposition du sujet parlant, qui les met en place en fonction de ses besoins commu nicatifs.” 4 For instance, Rudolf Carnap (1968) wrote ingenuously that he was pleased by a Czech anthol ogy of his texts because understanding between Western and Eastern countries was very nec essary for making exploitation of one class by another impossible. https://doi.org/10.15388/SBOL.2021.21 Louis Hjelmslev and the Baltic countries general or realizable (which is possible) as Cigana (in press) put it,5 whereas other structuralists or neo-positivists did not: what is realized is not open for discussion because freedom is the real ized necessity as communists used to say in a pseudo-Hegelian manner; “voilà pourquoi, au moins de ce point de vue, l’interprétation essentialiste de cette démarche structurale nous semble échouer” (Cigana in press). Acknowledgements The preparation of this article was financed within the statutory activity of the Czech Language Institute of the Czech Academy of Sciences (RVO No. 6837809). I thank Mark Richard Lauersdorf (Lexington) for improving my English. References Blatt, Franz. 1965. Louis Hjelmslev død. [Louis Hjelmslev dead]. Politiken 1 (6), 6. Boček, Vít, Petr Malčík, eds. 2011–2016. Václav Machek: Korespondence I–III. [Vá l M h k C d I III] P h Lid é i Blatt, Franz. 1965. Louis Hjelmslev død. [Louis Hjelmslev dead]. Politiken 1 (6), 6. Boček, Vít, Petr Malčík, eds. 2011–2016. Václav Machek: Korespondence I–III. Blatt, Franz. 1965. Louis Hjelmslev død. [Louis Hjelmslev dead]. Politiken 1 (6), 6. Boček, Vít, Petr Malčík, eds. 2011–2016. Václav Machek: Korespondence I–III. [Václav Machek: Correspondence I–III]. Praha: Lidové noviny. [Václav Machek: Correspondence I–III]. Praha: Lidové noviny. Carnap, Rudolf. 1968. Úvodní slovo k českému překladu mých článků. [Introductory word to the Czech translation of my articles]. In Problémy jazyka vědy. Rudolf Carnap. Praha: Svoboda. 7–8.i Cigana, Lorenzo. 2014. La notion de « participation » chez Louis Hjelmslev: un fil rouge de la glossématique. Cahiers Ferdinand de Saussure 67, 191–202. Cigana, Lorenzo. in press. Le langage entre structure, description et possibilités. Louis Hjelmslev. Écrits choisis. Recueil d’essais publiés en traduction du danois, de l’anglais et de l’allemand. Lorenzo Cigana, ed. Lausanne: Sdvig Press. Gregersen, Frans. 1991. Sociolingvistikkens (u)mulighed. Videnskabshistoriske studier i Ferdinand de Saussures og Louis Hjelmslevs strukturalistiske sprog teorier. [The (im)possibility of sociolinguistics: Science-historical studies on Ferdinand de Saussure and Louis Hjelmslev’s structuralist language theories]. I–II. København: Tiderne skifter. 385 Hjelmslev, Louis. 1922. Indtryk fra Litauen. [Impressions from Lithuania]. Gads danske Magazin 16, 409–416, 456–464. Hjelmslev, Louis. 1930. Det litauiske Folk og dets Sprog. [The Lithuanian people and their language]. København: Henrik Koppels Forlag. Hjelmslev, Louis. 1932. Études baltiques. København: Levin & Munksgaard. Hjelmslev, Louis. 1936–1937. Accent, intonation, quantité. Studi baltici 6, 1–57. Hjelmslev, Louis. 1946. Estland, Letland og Litauen. [Estonia, Latvia, and Lithua nia]. Frit Danmark 5(19), 7, 10–11. Hjelmslev, Louis. 1956. Études de phonétique dialectale dans le domaine letto-litu anien. Scando-Slavica 2, 62–86. Janyšková, Ilona, Helena Karlíková, Eva Havlová, Radoslav Večerka, eds. 2011. Sebrané spisy Václava Machka. [The collected writings of Václav Machek]. Praha: Lidové noviny. Lowe, Keith. 2013. Savage Continent: Europe in the Aftermath of World War II. London: Penguin Books. Palionis, Jonas. 1978. L. Hjelmslevo lietuviški laiškai. [Hjelmslev’s Lithuanian letters]. Kalbotyra 29, 108–112. Rasmussen, Michael. 1992. Hjelmslevs sprogteori. Glossematikken i videnskab shistorisk, videnskabsteoretisk og erkendelsesteoretisk perpektiv. [Hjelmslev’s language theory: Glossematics in a science-historical, science-theoretical and epistemological perspective]. Odense: Odense Universitetsforlag. Registrant over Louis Hjelmslevs efterladte papirer. [Register of Louis Hjelmslev’s left documents]. References Available at: https://cc.au.dk/fileadmin/dac/Projekter/Infras trukturalisme/Billeder_til_Infrastrukturalisme/Louis_Hjelmslev_-_arkiv_paa_ KB.pdf. Accessed: 31. 12. 2020. Sabaliauskas, Algirdas. 1979. Lietuvių kalbos tyrinėjimo istorija (iki 1940. m.). [A history of research on the Lithuanian language (until 1940)]. Vilnius: Mokslas. Sabaliauskas, Algirdas. 2008. Hjelmslev Louis Trolle. Lietuvių kalbos enciklope dija. [Encyclopaedia of the Lithuanian language]. 2nd edition. Kazys Morkūnas, ed. Vilnius: Mokslo ir enciklopedijų leidybos institutas. 212–213. Vykypěl, Bohumil. 2005. Hjelmslevo glosematika ir baltų kalbų fonologija. [Hjelmslev’s glossematics and the phonology of the Baltic languages]. Vilnius: Lietuvių kalbos institutas (Bibliotheca Salensis 1). Submitted: 10 January 2021 386
https://openalex.org/W4389685521	https://aacr.figshare.com/articles/journal_contribution/Supplementary_Figure_S4_from_Spatial_and_Temporal_Mapping_of_Breast_Cancer_Lung_Metastases_Identify_TREM2_Macrophages_as_Regulators_of_the_Metastatic_Boundary/24799474/1/files/43620976.pdf	English	null	Supplementary Figure S4 from Spatial and Temporal Mapping of Breast Cancer Lung Metastases Identify TREM2 Macrophages as Regulators of the Metastatic Boundary	null	2,023	cc-by	2,160	Supplementary Figure 4 0 5 10 15 20 PD1+ [% of CD8 T cells] EO771 model * 0 10 20 30 4T1 model Distal normal tissue Metastasis * NS. 0 10 20 30 40 4T1 model * 0 10 20 30 EO771 model TREM2+ [% of CD11b+LY6G-] Distal normal tissue Metastasis Control Pre-MET Control Distal normal tissue Metastasis ns * * ns T CD4 Lef1 T CD8 Gzmk 0 10 20 30 40 0 20 40 60 % of T cells % of T cells ns ns ns ns ns ns T CD8 Gzma T CD8 Lag3 Treg 0 10 20 30 0 20 40 60 0 5 10 15 C % Control D 3,1 0 103 104 105 0 103 104 105 Comp-APC-A :: PD1 Comp-APC-A :: PD1 19,7 0 103 104 105 Comp-PE-Cy7-A :: CD8a Distal normal Metastas PD1+ CD8 T cells D 3,1 0 103 104 105 0 103 104 105 Comp-APC-A :: PD1 Comp-APC-A :: PD1 19,7 0 103 104 105 Comp-PE-Cy7-A :: CD8a Distal normal Metastasis PD1+ CD8 T cells NS. 0 5 10 15 20 PD1+ [% of CD8 T cells] EO771 model * 0 10 20 30 4T1 model Distal normal tissue Metastasis C D NS. Supplementary Figure 4 Supplementary Figure 4 Supplementary Figure 4 * ns ns * ns Neutrophil NK T 0 20 40 0 10 20 Supplementary Figure 4 A T Mon Ace Mon Fn1 Mon Thbs1 Mac Cd81 Mac Cd9 DC cDC1 DC cDC2 Neut Ptgs2 T CD4 Lef1 T CD4 S100a4 0 20 40 60 80 0 25 50 75 % Control %Relapse free % 20 40 60 80 0.001 0.010 0.100 pvalue Control Distal normal tissue Metastasis *** ns ns ns ns * *** ns * ns ns * ns B DC Macrophage Monocyte Neutrophil NK T 0 20 40 0 10 20 0 25 50 75 0 20 40 60 80 0 10 20 30 40 50 0 3 6 9 12 0.0 2.5 5.0 7.5 10.0 12.5 % of CD45+ B *** ns ns ns ns * * ns B DC Macrophage Monocyte 0 25 50 75 0 20 40 60 80 0 10 20 30 40 50 0 3 6 9 12 0.0 2.5 5.0 7.5 10.0 12.5 % of CD45+ B B A Control Distal normal tissue Metastasis * ns ns ns ns * *** ns * ns ns * ns B DC Macrophage Monocyte Neutrophil NK T 0 20 40 0 10 20 0 25 50 75 0 20 40 60 80 0 10 20 30 40 50 0 3 6 9 12 0.0 2.5 5.0 7.5 10.0 12.5 % of CD45+ Control Distal normal tissue Metastasis ns * * ns T CD4 Lef1 T CD8 Gzmk 0 10 20 30 40 0 20 40 60 % of T cells % of T cells F K Arg1 Ccl12 Ccl2 Ccl7 Ccr5 Cd274 Cd74 Cxcl10 Cxcl16 Cxcl9 H2−Aa Hilpda Ifi209 Ifit3 Isg15 Lilr4b Mmp12 Ms4a7 Rgs1 Rsad2 Stab1 Thbs1 Vegfa Cd300a Hp Klf2 Klf9 Lyz1 Mmp8 Ramp1 Retnlg Tmem176b Tnfrsf21 0 1 2 3 4 5 0 2 4 Mon Fn1 log2 expression Mon Thbs1 log2 expression G L Mon Fn1 Mon Thbs1 Oxidative phosphorylation Respiratory chain Fatty acid catabolic process Lipid oxidation Cellular lipid catabolic process Pos’ reg' of type I interferon Pos' reg' of adaptive immunity Negative reg' of vasculature develop Granulocyte migration Pos' reg' of interleukin 6 production Pos' reg' of interferon gamma Pos' reg' of inflammatory response Monocyte differentiation Glucose catabolic process Pos' reg' of Il1β Lymphocyte migration −2 −1 0 1 2 NES 1 2 34 56 7 8 −5 −2.5 0 2.5 5 −5 0 5 UMAP 1 UMAP 2 Mon Ace Mon Thbs1 Mon Fn1 Mreg Mac Isg20 1 −5 0 5 UMAP 1 0 5 10 15 20 pseudotime ns ns ns ns ns ns T CD8 Gzma T CD8 Lag3 Treg 0 10 20 30 0 20 40 60 0 5 10 15 ns ns ** ns ns * *** ns Neut Ifit3 Neut Lcn2 Neut PMN−MDSC Neut Ptgs2 0 25 50 75 100 0 4 8 12 0 25 50 75 100 0 10 20 30 40 % of Neutrophils Control Distal normal tissue Metastasis A C D E B Adam8 Arg1 Ccl12 Ccl2 Clec4n Gpnmb Hilpda Hmox1 Ifnar2 Il1b Isg20 Maf Mif Msr1 Ninj1 Rsad2 Spp1 Vcan Vegfa Car4 Cd81 Chil3 Ear1 Ear2 Fabp1 Mrc1 Selenop Tmem176b Wfdc21 0 2 4 6 0 2 4 Distal normal Macrophages log2 expression Metastasis Macrophages log2 expression H I NES cytokine mediated signaling pathway reg' of interleukin 1 production negative reg' of leukocyte proliferation reg' of innate immune response pos' reg' of adaptive immune response activation of innate immune response cellular response to interferon gamma pos' reg' of proteolysis carboxylic acid catabolic process cellular lipid catabolic process −2 −1 0 1 2 Distal normal Metastasis T Mon Ace Mon Fn1 Mon Thbs1 Mac Cd81 Mac Cd9 DC cDC1 DC cDC2 Neut Ptgs2 T CD4 Lef1 T CD4 S100a4 0 20 40 60 80 0 25 50 75 % Control %Relapse free % 20 40 60 80 0.001 0.010 0.100 pvalue 3,1 0 103 104 105 0 103 104 105 Comp-APC-A :: PD1 Comp-APC-A :: PD1 19,7 0 103 104 105 Comp-PE-Cy7-A :: CD8a Distal normal Metastasis J PD1+ CD8 T cells 0 -102 102 103 104 105 Comp-PE-A :: TREM2 32,7 0 50K 100K 150K 200K 250K FSC-A 0 50K 100K 150K 200K 250K FSC-A 8,1 TREM2+ macrophages Distal normal Metastasis NS. Supplementary Figure 4 0 5 10 15 20 PD1+ [% of CD8 T cells] EO771 model * 0 10 20 30 4T1 model Distal normal tissue Metastasis F Arg1 Ccl12 Ccl2 Ccl7 Ccr5 Cd274 Cd74 Cxcl10 Cxcl16 Cxcl9 H2−Aa Hilpda Ifi209 Ifit3 Isg15 Lilr4b Mmp12 Ms4a7 Rgs1 Rsad2 Stab1 Thbs1 Vegfa Cd300a Hp Klf2 Klf9 Lyz1 Mmp8 Ramp1 Retnlg Tmem176b Tnfrsf21 0 1 2 3 4 5 0 2 4 Mon Fn1 log2 expression Mon Thbs1 log2 expression G ns ns ** ns ns * *** ns Neut Ifit3 Neut Lcn2 Neut PMN−MDSC Neut Ptgs2 0 25 50 75 100 0 4 8 12 0 25 50 75 100 0 10 20 30 40 % of Neutrophils Control Distal normal tissue Metastasis E F G Mon Fn1 Mon Thbs1 Oxidative phosphorylation Respiratory chain Fatty acid catabolic process Lipid oxidation Cellular lipid catabolic process Pos’ reg' of type I interferon Pos' reg' of adaptive immunity Negative reg' of vasculature develop Granulocyte migration Pos' reg' of interleukin 6 production Pos' reg' of interferon gamma Pos' reg' of inflammatory response Monocyte differentiation Glucose catabolic process Pos' reg' of Il1β Lymphocyte migration −2 −1 0 1 2 NES E Adam8 Arg1 Ccl12 Ccl2 Clec4n Gpnmb Hilpda Hmox1 Ifnar2 Il1b Isg20 Maf Mif Msr1 Ninj1 Rsad2 Spp1 Vcan Vegfa Car4 Cd81 Chil3 Ear1 Ear2 Fabp1 Mrc1 Selenop Tmem176b Wfdc21 0 2 4 6 0 2 4 Distal normal Macrophages log2 expression Metastasis Macrophages log2 expression H NES cytokine mediated signaling pathway reg' of interleukin 1 production negative reg' of leukocyte proliferation reg' of innate immune response pos' reg' of adaptive immune response activation of innate immune response cellular response to interferon gamma pos' reg' of proteolysis carboxylic acid catabolic process cellular lipid catabolic process −2 −1 0 1 2 Distal normal Metastasis I H Metastasis Macrophages log2 expression K L 1 2 34 56 7 8 −5 −2.5 0 2.5 5 −5 0 5 UMAP 1 UMAP 2 Mon Ace Mon Thbs1 Mon Fn1 Mreg Mac Isg20 K L 1 2 34 56 7 8 −5 −2.5 0 2.5 5 −5 0 5 UMAP 1 UMAP 2 Mon Ace Mon Thbs1 Mon Fn1 Mreg Mac Isg20 1 −5 0 5 UMAP 1 0 5 10 15 20 pseudotime p g g p J 0 -102 102 103 104 105 Comp-PE-A :: TREM2 32,7 0 50K 100K 150K 200K 250K FSC-A 0 50K 100K 150K 200K 250K FSC-A 8,1 TREM2+ macrophages Distal normal Metastasis * NS. Supplementary Figure 4 In boxplots, the center line represents the median, the box limits denote the 25th to the 75th percentile, and the whiskers represent the minimum and maximum values. Differentially expressed genes (DEGs) are colored in red and leading DEGs are labeled. Normalized GO term enrichment scor (NES) is shown on x-axis For all terms padj < 0 05 yp A. Fractions of cells belonging to different immune lineages (from total) or subtypes (from their respective lineage), averaged over control (x axis) or relapse-free (y axis) samples. Size indicates the average of x and y. Color depicts p-value of two-sided t-test between x and y, accounting for sample variation. A. Fractions of cells belonging to different immune lineages (from total) or subtypes (from their respective lineage), averaged over control (x axis) or relapse-free (y axis) samples. Size indicates the average of x and y. Color depicts p-value of two-sided t-test between x and y, accounting for sample variation. Color depicts p-value of two-sided t-test between x and y, accounting for sample variation. B. Fractions of indicated cell types out of total CD45+ cells. C. Fractions of indicated T cell subtypes from total T cells. D. Representative flow cytometry plots and quantification of PD1+ CD8 T cells in EO771 or 4T1 distal normal and metastatic lung tissues. Gating was performed on live, CD45+, TCRb+, CD8+ cells. E. Fractions of indicated neutrophil subtypes from total neutrophils. E. Fractions of indicated neutrophil subtypes from total neutrophils. F. Comparison of monocyte subtypes gene expression (log2 normalized). G. Enriched gene ontology terms in monocyte subtypes. G. Enriched gene ontology terms in monocyte subtypes. H. Comparison between gene expression (log2 normalized) of macrophages from distal normal and metasta- sis tissue samples. I. Enriched gene ontology terms in macrophages distal normal and metastasis tissue samples. Normalized GO term enrichment score (NES) is shown on x-axis. For all terms, padj < 0.05. J. Representative flow cytometry plots and quantification of TREM2+ macrophages in EO771 distal normal and metastatic lung tissues (left), or 4T1 control, Pre-MET, distal normal, and metastatic lung tissues (right). Gating was performed on live, CD45+, CD11b+ cells. K. Monocyte-macrophage pseudotime trajectory (Monocle 3, Methods) projected on UMAP (Mon Ace defined as the source, cells colored by subpopulation). L. Pseudotime analysis of (G), cells colored by pseudotime. Two-tailed Student's t-test was used. Supplementary Figure 4 0 10 20 30 40 4T1 model * 0 10 20 30 EO771 model TREM2+ [% of CD11b+LY6G-] Distal normal tissue Metastasis Control Pre-MET 1 −5 0 5 UMAP 1 0 5 10 15 20 pseudotime L J L J 0 -102 102 103 104 105 Comp-PE-A :: TREM2 32,7 0 50K 100K 150K 200K 250K FSC-A 0 50K 100K 150K 200K 250K FSC-A 8,1 TREM2+ macrophages Distal normal Metastasis *** 0 10 20 30 EO771 model TREM2+ [% of CD11b+LY6G-] Control Pre-M K 5 Supplementary Figure 4. Progression to lung metastasis is associated with infiltration by unconventional immune cell subtypes. Supplementary Figure 4. Progression to lung metastasis is associated with infiltration by unconventional immune cell subtypes. Supplementary Figure 4. Progression to lung metastasis is associated with infiltration by unconventional immune cell subtypes. A. Fractions of cells belonging to different immune lineages (from total) or subtypes (from their respective lineage), averaged over control (x axis) or relapse-free (y axis) samples. Size indicates the average of x and y. Color depicts p-value of two-sided t-test between x and y, accounting for sample variation. B. Fractions of indicated cell types out of total CD45+ cells. C. Fractions of indicated T cell subtypes from total T cells. D. Representative flow cytometry plots and quantification of PD1+ CD8 T cells in EO771 or 4T1 distal normal and metastatic lung tissues. Gating was performed on live, CD45+, TCRb+, CD8+ cells. E. Fractions of indicated neutrophil subtypes from total neutrophils. F. Comparison of monocyte subtypes gene expression (log2 normalized). G. Enriched gene ontology terms in monocyte subtypes. H. Comparison between gene expression (log2 normalized) of macrophages from distal normal and metasta sis tissue samples. I. Enriched gene ontology terms in macrophages distal normal and metastasis tissue samples. Normalized GO term enrichment score (NES) is shown on x-axis. For all terms, padj < 0.05. J. Representative flow cytometry plots and quantification of TREM2+ macrophages in EO771 distal norma and metastatic lung tissues (left), or 4T1 control, Pre-MET, distal normal, and metastatic lung tissues (right). Gating was performed on live, CD45+, CD11b+ cells. K. Monocyte-macrophage pseudotime trajectory (Monocle 3, Methods) projected on UMAP (Mon Ace defined as the source, cells colored by subpopulation). L. Pseudotime analysis of (G), cells colored by pseudotime. Two-tailed Student's t-test was used. Supplementary Figure 4 In boxplots, the center line represents the median, the box limits denote the 25th to the 75th percentile, and the whiskers represent the minimum and maximum values. Differentially expressed genes (DEGs) are colored in red and leading DEGs are labeled. Normalized GO term enrichment score (NES) is shown on x-axis. For all terms, padj < 0.05.
https://openalex.org/W3188814757	https://www.researchsquare.com/article/rs-775793/latest.pdf	English	null	Open-Source Software Library For Real-Time Inertial Measurement Unit Data-Based Inverse Kinematics Using OpenSim	Research Square (Research Square)	2,021	cc-by	6,785	Open-Source Software Library For Real-Time Inertial Measurement Unit Data-Based Inverse Kinematics Using OpenSim Jere Lavikainen (  jere.lavikainen@uef.fi ) University of Eastern Finland https://orcid.org/0000-000 Paavo Vartiainen University of Eastern Finland Lauri Stenroth University of Eastern Finland Pasi Karjalainen University of Eastern Finland Jere Lavikainen (  jere.lavikainen@uef.fi ) University of Eastern Finland https://orcid.org/0000-0002-7636-8229 Introduction Inertial measurement units (IMUs) are measurement devices that contain triaxial magnetometers, gyroscopes and accelerometers. IMUs used in biomechanics are usually packed into cases that fit on a human palm. Sensor fusion algorithms such as Kalman filters are used to estimate the three-dimensional orientation of the IMUs in space. [1] IMUs can be used as an alternative to marker-based optical motion tracking systems to perform motion analysis. Compared with optical motion tracking systems, IMUs are cheaper, can be attached to the subject without the palpation of anatomical landmarks, do not suffer from line-of-sight issues, are not limited to a specific target volume and can be used in field conditions. This comes at a small cost of accuracy compared with optical motion tracking systems, and IMU-specific error sources such as drifting. [2] ernative to marker-based optical motion tracking systems to perform motion analysis. Although IMUs have been used in the study of human motion and biomechanics for less than two decades, they have been proven to be a good solution to overcome the shortcomings of optical motion tracking systems and their popularity is increasing. Moreover, sensors in which IMU is coupled with electromyography (EMG) electrodes further enhances the versatility of these sensors for analyzing human movement in sports and clinical applications [3–5]. OpenSim [6] is an open-source software for analyzing the kinematics and dynamics of musculoskeletal systems. The inverse kinematics (IK) tool is commonly used in OpenSim. It takes motion tracking data, which is often marker-based, as its input and solves the motion of the model (i.e., joint angles) by minimizing the sum of squares of the difference between experimental and model markers positions. As of OpenSim 4.1 [7], inverse kinematics can also be calculated using IMU data as the input. Here, the differences between the model and experimental sensor orientations are minimized. IK analyses are typically done offline so that the IK is analyzed after the measurement is finished. Recent years have seen progress in some real-time or online analysis solutions and systems, but these studies [8–12] have mostly focused on specific marker sets and models and their generalization to arbitrary measurement setups is difficult. For example, in real-time IMU-based applications, Bonnet et al. [8] estimated the real-time inverse kinematics (RTIK) of the trunk and lower limbs using a single IMU located at the lower back. Research Article Keywords: OpenSim, real-time, inverse kinematics, inertial measurement unit, motion analysis Posted Date: October 19th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-775793/v2 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/13 Page 1/13 Abstract An open-source software library for multithreaded real-time inverse kinematical (IK) analysis of inertial measurement unit (IMU) data using OpenSim was developed. Its operation delays and throughputs were measured with a varying number of IMUs and parallel computing IK threads using two different musculoskeletal models, one a lower-body and torso model and the other a full-body model. Full-body inverse kinematics with data from 12 IMUs could be calculated in real-time with a mean delay below 100 ms and at more than 900 samples per second. Live visualization of IK is an option but results in limited IK throughput. The effect of this limitation was assessed by comparing the range of motion (ROM) of each joint from visualized real-time IK to the ROM from offline IK at IMU sampling frequency, resulting in mean ROM differences below 0.3 degrees. The software library enables real-time inverse kinematical analysis with different numbers of IMUs and customizable musculoskeletal models, making it possible to do subject-specific full-body motion analysis outside the motion laboratory in real-time. Introduction In another example, sensors containing IMUs and EMG electrodes, which were placed on muscles, were used in a recent study [12] to calculate the real-time kinematics and kinetics of the lower limb. However, these studies rely on complex computational methods, making it difficult for others to repeat and adapt the experiment without knowledge of sensor fusion or deep learning. Page 2/13 Page 2/13 It has been shown that a full biomechanical analysis of joint and muscle function can be obtained in real-time with a C/C + + software library [9]. The software library was capable of reading marker data and performing IK and inverse dynamics on a full-body model at more than 120 samples per second. However, the software is commercial and relies on a single predefined model rather than a subject-specific or user-defined musculoskeletal model, limiting its usefulness in research. Additionally, Falisse et al. [11] showed that comparing its outputs with those of OpenSim 3.3 resulted in statistically significant differences in joint kinematics, kinetics and muscle forces, highlighting the dependency of the output on the selected model. Hence, it is invaluable that the user can select or generate a model that best fits to the application or research question. A software architecture developed by Pizzolato et al. for RTIK and inverse dynamics of human motion [13] utilizes OpenSim to enable analysis with different marker placement schemes and musculoskeletal models. However, it is based on a now-outdated version of the OpenSim application programming interface (API) and does not support IMU-based motion data. Consequently, reliance on optical motion tracking limits its applicability to laboratory settings. While this study was being reported, Stanev et al. [14] published an open-source software framework that allows the kinematical and dynamical analysis of subject-specific musculoskeletal models using the OpenSim API. Their software performs the analysis in real-time and supports both IMU- and marker-based data. The differences between it and the one developed in this study are evaluated in Conclusions. The aim of this study was to develop a freely available software library that reads orientation data from IMUs and calculates the IK on a user-given musculoskeletal model in real-time using OpenSim 4.1 API. Introduction Further, the aim was to quantify the software’s IK execution time and throughput with different numbers of processor threads calculating the IK and different numbers of IMUs, and to determine if lowered input data frequency resulting from live visualization meaningfully affects calculated ranges of motion (ROMs). Methods Working principles of the software A software library for reading real-time IMU orientation data as quaternions and processing the data to calculate the IK of a musculoskeletal model was developed using C + + and published on GitHub. The software utilizes OpenSim 4.1 API to invoke methods that calibrate the musculoskeletal model and perform the IK to solve joint angles using quaternion- based orientation data from live measurements. The software library supports Xsens MTw Awinda™ (Xsens Technologies B.V., Enschede, Netherlands) and Delsys Trigno (Delsys Inc., Natick, Massachusetts, USA) IMUs. The open-source nature of the software library allows others to add support for other devices. IMU orientations are received wirelessly as quaternions using the Xsens Device API or individual quaternion elements are read from a byte stream via socket communication sent by Trigno Control Utility. Information about which IMU corresponds to which body on the musculoskeletal model is read from an XML file. Instead of reading orientation data from the actual IMUs, an option to generate randomized quaternion orientations for testing purposes without IMUs is available. The orientation information from IMUs is combined in a time series table that contains only one sample, i.e., time point. The time series table is given to OpenSim’s IK solver object, which solves the IK for that time point. The process is repeated for each sample. The resulting time series of joint angles can be saved in a text file in .mot format, which allows the output to be viewed using the OpenSim graphical user interface. The read quaternions, and also EMG time series in the case of Trigno Avanti sensors, can be saved in a text file for later offline analysis. Page 3/13 Page 3/13 The working principle of the software library is straightforward. Producer-consumer thread synchronization is used to get orientation data from IMUs. A producer thread and a consumer thread run concurrently. The producer reads orientation data as quaternions from the IMUs and saves it into a buffer that is shared between the threads. The consumer reads and removes data from the buffer and creates a new IK thread in a thread pool. This way there can be several concurrent IK calculations, improving the throughput of the program. The maximum number of concurrent IK threads is defined by the user. Methods If that number is already active when the consumer thread starts another IK thread, the consumer thread will wait for one of the IK threads in the thread pool to finish. A diagram illustrating the workflow is shown in Fig. 1. The software library has been tested to work on 64-bit Windows 7 and Windows 10 operating systems. The source code of the software library is available on GitHub at https://github.com/jerela/OpenSimLive. Execution times of the IK operation The execution time of a single IK operation was determined by calculating the duration between its starting and finishing points. The mean, standard deviation (STD) and 95% confidence interval of the execution time was calculated for 10 000 IK operations. Because the orientation retrieval time of IMUs was negligibly small compared with the execution time of the computationally heavier IK and varied between Xsens and Delsys IMUs, only IK execution time was studied. The execution time represents the delay between receiving orientation data from IMUs and retrieving the corresponding IK output. For the orientation input for IK, unit quaternions were randomly generated instead of using real data. The use of random quaternions ensured that execution times were calculated for a large variety of different orientation combinations. Execution time was calculated separately using the Gait2392 lower extremities and torso model (23 degrees of freedom, DOFs; referenced from here on as the lower body model) [15–18] with 1 and 7 IMUs and the Hamner full-body model (29 DOFs, referenced from here on as the full-body model) [19] with 1, 7 and 12 IMUs. With 1 IMU, the corresponding body segment was the pelvis. With 7 IMUs, they were the pelvis, thighs, shanks and the feet. With 12 IMUs, they were the torso, upper arms and lower arms in addition to the segments of the setup with 7 IMUs. Throughputs We evaluated the effect of this frame drop by comparing the ranges of motion (ROMs) of the joint angles solved with RTIK to ROMs of the joint angles solved with offline IK at the full sampling frequency of the IMUs. When live visualization is disabled, the typical RTIK solving frequency is higher than the sampling frequency of the IMUs and therefore the IK solution includes data at the sampling frequency of the IMUs. However, when live visualization is enabled during RTIK, its throughput drops leading to RTIK solving frequency that is lower than the sampling frequency of the IMUs. In this case, samples are skipped which may negatively affect the accuracy of gait parameters derived from the IK solution. We evaluated the effect of this frame drop by comparing the ranges of motion (ROMs) of the joint angles solved with RTIK to ROMs of the joint angles solved with offline IK at the full sampling frequency of the IMUs. To measure how accurately ROM can be analyzed in real time with and without visualization, subject motion during walking on an instrumented treadmill (Motek Medical B.V., Amsterdam, Netherlands) was measured with IMUs and RTIK was calculated. The live visualization of RTIK was enabled, which reduced the frequency of RTIK from the 60 Hz sampling frequency of the IMUs to 45 Hz on average when using the desktop computer. The time series of IMU orientations as quaternions were saved to a file at the full 60 Hz sampling frequency after each trial and analyzed to obtain offline IK. Ten walking trials, each containing approximately a minute of walking data at a speed of 1.5 m/s, were measured with Xsens MTw Awinda IMUs. The full-body model was used in the analysis and the measured motion exerted 26 of its DOFs. The subject was 25 years old, had a height of 181 cm and a weight of 85 kg and gave his written consent to participate in the study. A total of 12 IMUs were strapped on the subject’s upper arms, forearms, chest, pelvis, thighs, shanks and feet, as shown in supplementary information (Online Resource 1). The subject was instructed to take the standard anatomical position at the beginning of each trial to calibrate the IMUs on the musculoskeletal model as per the standard IMU calibration procedure of OpenSim. Throughputs RTIK and offline IK were used to calculate ROM for all joints of the model from 715 gait cycles. The ROMs were compared between RTIK and offline IK to determine effect of the lower IK sampling frequency due to the visualization on the ROMs. ROM was selected as the variable of interest since it indicates if the lower sampling frequency has a major effect on the IK solution. The results were reported as the mean absolute error (MAE) and 95% confidence interval between offline IK and RTIK ROM. Throughputs The throughput of the IK, defined as IK operations per second, was measured by calculating how many IK operations the software library could finish in one minute. In addition to the IK calculation itself, the measurement workflow included the producer thread pushing identity quaternions and their time points to the shared buffers and the consumer thread reading them. The measurements were done with both the lower body model and the full-body model using 1, 2, 4, 6, 8, 12, 16 and 24 threads for concurrent IK. For the lower body model, the measurements were done with 1 and 7 IMUs, and with 1, 7 and 12 IMUs for the full-body model, similarly to how the execution time measurements were measured. Instead of measuring real human motion, an identity quaternion was used as the orientation for each IMU, ensuring that the IK throughputs are comparable between repetitions and that the throughputs represent ideal performance. Instead of measuring real human motion, an identity quaternion was used as the orientation for each IMU, ensuring that the IK throughputs are comparable between repetitions and that the throughputs represent ideal performance. The execution time and throughput tests were run with two computers, one a desktop (Fujitsu Celsius W550 Power: Windows 10 Education 64-bit, Intel Core i7-6700 3.40 GHz 8-CPU processor, 32768 MB RAM) and the other a laptop computer (HP EliteBook 8570w: Windows 10 Education 64-bit, Intel Core i7-3740QM 2.70 GHz 8-CPU processor, 8192 MB RAM). The desktop results represent the performance of the software library in a stationary laboratory setup while the laptop results illustrate how the software library can perform in portable measurement settings outside a laboratory. The live visualization of RTIK was disabled during both performance tests. Error comparison of joint angles Page 4/13 When live visualization is disabled, the typical RTIK solving frequency is higher than the sampling frequency of the IMUs and therefore the IK solution includes data at the sampling frequency of the IMUs. However, when live visualization is enabled during RTIK, its throughput drops leading to RTIK solving frequency that is lower than the sampling frequency of the IMUs. In this case, samples are skipped which may negatively affect the accuracy of gait parameters derived from the IK solution. Error comparison of joint angles The mean ROM error for all DOFs was 0.0675 degrees. The greatest MAE in ROM was observed in ankle joints (up to 360% of the mean for all joints), followed by the left hip joint (Fig. 3). Pelvis and upper extremities had the smallest ROM error. All MAEs remained below 0.3 degrees. Throughputs Throughputs higher than 1100 and 700 operations per second were reached on the desktop and the laptop, respectively (Fig. 2). The throughput of the desktop was 30–50% greater than that of the laptop when more than one IK thread was used. When one IK thread was used, the throughput of the desktop was 630–1560% greater than that of the laptop, with the use of fewer IMUs leading to a greater difference in throughput. The lower body model performed faster than the full- body model, and model selection had a higher impact on throughput than the number of IMUs. Maximum CPU utilization is seen as flattening of the curve after eight threads for the desktop and six threads for the laptop. Error comparison of joint angles Execution times of the IK operation Execution times of the IK operation The means and standard deviations of execution times (operation delays) increased with increasing number of IMUs similarly on both the desktop and the laptop (Table 1). With one IMU, the full-body model (Hamner) was 60–65% slower than the lower body model (Gait2392) and had more variation in execution times. The mean execution times were approximately 25% longer on the laptop than on the desktop. Page 5/13 Page 5/13 Page 5/13 Table 1 Table 1 Mean, standard deviation (STD) and 95% confidence interval (CI) of execution times of a single inverse kinematics (IK) operation. The values are calculated over 10 000 IK operations for two different musculoskeletal models, two different computers and 1, 7 or 12 inertial measurement units (IMUs). Randomly selected unit quaternions were used as IMU p , ( ) y q orientations. Gait2392, 1 IMU Gait2392, 7 IMUs Hamner, 1 IMU Hamner, 7 IMUs Hamner, 12 IMUs Desktop Laptop Desktop Laptop Desktop Laptop Desktop Laptop Desktop Laptop Mean time (ms) 6.97 8.52 43.55 56.96 11.52 13.77 48.12 64.01 80.42 95.24 STD (ms) 2.31 3.62 22.39 27.19 6.08 7.05 34.03 46.36 29.49 42.00 95% CI (ms) 0.05 0.07 0.44 0.53 0.12 0.14 0.67 0.91 0.58 0.82 Throughputs Discussion With our software library, we aimed to achieve this target by using multithreading and the IK algorithm of OpenSim 4.1. We investigated the execution times and throughputs of the IMU-based IK to determine if the output can be considered real-time. Pizzolato et al. [13] used an execution time of 75 ms as the threshold for a real-time system. It was based on a study by Kannape and Blanke [21] in which the subjects were able to identify the displayed motion as self-generated in real-time in over 80% of the cases if the delay in motion display was less than 75 ms. Even with a delay of 210 ms, subjects identified the visualized motion as self-generated in real-time in 50% of the cases. Borbély and Szolgay [10] noted that the IK algorithm of OpenSim 3.3 had an execution time of about 145 ms, thus calculating IK at about 7 Hz and “falling behind the generally accepted practice in human movement recording of at least 50 Hz”. Therefore, a real- time application should achieve IK throughput of 50 operations per second with an execution time below 75 ms for any single operation. With our software library, we aimed to achieve this target by using multithreading and the IK algorithm of OpenSim 4.1. Another interesting finding by Kannape and Blanke [21] was that subjects modulated their stride based on the delay between the motion and its visualization. Therefore, it is important to minimize the delay when preparing a real-time measurement setup to prevent subjects from altering their gait characteristics based on delayed visual feedback. Live visualization is unnecessary in applications where IK is an intermediate output that is used to estimate contact forces, instruct a robot arm in rehabilitation applications or calculate gait parameters, to name a few examples. Thus, the performance tests were designed so that they evaluate only the performance of IK, which is the core feature of the software library. Real motion, such as walking, contains a combination of different orientations, most of which are within a typical model’s joint angle boundaries. The constant identity quaternions used in throughput tests represent the calibration pose reoccurring repeatedly, while randomly generated unit quaternions used in execution time tests often result in unrealistic poses. Discussion We present an open-source software library for the real-time inverse kinematical analysis of IMU data with user-defined musculoskeletal models using OpenSim 4.1. Full-body IK can be calculated for a single sample in less than 100 ms. On a desktop computer, the software library can solve RTIK at more than 1100 samples per second while tracking the pelvis and the lower extremities and more than 900 samples per second while tracking the full-body kinematics. On a laptop computer, the corresponding throughputs were 700 and 600 samples per second, respectively. Using 12 IMUs to track walking and visualizing the results on a full-body running model, RTIK was solved at 45 samples per second. The drop from the IMU output sampling rate of 60 Hz resulted in a minimal difference in calculated joint ROMs (< 0.3 degrees). The software library allows the use of RTIK virtually without limitations due to location or environment. This opens possibilities for a variety of applications including rehabilitation, ergonomics and human-machine interfaces for controlling collaborative robots. Moreover, it has been shown that a laboratory setting may affect how a person moves [20] and thus it is beneficial that the movement of interest can be observed in the real environment of that movement or behavior. Execution times and throughputs Page 6/13 We investigated the execution times and throughputs of the IMU-based IK to determine if the output can be considered real-time. Pizzolato et al. [13] used an execution time of 75 ms as the threshold for a real-time system. It was based on a study by Kannape and Blanke [21] in which the subjects were able to identify the displayed motion as self-generated in real-time in over 80% of the cases if the delay in motion display was less than 75 ms. Even with a delay of 210 ms, subjects identified the visualized motion as self-generated in real-time in 50% of the cases. Borbély and Szolgay [10] noted that the IK algorithm of OpenSim 3.3 had an execution time of about 145 ms, thus calculating IK at about 7 Hz and “falling behind the generally accepted practice in human movement recording of at least 50 Hz”. Therefore, a real- time application should achieve IK throughput of 50 operations per second with an execution time below 75 ms for any single operation. Discussion This makes the IK based on identity quaternions simpler and the IK based on randomized unit quaternions heavier to calculate than the average orientations during walking, or any typical human motion. Therefore, the execution times can be interpreted as the worst performance and the throughput results as the best performance when analyzing human motion without live visualization. Throughputs Figure 2 shows that increasing the number of concurrent processor threads increased the throughput until about eight threads, which was the maximum CPU core number for both computers. Increasing the number of IK threads further had no meaningful effect on the throughput, which was also observed in an earlier study on RTIK [13]. It was observed during the testing that CPU utilization reached 100% at six and eight concurrent threads with the desktop and the laptop, respectively, while memory utilization did not get close to its maximum capacity. In terms of CPU clock rate, the desktop was 26% faster than the laptop. The closest match to this performance difference in throughput is found with two IK threads (30–33% faster). The increase in throughput by multithreading is especially large when a low number of threads are used. For example, throughput increases from less than 50 to approximately 400 when the number of IK threads increases from one to two on the laptop. Doubled computational capacity alone cannot explain the increase in throughput. The effect is less pronounced but still present on the desktop. Furthermore, the relationship between the throughput and the number of IK threads is clearly nonlinear whereas an earlier RTIK study found it almost linear [13]. No explanation for this phenomenon was found, but it should be addressed in the future development of the software library. For any number of IMUs and 4 or more concurrent threads, the lower body model with 23 DOFs performed approximately 10% faster than the full-body model with 29 DOFs. Therefore, model selection has a noticeable effect on the performance of RTIK and the model with the smallest sufficient number of DOFs should be chosen to reach maximal RTIK performance. The number of IMUs had a smaller effect on throughput than model selection and computer hardware. Although it was not reported in the results, the number of IMUs was observed to affect the throughput more significantly when joint angle boundaries of the model are exceeded. Furthermore, in that situation the performance effect of model selection increased and became more important than computer hardware. The software library is clearly capable of calculating IK at a higher rate than the lower limit of 50 Hz named by Borbély and Szolgay [10], but requires multithreading to reach it with complex musculoskeletal models. Execution times of the IK operation Execution times of the IK operation Table 1 shows that with one and seven IMUs, the execution times are shorter and vary less for the lower body model (Gait2392) than for the full-body model (Hamner), which implies that the execution time of the lower body model is more consistent than that of the full-body model. Both the mean execution times and the standard deviations are smaller on the desktop than on the laptop. However, the execution times vary less with 12 than with seven IMUs on the full-body model. For both models, the standard deviations of the execution times are on the same scale as the mean execution times, implying that there is great variation in the execution time. The randomized nature of the used quaternion orientations is a likely contributor to the high standard deviation, because randomized orientations occasionally lead to strange segment orientation combinations that do not reflect valid human motion and take the IK algorithm a varying amount of time to solve. During the development of the test program, it was noticed that the results varied greatly, implying that more than 10 000 IK operations are required to draw lasting conclusions. However, running the test even with 10 000 operations could take up to 20 minutes, so the number of operations was left as it was. The 95% confidence intervals of execution times are roughly 1% of the mean execution time in all cases, meaning that the execution times stay consistently below 75 ms except when 12 IMUs are used. In that case, the execution times stay consistently below 100 ms. Although measuring full-body motion with 12 IMUs fails to meet the best criterion for delay, it is less than half of the 210 ms delay that marks 50% confidence in perceiving motion as real-time [21]. Therefore, Page 7/13 while the execution times with 12 IMUs are not ideal, they are still acceptable. Because the execution times represent the minimum delay from the orientation data retrieval to the moment we can visualize or further analyze the IK output, the number of IMUs in a real-time measurement should be chosen considering the delays that are acceptable for the application. while the execution times with 12 IMUs are not ideal, they are still acceptable. Execution times of the IK operation Because the execution times represent the minimum delay from the orientation data retrieval to the moment we can visualize or further analyze the IK output, the number of IMUs in a real-time measurement should be chosen considering the delays that are acceptable for the application. Throughputs The throughputs reported in this study should be sufficient to match the sampling frequency of most IMUs because they typically have a maximal sampling frequency well below 500 Hz. For instance, the maximum sampling rate of Xsens MTw Awinda IMUs is 120 Hz [1]. Availability of data and material Not applicable. Code availability The software library described in this article is published open-source at https://githu The software library described in this article is published open-source at https://github.com/jerela/OpenSimLive. The software library described in this article is published open-source at https://github.com/jerela/OpenSimLive. Error comparison Because loss of frequency may lead to reduced accuracy in measuring sharp peaks in joint angles, joints where motion direction changes fast are likely to have high ROM error (Fig. 3). During walking ankle flexion (ankle_angle_r and ankle_angle_l) undergoes fast changes, which explains why its ROM error stands out. However, because all ROM errors remain consistently below 0.3 degrees, the effect of the drop in visualized IK from 60 to 45 Hz on ROM is very small. The ROM error of left hip adduction stands out because it is visibly higher than that of the right hip. The error is caused by an artifact in IMU signal that caused the left leg to be violently jerked to the right after the left toe-off phase. The artifact is probably caused by the distortion of magnetic fields near the ferromagnetic laboratory hardware, which the left leg was closer to. Page 8/13 Page 8/13 Funding This work was supported by the European Union (European Regional Development Fund) and the University of Eastern Finland under the projects Human measurement and analysis research and innovation laboratories (HUMEA, project IDs A73200 and A73241) and Digital Technology RDI Environment (Digi Center, project IDs A74338 and A74340). Conflicts of interest The authors have no relevant financial or non-financial interests to disclose. Conclusion An open-source software library that builds upon the widely used OpenSim software was developed and published for IMU-based RTIK. This allows the joint angles of any OpenSim-compatible musculoskeletal model to be analyzed in real- time. While another real-time solution was concurrently and independently developed by Stanev et al. [14], its IK calculation does not utilize multithreading, which may limit its throughput, although its IK calculation relies on lower- level API classes that are faster than those used by the software library developed in this study when a single thread is used. The authors encourage others to contribute to the open-source project. The development of the software library will closely follow the development of OpenSim to utilize its built-in functionality for processing live data. The software library could be utilized in real-time estimation of joint moments, muscle forces and joint contact forces based only on IMU data. Ground reaction forces and moments and kinematics are required for solving the equations of motion for the musculoskeletal model using inverse dynamics. It has been shown that ground reaction forces and moments can be predicted from IMU-derived kinematics [14, 22]. Moreover, estimation of muscle forces using optimization techniques uses kinematics and inverse dynamics estimates of joint moments as inputs and estimates of joint contact forces can be derived based on kinematics, inverse dynamics and muscle forces. Hence, IMUs could be potentially used for the real- time estimation of musculoskeletal dynamics outside the laboratory and implemented in the software library in the future. Another interesting future application is the use of RTIK output together with EMG. Thus, combining IK output with EMG in real-time may provide interesting possibilities for estimating muscle forces and musculoskeletal loading using EMG driven musculoskeletal simulations [23], for biofeedback to optimize rehabilitation or ergonomics or for biosignal-based operating systems. Ethics approval Not applicable. Consent to participate The subject signed a participation form with their informed consent to participate in t Consent for publication The subject signed a participation form with their informed consent to allow the publication of their data recorded during this study. Author’s contributions JL conceptualized and developed the software library and wrote the original manuscript draft and planned and conducted the measurements and their analysis. Page 9/13 PV conceptualized the software library and planned and conducted the measurements. He also provided resources during the research, supervised the research and reviewed and edited the manuscript. PV conceptualized the software library and planned and conducted the measurements. He also provided resources during the research, supervised the research and reviewed and edited the manuscript. LS conceptualized the software library and planned the measurements. He also provided resources during the research and reviewed and edited the manuscript. PAK supervised and administered the research and acquired funding for it. References Page 10/13 References 1. Paulich M, Schepers M, Rudigkeit N, Bellusci G (2018) Xsens MTw Awinda: Miniature Wireless Inertial Motion Tracker for Highly Accurate 3D Kinematic Applications 2. Saber-Sheikh K, Bryant EC, Glazzard C et al (2010) Feasibility of using inertial sensors to assess human movement. Man Ther 15:122–125. https://doi.org/10.1016/j.math.2009.05.009 3. Cometa, Srl, Cometa Systems \| Wireless EMG and IMU solutions. https://www.cometasystems.com/. Accessed 10 Feb 2021 4. Delsys Incorporated, Trigno™ Avanti Platform - Delsys. https://www.delsys.com/trigno/. Accessed 10 Feb 2021 5. Noraxon USA, Ultium EMG, \| Noraxon USA. https://www.noraxon.com/our-products/ultium-emg/. Accessed 10 Feb 2021 6. Delp SL, Anderson FC, Arnold AS et al (2007) OpenSim: Open-source software to create and analyze dynamic simulations of movement. IEEE Trans Biomed Eng 54:1940–1950. https://doi.org/10.1109/TBME.2007.901024 7. Seth A, Hicks JL, Uchida TK et al (2018) OpenSim: Simulating musculoskeletal dynamics and neuromuscular control to study human and animal movement. PLoS Comput Biol 14:. https://doi.org/10.1371/journal.pcbi.1006223 8. Bonnet V, Mazza C, Fraisse P, Cappozzo A (2013) Real-time estimate of body kinematics during a planar squat task using a single inertial measurement unit. IEEE Trans Biomed Eng 60:1920–1926. https://doi.org/10.1109/TBME.2013.2245131 9. Van Den Bogert AJ, Geijtenbeek T, Even-Zohar O et al (2013) A real-time system for biomechanical analysis of human movement and muscle function. Med Biol Eng Comput 51:1069–1077. https://doi.org/10.1007/s11517- 013-1076-z 1. Paulich M, Schepers M, Rudigkeit N, Bellusci G (2018) Xsens MTw Awinda: Miniature Wireless Inertial Motion Tracker for Highly Accurate 3D Kinematic Applications 2. Saber-Sheikh K, Bryant EC, Glazzard C et al (2010) Feasibility of using inertial sensors to assess human movement. Man Ther 15:122–125. https://doi.org/10.1016/j.math.2009.05.009 3. Cometa, Srl, Cometa Systems \| Wireless EMG and IMU solutions. https://www.cometasystems.com/. Accessed 10 Feb 2021 3. Cometa, Srl, Cometa Systems \| Wireless EMG and IMU solutions. https://www.cometasystems.com/. Accessed 10 Feb 2021 4. Delsys Incorporated, Trigno™ Avanti Platform - Delsys. https://www.delsys.com/trigno/. Accessed 10 Feb 2021 4. Delsys Incorporated, Trigno™ Avanti Platform - Delsys. https://www.delsys.com/trig 5. Noraxon USA, Ultium EMG, \| Noraxon USA. https://www.noraxon.com/our-products/ultium-emg/. Accessed 10 Feb 2021 6. Delp SL, Anderson FC, Arnold AS et al (2007) OpenSim: Open-source software to create and analyze dynamic simulations of movement. IEEE Trans Biomed Eng 54:1940–1950. https://doi.org/10.1109/TBME.2007.901024 7. Seth A, Hicks JL, Uchida TK et al (2018) OpenSim: Simulating musculoskeletal dynamics and neuromuscular control to study human and animal movement. PLoS Comput Biol 14:. https://doi.org/10.1371/journal.pcbi.1006223 8. References Bonnet V, Mazza C, Fraisse P, Cappozzo A (2013) Real-time estimate of body kinematics during a planar squat task using a single inertial measurement unit. IEEE Trans Biomed Eng 60:1920–1926. https://doi.org/10.1109/TBME.2013.2245131 8. Bonnet V, Mazza C, Fraisse P, Cappozzo A (2013) Real-time estimate of body kinematics during a planar squat task using a single inertial measurement unit. IEEE Trans Biomed Eng 60:1920–1926. https://doi.org/10.1109/TBME.2013.2245131 9. Van Den Bogert AJ, Geijtenbeek T, Even-Zohar O et al (2013) A real-time system for biomechanical analysis of human movement and muscle function. Med Biol Eng Comput 51:1069–1077. https://doi.org/10.1007/s11517- 013-1076-z 9. Van Den Bogert AJ, Geijtenbeek T, Even-Zohar O et al (2013) A real-time system for biomechanical analysis of human movement and muscle function. Med Biol Eng Comput 51:1069–1077. https://doi.org/10.1007/s11517- 013-1076-z Page 10/13 10. Borbély BJ, Szolgay P (2017) Real-time inverse kinematics for the upper limb: A model-based algorithm using segment orientations. Biomed Eng Online 16:. https://doi.org/10.1186/s12938-016-0291-x 11. Falisse A, Van Rossom S, Gijsbers J et al (2018) OpenSim versus human body model: A comparison study for the lower limbs during gait. J Appl Biomech 34:496–502. https://doi.org/10.1123/jab.2017-0156 12. Yi C, Jiang F, Bhuiyan MZA et al (2021) Smart healthcare-oriented online prediction of lower-limb kinematics and kinetics based on data-driven neural signal decoding. Futur Gener Comput Syst 114:96–105. https://doi.org/10.1016/j.future.2020.06.015 13. Pizzolato C, Reggiani M, Modenese L, Lloyd DG (2017) Real-time inverse kinematics and inverse dynamics for lower limb applications using OpenSim. Comput Methods Biomech Biomed Engin 20:436–445. https://doi.org/10.1080/10255842.2016.1240789 13. Pizzolato C, Reggiani M, Modenese L, Lloyd DG (2017) Real-time inverse kinematics and inverse dynamics for lower limb applications using OpenSim. Comput Methods Biomech Biomed Engin 20:436–445. https://doi.org/10.1080/10255842.2016.1240789 14. Stanev D, Filip K, Bitzas D et al (2021) Real-time musculoskeletal kinematics and dynamics analysis using marker- and imu-based solutions in rehabilitation. Sensors 21:1–20. https://doi.org/10.3390/s21051804 14. Stanev D, Filip K, Bitzas D et al (2021) Real-time musculoskeletal kinematics and dynamics analysis using marker- and imu-based solutions in rehabilitation. Sensors 21:1–20. https://doi.org/10.3390/s21051804 15. Yamaguchi GT, Zajac FE (1989) A planar model of the knee joint to characterize the knee extensor mechanism. J Biomech 22:1–10. https://doi.org/10.1016/0021-9290(89)90179-6 15. Yamaguchi GT, Zajac FE (1989) A planar model of the knee joint to characterize the knee extensor mechanism. J Biomech 22:1–10. https://doi.org/10.1016/0021-9290(89)90179-6 16. Delp SL, Loan JP, Hoy MG et al (1990) An Interactive Graphics-Based Model of the Lower Extremity to Study Orthopaedic Surgical Procedures. IEEE Trans Biomed Eng 37:757–767. https://doi.org/10.1109/10.102791 16. References Delp SL, Loan JP, Hoy MG et al (1990) An Interactive Graphics-Based Model of the Lower Extremity to Study Orthopaedic Surgical Procedures. IEEE Trans Biomed Eng 37:757–767. https://doi.org/10.1109/10.102791 17. Anderson FC, Pandy MG (1999) A dynamic optimization solution for vertical jumping in three dimensions. Comput Methods Biomech Biomed Engin 2:201–231. https://doi.org/10.1080/10255849908907988 17. Anderson FC, Pandy MG (1999) A dynamic optimization solution for vertical jumping in three dimensions. Comput Methods Biomech Biomed Engin 2:201–231. https://doi.org/10.1080/10255849908907988 18. Anderson FC, Pandy MG (2001) Dynamic optimization of human walking. J Biomech Eng 123:381–390. https://doi.org/10.1115/1.1392310 18. Anderson FC, Pandy MG (2001) Dynamic optimization of human walking. J Biomech Eng 123:381–390. https://doi.org/10.1115/1.1392310 19. Hamner SR, Seth A, Delp SL (2010) Muscle contributions to propulsion and support during running. J Biomech 43:2709–2716. https://doi.org/10.1016/j.jbiomech.2010.06.025 19. Hamner SR, Seth A, Delp SL (2010) Muscle contributions to propulsion and support during running. J Biomech 43:2709–2716. https://doi.org/10.1016/j.jbiomech.2010.06.025 20. Friesen KB, Zhang Z, Monaghan PG et al (2020) All eyes on you: how researcher presence changes the way you walk. Sci Rep 10:. https://doi.org/10.1038/s41598-020-73734-5 20. Friesen KB, Zhang Z, Monaghan PG et al (2020) All eyes on you: how researcher presence changes the way you walk. Sci Rep 10:. https://doi.org/10.1038/s41598-020-73734-5 21. Kannape OA, Blanke O (2013) Self in motion: Sensorimotor and cognitive mechanisms in gait agency. J Neurophysiol 110:1837–1847. https://doi.org/10.1152/jn.01042.2012 21. Kannape OA, Blanke O (2013) Self in motion: Sensorimotor and cognitive mechanisms in gait agency. J Neurophysiol 110:1837–1847. https://doi.org/10.1152/jn.01042.2012 22. Karatsidis A, Bellusci G, Schepers HM et al (2017) Estimation of ground reaction forces and moments during gait using only inertial motion capture. Sensors (Switzerland) 17:. https://doi.org/10.3390/s17010075 22. Karatsidis A, Bellusci G, Schepers HM et al (2017) Estimation of ground reaction forces and moments during gait using only inertial motion capture. Sensors (Switzerland) 17:. https://doi.org/10.3390/s17010075 23. Sartori M, Reggiani M, Lloyd DG, Pagello E (2011) A neuromusculoskeletal model of the human lower limb: Towards EMG-driven actuation of multiple joints in powered orthoses. IEEE Int Conf Rehabil Robot. https://doi.org/10.1109/ICORR.2011.5975441 23. Sartori M, Reggiani M, Lloyd DG, Pagello E (2011) A neuromusculoskeletal model of the human lower limb: Towards EMG-driven actuation of multiple joints in powered orthoses. IEEE Int Conf Rehabil Robot. https://doi.org/10.1109/ICORR.2011.5975441 Figures Figures Page 11/13 Figure 1 Figure 3 Mean absolute error (MAE) between real-time inverse kinematics and offline inverse kinematics ranges of motion (ROMs) of the exerted degrees of freedom of the musculoskeletal model and the 95\% confidence interval of the error. Solid bars show MAEs of ROMs for each exerted degree of freedom. Confidence intervals are shown as error bars centered on the top of the MAE bars. Figure 1 A diagram illustrating the working principle of the inverse kinematics (IK) workflow. Orientation data of inertial measurement units (IMUs) is read as quaternions by the producer thread and saved to a buffer. Time values are saved to another buffer. The consumer thread reads data from both buffers and initiates new threads that calculate IK based on the data. IK threads output joint angle values for the model, which can be sent to a visualizer window. When the program finishes, the IK output frames can be sorted in a time-ascending order and saved to file. Page 12/13 Page 12/13 Figure 2 Inverse kinematics (IK) throughput with respect to the number of IK threads used, measured on a desktop computer (left) and a laptop computer (right). The throughput is calculated as mean operations per second over a one-minute trial. The measurements were repeated with two different musculoskeletal models, and using one, seven or 12 inertial measurement units (IMUs). Identity quaternions were used as IMU orientations. Onlineresource1.pdf Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Onlineresource1.pdf Page 13/13 Page 13/13
https://openalex.org/W2192933481	https://europepmc.org/articles/pmc4749556?pdf=render	English	null	Human observers have optimal introspective access to perceptual processes even for visually masked stimuli	eLife	2,015	cc-by	16,185	Human observers have optimal introspective access to perceptual processes even for visually masked stimuli Megan A K Peters1, Hakwan Lau1,2 1Department of Psychology, University of California, Los Angeles, Los Angeles, United States; 2Brain Research Institute, University of California, Los Angeles, Los Angeles, United States Abstract Many believe that humans can ‘perceive unconsciously’ – that for weak stimuli, briefly presented and masked, above-chance discrimination is possible without awareness. Interestingly, an online survey reveals that most experts in the field recognize the lack of convincing evidence for this phenomenon, and yet they persist in this belief. Using a recently developed bias-free experimental procedure for measuring subjective introspection (confidence), we found no evidence for unconscious perception; participants’ behavior matched that of a Bayesian ideal observer, even though the stimuli were visually masked. This surprising finding suggests that the thresholds for subjective awareness and objective discrimination are effectively the same: if objective task performance is above chance, there is likely conscious experience. These findings shed new light on decades-old methodological issues regarding what it takes to consider a neurobiological or behavioral effect to be ’unconscious,’ and provide a platform for rigorously investigating unconscious perception in future studies. DOI 10 7554/ L f 09651 001 RESEARCH ARTICLE Human observers have optimal introspective access to perceptual processes even for visually masked stimuli Introduction These individuals could respond correctly to questions about the shape and location of objects in their visual field, even though they could no longer see the objects. This phenomenon became known as ’blindsight’, and it is regarded as a classic example of perception in the absence of conscious awareness. eLife digest In the 1980s, psychologists made an unexpected discovery while working with individuals who had become blind after sustaining damage to areas of the brain required for vision. These individuals could respond correctly to questions about the shape and location of objects in their visual field, even though they could no longer see the objects. This phenomenon became known as ’blindsight’, and it is regarded as a classic example of perception in the absence of conscious awareness. Many researchers who study consciousness believe that everyone is capable of subliminal or unconscious perception: that is, of detecting and processing stimuli without being consciously aware of them. However, studies investigating this phenomenon have produced contradictory results. Peters and Lau have now tested unconscious perception directly, using a recently developed method that overcomes some of the problems faced by previous studies. Human volunteers took part in several trials, in which they were shown two images. Each image was ‘masked’ to prevent the volunteers from consciously registering them. After each image was shown, the volunteers had to state whether a patch of gray and white stripes in the masked image was tilted to the left or to the right. However, one of the two images did not include a gray and white patch. After seeing both images in a trial, the volunteers also had to indicate which of their answers they were most confident about. If the volunteers could perceive the patches without being consciously aware of doing so, their response should show two features. The volunteers should correctly state the tilt direction of the stripes more often than would be expected if they were guessing at random. However, they should also feel no more confident in their responses for the images that did feature a striped patch than for the ‘no patch’ ones. Peters and Lau found no such evidence of unconscious perception. Nevertheless, the volunteers were consistently better at correctly stating the direction the stripes were tilted in than their confidence ratings would suggest. Does this indicate some degree of perception without awareness? Introduction For correspondence: meganakpeters@ucla.edu Above-chance performance without awareness in perceptual discrimination tasks is a strong form of unconscious perception. In these demonstrations (e.g., blindsight: Weiskrantz, 1986) the subjective threshold for awareness (when a stimulus is consciously ‘seen’) seems well above the objective threshold for forced-choice discrimination (when a stimulus can be correctly identified): subjects can discriminate a target above chance performance, yet report no awareness of the target. Many researchers believe normal, healthy subjects can also directly discriminate near-threshold, low-inten- sity targets without subjective awareness (e.g., Boyer et al., 2005; Charles et al., 2013; Merikle et al., 2001; but see Snodgrass et al., 2004 for an opposing view). Competing interests: The authors declare that no competing interests exist. Received: 24 June 2015 Accepted: 02 October 2015 Published: 03 October 2015 We conducted an informal survey to confirm this popular belief, which also revealed that many believe convincing evidence for this phenomenon is lacking. We asked survey participants three key questions: (1) "Do you believe in subliminal perception?" (2) "Do you believe that the subjective threshold for awareness is above the objective discrimination threshold?" and (3) "If ‘yes’, do you believe this has been convincingly demonstrated in the literature?" Most respondents reported believing that subliminal processing exists (94%), but also that they did not believe it had been con- vincingly demonstrated in the literature (64%). These belief patterns were shown even among those who reported having published on subliminal or unconscious perception (94% and 61%, respec- tively). See Appendix 1 for full text of questions and detailed survey results. Reviewing editor: Matteo Carandini, University College London, United Kingdom Copyright Peters and Lau. This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited. A primary culprit in this controversy is the problem of criterion bias: an observer’s report of ‘unseen’ doesn’t necessarily imply complete lack of awareness, only that the stimulus’ strength fell below some arbitrary boundary for reporting ‘seen’ (Eriksen, 1960; Hannula et al., 2005; Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 1 of 30 Research article Research article Research article Neuroscience eLife digest In the 1980s, psychologists made an unexpected discovery while working with individuals who had become blind after sustaining damage to areas of the brain required for vision. Introduction DOI: 10.7554/eLife.09651 2 of 30 Research article Research article Research article Research article Neuroscience they felt more confident in. This approach has several advantages. First, 2IFC tasks depend little on response bias compared to multi-point confidence-rating scales. Maintaining the criteria for exten- Figure 1. Stimuli and procedures for the 2IFC confidence-rating task. (A) Targets consisted of oriented (45˚ left- or right-tilted from vertical) Gabor patches presented at multiple near-threshold contrast levels; masks consisted of bandpass-noise filtered random RGB values (see Materials and methods). (B) Each trial consists of two intervals of discrimination in which the target stimulus (T) was forward- and backward-masked (M). Gabor patch targets were presented only in target-present (TP) intervals; in target-absent (TA) intervals, the target was replaced with blank frames. Otherwise timings of stimuli were matched between the two intervals. (C) Experimental tasks. Experiment 1 required subjects to bet on which discrimination they felt more confident before they indicated their orientation discrimination choices (left or right tilt of the Gabor) sequentially for both intervals. Shown is an example trial in which TP is presented before TA; in the experiment this order varied randomly from trial to trial. In Experiment 2, subjects bet on the more confident interval after the discriminations, and feedback was given. (See Materials and methods for more details.) DOI: 10.7554/eLife.09651.003 Figure 1. Stimuli and procedures for the 2IFC confidence-rating task. (A) Targets consisted of oriented (45˚ left- or right-tilted from vertical) Gabor patches presented at multiple near-threshold contrast levels; masks consisted of bandpass-noise filtered random RGB values (see Materials and methods). (B) Each trial consists of two intervals of discrimination in which the target stimulus (T) was forward- and backward-masked (M). Gabor patch targets were presented only in target-present (TP) intervals; in target-absent (TA) intervals, the target was replaced with blank frames. Otherwise timings of stimuli were matched between the two intervals. (C) Experimental tasks. Experiment 1 required subjects to bet on which discrimination they felt more confident before they indicated their orientation discrimination choices (left or right tilt of the Gabor) sequentially for both intervals. Shown is an example trial in which TP is presented before TA; in the experiment this order varied randomly from trial to trial. In Experiment 2, subjects bet on the more confident interval after the discriminations, and feedback was given. (See Materials and methods for more details.) they felt more confident in. This approach has several advantages. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Introduction Peters and Lau argue that it does not, because a computer model designed to perform the task showed a similar level of performance to the volunteers. These findings suggest that previous reports of unconscious perception may have been contaminated by the problems that Peters and Lau controlled for, and that perhaps unconscious perception doesn’t occur in people without brain damage. Researchers will now need to do more studies using similar approaches to determine whether observers without brain damage can truly experience unconscious perception, and how such unconscious perception might be represented in the brain. Lloyd et al., 2013; Merikle et al., 2001). Unfortunately, most methods of studying unconscious per- ception suffer from this ‘criterion problem’ (e.g., Charles et al., 2013; Jachs et al., 2015; Ramsøy and Overgaard, 2004). With such methods, one could argue that reports of ‘unawareness’ may only mean some stimuli are relatively hard to perceive compared to those that are clearly visible. To avoid this criterion problem, several groups (Kolb and Braun, 1995; Kunimoto et al., 2001) sought to identify conditions in which confidence was uncorrelated with accuracy, which they argued would indicate no subjective awareness of the target. Unfortunately, some of these efforts were not replicable (Morgan et al., 1997; Robichaud and Stelmach, 2003). Others revealed that estimating the correspondence between confidence and accuracy requires mathematical considerations more complicated than originally envisaged (Evans and Azzopardi, 2007; Galvin et al., 2003; Maniscalco and Lau, 2012). Importantly, the conceptual link between metacognitive sensitivity (i.e., correlation between confidence and accuracy) and conscious awareness is itself controversial (Charles et al., 2013; Fleming and Lau, 2014; Jachs et al., 2015). Here, we employ a recently-developed confidence-rating method to address this problem (Barthelme´ et al., 2009; de Gardelle and Mamassian, 2014). Subjects discriminated two stimulus intervals, only one of which contained a target, and indicated confidence in their decisions using a 2- interval forced-choice procedure (2IFC), that is, indicating which of the two discrimination decisions Here, we employ a recently-developed confidence-rating method to address this problem (Barthelme´ et al., 2009; de Gardelle and Mamassian, 2014). Subjects discriminated two stimulus intervals, only one of which contained a target, and indicated confidence in their decisions using a 2- interval forced-choice procedure (2IFC), that is, indicating which of the two discrimination decisions Peters and Lau. eLife 2015;4:e09651. Introduction First, 2IFC tasks depend little on response bias compared to multi-point confidence-rating scales. Maintaining the criteria for exten- sive confidence scales may also be demanding, leading subjects to respond somewhat randomly in conditions of vague awareness and thereby producing the negative result Kolb and Braun (1995) observed (Morgan et al., 1997). Second, the interpretation of 2IFC confidence-rating in this context is straightforward: ‘Performance without Awareness’ would mean subjects can perform the target discrimination yet fail to place bets appropriately to distinguish this performance from discrimination of a blank stimulus (which guarantees chance performance). That is, following psychophysics tradi- tions (Kolb and Braun, 1995; Peirce and Jastrow, 1884), if a certain above-chance discrimination seems introspectively no different from a random guess based on no stimulus at all (as reflected by betting behavior), we interpret the discrimination to be unconscious. Here, we explored whether such Performance without Awareness occurs in normal observers in two behavioral experiments, and compared these results to predictions of a Bayesian ideal observer. they felt more confident in. This approach has several advantages. First, 2IFC tasks depend little on response bias compared to multi-point confidence-rating scales. Maintaining the criteria for exten- sive confidence scales may also be demanding, leading subjects to respond somewhat randomly in conditions of vague awareness and thereby producing the negative result Kolb and Braun (1995) observed (Morgan et al., 1997). Second, the interpretation of 2IFC confidence-rating in this context is straightforward: ‘Performance without Awareness’ would mean subjects can perform the target discrimination yet fail to place bets appropriately to distinguish this performance from discrimination of a blank stimulus (which guarantees chance performance). That is, following psychophysics tradi- tions (Kolb and Braun, 1995; Peirce and Jastrow, 1884), if a certain above-chance discrimination seems introspectively no different from a random guess based on no stimulus at all (as reflected by betting behavior), we interpret the discrimination to be unconscious. Here, we explored whether such Performance without Awareness occurs in normal observers in two behavioral experiments, and compared these results to predictions of a Bayesian ideal observer. 3 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Research article Research article Neuroscience Figure 2. Schematic explanation of predictions of the experiments. (A) A ‘Performance without Awareness’ pattern of behavior, in which subjects are able to discriminate the target above chance while betting on the target-present interval at chance. Introduction (B) A ‘Performance > Awareness’ pattern of behavior, in which subjects are less able to bet on their discrimination decisions than they are able to correctly discriminate the target. In both (A) and (B), the diagonal dashed line indicates where rate of betting on the target-present interval equals objective discrimination performance. DOI: 10.7554/eLife.09651.004 Figure 2. Schematic explanation of predictions of the experiments. (A) A ‘Performance without Awareness’ pattern of behavior, in which subjects are able to discriminate the target above chance while betting on the target-present interval at chance. (B) A ‘Performance > Awareness’ pattern of behavior, in which subjects are less able to bet on their discrimination decisions than they are able to correctly discriminate the target. In both (A) and (B), the diagonal dashed line indicates where rate of betting on the target-present interval equals objective discrimination performance. DOI: 10 7554/eLife 09651 004 DOI: 10.7554/eLife.09651.004 Behavioral experiments However, the ideal observer mode such behavior (panel G), indicating that it is not suboptimal at all but arises from the 2IFC nature of the confidence task (see Baye Model results section and Figure 2 caption for explanation). Horizontal gray lines in panels A, D, and G indicate chance-level bet target-present (TP) interval. Panels B, E, and H show rising Type 2 hit rate (‘HR’; when subjects bet on a correct orientation discrim relatively flat Type 2 false alarm rate (‘FAR’; when subjects bet on an incorrect orientation discrimination choice), and panels C, F, orientation discrimination accuracy when the target-present (TP) interval is bet on; these patterns suggest that human subjects an observer were rating confidence via assessing their probability of correctly discriminating orientation, rather than target presence The model demonstrates good explanatory power for the data across all participants (mean proportion of variance accounted for 0.565). Error bars for behavioral data indicate the standard error of the mean across subjects with data in each bin. DOI: 10.7554/eLife.09651.005 Figure 3. Group-level results of behavioral experiments (rows 1 and 2), presented in comparison to the predictions of the Bayesian ideal observer model (row 3; see Materials and methods - Computational Model). In both experiments, human observers displayed no evidence of Performance without Awareness, but appeared to demonstrate Performance > Awareness (panels A and D). However, the ideal observer model also demonstrated such behavior (panel G), indicating that it is not suboptimal at all but arises from the 2IFC nature of the confidence task (see Bayesian Ideal Observer Model results section and Figure 2 caption for explanation). Horizontal gray lines in panels A, D, and G indicate chance-level betting (50%) on the target-present (TP) interval. Panels B, E, and H show rising Type 2 hit rate (‘HR’; when subjects bet on a correct orientation discrimination choice) but relatively flat Type 2 false alarm rate (‘FAR’; when subjects bet on an incorrect orientation discrimination choice), and panels C, F, and I show higher orientation discrimination accuracy when the target-present (TP) interval is bet on; these patterns suggest that human subjects and the Bayesian ideal observer were rating confidence via assessing their probability of correctly discriminating orientation, rather than target presence versus absence only. The model demonstrates good explanatory power for the data across all participants (mean proportion of variance accounted for by the model, R2 = 0.565). Behavioral experiments Nine human observers participated in two experiments of our 2IFC confidence-rating paradigm (Fig- ure 1). In both experiments, participants viewed two intervals in which they were required to dis- criminate the orientation (right or left tilt) of a Gabor patch target embedded in forward- and backward-masks (Figure 1A,B), and judged which of the discrimination choices they felt more confi- dent in. Crucially, in one of the intervals the target was absent (Figure 1B), such that above-chance discrimination performance was impossible. We performed two experiments to assess the potential contributions of question order, receipt of feedback, and a priori knowledge of the presence of a target-absent interval (Figure 1C). In Experiment 1, participants judged which decision they felt more confident in and then indicated their orientation decisions for both intervals, while in Experi- ment 2 they indicated their orientation discrimination decisions before selecting the more-confident interval. In Experiment 2, we also provided feedback on the confidence decision, and told partici- pants that one interval contained no target; this information was withheld from participants in Exper- iment 1. Stimuli, timing details, and order of question prompts in the two experiments are also discussed in greater detail in the Methods section. For both experiments, we evaluated whether participants exhibited Performance without Aware- ness (Figure 2A) or Performance > Awareness (Figure 2B). In both cases, the response pattern of interest can be visualized as percent of time betting on the target-present interval as a function of percent correct orientation discrimination in the target-present interval. ‘Performance without Awareness’ (Figure 2A) would be supported if observers can discriminate the target above chance (>50% accuracy) while being unable to bet on their choices more often than betting on the target- absent interval (which necessarily yields chance-level performance). That is, observers correctly dis- criminate the target’s orientation more than 50% of the time, but bet on the target-present interval 50% of the time (i.e., they bet randomly on the target-present versus target-absent interval), Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 4 of 30 Research article Research article R Neuroscience Figure 3. Group-level results of behavioral experiments (rows 1 and 2), presented in comparison to the predictions of the Bayesia model (row 3; see Materials and methods - Computational Model). In both experiments, human observers displayed no evidence without Awareness, but appeared to demonstrate Performance > Awareness (panels A and D). Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Behavioral experiments We call this possibility ‘Performance > Awareness’. threshold for confidence might be above the objective threshold for discrimination. In other words, observers may bet on the target-present interval less often than they get the discrimination correct, but still above chance. This would occur because the orientation discrimination choice requires eval- uation of only one interval (the one with the target in it) and therefore is subject to only one source of uncertainty, but the ‘betting’ choice requires evaluation of both intervals, and therefore has two potential sources of uncertainty. This pattern of behavior (Figure 2B) may occur even if subjects do not display Performance without Awareness, and would be characterized by a pattern of responses that fall below the identity line (diagonal dashed line). We call this possibility ‘Performance > Awareness’. We discuss the results of both experiments together for ease of interpretation, and because the results are very similar (Figure 3A–F). To anticipate, we found no evidence of Performance without Awareness. Although we found strong evidence of Performance > Awareness across the experi- ments (Figure 3A,D), subsequent computational modeling (Bayesian Ideal Observer Model section) suggests that this is somewhat trivial: even an ideal observer is expected to show Performance > Awareness (Figure 3G; see Bayesian Ideal Observer Model section for further explanation). To look for evidence of Performance without Awareness, we first plotted percent of trials in which observers bet on the target-present interval against orientation discrimination accuracy for both experiments (Figure 3A,D). In contrast to what might have been suggested based on previous results (e.g., Boyer et al., 2005; Charles et al., 2013; Merikle et al., 2001; but see Snodgrass et al., 2004), visual inspection alone clearly reveals no evidence for Performance without Awareness in either experiment: it looks as though observers could bet on the target-present inter- val above chance as soon as they were able to discriminate the target above chance, and there is no hint of the Performance without Awareness pattern. We quantitatively assessed the possibility of Performance without Awareness using a Bayesian observer model (see Modeling Results, below), but found no evidence that a Performance without Awareness pattern could capture human behav- ior. Individual subjects’ performance closely resembles group data and averages (Appendix 2). Because thresholds can be defined in psychophysical terms (75% performance) rather than abso- lute terms (>50%), we also evaluated the possibility of Performance > Awareness. Behavioral experiments Error bars for behavioral data indicate the standard error of the mean across subjects with data in each bin. DOI: 10.7554/eLife.09651.005 indicating they are not aware of the information that contributed to their discrimination decision. If this were to occur, it would most likely happen at low discrimination performance levels, yielding a pattern of behavior similar to that presented in Figure 2A. However, in psychophysics, thresholds can also be defined as midway between ceiling and floor performance (Macmillan and Creelman, 2004), such that threshold discrimination performance is defined as 75% accuracy rather than >50% (chance level). This concept can also be applied to sub- jective betting data in the sense that betting on the target-present interval could be considered ‘cor- rect’ or ‘advantageous’ betting. In this sense (threshold = 75% correct performance), the subjective 5 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience Table 1. Individual values, means, standard deviations, and p-values for t-tests showing that Performance > Awareness occurs across both experiments. Results from Experiment 2 show that the pattern does not change with different question order or feedback. Table 1. Individual values, means, standard deviations, and p-values for t-tests showing that Performance > Awareness occurs across both experiments. Results from Experiment 2 show that the pattern does not change with different question order or feedback. Expt Subject p(choose TP interval) at p(correct) = 0.75 1 1 AVT 0.676 2 AM 0.714 3 JDK 0.716 4 SH 0.682 5 MM 0.684 6 AC 0.685 7 MR 0.674 8 MK 0.658 9 RA 0.619 2 1 AVT 0.666 2 AM 0.713 3 JDK 0.746 Mean (s) 0.686 (0.033) t(11) 6.718 p 0.00003 DOI: 10.7554/eLife.09651.006 threshold for confidence might be above the objective threshold for discrimination. In other words, observers may bet on the target-present interval less often than they get the discrimination correct, but still above chance. This would occur because the orientation discrimination choice requires eval- uation of only one interval (the one with the target in it) and therefore is subject to only one source of uncertainty, but the ‘betting’ choice requires evaluation of both intervals, and therefore has two potential sources of uncertainty. This pattern of behavior (Figure 2B) may occur even if subjects do not display Performance without Awareness, and would be characterized by a pattern of responses that fall below the identity line (diagonal dashed line). 2IFC detection? 2IFC detection? One possible concern is that subjects were not rating confidence but instead engaging in 2IFC detection of the target-present interval. To confirm that subjects were indeed rating confidence, we plotted Type 2 hit rate and Type 2 false alarm rate against orientation discrimination accuracy (Figure 3B,E). A Type 2 hit is defined as placing a bet on a correct orientation discrimination deci- sion, whereas a Type 2 false alarm is defined as placing a bet on an incorrect orientation discrimina- tion decision. These are in contrast to Type 1 hits and false alarms, which can be defined as saying ‘left’ when a left-tilted Gabor was presented and saying ‘left’ when a right-tilted Gabor was pre- sented, respectively, according to standard signal detection theoretic definitions (Green and Swets, 1966; Macmillan and Creelman, 2004). Subjects displayed increasing Type 2 hit rate as a function of orientation discrimination accuracy, whereas Type 2 false alarm rate remained relatively flat at around 50% (chance level) across increas- ing orientation discrimination accuracy. In other words, subjects did not bet on orientation discrimi- nation choices they expected to get wrong, even at high performance (i.e. high contrast) levels. Thus, they were probably truly rating confidence and not simply engaging in 2IFC detection. In keeping with this observation, we also plotted orientation discrimination accuracy conditional upon subjects’ selection of the target-present interval, i.e. p(correctorientationDiscrimination \| target-present selected) and p(correctorientationDiscrimination \| target-present not selected) (Figure 3C,F). This visualiza- tion revealed that subjects were worse at orientation discrimination when they did not select the tar- get-present interval. This result is in keeping with typical observations of worse objective performance for low confidence trials, since not betting on the target-present interval is essentially an indication of low confidence in that discrimination choice. See also the ’Unconscious ‘hunches’?’ section, below. Notably, the similarity in participants’ behavior between Experiments 1 and 2 reveals that receipt of feedback on confidence judgments, knowledge that one interval is physically blank, question order, and ability to monitor reaction time do not affect behavioral outcomes. Unconscious ‘hunches’? Throughout this report, we define conscious awareness of the target to occur when introspective assessment of the correctness of an orientation discrimination choice can differentiate between a tar- get being present or not. Behavioral experiments We used kernel smoothing regression (see Materials and methods) to interpolate each individual subject’s data in order to estimate how often subjects bet on the target-present interval when they were performing at 75% correct on orientation discrimination. Because results are very similar across the two Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 6 of 30 Research article Research article Neuroscience experiments, we combined results from both and performed a two-tailed one-sample t-test to assess whether this predicted percentage betting on the target-present interval significantly diverged from 75%. This analysis revealed that observers bet on the target-present interval significantly less than 75% of the time at 75% correct orientation discrimination accuracy (Figure 3A,D, Table 1). Thus, observers exhibited Performance > Awareness (but see also Modeling Results, below). 2IFC detection? Thus, even when the 2IFC task was visibility judgment rather than confidence, subjects’ behavior was inconsistent with the Performance without Awareness pattern – suggesting there is also no Performance without Visual Awareness. See Appendix 3 for details of the control study. 2IFC detection? On each simulated trial, the model ‘sees’ two samples, one drawn from a source distribution Si to represent the target-present interval (dTP) and the other from the noise distribution to represent the target-absent interval (dTA). It marginalizes across all contrast evidence levels to guess the orientations of both samples according to the posterior probabilities of left- and right-tilted sources. Then, it compares the posterior probabilities of the chosen orientations in each interval to select the interval with higher confidence (p(correct)) (see Materials and methods - Bayesian ideal observer model). DOI: 10.7554/eLife.09651.007 Neuroscience Figure 4. Illustration of the Bayesian ideal observer’s 2-dimensional representation space, following standard 2- dimensional signal detection theory (King and Dehaene, 2014; Macmillan and Creelman, 2004). (a) Distributions Sleft and Sright lie on orthogonal axes cleft and cright representing left- and right-tilted targets, respectively, and the noise distribution lies at the origin. On each simulated trial, the model ‘sees’ two samples, one drawn from a source distribution Si to represent the target-present interval (dTP) and the other from the noise distribution to represent the target-absent interval (dTA). It marginalizes across all contrast evidence levels to guess the orientations of both samples according to the posterior probabilities of left- and right-tilted sources. Then, it compares the posterior probabilities of the chosen orientations in each interval to select the interval with higher confidence (p(correct)) (see Materials and methods - Bayesian ideal observer model). DOI: 10 7554/eLife 09651 007 Figure 4. Illustration of the Bayesian ideal observer’s 2-dimensional representation space, following standard 2- dimensional signal detection theory (King and Dehaene, 2014; Macmillan and Creelman, 2004). (a) Distributions Sleft and Sright lie on orthogonal axes cleft and cright representing left- and right-tilted targets, respectively, and the noise distribution lies at the origin. On each simulated trial, the model ‘sees’ two samples, one drawn from a source distribution Si to represent the target-present interval (dTP) and the other from the noise distribution to represent the target-absent interval (dTA). It marginalizes across all contrast evidence levels to guess the orientations of both samples according to the posterior probabilities of left- and right-tilted sources. Then, it compares the posterior probabilities of the chosen orientations in each interval to select the interval with higher confidence (p(correct)) (see Materials and methods - Bayesian ideal observer model). DOI: 10.7554/eLife.09651.007 chance. 2IFC detection? In this sense, observers are unconscious of the information contributing to their decision if they can discriminate a target above chance, but doing so feels no different intro- spectively from discriminating (or guessing about) nothing at all. However, one concern might be that subjects are able to meaningfully rate confidence despite no subjective visual experience of the stimulus due to some sort of non-visual ‘hunch’ or ‘feeling’. Indeed, such metacognitive insights (the ability to introspectively distinguish between correct and incorrect responses) have recently been reported even in the absence of objective task performance sensitivity, although not in the context of perception (e.g., Scott et al., 2014). We think this issue is essentially one of terminology; our definition of conscious awareness follows a long history in psychology and psychophysics traditions in relating the ability to meaningfully rate confidence to subjective awareness (c.f. Kolb and Braun, 1995; Peirce and Jastrow, 1884), accord- ing to which, strictly speaking, a non-visual hunch is also defined as conscious so long as it meaning- fully tracks visual processes; regardless of whether such ‘hunches’ are visual in nature, it is still meaningful to distinguish between having such introspective insight versus having no insight whatso- ever. However, we also ran a control study in which the subjective task was to indicate which interval appeared more visible rather than confidence in the corresponding discrimination. In other words, it was akin to a 2IFC detection task rather than a metacognitive judgment. Results of this control study (Appendix 3) mirrored those of the main experiments: as soon as participants were able to discrimi- nate the target above chance, they were able to indicate which interval contained the target above Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 7 of 30 Research article Research article Neuroscience chance. Thus, even when the 2IFC task was visibility judgment rather than confidence, subjects’ behavior was inconsistent with the Performance without Awareness pattern – suggesting there is Figure 4. Illustration of the Bayesian ideal observer’s 2-dimensional representation space, following standard 2- dimensional signal detection theory (King and Dehaene, 2014; Macmillan and Creelman, 2004). (a) Distributions Sleft and Sright lie on orthogonal axes cleft and cright representing left- and right-tilted targets, respectively, and the noise distribution lies at the origin. Bayesian ideal observer model We developed a Bayesian ideal observer model utilizing a similar representation space as standard 2-dimensional signal detection theory (Figure 4) (King and Dehaene, 2014; Macmillan and Creel- man, 2004). The primary finding is that even an ideal observer model exhibits Performance > Aware- ness, as depicted in Figure 1B. Intuitively, this effect occurs because the orientation discrimination choice requires evaluation of only one interval (the one with the target in it) and therefore is cor- rupted by only one source of noise, but the ‘betting” choice requires evaluation of both intervals, and therefore has two potential sources of noise. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 8 of 30 Research article Table 2. R2 values quantifying goodness of fit for ideal observer (sd = 0) and three alternative decisional noise magnitudes (sd >0) which cause increasing degrees of Performance without Awareness. Decisional noise greater than 0 – i.e., increased level of Performance without Awareness – causes a drop in goodness of fit between model and human data. See Methods and Appendix 4 for more details. ble 2. R2 values quantifying goodness of fit for ideal observer (sd = 0) and three alternative cisional noise magnitudes (sd >0) which cause increasing degrees of Performance without Table 2. R2 values quantifying goodness of fit for ideal observer (sd = 0) and three alternative decisional noise magnitudes (sd >0) which cause increasing degrees of Performance without Expt Subject Decisional noise sd 0 (Ideal observer) 0.1 0.2 0.3 1 1 0.465 0.459 0.456 0.447 2 0.580 0.578 0.565 0.544 3 0.470 0.464 0.448 0.428 4 0.396 0.392 0.381 0.363 5 0.649 0.655 0.645 0.628 6 0.480 0.473 0.458 0.434 7 0.453 0.452 0.444 0.427 8 0.602 0.595 0.583 0.563 9 0.503 0.509 0.512 0.504 2 1 0.624 0.624 0.622 0.612 2 0.783 0.780 0.775 0.766 3 0.777 0.778 0.767 0.753 Mean R2 (s) 0.565 (0.126) 0.563 (0.128) 0.555 (0.129) 0.539 (0.131) DOI: 10.7554/eLife.09651.008 The model ‘performs’ a 2IFC confidence discrimination by comparing the posterior probability of left- or right-tilted source distributions given the data to perform the orientation discrimination task on each of the two intervals on each trial. Bayesian ideal observer model Then, it uses the posterior probability of the choice it made on each interval as a measure of confidence (i.e., p(correct)), and compares this measure between the two intervals to select the choice it is more confident in (see Figure 4 and Materials and methods – Bayesian ideal observer model). We also explored several model variants to establish the robustness of the model’s performance; see Appendix 4 for details on model variants. Unsurprisingly, the Bayesian ideal observer did not display signs of Performance without Aware- ness. We next evaluated whether causing the model to exhibit Performance without Awareness (Figure 2A) by degrading the 2IFC confidence judgment could produce better fit to participants’ data. We tested three levels of increasing decisional noise (sd; see Materials and methods) to cause the model to exhibit increasing Performance without Awareness as described in Figure 2A, and assessed the goodness of fit (R2) for each subject for each decisional noise value. We found that causing the model to exhibit increasing Performance without Awareness behavior resulted in increas- ingly worse R2 values (Table 2). To confirm this trend, we conducted a 12 (subjects; subjects 1–3 who completed both experiments are treated independently) x 4 (decisional noise magnitude) repeated measures ANOVA on the R2 values. This analysis revealed a main effect of decisional noise (F(3,33) = 19.301, p <0.001), indicating that the ideal observer model (sd = 0) best captures human performance, and that any suboptimal Performance without Awareness (sd >0) pattern fits human data more poorly than the ideal observer behavior – even without punishing the decisional noise model for having an additional parameter. Crucially, however, the ideal observer does exhibit Performance > Awareness (Figure 3G), and to a similar extent as our human participants (R2 = 0.565; see Appendix 4 for details of goodness of fit metrics); trends for Type 2 hit and false alarm rates (Figure 3H), and percent correct conditional upon having bet on the target-present versus target-absent interval (Figure 3I), also match human data. That the ideal observer exhibits behavior that may seem suboptimal, and in the same pattern as human observers, confirms that this perhaps counterintuitive but optimal behavior arises from the confidence-comparison nature of the 2IFC confidence-rating task: the decision about orientation in Peters and Lau. eLife 2015;4:e09651. Discussion Blindsight (Weiskrantz, 1986) is the intriguing demonstration of Performance without Awareness in neurological patients. Despite widely held beliefs by experts, here we found no evidence that it occurs in normal observers. Importantly, although the measured psychophysical threshold (75%) for awareness seemed to be above the objective discrimination threshold, computational analysis revealed that the actual effective thresholds are essentially the same; people’s subjective ratings are close to ideal, given their objective performance levels. This challenges longstanding beliefs regard- ing the nature of subjective versus objective thresholds in perceptual studies (Merikle et al., 2001; see survey results in Appendix 1). Our findings cannot rule out all forms of unconscious perception, such as subliminal priming, in which the evidence for unconscious processing is typically indirect benefits in reaction times (Hannula et al., 2005). However, our findings bear upon those studies, too. Traditionally, interpret- ing such effects as unconscious required that the relevant stimuli yield zero sensitivity in a direct task (d’ = 0). Recently, many have relaxed this requirement and considered subjectively reported lack of awareness as sufficient (Pessiglione et al., 2009; Soto et al., 2011), presumably because we (wrongly) believed that certain stimuli might surpass the objective threshold while still being below the subjective one. One may also argue that while objective threshold requirements are rigorous, the valid and meaningful measure is the subjective threshold (Charles et al., 2013; Merikle et al., 2001). Our results suggest this reasoning is flawed. If a stimulus surpasses the objective threshold, there is likely conscious experience; subjects likely report lack of awareness because they interpret the response options in relative terms in the context of stimuli of various strengths. This undermines claims that higher-cognitive phenomena – e.g. working memory, error detection, or motivation – can really operate unconsciously, if assessed with reference to subjective rather than objective thresholds (Charles et al., 2013; Pessiglione et al., 2009; Soto et al., 2011). Although the 2IFC confidence-rating procedure bypasses the response bias problem, interpreting the subjective vs. objective function is non-trivial: to determine whether participants’ Performance > Awareness behavior was optimal required detailed computational analysis. An alternative approach, which may be simpler, would be to compare the objective and subjective functions between task conditions, in a rationale similar to Lau and Passingham (2006). Although we found no evidence of ‘blindsight’ in normal observers, our study lays out the logic of what would be required to demonstrate it unequivocally. Bayesian ideal observer model DOI: 10.7554/eLife.09651 9 of 30 Research article Research article Neuroscience the target-present interval is limited by one source of noise (the single target-present interval), but the comparison of confidence is limited by the system’s noise in both intervals. So even if confidence monotonically increases with accuracy for the target-present interval, there will be trials in which – by chance – the discrimination choice for the blank (target-absent) interval happens to seem more confident, that is, its posterior probability is larger. This will happen sometimes even on trials in which the observer gets the target-present orientation discrimination correct. In these trials, the observer (human or simulated) will select the target-absent interval. This process will lead to the appearance of what we called Performance > Awareness, as displayed by our human participants and ideal observer (refer also to Figure 2 for additional explanation). Thus, the subjective ratings by human participants are already close to ideal, as if the actual effective threshold for subjective awareness is no different from the objective threshold for discrimination. Importantly, this is true despite the apparent measured differences in psychophysically defined thresholds (75%). Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Procedure – Experiment 1 Nine subjects participated in Experiment 1. Subjects were seated with their chins in a chinrest at a viewing distance of 42 cm from the screen. Targets and masks (Figure 1A) were presented for two to three frames (33–40 ms) each (jittered timing, with equal probability for two or three frames), with 33-40 ms ISI between masks and 0ms ISI for target-mask or mask-target transitions, in a forward- and backward-masking paradigm in which three masks were presented before and three after the target presentation (i.e., the target was ‘sandwiched’ between mask presentations) (Figure 1B). The trial structure extends the two-by-two forced-choice (2x2FC) paradigm first introduced by Nachmias and Weber (1975) and subsequently employed to explore the relationship between detection and identification (e.g., Thomas et al., 1982; Watson and Robson, 1981), and more recently applied to research on confidence (Barthelme´ et al., 2009, 2010; de Gardelle and Mamas- sian, 2014). We combined these procedure types. In our procedures, each trial consists of two time intervals, within only one of which the target is presented. In target-absent intervals, the target pre- sentation was replaced with blank frames, similar to the blank frames between masks, to maximize phenomenological similarity between target-present (TP) and target-absent intervals (TA) (Figure 1B). Unlike previous usage of the 2x2FC, however, we required observers to indicate target orientation on both target-present and target-absent intervals within a trial in addition to the final judgment type, despite the fact that there was a target in only one of the intervals. In target-present intervals, targets were presented at 45˚ tilted right or left from vertical at one of the possible contrasts. Following presentation of both intervals, observers pressed a key indicating which discrimination decision they would like to bet on (a measure of confidence; Type 2 judgment), and then indicated their discrimination choices for both intervals in order (leftward or rightward tilt; Type 1 judgment) (Figure 1C). In target-absent intervals, participants’ answers were coded as ‘cor- rect’ with 50% probability. No feedback was provided on a trial-by-trial basis. To motivate subjects, we informed them that a target was present in both intervals, but that one might be harder to dis- criminate than the other. Stimuli and apparatus Targets consisted of Gabor patches (sinusoidal gratings) at a spatial frequency of 0.025 cycles/pixel, tilted by 45˚ to the right or the left of vertical. Gratings and subtended 500 pixels, or ~111 visual degrees, and were presented in a circular annulus with a Gaussian hull spatial constant of 100. On each trial, targets could take on one of thirteen possible contrast levels drawn from the range 15– 90%. Masks consisted of white noise patches of random RGB values bandpass-filtered to a range of spatial frequencies immediately surrounding the spatial frequency of the target. They were pre- sented in a circular annulus of identical size to the spatial envelope of the Gabor patch targets. All stimuli were displayed via a custom Matlab R2013a (Natuck, MA) script utilizing PsychToolbox 3.0.12 on a gamma-corrected Dell E773c CRT monitor with a refresh rate of 75 Hz. Behavioral experiments Subjects Twelve subjects (two women, ages 19–32, ten right-handed) gave written informed consent to par- ticipate in our behavioral experiments. All subjects had normal or corrected-to-normal eyesight, and wore the same corrective lenses for all sessions, if applicable. Behavioral experiments were con- ducted in accordance with the Declaration of Helsinki and were approved by the UCLA Institutional Review Board. Stimuli and apparatus Discussion For example, it has recently been argued that TMS-induced ‘blindsight’ (Boyer et al., 2005) is contaminated by criterion bias (Lloyd et al., 2013). 2IFC confidence-rating may help resolve such issues without invoking theoretically compli- cated problems concerning signal detection theory (e.g., Heeks and Azzopardi, 2015). Thus, despite their negative nature, our findings may beget fruitful lines of inquiry to address which stimuli, procedures, or brain stimulation techniques can selectively impair subjective conscious experience, beyond impacting sheer objective processing sensitivity. 10 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience Neuroscience Procedure – Experiment 2 Procedure – Experiment 2 Three subjects who had participated in Experiment 1 also participated in Experiment 2. Procedures for Experiment 2 were identical to those described above for Experiment 1, except for the feedback structure, observer’s knowledge about target-present versus target-absent intervals, and order of questions (Figure 1B). In Experiment 2, we wanted to motivate subjects to bet on the target-present interval as much as possible, to maximize the possibility of observers performing optimally (i.e., to alleviate any Performance > Awareness). So, we defined a ‘correct’ Type 2 judgment for the pur- poses of feedback only as a Type 2 hit, i.e. trials in which the observer correctly discriminated the target-present interval and bet on the target-present interval. Subjects were also informed that in one of the intervals the target was physically absent, and that betting on that interval would not earn them a point even if they ‘discriminated’ its orientation correctly (as before, they still had a 50% chance to earn a point for ‘correctly discriminating’ the target-absent interval; subjects were made aware of this structure). Additionally, we provided ‘correct/incorrect’ feedback on the Type 2 responses to further encourage betting on the target-present interval. Finally, we altered the ques- tion order such that after each interval was presented, subjects pressed a button to discriminate the interval, and then only after both intervals had been presented did they indicate which choice they would like to bet on. In this way, subjects were allowed the ability to monitor their own reaction times, which ought to be faster for target-present intervals on average (as target-absent intervals are simply guesses by definition); this would provide another source of potential information to contrib- ute to confidence judgments, as it has been shown that subjects use reaction time monitoring to inform confidence judgments (Kiani et al., 2014). Points were awarded as in Experiment 1, and the same bonus payment motivation was employed. Also as before, participants completed five behav- ioral sessions each for Experiment 2, and were paid $10 per session. All analyses were carried out in Matlab R2013a (Natuck, MA) and SPSS Version 22 (IBM Corporation; Armonk, NY). Statistical analyses For each subject in each experiment, data were collapsed across tilt (left/right), interval presentation order (first/second), and session for each contrast level. At each contrast level for each subject, we next calculated (a) percent correct orientation discrimination, (b) percent of trials in which the target- present interval was chosen, (c) Type 2 hit rate and Type 2 false alarm rate according to standard Type 2 signal detection theoretic definitions (Type 2 hit: correct orientation discrimination and bet on target-present interval; Type 2 false alarm: incorrect orientation discrimination and bet on target- present interval) (Fleming and Lau, 2014; Maniscalco and Lau, 2012), and (d) percent correct orien- tation discrimination conditional on having chosen the target-present versus target-absent interval. Group-level analyses and graphical presentation were conducted by binning subjects’ data into ten equally-spaced bins of percent correct orientation discrimination performance in the range 0.5 – 1 and calculating the mean and standard deviation of each of the above statistics for each bin. To interpolate between discrete data points, we fitted a kernel smoothing regression function to each observer’s data, which is a non-parametric approach to estimate the conditional expectation of a random variable, E Y X j Þ ¼ f X ð Þ ð where f is a non-parametric function. This approach is based on kernel density estimation, implementing Nadaraya-Watson kernel regression (Nadaraya, 1964; Wat- son, 1964) via ^f x;K;h ð Þ ¼ Pn i¼1 Kh x xi ð Þyi Pn i¼1 Kh x xi ð Þ (1) (1) where K is a Gaussian kernel with bandwidth h. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 where K is a Gaussian kernel with bandwidth h. Research article Procedure – Experiment 1 Subjects were informed that they would be awarded a point for every cor- rect discrimination (Type 1 judgment), and an additional point every time they bet on an interval they discriminated correctly (Type 2 judgment), and total points were displayed at the end of the experiment; they were also told that if they earned more points than the previous participant, they would be paid an additional $10 bonus at the end of all sessions. In each behavioral session, trials were presented in a randomized full factorial design, counterbal- ancing interval order, in ten blocks of 52 trials per block. Every subject undertook five 60-minute ses- sions, for a total of 2600 trials spread across up to thirteen contrast levels, two orientations, and two interval presentation orders. Levels of contrast presented to each participant were titrated across Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 11 of 30 Research article Research article Neuroscience sessions to ensure performance spanning approximately evenly from chance (50% correct) to 100% correct, resulting in no fewer than 200 trials per contrast level (10 trials per condition x 2 orientations x 2 interval orders x 5 sessions). Subjects were paid $10 per session. Inference process Our ideal observer employs Bayesian inference in which each interval’s sample (i.e., evidence pair) d is first categorized as belonging to S1 or S2 on the basis of the posterior probabilities of each, and then uses the posterior probability of the chosen orientation as a measure of confidence in each dis- crimination decision. We assume that each generating stimulus category, S, is dependent on the evidence in favor (or contrast) of the presented stimulus, c, and can be represented by a bivariate Gaussian distribution such that Scleft~ N c; 0 ½ ; P ð Þ for a ‘left’ tilt and Scright~ N 0; c ½ ; P ð Þ for a ‘right’ tilt (Figure 5). Addi- tionally, in the most basic formulation we define P ¼ 1 0 0 1 (although we explore other poten- tially more biologically plausible variants; see Appendix 4). We also assume the cleft and cright axes (left and right tilt) to be orthogonal, although this constraint is not necessary for the model to cap- ture behavioral performance (see Appendix 4). Importantly, c – the contrast or evidence level along each axis that gave rise to the data sample the observer sees – is unknown to the observer. So, because contrast evidence is a secondary (or nuisance) variable to the primary variable of interest – in this case, the orientation of the Gabor patch – the observer ‘integrates out’ or marginalizes over all possible contrast evidence levels to produce the posterior probability estimate of each tilt (Yuille and Bu¨lthoff, 1996). Thus, the joint probability of each orientation and contrast evidence level is estimated through Bayes’ rule pðS;cjdÞ ¼ pðdjS;cÞ p ðS;cÞ p ðdÞ (2) (2) and then the secondary variable is integrated out, leaving estimation of the posterior probability of each orientation S via the marginal distribution (Yuille & Bu¨lthoff, 1996) and then the secondary variable is integrated out, leaving estimation of the posterior probability of each orientation S via the marginal distribution (Yuille & Bu¨lthoff, 1996) p ðSjdÞ ¼ ð p ðS; cjdÞdc (3) (3) In the simplest form, both orientations have equal prior probability of 0.5. where K is a Gaussian kernel with bandwidth h. All analyses were carried out in Matlab R2013a (Natuck, MA) and SPSS Version 22 (IBM Corporation; Armonk, NY). 12 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience Bayesian ideal observer model Model space p Our model representation space extends Macmillan and Creelman’s (2004) two-dimensional signal detection theory (SDT) and related Bayesian (King and Dehaene, 2014) framework, in which stimu- lus categories are represented by bivariate Gaussian distributions centered along the axes in a Car- tesian plane, and ‘noise’ (or a blank stimulus) is represented by a similar bivariate Gaussian centered at the origin (Figure 4). Although for this particular task we could have used a 1-dimensional space alternative (see e.g. Sridharan et al., 2014), to facilitate additional model variants (see Appendix 4) and possible future applications to stimuli that contain a mixture of multiple stimulus categories, we elected to present the model in a two-dimensional format. To accomplish both the orientation dis- crimination and 2IFC confidence judgments, on each simulated two-interval trial, two pairs of evi- dence values (representing the evidence in favor of a left- or right-tilted target) of the form d = [dleft, dright] are drawn: one sample is drawn from one of the signal distributions S (dTP, target-present intervals), and the other drawn from the noise distribution (dTA, target-absent intervals) (Figure 4). Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Evaluation of model performance We examined the relative agreement between our model’s predictions and collected behavioral data by calculating the multinomial likelihood of the model given the observed data, which has pre- viously been used within a signal detection framework. Details of goodness of fit calculations are described in Appendix 4. To evaluate whether human participants exhibited Performance without Awareness, we needed to cause the model to also exhibit Performance without Awareness. We therefore degraded the 2IFC confidence judgment process in the following way: On each trial, after the orientation decision had been reached, we programmed an added decisional noise parameter, sd, such that the decision variable D calculated as in Equation 5 was corrupted by additive Gaussian noise with mean 0 and standard deviation sD, such that D ¼ logðpðSchosen;TPjdTPÞ pðSchosen;TAjdTAÞÞ þ sd (6) (6) This causes the model to perform closer to chance at higher levels of orientation discrimination performance, i.e. to exhibit Performance without Awareness at increasing objective performance lev- els (Figure 5). We tested three decisional noise magnitudes – 0.1, 0.2, and 0.3 – and calculated the goodness of fit (see Appendix 4) for each sd for each subject. Inference process The observer then makes its orientation decision (for each interval) via Schosen ¼ arg max i p Si dj Þ ð (4) (4) To determine which interval’s choice the observer is more confident in, the model refers to the magnitude of the posterior probabilities of each Schosen in each interval as a measure of the probabil- ity of having made a correct orientation discrimination choice, i.e. p(correct) = p(Schosen\|d). Then, the observer compares these posterior probabilities for the target-present (TP) and target-absent (TA) intervals by computing a decision variable D via D ¼ logðpðSchosen;TPjdTPÞ pðSchosen;TAjdTAÞÞ (5) (5) 13 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience Research article Research article Figure 5. Illustration of increasing values for sd on the appearance of Performance without Awareness behavior, used to evaluate the possibility that human participants may have exhibited Performance without Awareness. Increasing sd values resulted in increasingly poor R2 values (see Results), indicating that the ideal observer (which displays no performance without awareness) produces the best fit to human data. DOI: 10.7554/eLife.09651.009 Neuroscience Figure 5. Illustration of increasing values for sd on the appearance of Performance without Awareness behavior, used to evaluate the possibility that human participants may have exhibited Performance without Awareness. Increasing sd values resulted in increasingly poor R2 values (see Results), indicating that the ideal observer (which displays no performance without awareness) produces the best fit to human data. DOI: 10.7554/eLife.09651.009 The observer bets on the interval with the higher probability of being correct: if this decision vari- able D is greater than 0, the observer selects the target-present interval to ‘bet’ on; if it is less than 0, the observer selects the target-absent interval. Sample code for this Bayesian ideal observer is included in Source code 1. Ethics Human subjects: Eleven subjects (two women, ages 19-32, ten right-handed) gave written informed consent to participate in our behavioral experiments. All subjects had normal or corrected-to-normal eyesight, and wore the same corrective lenses for all sessions, if applicable. Behavioral experiments were conducted in accordance with the Declaration of Helsinki and were approved by the UCLA Institutional Review Board. Eighty-seven respondents replied to our informal online survey. Survey procedures were conducted in accordance with the Declaration of Helsinki and were approved by the UCLA Institutional Review Board. Thus, all survey respondents provided informed consent to participate in the informal online survey, and behavioral subjects provided written informed consent to participate in the behavioral experiments. Additional files Supplementary files . Source code 1. Ideal observer model. DOI: 10.7554/eLife.09651.010 Additional files Supplementary files . Source code 1. Ideal observer model. DOI: 10.7554/eLife.09651.010 Alternative models We also examine three other possible contributing factors: correlated noise/non-orthogonal source distributions, signal-dependent (multiplicative), and signal-independent (additive) noise (see Appen- dix 4). These factors do not affect the qualitative trend of the model’s performance. For completeness, we also examine two other decision rules, detailed in Appendix 5: a heuristic observer which does not ignore contrast evidence as above, but explicitly estimates the most likely contrast level via hierarchical Bayesian inference (Yuille and Bu¨lthoff, 1996); and a heuristic likeli- hood comparison observer (similar to Barthelme´ et al., 2009). Importantly, the hierarchical model produced behavior similar to the ideal observer, indicating that such behavior is not idiosyncratic or Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 14 of 30 Research article Research article Neuroscience Neuroscience specific only to the ideal observer presented above. The likelihood-only model, on the other hand, failed to produce predictions that matched collected behavioral data, either qualitatively or quantitatively. specific only to the ideal observer presented above. The likelihood-only model, on the other hand, failed to produce predictions that matched collected behavioral data, either qualitatively or quantitatively. Acknowledgements This work was supported by the National Institute of Health (US) to HL (grant number R01NS088628). We thank Brian Maniscalco, Dobromir Rahnev, Hongjing Lu, and Zili Liu for helpful comments. Research article Additional information Funding Funder Grant reference number Author National Institutes of Health R01NS088628 Hakwan Lau The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication. References Consciousness a C i i 32 68 78 d i 10 1016/j 2014 09 009 Heeks F, Azzopardi P. 2015. Thresholds for detection and awareness of masked facial stimuli. Consciousness a Cognition 32:68–78. doi: 10.1016/j.concog.2014.09.009 Cognition 32:68 78. doi: 10.1016/j.concog.2014.09.009 Jachs B, Blanco MJ, Grantham-Hill S, Soto D. 2015. On the independence of visual awareness and metacognition: a signal detection theoretic analysis. Journal of Experimental Psychology 41:269–276. doi: 10. 1037/xhp0000026 p Kiani R, Corthell L, Shadlen MN. 2014. Choice certainty is informed by both evidence and decision time. Neuro 84:1329–1342. doi: 10.1016/j.neuron.2014.12.015 King J-R, Dehaene S. 2014. A model of subjective report and objective discrimination as categorical decisions in a vast representational space. Philosophical Transactions of the Royal Society B: Biological Sciences 369: 20130204. doi: 10.1098/rstb.2013.0204 Kolb FC, Braun J. 1995. Blindsight in normal observers. Nature 377:336–338. doi: 10.1038/377336a0 Kunimoto C, Miller J, Pashler H. 2001. Confidence and accuracy of near-threshold discrimination responses. Consciousness and Cognition 10:294–340. doi: 10.1006/ccog.2000.0494 Lagarias JC, Reeds JA, Wright MH, Wright PE. 1998. Convergence properties of the nelder–mead simplex method in low dimensions. SIAM Journal on Optimization 9:112–147. doi: 10.1137/S1052623496303470 Lau HC, Passingham RE. 2006. Relative blindsight in normal observers and the neural correlate of visual consciousness. Proceedings of the National Academy of Sciences of the United States of America 103:18763 18768. doi: 10.1073/pnas.0607716103 p Lloyd DA, Abrahamyan A, Harris JA, Antal A. 2013. Brain-stimulation induced blindsight: unconscious vision or response bias? PLoS One 8:e82828–16. doi: 10.1371/journal.pone.0082828.g006 p j p g Macmillan NA, Creelman CD. 2004. Detection Theory: A User’s Guide. New York:Taylor & Francis. Macmillan NA, Creelman CD. 2004. Detection Theory: A User’s Guide. New York:Taylor & Francis. Maniscalco B Lau H 2012 A signal detection theoretic approach for estimating metacognitive sensitivity from Macmillan NA, Creelman CD. 2004. Detection Theory: A User s Guide. New York:Taylor & Francis. Maniscalco B, Lau H. 2012. A signal detection theoretic approach for estimating metacognitive sensitivity from confidence ratings. Consciousness and Cognition 21:422–430. doi: 10.1016/j.concog.2011.09.021 confidence ratings. Consciousness and Cognition 21:422–430. doi: 10.1016/j.concog.2011.09.021 Merikle P, Smilek D, Eastwood JD. 2001. Perception without awareness: perspectives from cognitive psychology. Cognition 79:115–134. doi: 10.1016/S0010-0277(00)00126-8 g g j g Merikle P, Smilek D, Eastwood JD. 2001. Perception without awareness: perspectives from cognitive psychology. Cognition 79:115–134. doi: 10.1016/S0010-0277(00)00126-8 g ( ) Morgan MJ, Mason AJS, Solomon JA. 1997. Blindsight in normal subjects? Nature 385:401–402. doi: 10.1038/ 385401b0 Nachmias J, Weber A. 1975. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 References References Barthelme´ S, Mamassian P. 2010. Flexible mechanisms underlie the evaluation of visual confidence. Proceedings of the National Academy of Sciences of the United States of America 107:20834–20839. doi: 10.1073/pnas. 1007704107 Barthelme´ S, Mamassian P, Kording K. 2009. Evaluation of objective uncertainty in the visual system. PLoS Computational Biology 5:e1000504 doi: 10.1371/journal.pcbi.1000504.s001 Barthelme´ S, Mamassian P. 2009. Evaluation of objective uncertainty in the visual system. PLoS Computational Biology 5:e1000504 doi: 10.1371/journal.pcbi.1000504 Boyer JL, Harrison S, Ro T. 2005. Unconscious processing of orientation and color without primary visual cortex. Proceedings of the National Academy of Sciences of the United States of America 102:16875–16879. doi: 10. 1073/pnas.0505332102 Charles L, van Opstal F, Marti S, Dehaene S. 2013. Distinct brain mechanisms for conscious versus subliminal error detection. NeuroImage 73:80–94. doi: 10.1016/j.neuroimage.2013.01.054 15 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 15 of 30 Research article Research article de Gardelle V, Mamassian P. 2014. Does confidence use a common currency across two visual tasks? Psychological Science 25:1286–1288. doi: 10.1177/0956797614528956 y g Dorfman DD, Alf E. 1969. Maximum-likelihood estimation of parameters of signal-detection theory and determination of confidence intervals—rating-method data. Journal of Mathematical Psychology 6:487–496. doi: 10.1016/0022-2496(69)90019-4 ( ) Dosher BA, Lu Z-L. 1998. Perceptual learning reflects external noise filtering and internal noise reduction through channel reweighting. Proceedings of the National Academy of Sciences of the United States of America 95: 13988–13993. doi: 10.1073/pnas.95.23.13988 Eriksen CW. 1960. Discrimination and learning without awareness: a methodological survey and evaluation. Psychological Review 67:279–300. doi: 10.1037/h0041622 y g Evans S, Azzopardi P. 2007. Evaluation of a ’bias-free’ measure of awareness. Spatial Vision 20:61–77. doi: 10. 1163/156856807779369742 Galvin SJ, Podd JV, Drga V, Whitmore J. 2003. Type 2 tasks in the theory of signal detectability: discrimination between correct and incorrect decisions. Psychonomic Bulletin & Review 10:843–876. doi: 10.3758/ BF03196546 Green DM, Swets JA. 1966. Signal Detection Theory and Psychophysics. New York: John Wiley & Sons, Inc. Green DM, Swets JA. 1966. Signal Detection Theory and Psychophysics. New York: John Wiley & Sons, Inc. Hannula DE, Simons DJ, Cohen NJ. 2005. Imaging implicit perception: promise and pitfalls. Nature Reviews. Neuroscience 6:247–255. doi: 10.1038/nrn1630 , g y y p y y , Hannula DE, Simons DJ, Cohen NJ. 2005. Imaging implicit perception: promise and pitfalls. Nature Reviews. Neuroscience 6:247–255. doi: 10.1038/nrn1630 Neuroscience 6:247–255. doi: 10.1038/nrn1630 Heeks F, Azzopardi P. 2015. Thresholds for detection and awareness of masked facial stimuli. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Sridharan D, Steinmetz NA, Moore T, Knudsen EI. 2014. Distinguishing bias from sensitivity effects in multialternative detection tasks. Journal of Vision 14:16. doi: 10.1167/14.9.16 References Discrimination of simple and complex gratings. Vision Research 15:217–223. doi: 10.1016/0042-6989(75)90210-2 Nadaraya EA. 1964. On estimating regression. Theory of Probability & Its Applications 9:141–142. doi: 10.113 1109020 Nagelkerke NJD. 1991. A note on a general definition of the coefficient of determination. Biometrika 78:691– 692. doi: 10.1093/biomet/78.3.691 Peirce CS, Jastrow J. 1884. On small differences in sensation. Memoirs of the National Academy of Sciences 3 Peirce CS, Jastrow J. 1884. On small differences in sensation. Memoirs of the National Academy of Sciences 3. Pessiglione M, Schmidt L, Draganski B, Kalisch R, Lau H, Dolan RJ, Frith CD. 2007. How the brain translates money into force: a neuroimaging study of subliminal motivation. Science 316:904–906. doi: 10.1126/science. 1140459 Ramsøy Thomas Zoe¨ ga, Overgaard M. 2004. Introspection and subliminal perception. Phenomenology and th Cognitive Sciences 3:1–23. doi: 10.1023/B:PHEN.0000041900.30172.e8 Robichaud L, Stelmach LB. 2003. Inducing blindsight in normal observers. Psychonomic Bulletin & Review 10: 206–209. doi: 10.3758/BF03196486 Scott RB, Dienes Z, Barrett AB, Bor D, Seth AK, A. K. 2014. Blind insight: metacognitive discrimination despite chance task performance. Psychological Science 25:2199–2208. doi: 10.1177/0956797614553944 Shabel SJ, Murphy RT, Malinow R. 2014. Negative learning bias is associated with risk aversion in a genetic animal model of depression. Frontiers in Human Neuroscience 8:1–9. doi: 10.3389/fnhum.2014.00001 Snodgrass M, Bernat E, Shevrin H. 2004. Unconscious perception: a model-based approach to method and evidence. Perception & Psychophysics 66:846–867. doi: 10.3758/BF03194978 S D M l T Sil J 2011 W ki i h i C Bi l 21 R912 R913 evidence. Perception & Psychophysics 66:846–867. doi: 10.3758/BF03194978 Soto D, Ma¨ntyla¨ T, Silvanto J. 2011. Working memory without consciousness. Current Biology 21:R912–R913. doi: 10.1016/j.cub.2011.09.049 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 16 of 30 Research article Research article Sridharan D, Steinmetz NA, Moore T, Knudsen EI. 2014. Distinguishing bias from sensitivity effects in multialternative detection tasks. Journal of Vision 14:16. doi: 10.1167/14.9.16 Thomas JP, Gille J, Barker RA. 1982. Simultaneous visual detection and identification: theory and data. Journal of the Optical Society of America 72:1642–1651. doi: 10.1364/JOSA.72.001642 Thomas JP. 1987. Effect of eccentricity on the relationship between detection and identification. Journal of the Optical Society of America A 4:1599–1605. doi: 10.1364/JOSAA.4.001599 Watson AB, Robson JG. 1981. Discrimination at threshold: labelled detectors in human vision. Vision Research 21:1115–1122. doi: 10.1016/0042-6989(81)90014-6 Watson GS. 1964. Smooth regression analysis. Sankya: The Indian Journal of Statistics, Series A 26:359–372. Weiskrantz L. 1986. Blindsight: A Case Study and Implications. y y p y Perception as bayesian inference. Cambridge: Cambridge University Press; 123–162. do CBO9780511984037.006 Subjects Eighty-seven respondents replied to our informal online survey. Survey procedures were conducted in accordance with the Declaration of Helsinki and were approved by the UCLA Institutional Review Board. Survey questions All questions were presented through the online software SurveyMonkey. The survey was advertised through use of social media (e.g. Facebook, Twitter), and through placement of an advertisement soliciting volunteer participation in the Association for the Scientific Study of Consciousness (ASSC) Monthly Newsletter. Respondents were not paid for their answers. This survey was informal, in that it did not involve random sampling or counterbalancing of question order. We asked three critical questions of respondents: 1. “Do you personally believe that it is possible for a stimulus to be perceived subliminally, i.e., to exert influence on neural processing and behavior without the relevant feature being perceivable at all?” 2. “Do you personally believe that for some stimulus, there is a subjective threshold (for the subjective, conscious experience of seeing to occur) above an objective threshold (for one to be able to discriminate or identify the stimulus)? That is, do you think there exists a certain contrast level or duration of presentation such that subjects will be able to discriminate or identify the stimulus even though they cannot subjectively see the stimulus?” 3. “If you answered ’yes’ to the last question, do you think it has been demonstrated unequivocally in the literature?” Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 References Oxford: Oxford University Press. Yuille AL, Bu¨ lthoff HH, Knill DC, Richards W. (eds) 1996. Bayesian decision theory and psychophysics. In: Perception as bayesian inference. Cambridge: Cambridge University Press; 123–162. doi: 10.1017/ CBO9780511984037.006 Neuroscience Neuroscience Sridharan D, Steinmetz NA, Moore T, Knudsen EI. 2014. Distinguishing bias from sensitivity effects in multialternative detection tasks. Journal of Vision 14:16. doi: 10.1167/14.9.16 Sridharan D, Steinmetz NA, Moore T, Knudsen EI. 2014. Distinguishing bias from sensitivity effects in multialternative detection tasks. Journal of Vision 14:16. doi: 10.1167/14.9.16 Thomas JP, Gille J, Barker RA. 1982. Simultaneous visual detection and identification: theory and data. Journa the Optical Society of America 72:1642–1651. doi: 10.1364/JOSA.72.001642 Thomas JP. 1987. Effect of eccentricity on the relationship between detection and identification. Journal of the Optical Society of America A 4:1599–1605. doi: 10.1364/JOSAA.4.001599 Watson AB Robson JG 1981 Discrimination at threshold: labelled detectors in human vision Vision Research Optical Society of America A 4:1599–1605. doi: 10.1364/JOSAA.4.001599 tson AB, Robson JG. 1981. Discrimination at threshold: labelled detectors in human vision. Vision Research 1 1115 1122 d i 10 1016/0042 6989(81)90014 6 21:1115 1122. doi: 10.1016/0042 6989(81)90014 6 Watson GS. 1964. Smooth regression analysis. Sankya: The Indian Journal of Statistics, Series A 26:359–372. Weiskrantz L. 1986. Blindsight: A Case Study and Implications. Oxford: Oxford University Press. Yuille AL Bu¨ lthoff HH Knill DC Richards W (eds) 1996 Bayesian decision theory and psychophysics In: 17 of 30 Research article Neuroscience Appendix 1: Survey Appendix 1: Survey Results (A) Most respondents reported believing that some forms of subliminal perception exists (Q1), and that specifically the subjective threshold for conscious awareness is above the objective threshold for direct discrimination (Q2). However, most respondents also reported that they did not think this phenomenon (Q2) had been convincingly demonstrated in the literature (Q3). (B) The pattern of results was similar for the subset of respondents who had at least one relevant publication. Results Eighty-seven individuals responded to our informal survey. We collected demographic data on the following: age, highest degree earned, year of highest degree, field of degree earned, current occupation/position, and number of publications related to subliminal perception. Most respondents held a research doctorate (66%), mostly in Psychology (39%), Philosophy (19%), or Neuroscience (18%). About half of respondents (52%) received their degree in the last five years, and just over a third (38%) reported having one or more relevant publications on subliminal/unconscious perception. See Appendix 2—Figure 1 for detailed demographic information. 18 of 30 Research article Research article Research article Research article Research article Appendix 1—Figure 1. Demographics of survey respondents (n = 87). DOI: 10.7554/eLife.09651.011 Appendix 1—Figure 2. Answers to the three survey questions of interest. (A) Most respondents reported believing that some forms of subliminal perception exists (Q1), and that specifically the subjective threshold for conscious awareness is above the objective threshold for direct discrimination (Q2). However, most respondents also reported that they did not think this phenomenon (Q2) had been convincingly demonstrated in the literature (Q3). (B) The pattern of results was similar for the subset of respondents who had at least one relevant publication. DOI: 10.7554/eLife.09651.012 Most respondents reported believing that subliminal processing exists (Q1) (94%), and that the subjective threshold is above the objective one (Q2) (89%). Of those who responded ‘Yes’ Q i 2 h l b hi d (36%) d b li i P f i h Neuroscience Neuroscience Appendix 1—Figure 1. Demographics of survey respondents (n = 87). DOI: 10.7554/eLife.09651.011 Appendix 1—Figure 1. Demographics of survey respondents (n = 87). DOI: 10.7554/eLife.09651.011 Appendix 1—Figure 2. Answers to the three survey questions of interest Appendix 1—Figure 2. Answers to the three survey questions of interest. Appendix 1—Figure 2. Answers to the three survey questions of interest. Appendix 1—Figure 2. Answers to the three survey questions of interest. (A) Most respondents reported believing that some forms of subliminal perception exists (Q1), and that specifically the subjective threshold for conscious awareness is above the objective threshold for direct discrimination (Q2). However, most respondents also reported that they did not think this phenomenon (Q2) had been convincingly demonstrated in the literature (Q3). (B) The pattern of results was similar for the subset of respondents who had at least one relevant publication. Appendix 1 Figure 2. Answers to the three survey questions of interest. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 DOI: 10.7554/eLife.09651.012 Most respondents reported believing that subliminal processing exists (Q1) (94%), and that the subjective threshold is above the objective one (Q2) (89%). Of those who responded ‘Yes’ to Question 2, however, only about a third (36%) reported believing Performance without Awareness has been unequivocally demonstrated in the literature (Q3) (Appendix 2—Figure 2A). Interestingly, the pattern of responses did not change for the subset of respondents who reported having at least one relevant publication (Q1: 94%, Q2: 88%, Q3: 39%; Appendix 2— Figure 2B). Thus, although a majority of scholars in this field report believing Performance without Awareness can exist, most also recognize that demonstrating it convincingly has been difficult. 19 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Neuroscience Appendix 2: Individual behavioral results: Experiments 1 and 2 Appendix 2: Individual behavioral results: Experiments 1 and 2 Appendix 2: Individual behavioral results: Experiments 1 and 2 Appendix 2—Figure 1. Individual subjects’ data for percent betting on target-present (T interval as a function of orientation discrimination percent correct. Individual subjects’ data closely resembles grouped data: all subjects exhibited Performa Awareness, but no subjects exhibited Performance without Awareness. The same three subjects participated in both Experiments 1 and 2. Appendix 2—Figure 1. Individual subjects’ data for percent betting on target-present (TP) interval as a function of orientation discrimination percent correct. Appendix 2—Figure 1. Individual subjects’ data for percent betting on target-present (TP) interval as a function of orientation discrimination percent correct. Individual subjects’ data closely resembles grouped data: all subjects exhibited Performance > Awareness, but no subjects exhibited Performance without Awareness. The same three subjects participated in both Experiments 1 and 2. DOI: 10.7554/eLife.09651.013 Appendix 2—Figure 1. Individual subjects’ data for percent betting on target-present (TP) interval as a function of orientation discrimination percent correct. Individual subjects’ data closely resembles grouped data: all subjects exhibited Performance > Awareness, but no subjects exhibited Performance without Awareness. The same three subjects participated in both Experiments 1 and 2. DOI: 10.7554/eLife.09651.013 p Individual subjects’ data closely resembles grouped data: all subjects exhibited Performance > Awareness, but no subjects exhibited Performance without Awareness. The same three subjects participated in both Experiments 1 and 2. 20 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Research article Neuroscience Appendix 2—Figure 2. Individual subjects’ data for Type 2 hit rate (T2HR) and Type 2 fa alarm rate (T2FAR). Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 DOI: 10.7554/eLife.09651.012 Individual subjects’ data closely resembles grouped data. The same three subjects participated in both Experiments 1 and 2. Appendix 2—Figure 2. Individual subjects’ data for Type 2 hit rate (T2HR) and Type 2 false alarm rate (T2FAR). Appendix 2—Figure 2. Individual subjects’ data for Type 2 hit rate (T2HR) and Type 2 false alarm rate (T2FAR). Individual subjects’ data closely resembles grouped data. The same three subjects participated in both Experiments 1 and 2. Appendix 2—Figure 2. Individual subjects’ data for Type 2 hit rate (T2HR) and Type 2 false alarm rate (T2FAR). Individual subjects’ data closely resembles grouped data. The same three subjects i i d i b h E i 1 d 2 21 of 30 Research article Research article Research article Appendix 2—Figure 3. Individual subjects’ data for percent correct orientation discrimina conditional on having chosen to bet on the target-present (TP) interval or not. Individual subjects’ data closely resembles grouped data. Appendix 2—Figure 3. Individual subjects’ data for percent correct orientation discrimination conditional on having chosen to bet on the target-present (TP) interval or not. Individual subjects’ data closely resembles grouped data. The same three subjects participated in both Experiments 1 and 2. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 DOI: 10.7554/eLife.09651.015 We also found that subjects were overall unbiased in their choices of left- versus right-tilted Gabor patches, justifying use of percent correct rather than the signal detection theory metric d’. We calculated the mean criterion used by each subject across all levels of contrast according to signal detection theory, assuming equal variance for distributions representing left- versus right-tilted stimuli, i.e. c ¼ 1 2 z HR ð Þ þ z FAR ð Þ ð Þ (A2 1) c ¼ 1 2 z HR ð Þ þ z FAR ð Þ ð Þ (A2 1) 22 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience where HR refers to the hit rate (i.e., the participant said ‘left’ when a left-tilted Gabor was presented) and FAR refers to the false alarm rate (i.e., the participant said ‘left’ when a right- tilted Gabor was presented) (Macmillan and Creelman, 2004). Individual subjects’ criterion values are presented in Appendix 2–Table 1. A two-tailed t-test (treating each subject as an independent sample) revealed these values to not be significantly different from 0 (Appendix 2–Table 1). Appendix 2–Table 1. Mean value of criterion for each subject, demonstrating that subjects did not display biases to say ‘left’ versus ‘right’ in the orientation discrimination task. Experiment Subject c 1 1 0.047 2 0.074 3 -0.039 4 -0.029 5 -0.013 6 -0.030 7 0.0267 8 -0.039 9 0.003 2 1 0.229 2 0.004 3 0.035 Mean (s) 0.022 t(11) 1.04 p 0.3215 DOI: 10.7554/eLife.09651.016 Experiment Subject c 1 1 0.047 2 0.074 3 -0.039 4 -0.029 5 -0.013 6 -0.030 7 0.0267 8 -0.039 9 0.003 2 1 0.229 2 0.004 3 0.035 Mean (s) 0.022 t(11) 1.04 p 0.3215 DOI: 10.7554/eLife.09651.016 23 of 30 Research article Research article Research article Neuroscience Appendix 3: Control study Appendix 3: Control study We ran a control study to assess whether the lack of Performance without Awareness observed in Experiments 1 and 2 may have occurred because subjects have an unconscious ‘hunch’ of confidence despite having absolutely no phenomenal experience of the stimulus. Three observers (all male, ages 19-32, all right-handed) gave written informed consent to participate. One observer had participated in Experiments 1 and 2; the other two were naive. All procedures were identical to Experiment 2, with two exceptions: (1) observers were not told that one interval was blank; and (2) observers were asked to indicate which interval was ‘more visible’ rather than which discrimination they believed they were more likely to get correct. Thus, observers engaged in a task more akin to a 2IFC detection task rather than a metacognitive judgment. Despite these manipulations, the results of the control experiment closely mirror the results of Experiments 1 and 2. Subjects were able to detect the target in the TP interval above chance as soon as they were able to discriminate the target above chance, and individual subjects’ data closely resembled the group data (Figure S3). Note that the computational model described in the main text does not apply to these data, as participants were doing 2IFC detectability discrimination rather than a 2IFC metacognitive judgment. This means that even though participants’ data lie near the identity line of equal orientation percent correct and percent betting on the TP interval (dashed diagonal line) – above the behavior predicted by the confidence model – this should not be taken to mean they are displaying any sort of ‘supra-optimal’ behavior, as the tasks are fundamentally different. The position of responses on this graph for this task will depend on the precise relationship between the detectability and discriminability of the stimulus for each individual subject (Thomas et al., 1982; Thomas, 1987). Appendix 3—Figure 1. Results of control experiment. Psychometric functions were fitted as in Experiments 1 and 2 (see Materials and methods — Statistical analyses).Group data is representative of individual subjects’ responses (left panel versus right three panels). As before, subjects displayed behavior inconsistent with Performance without Awareness. If anything, participants were able to bet on the target- present interval more often when the interval judgment was one of detection rather than Type 2 or metacognitive assessment of the correctness of the Type 1 discrimination. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Appendix 3: Control study However, this should not be taken to indicate ‘supra-optimal’ behavior with reference to the ideal observer model of confidence, as the task here is not confidence but a task akin to 2IFC detection, meaning it would require a different computational analysis for assessment of optimality. DOI 10 7554/ Lif 09651 017 Appendix 3—Figure 1. Results of control experiment. Psychometric functions were fitted as in Experiments 1 and 2 (see Materials and methods — Statistical analyses).Group data is representative of individual subjects’ responses (left panel versus right three panels). As before, subjects displayed behavior inconsistent with Performance without Awareness. If anything, participants were able to bet on the target- present interval more often when the interval judgment was one of detection rather than Type 2 or metacognitive assessment of the correctness of the Type 1 discrimination. However, this should not be taken to indicate ‘supra-optimal’ behavior with reference to the ideal observer model of confidence, as the task here is not confidence but a task akin to 2IFC detection, meaning it would require a different computational analysis for assessment of optimality. DOI: 10.7554/eLife.09651.017 24 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience Appendix 4: Model details and variants Adjustments for non-orthogonal signal dimensions j g g In its simplest form, the ideal observer model assumes independence/orthogonality of the stimuli – graphically, that the Sleft and Sright distributions lie on the x and y axes, respectively. However, it is conceivable that the stimuli are correlated or anti-correlated. Therefore, we explored modifying the cright axis to define it as a vector specified by , the angle between the x axis and the vector (Appendix 4—Figure 1). Appendix 4—Figure 1. Graphical intuition of correlated stimuli (A) and anti-correlated stimuli (B). DOI: 10.7554/eLife.09651.018 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Comparison of model to behavioral data Comparison of model to behavioral data We examined the relative agreement between our model’s predictions and collected behavioral data by calculating the multinomial likelihood of the model given the observed data (Dorfman and Alf, 1969), which has previously been used within a signal detection framework. Formally, the likelihood of a certain model m (with a given set of parameters ’) can be expressed as, LmðfjdataÞ ¥ Y ijPf ðRijSjÞndataðRijSjÞ LmðfjdataÞ ¥ Y ijPf ðRijSjÞndataðRijSjÞ where each Ri is a behavioral response a subject may produce on a given trial, and each Sj is a type of stimulus that might be shown on that trial. The expression ‘ndata(Ri Sj)’ is a count of how many times a subject actually produced Ri after being shown Sj. The expression ‘Pf(Ri Sj)’ denotes the probability with which the subject produces the response Ri after being presented with Sj, according to the model specified with parameters f. This corresponds to the percentage of time each of the models described above produced response Ri after having been ‘presented’ with stimulus Sj. Note that this approach does not examine the performance of a model relative to the behavioral data with reference to any summary statistics, but instead allows for fitting of the full distribution of probabilities of each response type contingent on each stimulus type. Given that the model cannot meaningfully represent the physical contrast presented to a human participant, we simulated all possible contrast evidence values on a scale of 0 to 5 in steps of 0.2, and calculated the percent correct orientation discrimination performance predicted by the model at each of these contrast evidence levels. We then matched each participant’s data to the closest ‘contrast evidence’ level in the model by identifying ‘contrast’ levels at which each model performed similarly to the participant on the orientation discrimination task, i.e. matching the percent correct orientation discrimination performance between the model and the data for each contrast level presented to a human participant. The remainder of the summary statistics used in the Behavioral Results analyses – percent of time TP interval was chosen, Type 2 Hit Rate and False Alarm Rate, and percent correct orientation discrimination conditional on TP chosen or unchosen – can then all be derived from this full behavioral profile of stimulus-contingent response probabilities, allowing for comprehensive comparison between model predictions and behavioral data. DOI: 10.7554/eLife.09651.018 Formally, instead of cright = [0, c], we define cright ¼ c cos ð Þ;c sin ð Þ ½ (A4 1) cright ¼ c cos ð Þ;c sin ð Þ ½ cright ¼ c cos ð Þ;c sin ð Þ ½ The parameter thus allows control of the degree of correlation (or anti-correlation) between the Sleft and Sright distributions. This formulation is mathematically equivalent to specifying non-zero covariance in the definition of the variance-covariance matrix S (Macmillan & Creelman, 2004), but we favor using here because it provides a graphically intuitive analog to changing the amount of ‘tilt’ between the left- and right-tilted Gabor patch targets, that is, the relationship between the detectability and discriminability of a stimulus (see also King and Dehaene, 2014). For example, Gabor patches that are ± 45˚ tilted from vertical ought to demand a larger value than other, less discriminable versions, such as ± 5˚ tilt. To explore whether this factor was important to model fits, we fitted the value for to the three participants’ data who had completed both main experiments (see Comparison of model to behavioral data, below) and compared model fits with measures of R2 (see Goodness of fit, below) and the Bayesian information criterion (BIC). Best-fitting values for were quite consistent across participants, and indicate a slight anti- correlation between the Sleft and Sright distributions ( = 97.27º, s = 18.79º). The goodness of fit was evaluated for these three participant’s data individually for each experiment given these best-fitting values, and was found to be very good for all across two different null models (R2 = 0.636 ± 0.172; BIC = 4669.31 ± 960.94). However, the single free parameter does not provide any significant improvement in the fit of the model to the data: with no free parameters, R2 = 0.617 ± 0.141 and BIC = 4657.10 ± 960.73. Thus, we present the parameter- free model in the main text. 25 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Neuroscience Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Comparison of model to behavioral data We used this measure both to explore the goodness of fit (see below) of the model given no free parameters, and also to evaluate the ideal observer’s dependence on the free parameter to the observed behavioral data in all experiments on a subject-by-subject basis, by maximizing the likelihood of the model parameters given the data (Equation A4-2) for each model-subject combination. To compare between two models with unequal numbers of parameters, it is standard to use the Bayesian information criterion (BIC), which penalizes more complicated models according to the number of free parameters to avoid overfitting. The BIC is calculated as maximizing the likelihood of the model parameters given the data (Equation A4-2) for each model-subject combination. To compare between two models with unequal numbers of parameters, it is standard to use the Bayesian information criterion (BIC), which penalizes more complicated models according to the number of free parameters to avoid overfitting. The BIC is calculated as BIC ¼ 2 ln Lm ; data j Þ ð Þ þ klnðnÞ ð (A4 3) (A4 3) (A4 3) BIC ¼ 2 ln Lm ; data j Þ ð Þ þ klnðnÞ ð where Lm ; data j Þ ð is defined as in Equation A4-2, k is the number of free parameters, and n is the number of data points. where Lm ; data j Þ ð is defined as in Equation A4-2, k is the number of free parameters, and n is the number of data points. Maximum likelihood fits were accomplished using a customized Nelder-Mead simplex search algorithm (Lagarias et al., 1998) to minimize the negative log-likelihood (Equation A4-2). Model predictions were accomplished through Monte Carlo simulation, with 10,000 paired- interval ‘trials’ at each simulated stimulus contrast level. 26 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 26 of 30 26 of 30 Research article Neuroscience Goodness of fit To quantify the goodness of fit for the parameter-free ideal observer model presented in the main text, as well as for the model utilizing the best-fitting value of for each participant, we calculated the generalized coefficient of determination, R2, as described in To quantify the goodness of fit for the parameter-free ideal observer model presented in the main text, as well as for the model utilizing the best-fitting value of for each participant, we calculated the generalized coefficient of determination, R2, as described in (Nagelkerke, 1991). This provides a continuous measure ranging from zero (corresponding to worst fit) to one (corresponding to best fit) based on the maximum likelihood criteria of fit. The formula used is calculated the generalized coefficient of determination, R2, as described in (Nagelkerke, 1991). This provides a continuous measure ranging from zero (corresponding to worst fit) to one (corresponding to best fit) based on the maximum likelihood criteria of fit. The formula used is (Nagelkerke, 1991). This provides a continuous measure ranging from zero (corresponding to worst fit) to one (corresponding to best fit) based on the maximum likelihood criteria of fit. The formula used is R2 ¼ 1 exp 2 n l b ^ l 0 ð Þ (A4 4) where n is the number of data points, and l b^ ð Þ and l 0 ð Þ denote the log-likelihoods of the fitted and the null model, respectively. We explored two null models: (1) A model which produces every response type with equal probability (i.e., randomly guesses); and (2) a model which knows the correct answer with 100% probability. The generalized R2 is interpreted as the proportion of variance in the data that is explained by the model. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Appendix 5: Alternative models For completeness, we detail two other models considered here. Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Alternative model 1: heuristic – hierarchical inference To evaluate whether the behavioral pattern predicted by the Bayesian ideal observer is idiosyncratic or robust to different decision strategies, we explored an alternative model which does not ‘ignore’ (marginalize over) contrast as a secondary or nuisance variable. Instead, this Bayesian heuristic observer engages in hierarchical inference: it first estimates the most likely contrast to have produced the current stimulus as a secondary or hidden variable, then uses it to inform the decision about the primary variable of interest, that is, the most likely orientation of the stimulus (Yuille et al., 1996). Upon seeing a sample d, this observer first makes a guess ^ci at the most probable c to have generated the current data point d, along both the cleft and cright dimensions. It does so via maximizing the likelihood of each data point d given the possible source distributions centered at all possible contrast values c along the axes: ^cleft ¼ argmax cleft p d ^Scleft ^cleft ¼ argmax cleft p d ^Scleft and ^cright ¼ argmax cright p d ^Scright (A5 1) (A5 1) where ^Scleft~N cleft; 0 ; P and ^Scright~N 0; cright ; P . This can easily be calculated by a projection of each d ¼ dleft; dright onto the cleft and cright axes, that is ^cleft ¼ dleft and ^cright ¼ dright (Appendix 5—Figure 1). In other words, the model decomposes the stimulus in each interval into the most likely ’left tilt contrast’ (cleft) and ’right tilt contrast’ (cright) to have given rise to what the model is ’seeing’. Values for ^cleftand ^cright are not constrained to be positive; negative values represent contrast evidence for a left or right tilt that falls below the ’medium gray’ of the ’nothing’ distribution. Note that one could also assess this for values of other than 90º. Appendix 5—Figure 1. Illustration of the hierarchical inference process. The observer first makes a best guess about the most likely contrast to have given rise to the evidence it is seeing, and then uses that inferred contrast evidence level to conduct the rest of Appendix 5—Figure 1. Illustration of the hierarchical inference process. Appendix 5—Figure 1. Illustration of the hierarchical inference process. Biological plausibility We also considered several other modifications to make our model more biologically plausible Multiplicative noise is considered due to the observation that neurons often display constant or near-constant Fano factor F, where F ¼ s2W W , where s2W is the variance and W the mean of the neuron’s firing rates in some time window W. In other words, the variance increases as mean firing increases, leading to increased neuronal variability with stronger stimulus inputs. Additionally, both multiplicative and additive noise are considered elements of the noise in neural firing rates and behavioral responses (e.g., Dosher and Lu, 1998). We thus considered two additional free parameters, individually and in conjunction with one another: 1. f - signal-dependent (multiplicative) noise, or correction for near-constant Fano factor: the g p p variance-covariance matrix is modified to be P ¼ cf 0 0 cf variance-covariance matrix is modified to be P ¼ cf 0 0 cf 2. " - additive noise: the variance-covariance matrix is modified to be P ¼ " 0 0 " Notably, neither of these parameters – individually or in conjunction with one another – produced any significant effect on qualitative model trends or fit (Equation A4-2,A4-3). 27 of 30 Research article Neuroscience Appendix 5: Alternative models Alternative model 1: heuristic – hierarchical inference The observer first makes a best guess about the most likely contrast to have given rise to the evidence it is seeing, and then uses that inferred contrast evidence level to conduct the rest of Appendix 5—Figure 1. Illustration of the hierarchical inference process. The observer first makes a best guess about the most likely contrast to have given rise to the evidence it is seeing, and then uses that inferred contrast evidence level to conduct the rest of pp g p The observer first makes a best guess about the most likely contrast to have given rise to the evidence it is seeing, and then uses that inferred contrast evidence level to conduct the rest of Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 28 of 30 Research article Research article Neuroscience the inference process. It does this for both the target-present interval (A) and the target- absent interval (B). DOI: 10.7554/eLife.09651.019 Once the most likely values for ^cleftand ^cright have been obtained, the observer refers to these new ‘best guess’ source distributions as ^S^cleftand ^S^cright (Appendix 5—Figure 1). The Bayesian heuristic observer determines the posterior probability of each stimulus orientation source distribution S _ c _ ifor each data sample d via Bayes’ Rule, pð ^S^cIjdÞ ¼ pðdj ^S^cIÞpð ^S ^cIÞ pðdÞ (A5 2) with each orientation ^S ^cI having equal prior probability of 0.5. The observer then makes its orientation decision via ^Schosen ¼ arg max i pð ^S^cIjdÞ (A5 3) ^Schosen ¼ arg max i pð ^S^cIjdÞ To determine which choice the observer is more confident in, the model refers to the magnitude of the posterior probabilities of each ^Schosen in each interval as a measure of the probability of having made a correct orientation discrimination choice, i.e. pðcorrectÞ ¼ p ^Schosen dj Þ . Then the observer compares these posterior probabilities for the Target Present (TP) and Target Absent (TA) intervals by computing the decision variable D via D ¼ logðpð ^Schosen;TPjdTPÞ pð ^Schosen;TAjdTAÞÞ (A5 4) As before, if this decision variable is greater than 0, the observer selects the TP interval; if it is smaller than 0, the observer selects the TA interval as being more confident. This model produces the same behavior as the Bayesian ideal observer described in the main text (Appendix 5—Figure 2, top row). Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Alternative model 2: heuristic – likelihood comparison p Barthelme´ et al., 2009 used a variant of the two-by-two forced-choice paradigm (Nachmias and Weber, 1975), in which two threshold-level stimuli are simultaneously rather than sequentially presented. The observer is asked to choose which stimulus he would be more confident discriminating, and then to discriminate that stimulus. Importantly, in their task, templates are provided on-screen, such that observers may match the stimulus to a provided template rather than a remembered or inferred stimulus dimension. In that study, the authors compare the performance of a series of models in predicting discrimination and confidence decision, including a Bayesian posterior for each template given the stimulus, and a Bayesian likelihood of each stimulus given each template. The authors report that the likelihood model fits their data better than the posterior model, and conclude that the likelihood method of evaluating confidence describes how human observers judge confidence (despite it being suboptimal). Here, we evaluate the performance of a similar likelihood-only model, defining the decision variable with reference to the most likely contrast level to have given rise to the current sample, i.e. 29 of 30 Peters and Lau. eLife 2015;4:e09651. DOI: 10.7554/eLife.09651 Research article Research article Neuroscience D ¼ logðpðdTPj ^Schosen;TPÞ pðdTAj ^Schosen;TAÞÞ (A5 5) This decision variable is evaluated as described for Equation A5-5 in the main text. This model failed to produce behavior consistent with human observers’ responses (Appendix 5—Figure 2, bottom row). Appendix 5—Figure 2. Sample of predicted behavior from Alternative Models 1 (Heuristic - hierarchical Inference) and 2 (Heuristic - likelihood comparison) (at = 90º; see above). While the hierarchical observer produces behavior similar to the ideal observer, the likelihood comparison strategy does not. DOI: 10.7554/eLife.09651.020 Appendix 5—Figure 2. Sample of predicted behavior from Alternative Models 1 (Heuristic - hierarchical Inference) and 2 (Heuristic - likelihood comparison) (at = 90º; see above). While the hierarchical observer produces behavior similar to the ideal observer, the likelihood comparison strategy does not. Appendix 5—Figure 2. Sample of predicted behavior from Alternative Models 1 (Heuristic - hierarchical Inference) and 2 (Heuristic - likelihood comparison) (at = 90º; see above). While the hierarchical observer produces behavior similar to the ideal observer, the likelihood comparison strategy does not. DOI: 10.7554/eLife.09651.020 30 of 30
https://openalex.org/W4386991625	https://www.bioinformation.net/019/97320630019833.pdf	English	null	Anti-arthritic efficacy of Bombax ceiba ethanolic extract in a murine model for rheumatoid arthritis using in vivo, in vitro and radiological analysis	Bioinformation	2,023	cc-by	5,429	Comments from readers: Articles published in BIOINFORMATION are open for relevant post publication comments and criticisms, which will be published immediately linking to the original article without open access charges. Comments should be concise, coherent and critical in less than 1000 words. License statement: License statement: This is an Open Access article which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. This is distributed under the terms of the Creative Commons Attribution License This is an Open Access article which permits unrestricted use, distribution, and reproduction in an properly credited. This is distributed under the terms of the Creative Commons Attribution Licens Declaration on Publication Ethics: The author’s state that they adhere with COPE guidelines on publishing ethics as described elsewhere at https://publicationethics.org/. The authors also undertake that they are not associated with any other third party (governmental or non-governmental agencies) linking with any form of unethical issues connecting to this publication. The authors also declare that they are not withholding any information that is misleading to the publisher in regard to this article. Declaration on official E-mail: Declaration on official E-mail: The corresponding author declares that lifetime official e-mail from their institution is not available for all authors The corresponding author declares that lifetime official e-mail from their institution is not availabl The corresponding author declares that lifetime official e-mail from their institution is not available for all authors License statement: IOINFORMATION Impact Factor (2023 release) is 1.9 with 2,198 citations from 2020 to 2022 across Declaration on Publication Ethics: ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ©Biomedical Informatics (2023) www.bioinformation.net Volume 19(8) Research Article Received August 1, 2023; Revised August 31, 2023; Accepted August 31, 2023, Published August 31, 2023 O 10 6026 Received August 1, 2023; Revised August 31, 2023; Accepted August 31, 2023, Published August 31, 2023 DOI: 10.6026/97320630019833 BIOINFORMATION Impact Factor (2023 release) is 1.9 with 2,198 citations from 2020 to 2022 across continents taken for IF calculations. Disclaimer: The views and opinions expressed are those of the author(s) and do not reflect the views or opinions of Bioinformation and (or) its publisher Biomedical Informatics. Biomedical Informatics remains neutral and allows authors to specify their address and affiliation details including territory where required. Bioinformation provides a platform for scholarly communication of data and information to create knowledge in the Biological/Biomedical domain. p p ( ) p ( ) er Biomedical Informatics. Biomedical Informatics remains neutral and allows authors to specify their address and affiliation details ng territory where required. Bioinformation provides a platform for scholarly communication of data and information to create dge in the Biological/Biomedical domain. Edited by P Kangueane Citation: Posinasetty et al. Bioinformation 19(8): 833-839 (2023) ti-arthritic efficacy of Bombax ceiba ethanolic extract a murine model for rheumatoid arthritis using in vo, in vitro and radiological analysis gavi Posinasetty1, Radhika Chikatipalli2, Santenna Chenchula3, Kuttiappan Anitha4, G. na2, G.N.A. Lakshmi2 & Audinarayana Nelavala2 tment of Clinical Data Management, Prometrika LLC, Cambridge – MA 02140, United States; 2Department of Pharmacology, Sri eswara College of Pharmacy, Chittoor - 517127, A.P., India; 3Department of clinical pharmacology, All India Institute of Medical Edited by P Kangueane Citation: Posinasetty et al. Bioinformation 19(8): 833-839 (2023) Edited by P Kangueane Citation: Posinasetty et al. Bioinformation 19(8): 833-839 (2023) Anti-arthritic efficacy of Bombax ceiba ethanolic extract in a murine model for rheumatoid arthritis using in vivo, in vitro and radiological analysis Bhargavi Posinasetty1, Radhika Chikatipalli2, Santenna Chenchula3, Kuttiappan Anitha4, G. Swapna2, G.N.A. Lakshmi2 & Audinarayana Nelavala2 1Department of Clinical Data Management, Prometrika LLC, Cambridge – MA 02140, United States; 2Department of Pharmacology, Sri Venkateswara College of Pharmacy, Chittoor - 517127, A.P., India; 3Department of clinical pharmacology, All India Institute of Medical Author contacts: Author contacts: Bhargavi Posinasetty – E-mail:Posinasettybhargavi@gmail.com Radhika Chikatipalli - E-mail: drradhika@svcop.in Santenna Chenchula – E-mail: csanten7@gmail.com Kuttiappan Anitha – E-mail: kanitha@gwa.amity.edu G. Swapna – E-mail: swapna@svcop.in G.N.A. Lakshmi – E-mail: alakshmi@svcop.in Audinarayana Nelavala – E-mail: audinarayana@svcop.in Plant collection and authentication: The Bombax ceiba plantswere purchased as leafy vegetables from the local market of Pileru, Chittoor District, Andhra Pradesh state of India in March 2017; the photograph of which is shown in (Figure 1). Authenticationof the plant was carried out by Dr. K. Madava Chetty, Asst. Professor, Dept. of Botany, Shri Venkateswara University, Tirupati, A.P., India. Avoucher specimen of theplant (Ref. No. 0610 dated 11/09/2017); has been preserved there forfuture reference. Background: Background: Rheumatoid arthritis (RA) is a systemic, chronic, progressive inflammatory autoimmune condition that manifests as symmetric polyarthritis with swelling and pain in several joints, most commonly in the hands and feet.The joint is the primary site of inflammatory tissue destruction in RA, which is a complicated, multi-system disease. RA is a multisystem autoimmune disease that causes discomfort, synovial membrane inflammation, peripheral joint inflammation, morning stiffness, articular tissue loss, and restricted joint movement [1]. RA pathophysiology and aetiology are complicated and unknown. Destructive alterations in cartilage and bone, as well as bone outgrowths, limit joint motion. Arthritis can cause severe disability and ultimately affects a person’s ability to carry out everyday tasks, restricts the quality of life and causes premature death [2].RA is the most common inflammatory disorder affecting about 1% of the global adult population; females are three times more prone to RA than males [3]. Traditional RA treatment with NSAIDs, corticosteroids, immune suppressants, and anti- rheumatic medicines (TNF-alpha and monoclonal antibodies) has several limitations [4]. Chronic use of the aforementioned medications causes major side effects such as GIT ulcers, and cardiovascular, haematological, and renal toxicity [5]. Patients with persistent autoimmune illnesses are encouraged to seek alternative symptomatic alleviation [6]. Bombax ceiba Linnaeus is a member of the Bombaceae family [7]. It is widely found in temperate Asia, Tropical Asia, Africa and Australia [8].Many components of the plant (root, stem bark, gum, leaf, prickles, flower, fruit, seed, and heartwood) are used to treat a range of diseases by tribal societies Anti-arthritic efficacy of Bombax ceiba ethanolic extract in a murine model for rheumatoid arthritis using in vivo, in vitro and radiological analysis Bhargavi Posinasetty1, Radhika Chikatipalli2, Santenna Chenchula3, Kuttiappan Anitha4, G. Swapna2, G.N.A. Lakshmi2 & Audinarayana Nelavala2 1Department of Clinical Data Management, Prometrika LLC, Cambridge – MA 02140, United States; 2Department of Pharmacology, Sri Venkateswara College of Pharmacy, Chittoor - 517127, A.P., India; 3Department of clinical pharmacology, All India Institute of Medical Bhargavi Posinasetty1, Radhika Chikatipalli2, Santenna Chenchula3, Kuttiappan Anitha4, G. Swapna2, G.N.A. Lakshmi2 & Audinarayana Nelavala2 1Department of Clinical Data Management, Prometrika LLC, Cambridge – MA 02140, United States; 2Department of Pharmacology, Sri Venkateswara College of Pharmacy, Chittoor - 517127, A.P., India; 3Department of clinical pharmacology, All India Institute of Medical 1Department of Clinical Data Management, Prometrika LLC, Cambridge – MA 02140, United States; 2Department of Pharmacology, Sri Venkateswara College of Pharmacy, Chittoor - 517127, A.P., India; 3Department of clinical pharmacology, All India Institute of Medical 833 ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ©Biomedical Informatics (2023) ©Biomedical Informatics (2023) Sciences, Bhopal, India, 462020; 4Amity Institute of Pharmacy, Amity University Madhya Pradesh (AUMP), Gwalior - 474005, Madhya Pradesh, India; Corresponding author Sciences, Bhopal, India, 462020; 4Amity Institute of Pharmacy, Amity University Madhya Prad Pradesh, India; Corresponding author Abstract: Rheumatoid arthritis (RA) is a multisystem autoimmune disease that causes discomfort, synovial membrane inflammation, peripheral joint inflammation, morning stiffness, articular tissue loss, and restricted joint movement. In the present study, we aim to explore the anti- arthritic efficacy of Bombax ceiba ethanolic extract in a Freund's Complete Adjuvant-induced arthritis, in murine model. The hot soxhlet method was used to extract dried aerial components of Bombax ceiba using an ethyl alcohol: water (70:30) ratio. Bombax ceiba ethanolic extract at two doses of 200 and 400 mg/kg was investigated in Wistar rats against Freund's full adjuvant-induced chronic immunological arthritis. Anti-arthritis efficacy was studied utilising morphological research (paw volume, paw diameter, and body weight). On the 28th day, the animals were sacrificed, and haematological parameters, pro-inflammatory cytokines (TNF-alpha, IL-6), cell culture, histological and radiological analysis were performed. BCEE inhibited paw oedema significantly (P 0.05) at a dose of 40mg/Kg, which was corroborated by paw volume and diameter, as well as haematological parameters, in Freund's complete adjuvant-induced arthritis model. The BCEE also significantly reduced pro-inflammatory cytokine levels and the histopathological changes caused by Freund's full adjuvant model. BCEE preserves synovial membranes by enhancing health and has shown a significant anti-arthritic activity. Thus, data confirms the traditional usage of Bombax ceiba for arthritis. Keywords: Bombax ceiba, Freund’s complete adjuvant-induced arthritis, haematological, histopathological and organ wt. Keywords: Bombax ceiba, Freund’s complete adjuvant-induced arthritis, haematological, histopatho and forest dwellers [9]. In the review of published literature on the traditional uses, biological activities, and isolated compounds from Bombax ceiba [10], pharmacological studies confirmed that the crude extracts or individual compounds from the plant showed Anti- diabetic activity, antimicrobial activity, anti urolithiatic activity, anti-inflammatory activities, hepato protective activity, and anti- hyperglycemic activity. Its stem bark extract can lower blood pressure [11]. Stem bark extract has antibacterial and antioxidant properties [12]. Therefore, it is of interest to document the arthritic efficacy of Bombax ceiba ethanolic extract in a murine model for rheumatoid arthritis using in vivo, in vitro and radiological analysis. Freund’s complete adjuvant-induced arthritic rats: Freund’s complete adjuvant-induced arthritic rats: Body wt. studies: Body weight was assessed before and after induction using an electronic balance (Shimadzu C054-E032S, Japan), and the mean difference in body weight was recorded. Measurement of pro-inflammatory cytokines (TNF-alpha and IL- 6): A total of 30 male albino Wistar rats weighing 180-200 g were chosen and divided into five groups of six rats each (n = 6). Group I served as the standard control. Group 2 was utilized for arthritis control, Group 3 for Diclofenac 10 mg/kg, Group 4 for treatment group BCEE 200 mg/kg, and Group 5 for treatment group BCEE 400 mg/kg The serum was isolated from the blood of experimental animals by clotting at room temperature for 30 minutes. TNF-alpha and IL-6 protein concentrations in serum were measured using ELISA kits and the method was carried out according to the manufacturer's instructions [17]. Experimental animals: The experiments were carried out on male Wistar rats weighing between 180 and 200 g. Sree Venkateshwara Enterprises in Bangalore, India provided all of the animals. The Institutional Animal Ethics Committee of Sree Vidyanikethan College of Pharmacy, Tirupati, Chittoor Dist., A.P., India (Approval No.: SVCP/IAEC/I-001/2018-19 dated 01/04/2019) approved all animal experiment protocols following the guidelines of the Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA). The animals were housed in Polypropylene cages and kept at 24°C 2°C on a 12-hour light/dark cycle, fed a normal pellet diet, and had free access to water. Organs wt. studies: Organs wt. studies: g On the 28thday, the spleen and thymus were removed, and the weight of the organs was recorded and corrected for 100 g body wt [16]. ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ©Biomedical Informatics (2023) On the 28th day, 3 mL of blood was obtained through retro-orbital puncture for the estimation of serum parameters (SGOT, SGPT, ALP, and Total protein) and blood parameters (ESR and%HB) using various diagnostic kits and conventional protocols. Later, on the 28th day, rats were euthanized with a strong dosage of halothane, and the ankle joints were used for additional radiological, histological, and organ weight examinations. ratio using the hot Soxhlet technique.The ethanolic extract was concentrated under reduced pressure in a rotary evaporator (Heidolph Instrument, Laborota 4000, Germany). The dried crude Bombax ceiba ethanolic extract (BCEE) was collected and stored at 4- 8 °C in an airtight glass jar until use. p g Paw volume and diameter studies: Rat paw volume and thickness were measured once every 7 days from the day '0' to '28' using a Plethysmometer (UGO Basile, Italy 7140) and a vernier calliper. MTT (3-(4, 5-dimethylthiazolyl-2)-2, 5-diphenyltetrazolium bromide) assay: Freund's complete adjuvant (FCA) comprises 10 mg/mL of heat- killed dead Mycobacterium tuberculosis bacteria in liquid paraffin [14]. Except for the rats in the normal control group, all rats were injected intradermally with 0.1 mL of FCA into the left hind paw on day '0'.The development of arthritis was given a 7-day interval. During this time, all of the animals acquired arthritic symptoms such as swelling, redness, and restricted movement [15]. The treatment came to an end on day 28. ) y The RAW 264.7 murine macrophage cell line was plated into a 96- well plate at a density of 1 104 in 100 L of DMEM media per well and incubated overnight until confluence was reached for treatment the next day. During treatment, the media was changed with different concentrations of plant extracts, including 12.5, 25, 50, 100, 200, 300, and 400 g/mL, and the cells were cultured for another 24 hours. After 24 hours of treatment incubation, 10 L of 5 mg/mL MTT solution was applied. The plate was centrifuged at that time, and the solution was removed, leaving the crystal in the bottom of the plate. The insoluble formazan salt was then dissolved with 100 L dimethyl sulfoxide [18]. The dish was gently stirred and placed in a dark room for about 30 minutes. The formazan produced was quantified using a micro plate reader with an absorbance value of 570nm. Cell Culture: Acute toxicity studies on the BCEE were not done because its safety of up to 200 mg/kg had already been documented [13]. Based on previous research, the two dosages of 200 and 400 mg/kg were chosen. RAW 264.7 cells were cultivated in 100 units/ml penicillin, 100 g/ml streptomycin, Foetal bovine serum, and 10% heat-inactivated at 37 ° C with 5% CO2. Cells were washed and detached using 0.25 percent EDTA-Trypsin in DMEM medium. The cells were re- suspended at a density of 2 106 cells/ml in a DMEM medium. Histopathological studies: The ankle joints were removed on the 28th day and fixed in 10% buffered formalin. The bones were decalcified in 5% formic acid, paraffin-embedded, sectioned at 5 m thickness, and stained with haemotoxylin-eosin before being examined under a light microscope for changes in the synovium, cartilage, and joint space. Extract preparation: The aerial plants were obtained by cutting the root sections, carefully washing them with tap water, air drying them in the shade, powdering them in a grinder, and passing them through Sieve No. 40 (ASTM). The dry powder was defatted with petroleum ether before being extracted with ethyl alcohol: water in a (70:30) 834 ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) Statistical analysis: The results were presented as Mean SEM (n = 6).Thestatistical significance was determined using the student t-test or one-way analysis of variance (ANOVA) followed by Dunnet's test, with P0.05, P0.01, and P0.001 considered statistically significant. On days 7th (p≤0.05), 14th (p≤0.05), 21st (p≤0.01), and 28th (p≤0.01), rats administered with BCEE (40 mg/kg) showed significant paw edema suppression. The paw diameter increased and then dropped somewhat until the 21st day of adjuvant induction. On day 14 (p0.01), 21 (p0.001) and 28 (p0.001) the diclofenac group revealed significant paw diameter inhibition. BCEE (200 mg/kg) inhibits paw diameter significantly on days 21 and 28 with (p0.01). Furthermore, rats treated with BCEE (400 mg/kg) showed a substantial inhibition in paw diameter on days 21 and 28 (p0.01).Table.1 shows the results. CFA-induced arthritic rats had considerably larger paw volumes than normal rats. When compared to arthritic rats, BCEE (low dose: 200mg/kg and high dose: 400 mg/kg b.w.) and diclofenac (10 mg/kg b.w.) treated arthritic rats demonstrated a substantial (p0.05) decrease in paw volume. Results and Discussion: Freund’s complete adjuvant-induced arthritic rats: Figure 1: Photographic analysis (changes in paw diameter) of CFA induced arthritis in Wister rats. Results and Discussion: Freund’s complete adjuvant-induced arthritic rats: Figure 2: Effect of BCEE on Paw diameterin FCA-induced arthritic rats Figure 1: Photographic analysis (changes in paw diameter) of CFA induced arthritis in Wister rats. Paw volume and diameter studies: Figure 2: Effect of BCEE on Paw diameterin FCA-induced arthritic rats Table 1 shows the effect of BCEE on paw volume and diameter in FCA-induced arthritic rats. The challenge with CFA (0.1 mL) results in the development of paw edema, which reaches a peak on the 21st day of injection. On days 7th (P0.05), 14th (P0.01), 21st (P0.001), and 28th (P0.001), the diclofenac-treated group significantly inhibits paw edema. BCEE (200mg/kg) inhibits paw edema significantly on days 21 and 28 with (P0.01). Furthermore, rats treated with BCEE (400 mg/kg) showed significant suppression of paw edema on days 7 (P0.05), 14 (P0.05), 21 (P0.01), and 28 (P0.01). Paw diameter grew until the 21st day of adjuvant induction when it slightly reduced. The diclofenac-treated group has significantly reduced paw diameter on days 14 (P0.01), 21 (P0.001), and 28 (P0.001). BCEE (200 mg/kg) inhibits paw diameter significantly on days 21 and 28 with (P0.01). Furthermore, rats treated with BCEE (400 mg/kg) showed significant inhibition in paw diameter on days 21 and 28 (P0.01). (Figure 1) depicts a morphological examination of rat paws. ©Biomedical Informatics (2023) Bioinformation 19(8): 833-839 (2023) Values are expressed as mean ± SEM (n=6). P<0.05,P<0.01,P<0.001 as compared with Arthritis control. (One-way ANOVA followed by Dunnet’s test). Body weight studies: y g Effect of BCEE on body weight in FCA-induced arthritic rats was tabulated in (Table 2) which indicates the increased body wt. during treatment of standard drugs and BCEE Table 2: Effect of BCEEon body wt. in FCA-induced arthritic rats Groups Mean Body wt. (gm) Mean Difference in Body wt Before Induction After Induction Normal control 178±1.23 178±1.23 -- Arthritis control 168 ± 3.13 188 ± 2.4 12 ± 1.26 Diclofenac 10 mg/kg 176 ± 2.24 207 ± 1.38 19 ± 1.11 BCEE200 mg/kg 174 ± 1.12 203 ± 3.21 17 ± 1.03* BCEE400 mg/kg 178 ± 3.65 208 ± 5.01 13 ± 1.07* Values are expressed as mean ± SEM (n=6); P<0.05, P<0.01 as compared with control followed by Student’s t-test. g/ g Values are expressed as mean ± SEM (n=6); P<0.05, *P<0.01 as compared with control followed by Student’s t-test. Radiological changes: Lower limb X-rays were collected using a Siemens Heliphos D X- ray machine, and joint alterations were measured using joint space and soft tissue swelling [19]. Haematological studies: 835 ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) Pro-inflammatory cytokines (TNF-alpha and IL-6): p ( ) P<0.05,P<0.01,P<0.001.as compared with control (One-way ANOVA followed by Dunnet’s test). y y ( p ) BCEE (400mg/kg) demonstrated a significant (P0.5) effect compared to arthritis control in a study of proinflammatory cytokines. The results revealed that proinflammatory cytokine inhibition was dosage dependent. When compared to BCEE (400mg/kg) and the arthritis control, conventional diclofenac demonstrated a substantial (P0.01) reduction in proinflammatory cytokines. Figure 4 depicts the results of TNF-Alpha and IL6 levels. Histopathological studies: Figure 3: Histopathological observation of the rat ankle tissues (A) Normal control (B) Arthritis control(C) Diclofenac (10 mg/kg)(D) BCEE (200 mg/kg) (E) BCEE (400 mg/kg) treated rats. Magnification: x100; thickness: 5 µm. Histopathological studies: demonstrated a substantial (P0.01) reduction in proinflammatory cytokines. Figure 4 depicts the results of TNF-Alpha and IL6 levels. Figure 4: Effect of BCEE on TNF-Alpha in FCA-induced arthritis rats Figure 5: Effect of on IL6 level in FCA-induced arthritis rats Radiographic study: Figure 6: Radiographic analysis of CFA-induced arthritis in Wister rats. (A) Control group; (B) Negative control; (C) standard group; (D) BCEE (200mg/kg) ; (E) BCEE (400 mg/kg). CFA i d d h i i h d i ibl i lli i Figure 4: Effect of BCEE on TNF-Alpha in FCA-induced arthritis rats Figure 5: Effect of on IL6 level in FCA-induced arthritis rats Figure 4: Effect of BCEE on TNF-Alpha in FCA-induced arthritis rats Figure 3: Histopathological observation of the rat ankle tissues (A) Normal control (B) Arthritis control(C) Diclofenac (10 mg/kg)(D) BCEE (200 mg/kg) (E) BCEE (400 mg/kg) treated rats. Magnification: x100; thickness: 5 µm. Figure 5: Effect of on IL6 level in FCA-induced arthritis rats In the arthritis control group, histopathological studies of the rat ankle joint tissues (Figure 3) show destructive lesions in connective tissue, vascularity into joint space, and granuloma development. The ankle joint in the normal control group had normal connective tissue structure and no necrosis. Diclofenac treatment revealed normal connective tissue in the ankle joint, as well as decreased oedema and a lack of necrosis. By lowering inflammation and necrosis, BCEE-treated rats provided knee joint protection compared to arthritis control group rats. BCEE (200mg/kg)-treated rats developed granuloma, oedema, and necrosis, with little inflammatory cells. BCEE (400mg/kg) treatment resulted in mild necrosis with oedema, but no granuloma in the ankle joint. Radiographic study: Figure 6: Radiographic analysis of CFA-induced arthritis in Wister rats. ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ©Biomedical Informatics (2023) Haemato-logical studies: Table 1: Effect of BCEE on paw volume in FCA-induced arthritic rats Groups Paw volume (mL) Day 0 Day 7 Day 14 Day 21 Day 28 Normal control 0.32±0.0 5 0.0.32±0.0 3 0.33±0.02 0.33±0.06 0.34±0.02 Arthritis control 0.36 ± 0.09 0.56 ± 0.09 1.12 ± 0.06 1.46 ± 0.04 2.09 ± 0.08 Diclofenac 10 mg/kg 0.37 ± 0.03 0.41± 0.21* 0.57±0.19* * 0.43±0.21* 0.39±0.15 * BCEE 20 mg/kg 0.35 ± 0.02 0.48 ± 0.26 0.62 ± 0.16* 0.59 ± 0.12 0.54 ± 0.12* BCEE 40 mg/kg 0.36 ± 0.05 0.44± 0.17* 0.60 ± 0.20 0.52 ± 0.21 0.48 ± 0.23* g Table 3 shows the effect of BCEE on various serum and blood parameters in FCA-induced arthritic rats. In the control group, CFA (0.1 mL) increases the levels of SGOT, SGPT, and ALP while decreasing the amount of total protein. The diclofenac-treated group has lower levels of SGOT (P0.01), SGPT (P0.01), ALP (P0.001), and higher levels of Total protein (P0.01). BCEE (20 mg/kg) has a substantial drop in SGPT (P0.05), ALP (P0.05), and an increase in Total protein (P0.05). The BCEE (400 mg/kg) treated group has lower levels of SGOT (P0.05), SGPT (P0.05), ALP (P0.01), and higher levels of Total protein (P0.05). 836 ©Biomedical Informatics (2023) Table 3: Effect of BCEE on haematological parameters in FCA-induced arthritic rats Groups Serum parameters Blood parameters SGOT (IU/L) SGPT (IU/L) ALP (IU/L) Total Protein (gm/dl) ESR (mm/hr) HB (gm %) Normal control 42.14±1.38 46.72 ± 1.70 139.21± 1.22 8.10 ± 0.09 12.54 ± 0.85 17.09 ± 0.66 Arthritis control 122.27±3.0 1 102.26 ± 1.87 262.36 ± 20.13 4.04 ± 0.22 39.11 ± 5.4 7.4 ± 0.6 Diclofenac 10mg/kg 51.42±2.55 52.18±2.04 ** 141.65±8.27* 7.18± 0.20 17.74±.5.04 15.5±0.81 BCEE200mg/kg 76 ±6.81 73.48 ±1.40* 196.81 ± 8.44* 5.84 ±0.47* 23.34 ± 3.12 11.2 ± 0.9** BCEE400mg/kg 68.93 ± 4.31* 63.63 ± 4.45* 171.18 ± 9.68** 6.42 ± 0.31* 18.26 ± 2.31* 13.3±1.01 ** Values are expressed as mean ± SEM (n=6). P<0.05,P<0.01,P<0.001.as compared with control (One-way ANOVA followed by Dunnet’s test). Organs weight studies: Organs weight studies: The mean thymus weight decreased while the mean spleen weight increased in the FCA-treated animals compared to the NC group (Table 5). When compared to FCA-treated rats, the rise in spleen weight was considerably (P 0.01) suppressed by BCEE (20 and 40 mg/kg) and diclofenac (10 mg/kg). Only treatment with 400 mg/kg of BCEE plus diclofenac significantly reduced thymus weight loss (P 0.01). ©Biomedical Informatics (2023) constriction or pseudo-widening of all joint areas. At the end of the BCEE therapy, rats administered with the usual medication diclofenac sodium showed reduced bone damage and produced arthritic animals. pain, bone and cartilage degradation, and severe impairment [21].Standard drug and hydro-alcoholic Bombax ceiba extract, on the other hand, can significantly suppress paw swelling and decrease paw volume in both acute and chronic phases, which may be due to the suppression of inflammatory mediators released due to Freund's adjuvant induction [22]. Though the exact mechanism of inflammation suppression is unknown, it can be connected with the presence of alkaloids and flavonoids in suppressing inflammation and antioxidant activity [23]. Changes in the body weights of the rats occurred throughout the experimental period as the incidence and severity of arthritis increased. Previous research suggests that absorption of 14C-glucose and 14C-leucine in the intestine of inflamed rats is reduced [24]; however, when treated with anti- inflammatory drugs, the decrease in absorption is nullified, indicating that anti-inflammatory drugs correct the decreased/deranged absorption capacity of the intestine during inflammation [25]. The increased body weight during therapy with the usual medicine, hydro-alcoholic Bombax ceiba extract, could be attributed to the restoration of intestinal absorption ability. The extract has also been shown to have a considerable influence on a variety of blood and serum parameters [26]. Formaldehyde- induced arthritis is a typical acute paradigm for evaluating the anti- arthritic properties of plant extracts [27-29]. Edema in the paw of the rat after injection of 0.1 mL of 2% v/v formaldehyde is caused by the production of histamine, serotonin, and prostaglandin-like chemicals at the injection site [30]. The substantial anti- inflammatory potential of BCEE may explain the inhibition of paw edema in formaldehyde-induced arthritis. MTT assay: Figure 7: Effect of BCEE on the percentage of cell viability in RAW 264. 7 macrophages Figure 7: Effect of BCEE on the percentage of cell viability in RAW 264. 7 macrophages In-vitro cell viability of previously cultivated RAW 264.7 cells was assessed in the presence of BCEE. The concentrations of BCEE were fixed at 10, 20, 40, 80, 160, and 320 g/mL. The results showed that BCEE at 160g/mL and 320g/mL significantly reduced RAW 264.7 viability to 71.23% 1.69% and 67.42%, 2.11%, respectively, as compared to the negative control. ©Biomedical Informatics (2023) Meanwhile, BCEE at concentrations ranging from 10 to 80 g/mL was not harmful to RAW 264.7 cells since it did not reduce cell viability, but rather increased RAW 264.7 cell viability in comparison to the negative control. Our investigation confirmed that BCEE concentration affects cell viability by increasing or lowering it, which can be useful in determining the ideal dose. Conclusion: Th l The results of the current study have shown that, the total flavonoid fraction (BCEE) of BCEE has shown to have promising anti-arthritic effects by lowering pro-inflammatory cytokine levels and maintaining spleen and thymus weight. We conclude that, BCEE can be used as a potential therapeutic agent to contro inflammation in both acute and chronic arthritic population. Conflicts of interests: Conflicts of interests: The authors have declared no conflicts of interest. Pro-inflammatory cytokines (TNF-alpha and IL-6): (A) Control group; (B) Negative control; (C) standard group; (D) BCEE (200mg/kg) ; (E) BCEE (400 mg/kg). Table 4: Effect of BCEE on thymus and spleen wtin FCA-induced arthritic rats Groups Spleen wt. (mg/100 g b.wt.) Thymus wt. (mg/100 g b.wt.) Normal control 189.53±3.12 100.5±1.01 Arthritis control 259.34±3.61 71.18±2.34 Diclofenac 10 mg/kg 199.83±4.20 91.00±1.46 BCEE 200 mg/kg 224.50±2.36 83.50±1.43 BCEE 400 mg/kg 210.00±2.34 85.75±1.53 Values are expressed as the mean ± SEM (n= 6); P<0.05, P<0.01 as compared with control (One-way ANOVA followed by Dunnet’s test). Table 4: Effect of BCEE on thymus and spleen wtin FCA-induced arthritic rats Groups Spleen wt. (mg/100 g b.wt.) Thymus wt. (mg/100 g b.wt.) Normal control 189.53±3.12 100.5±1.01 Arthritis control 259.34±3.61 71.18±2.34 Diclofenac 10 mg/kg 199.83±4.20* 91.00±1.46 BCEE 200 mg/kg 224.50±2.36 83.50±1.43 BCEE 400 mg/kg 210.00±2.34 85.75±1.53 Figure 6: Radiographic analysis of CFA-induced arthritis in Wister rats. (A) Control group; (B) Negative control; (C) standard group; (D) BCEE (200mg/kg) ; (E) BCEE (400 mg/kg). Values are expressed as the mean ± SEM (n= 6); P<0.05, *P<0.01 as compared with control (One-way ANOVA followed by Dunnet’s test). CFA-induced arthritic rats had visible tissue swelling, cystic expansion of bone, and widespread erosions, resulting in 837 ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) ©Biomedical Informatics (2023) Discussion: Th The most popular model is Freund's complete adjuvant (FCA) produced arthritis in rats. This preclinical model predicted the actions of a variety of drugs that are currently being investigated in clinical trials for the treatment of rheumatoid arthritis. The four phases of arthritis are based on biochemical parameters: Phase 1: Acute local inflammation of the liver and systemic effects (1-4 days); Phase 2: Remission of acute inflammation and periarthritis (7-12 days); Phase 3: Chronic inflammation, periarteritis, and osteogenic activity (12-28 days); and Phase 4: Permanent articular deformity and minimal inflammation (35 days and up). The purpose of this study was to determine the efficacy of a hydro- alcoholic extract of aerial portions of the indigenous herb Bombax ceiba against arthritis. Male Wistar strain rats were chosen for the current investigation because they develop persistent swelling in many joints due to the accumulation of inflammatory cells, erosion of joint cartilage, and bone deterioration. It closely resembles human rheumatoid illnesses [20]. The measurement of paw swelling appears to be a straightforward, sensitive, and rapid approach for assessing the level of inflammation and the therapeutic effects of medications. Freund's Adjuvant model was chosen because it causes chronic swelling in many joints due to the effect of inflammatory cells, resulting in joint cartilage erosion and bone deterioration. A variety of mediators, including cytokines (IL- 1B and TNF-alpha), GM-CSF, interferons, and PGDF, are released during chronic inflammation. These mediators are responsible for Acknowledgements: The authors are thankful to the Principal and Management of Sree Vidyanikethan College of Pharmacy, Tirupati for providing the necessary infrastructure and facilities to conduct this research work. References: References: [1] Elisha IL et al. BMC Complement Altern Med. 2016 16:307. [PMID: 27554099] [2] Ondua M et al. J Ethnopharmacol. 2019 234:27. [PMID: 30572091] [3] Ahmed AS et al. J Ethnopharmacol. 2014 154:339. [PMID: 24681040] [4] Serafini M et al. Proc Nutr Soc. 2010 69:273. [PMID: 20569521] [5] Jin JH et al. J Ethnopharmacol. 2010 127:589. [PMID: 20034551] [1] Elisha IL et al. BMC Complement Altern Med. 2016 16:307. [PMID: 27554099] [2] Ondua M et al. J Ethnopharmacol. 2019 234:27. [PMID: 30572091] [3] Ahmed AS et al. J Ethnopharmacol. 2014 154:339. [PMID: 24681040] [4] Serafini M et al. Proc Nutr Soc. 2010 69:273. [PMID: 20569521] [5] Jin JH et al. J Ethnopharmacol. 2010 127:589. [PMID: 20034551] 838 ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) Bioinformation 19(8): 833-839 (2023) ISSN 0973-2063 (online) 0973-8894 (print) ©Biomedical Informatics (2023) ©Biomedical Informatics (2023) Bioinformation 19(8): 833-839 (2023) [6] Shang JH et al. J Ethnopharmacol. 2010 129:174. [PMID: 20219658] [7] Kamble MA et al. J Pharm Pharmacogn Res 2017.5: 40-54. [8] Chakraborty DD et al. Int J Pharm Sci Res 2010. 1: 66. [9] Chaudhary PH et al. Syst. Rev. Pharm. 2019 10:20. [10] R K Shukla et al. Research J. Pharm. and Tech. 2020 13:5607. [11] Saleem R et al. Planta Med. 1999 65: 331. [12] Masood UR et al. Afr J Tradit Complement Altern Med. 2017 14:9. [PMID: 28573217] [13] H i E l J Ph R 2011 10 55 [6] Shang JH et al. J Ethnopharmacol. 2010 129:174. [PMID: 20219658] [19] Bondonno NP et al. Nat. Commun. 2019 10:3651. [PMID: 31409784] ] [20] Saeed N et al. Complement Altern Med. 2012 12:221. BMC . [PMID: 23153304] [7] Kamble MA et al. J Pharm Pharmacogn Res 2017.5: 40-54. g [8] Chakraborty DD et al. Int J Pharm Sci Res 2010. 1: 66. [21] Waghole RJ et al. Inflammopharmacol.2022 30 :291. [PMID: 35064430] [9] Chaudhary PH et al. Syst. Rev. Pharm. 2019 10:20. [22] Apaza Ticona et al. Rev. Bras. Farmacogn. 2021 31: 429. [11] Saleem R et al. Planta Med. 1999 65: 331. 1] Saleem R et al. Planta Med. 1999 65: 331. [12] Masood UR et al. Afr J Tradit Complement Altern Med. 2017 14:9. [PMID: 28573217] [23] Tang Y et al. Sci Rep. 2021 11: 17971. [PMID: 34504248] [24] Chang X et al. Chem. Biol. Interact. 2015 236:41. [PMID: 25935278] [13] Hossain E et al. J Pharm Res. 2011 10:55. [25] Gutiérrez-Rebolledo G.A et al. Rev. Bras. Farmacogn. 2018 28:198 [14] Hasan AU. Brazilian Journal of Pharmaceutical Sciences. 2019 20:55. [15] Panichayupakaranant P et al. J Ethnopharmacol. 2013 148:901. [PMID: 23743057] [26] Pandurangan K et al. Inflammo pharmacol. 2015 23: 307 . [PMID: 36670885] [ ] y p J p 148:901. [PMID: 23743057] [16] Tuñón MJ et al. Curr Drug Metab. 2009 10:256. [PMID: 19442088] [27] Santenna C et al. Journal of Experimental Pharmacology. 2019 2:85. [PMID: 31447593] [17] Waghole R.J et al. Inflammopharmacol 2022: 291. [PMID: 35064430] [28] Sadasivam B et al. Bioinformation. 2022 18:752. [PMID: 37426507] [18] Benavente-García O et al. J Agric Food Chem. 2008 56:6185. [PMID: 18593176] [29] Chenchula S et al. Int J Appl Res. 2015 1:198. [30] Chaves OA et al. International Journal of Biological Macromolecules 2022 222:1015. [PMID: 36183752] 839
https://openalex.org/W4243398791	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0127604&type=printable	English	null	Correction: Accurate Prediction of Severe Allergic Reactions by a Small Set of Environmental Parameters (NDVI, Temperature)	PloS one	2,015	cc-by	253	Correction: Accurate Prediction of Severe Allergic Reactions by a Small Set of Environmental Parameters (NDVI, Temperature) George Notas, Michail Bariotakis, Vaios Kalogrias, Maria Andrianaki, Kalliopi Azariadis, Errika Kampouri, Katerina Theodoropoulou, Katerina Lavrentaki, Stelios Kastrinakis, Marilena Kampa, Panagiotis Agouridakis, Stergios Pirintsos, Elias Castanas The following information is missing from the Funding section: This work was partially funded by the European Union Programs Regional Potential/Translational Potential Grant 285948 (to GN and EC). OPEN ACCESS Citation: Notas G, Bariotakis M, Kalogrias V, Andrianaki M, Azariadis K, Kampouri E, et al. (2015) Correction: Accurate Prediction of Severe Allergic Reactions by a Small Set of Environmental Parameters (NDVI, Temperature). PLoS ONE 10(4): e0127604. doi:10.1371/journal.pone.0127604 1. Notas G, Bariotakis M, Kalogrias V, Andrianaki M, Azariadis K, Kampouri E, et al. (2015) Accurate Pre- diction of Severe Allergic Reactions by a Small Set of Environmental Parameters (NDVI, Temperature). PLoS ONE 10(3): e0121475. doi: 10.1371/journal.pone.0121475 PMID: 25794106 Reference 1. Notas G, Bariotakis M, Kalogrias V, Andrianaki M, Azariadis K, Kampouri E, et al. (2015) Accurate Pre- diction of Severe Allergic Reactions by a Small Set of Environmental Parameters (NDVI, Temperature). PLoS ONE 10(3): e0121475. doi: 10.1371/journal.pone.0121475 PMID: 25794106 a1111 CORRECTION Published: April 29, 2015 Copyright: © 2015 Notas 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. 1 / 1 PLOS ONE \| DOI:10.1371/journal.pone.0127604 April 29, 2015 PLOS ONE \| DOI:10.1371/journal.pone.0127604 April 29, 2015
https://openalex.org/W2578339347	https://bmcgeriatr.biomedcentral.com/track/pdf/10.1186/s12877-017-0421-8	English	null	The use of physical restraints in long-term care in Spain: a multi-center cross-sectional study	BMC geriatrics	2,017	cc-by	6,412	© 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. * Correspondence: gabriel.estevez@ulpgc.es 1Nursing Department, University of Las Palmas de Gran Canaria, Tahiche, Spain 6Unidad Docente de Enfermería de Lanzarote. Facultad de Ciencias de la Salud. ULPGC, C/Rafael Alberti 50, Tahiche 35507, Spain Full list of author information is available at the end of the article Abstract Background: Physical restraint is a procedure used frequently in long-term care. It is a controversial practice because its use is associated with numerous complications and also affects freedom and individual autonomy. The objective of this study was to examine the use of physical restraint of long-term care residents with the ability to move voluntarily. Methods: We conducted a cross-sectional observational and correlational multi-center study. Nine centers agreed to participate. Of the 1,200 people present at the time of data collection, those without voluntary movement or in the facility for less than a month were excluded. Thus, the final sample was 920 residents. Data on the use of restraints was collected by direct observation. Information about the age, gender, length of stay, falls, mobility, cognition and functional status of residents was gathered by reviewing clinical records and interviewing nursing staff. A descriptive analysis of the data obtained was conducted. The generalized linear model was used, considering only the principal effects of each variable and using the logit link function. The model has been adjusted for clusters and for other possibly confounding factors. For all analyses, a confidence interval (CI) of 95% was estimated. Results: The prevalence of residents with at least one physical restraint was 84.9% (95% CI: 81.7–88.1), with variability between centers of 70.3 to 96.6% (p-value Kruskal Wallis test <0.001). Full-enclosure side rails were most often used (84.5; 95% CI: 81.1–87.9), but other types of restraints were also used frequently. Multivariate analysis showed that the degree of functional impairment increased the probability of the use of restraint. A significant association was also found between restraint use and the impaired cognitive status of residents. Conclusions: The prevalence was higher than in studies from other countries. The results emphasize the need to improve the training of nursing staff in the care of residents with impairments in functional and cognitive status. The use of alternative devices and nurse consultants need to be evaluated, and the introduction of specific laws considered. Keywords: Aged, Long-term care, Restraint, physical The use of physical restraints in long-term care in Spain: a multi-center cross-sectional study Gabriel J. Estévez-Guerra1,6* , Emilio Fariña-López1, Eduardo Núñez-González2, Manuel Gandoy-Crego3, Fernando Calvo-Francés4 and Elizabeth A. Capezuti5 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 DOI 10.1186/s12877-017-0421-8 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 DOI 10.1186/s12877-017-0421-8 Setting Information regarding the use of physical restraints in Spanish long-term care centers is scarce [16, 17]. This is a moderately aged country with 18.2% of the population 65 years and older. The resident profile is predominantly female, with a mean age of 81 years. In 2014, there were approximately 371,000 long-term care residents, 71% of whom were dependent [18]. All the public centers in the Canary Islands, Spain, with more than 80 beds assigned to long-term care were invited, by letter, to participate. All of them accepted, comprising a total of 1,238 beds in 30 units within nine centers. These centers had an average size of 153.5 ± 66.6 (ranging from 88 to 285 beds); one of them had a dementia care unit and two operated a protocol regulating the use of physical re- straint. The objectives, methodology and possible benefits of the study were explained to the center managers; a nurse was nominated by each center for the coordination of the study and to facilitate access to patient information. Data were collected between July 2013 and September 2014 due to the geographical dispersion of the islands. Spanish long term care centers comprise a wide range of institutions regarding ownership status, size, resident characteristics and the buildings themselves (eg the de- sign, architectural barriers, dimensions and distribution of the rooms and communal spaces) [18, 19]. There are also significant differences related to personnel ratios, which, on occasion, are insufficient; training is some- times poor and staff not always sufficiently qualified [20]. The centers are usually public, subsidized or privately owned, with 54% of all these receiving government funding [18]. The study was focused on public centers, as they tend to have similar characteristics. Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Page 2 of 7 Page 2 of 7 factors for using physical restraint [5]. Attitudes and training of professionals, staffing ratios, legal regulations and routine practices also influence its use [2, 10]. factors for using physical restraint [5]. Attitudes and training of professionals, staffing ratios, legal regulations and routine practices also influence its use [2, 10]. Accordingly, the objective of this study was to examine the prevalence rate of physical restraint of long-term care residents with the ability to move voluntarily. The association of physical restraint with demographic and clinical resident characteristics was also explored. According to international publications, the prevalence in the use of these devices in long-term care varies greatly among countries, ranging from 6 to 68% [1, 2, 11–13]. This large variability could be explained by the different definitions of physical restraint, data collection techniques or characteristics of the facility [14]. The most commonly used devices are full-enclosure side rails, belts and chairs with an attached table [2, 12, 15]. Study design A cross-sectional observational and correlational multi- center study was conducted. Participants The study population consisted of all residents who were present in the centers on the day of data collection. Those residents living in the center for less than a month were excluded from the study, as well as resi- dents with no voluntary movement. There is currently no national legislation regarding the use of physical restraint and thus the regulation of this practice is based on regional laws. In spite of this, legal regulations in the Canary Islands are similar to those of other regions. In general, these are undeveloped, allow- ing the use of restraint to protect the resident from harming himself or others. As a result, its application is at the discretion of professionals [21]. Background pressure sores, loss of muscle tone and reduced mobility are among those that have been reported [5]. It also increases the risk of accidents, which, in some cases, have resulted in the death of the patient [6, 7]. Negative psycho- logical effects, such as frustration, fear, loss of dignity, anger, aggression and decline in social interaction, have also been identified [8, 9]. Apart from that, physical restraints are considered a violation of the autonomy and freedom of the individual, and their right to take risks [4]. Physical restraint is used frequently in long-term care, mainly to prevent falls and, to a lesser extent, to reduce behavioural disturbances or avoid interference with pa- tient treatments [1, 2]. Physical restraint is a controversial practice throughout the world [3, 4]. Its use is associated with numerous adverse effects: contractures, incontinence, Some characteristics of the residents, such as their reduced ability to perform the activities of daily living, cognitive impairment or a previous history of falls, are contributing Variables and data collection In this study physical restraint was defined as “Any action or procedure that prevents a person’s free body movement to a position of choice and/or normal access to his/her body by the use of any method that is attached or adjacent to a person’s body and that he/she cannot control or remove easily” [22]. p An important issue in prevalence studies is when to de- termine if a device should be considered a physical re- straint. Full-enclosure side rails, for example, can be used to limit the freedom of movement of a person, preventing them from getting out of bed and thus, according to a re- cent international definition, are physical restraints [22]. Conversely, the use of full-enclosure side rails with patients suffering severe functional impairment who are unable to move voluntarily should not be considered as restraint. Other norms, such as US nursing home regulations, similarly do not consider full-enclosure side rails to be a restraint in these situations [23]. Our intent is to quantify the use of physical restraints and not simply to count the number of devices used. Using the latter method may lead to overestimating the number of devices used and might not give a true count of physical restraint use. Information on the use of physical restraint was col- lected by direct observation [12, 24, 25], as this method seems to be the most reliable [26]. Two trained nurse researchers visited the units on four separate occasions during the same day. The observations were conducted on weekdays at times when residents were either most likely to be active (10:00, 17:00) or when at rest (14:00, 21:00). Visits were determined randomly and not communicated to the facility staff or administrators. Members of the nursing staff accompanied the investiga- tors when the residents were in their own rooms. Prevalence was defined as the percentage of patients with at least one physical restraint documented during any of the observations. Page 3 of 7 Page 3 of 7 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 At the end of the study, the validity of the observations between the two investigators was assessed and the inter- rater reliability, using a random sample of 89 residents who were not part of the study, was found to be high (Cohen’s Kappa = 0,977; 95% Confidence Interval [CI]: 0,931 to 1). Ethical considerations h l The protocol was approved by the Ethics Committee for Human Research of the University of Las Palmas de Gran Canaria (CEIH 2013–10). Prior to data collection, verbal consent of the resident, or in cases of cognitive disorder, authorization of the family was obtained. In order to pre- serve the confidentiality of the data, the names of centers and residents were converted to numerical codes and only the investigators had access to the list, which was stored in a password-protected file. Mobility was measured using the items “activity” and “mobility” of the Braden Scale [28]. Residents’ mobility was graded between 2 points (no mobility) and 8 points (high mobility). Those with less than 4 were not included in the analysis [15]. Cognitive state data were extracted, when available, from the clinical record of each resident. This information was derived from the Mini Mental Status Examination (MMSE) [29]. It was categorized as severe, moderate, mild or intact. All the centers used the version validated by Lobo et al. [30]. Sample size No assumptions were made concerning prevalence of restraint use because of the scarcity of data available about Spain. Usually, an intra-cluster correlation coeffi- cient (ICCC) of between 0.01 and 0.02 is used in human studies [31]; in this work, an ICCC of 0.03 was applied to increase the sample size and decrease its error. As a result, the sample size would be 1,152 individuals (95% CI) in 9 centers. Although 1,200 participants were ini- tially recruited, the final sample, after applying exclusion criteria, comprised 920 residents. The cluster-adjusted prevalence of residents with at least one physical restraint was 84.9% (95% CI: 81.7–88.1), with variability between centers of 70.27 to 96.55% (p-value Kruskal Wallis test <0.001). When full-enclosure side rails were not included, the cluster-adjusted prevalence was 36.6% (95% CI: 33.2–40.0). Table 3 showed that the devices most used were full- enclosure side rails (84.5; 95% CI: 81.1–87.9), followed by belts in chair (26.9; 95% CI: 23.5–30.3) and belts in bed (9.9; 95% CI: 6.5–13.3). Results Data were obtained from 920 residents in 30 units within the nine centers. Since each resident received 4 observations, there were a total of 3,680 observations. The assessment of the reasons that led to the applica- tion of restraint was based on a review of the previous six months of clinical records. Given the paucity of these records, this data had to be supplemented with informa- tion provided by the registered nurses. The mean age of the study participants was 80.1 ± 11.09 years. Most were women (63.22%). Overall, 47.44% of the residents presented with total functional impairment and 41.76% with severe cognitive impairment (Table 1). People who were restrained were older (80.7 v. 76.1 years) and their length of stay in the centers was lower. They showed greater functional and cognitive deterioration than those who were not. They also had less mobility (Table 2). Variables and data collection restraints was adjusted for cluster, estimated the corres- ponding intra-cluster correlation coefficient (ICCC) and the design factor (DF). An explanatory model of physical restraints use, quanti- fying the probable associated factors (odds-ratio/adjusted odds-ratios), was estimated. The generalized linear model was used, considering only the principal effects of each variable and using the logit link function. The model has been adjusted for clusters and for other possibly con- founding factors (age and length of stay), although this did not improve the model. For all analyses, a confidence interval (CI) of 95% was estimated. These two researchers also examined the clinical records in order to obtain information regarding age, gender, length of stay, falls and cognition of the resi- dents. The registered nurses most knowledgeable of the resident evaluated functional status and mobility. The first of these was determined using the Barthel Index. The overall range could vary between 0 and 100, and the staff’s perception of performance was categorized in dif- ferent degrees of dependency according to their score: “Total” if <20, “Severe” if between 20 and 35, “Moderate” if between 40 and 55, “Mild” if ≥60, and “Independent” if the patient scored 100 (90 if using a wheelchair) [27]. Data analysis fKruskal-Wallis test Table 2 Bivariate relations between the use of physical restraints and characteristics of residents (n = 920)a use of restraint. A significant association was also found between restraint use and the impaired cognitive status of residents (Table 4). aNot cluster-adjusted. bMedian and interquartile range. cn = 25 missings not restrained and 143 restrained. dWilcoxon test. eChi-square test. fKruskal-Wallis test Data analysis The review of the clinical records and staff interviews confirmed that the major reason for the use of restraint was to prevent falls from a bed or a chair (94.2%). The use of side rails was rarely documented in the clinical notes. A descriptive analysis of the data obtained was conducted. Categorical variables were summarized in frequencies; numerical variables in mean and standard deviation. For comparison of distributions, the Wilcoxon test was used because the normality assumption was not met. The relationships for categorical variables were evaluated using the chi-square test and for multiple comparisons the non- parametric Kruskal-Wallis test was used. Prevalence of A model that integrates the variables described and the use of the restraints was used. The pattern of odds ratios, adjusted for clusters, indicated that the degree of functional impairment increased the probability of the Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Page 4 of 7 Table 1 Characteristics of residents (n = 920)a Variables Age (years) 80.1 (11.09) Length of stay (years) 4.22 (4.95) Gender Women 581 (63.22%) Men 339 (36.78%) Mobilityb 5 (±2.0) Fall previous 6 month 156 (16.9%) Dependence of ADL Total 436 (47.44%) Severe 175 (19.04%) Moderate 156 (16.86%) Mild 135 (14.69%) Independent 18 (1.95%) Cognitive impairmentc Severe 314 (41.76%) Moderate 213 (28.24%) Mild 104 (13.92%) Intact 91 (11.78%) Non-assessable 31 (4.28%) aNot cluster-adjusted. bMedian and interquartile range. cn = 167 missings Table 2 Bivariate relations between the use of physical restraints and characteristics of residents (n = 920)a Not restrained (n = 123) Restrained (n = 797) p-value Age (years) 76.10 (12.52) 80.70 (10.73) <0.001d Length of stay (years) 6.36 (6.38) 3.88 (4.6) <0.001d Gender Women 55 (44.71%) 526 (66.08%) <0.001e Men 68 (55.29%) 271 (33.92%) Mobilityb 8 (±2.0) 5 (±2.0) <0.001e Fall previous 6 month Yes 32 (26.0%) 124 (15.5%) <0.004e No 91 (74.0%) 673 (84.5%) Dependence of ADL Total 3 (2.44%) 433 (54.39%) <0.001f Severe 9 (7.31%) 166 (20.85%) Moderate 33 (26.83%) 123 (15.32%) Mild 62 (50.40%) 73 (9.17%) Independent 16 (13.01%) 2 (0.25%) Cognitive impairmentc Severe 15 (15.46%) 299 (45.69%) <0.001f Moderate 28 (28.86%) 185 (28.15%) Mild 17 (17.52%) 87 (13.38%) Intact 35 (35.05%) 56 (8.30%) Non-assessable 3 (3.09%) 28 (4.46%) aNot cluster-adjusted. bMedian and interquartile range. cn = 25 missings not restrained and 143 restrained. dWilcoxon test. eChi-square test. Discussion BMC Geriatrics (2017) 17:29 Page 5 of 7 Page 5 of 7 Table 4 Explanatory model of physical restraints use (cluster-adjusted) Factor/variable AOR (95% CI) p-value Age (years) 0.01 (0.00–0.04) 0.436 Length of stay (years) 0.001 (0.00–0.06) 0.979 Fall previous 6 mouth (yes/non) 0.49 (0.00–1.30) 0.236 Gender (woman) 0.86 (0.15–1.58) 0.018 Mobility (reference: high mobility = 8) 4 1.28 (0.17–2.74) 0.084 5 1.11 (0.19–2.41) 0.094 6 0.63 (0.40–1.65) 0.229 7 0.05 (0.92–1.01) 0.926 Dependence of ADL (reference: mild/indep.) Moderate 0.88 (0.11–1.65) 0.025 Severe 2.95 (1.75–4.15) <0.001 Total 4.66 (2.93–6.40) <0.001 Cognitive impairment (reference: intact) Mild 0.88 (0.17–1.93) 0.101 Moderate 1.32 (0.39–2.26) 0.006 Severe 1.52 (0.48–2.55) 0.004 It was observed that full-enclosure side rails were the most commonly used restraints, as reported in other studies [1, 2, 12, 13, 15]. Professionals often believe they are essential in the prevention of accidents and in the reduction of potential legal liability [40]. However, there is growing evidence to suggest that these devices may not be so benefi- cial, especially since they have been associated with fatal injuries [45]. Another important observation made in this study is that their use is not routinely recorded in the patient’s history; this suggests that nursing staff do not consider them as a form of restraint [36], which may help to explain their frequent use. However, full-enclosure side rails, when limiting the freedom of movement of the per- son, should be considered as a restraint. In other countries with more stringent regulations, documentation, including the rationale for their usage, is mandated [6]. Therefore, we believe that decisions about the use of physical restraint should be based on an individualized assessment that takes into account the benefits and risks, as well as possible damage to the dignity and autonomy of the person [4, 38]. Furthermore, their implementation should be agreed between professionals, residents and family mem- bers, documented and regularly reviewed [4]. g y Regarding the relationship between the variables and the application of physical restraint, the multivariate ana- lysis showed that low functional and cognitive status was associated with the use of restraint; these results are con- sistent with those reported in the literature [1, 12, 13, 25]. This loss of capacity may mean nurses see residents as fragile and dependent, to be protected from all risk, a perception that would lead to a pro-restraint attitude [36, 46]. Discussion common justification for the use of restraint, as also found in studies from other countries [1, 2]. However, there is in- sufficient evidence to support this reasoning and, in fact, it has been found that patients continue to fall despite the restraint [37, 38] and, moreover, their use may increase the risk of serious injury or even death [6, 7, 39]. The data showed that the use of physical restraints in the centers studied was higher than other studies using direct observation (84.9%). The highest usages reported in the lit- erature were in Northern Ireland, Canada and Taiwan, with values between 62 and 68% [1, 26, 32], and the lowest in Germany with 26.2% [12]. The Netherlands was situated between them, with a prevalence of around 50% [24, 25]. When the use of full-enclosure side rails was excluded, the prevalence dropped to 36.6%. Compared with other studies using a similar methodology, this rate was higher than those recorded in Canada (33.7%) and Singapore (23.3%) [26, 33], although it was far higher than that re- corded in Germany, where belts and other restraints are very rare [12]. Table 3 Frequency of physical restraints Cluster-adjusted prevalence (95% CI) ICCC DF Residents with at least one physical restraint 84.9% (81.7–88.1) 0.124 12.52 Full-enclosure side rails 84.5% (81.1–87.9) 0.137 13.60 Belt in chair 26.9% (23.5–30.3) 0.046 5.23 Belt in bed 9.9% (6.5–13.3) 0.050 5.60 Chair with attached table 6.2% (2.8–9.6) 0.044 5.05 Vest restraint 6.1% (2.7–9.5) 0.045 5.14 Wrist/ankle belt 1.2% (0–4.6) 0.052 5.78 Sleep suits 1.2% (0–4.6) 0.052 5.78 Values are cluster-adjusted percentages and intra-cluster correlation coefficients (ICCC). DF = Design factor Table 3 Frequency of physical restraints Table 3 Frequency of physical restraints There are several possible explanations for these differ- ences. The attitude of nursing staff towards the use of physical restraint is an important contributing factor since it will influence the process of decision-making [3]. While nurses often have negative feelings towards this procedure, in practice they use it whenever they consider it necessary, especially to address patient safety concerns [3, 34–36]. This finding was confirmed by reviewing clinical records and interviewing staff: the risk of falls was the most Estévez-Guerra et al. Discussion But it might also demonstrate that they do not have alternatives to the use of restraint in the care of this group of residents, and have to sacrifice the autonomy of these people for their own security [47]. However, it must be remembered that the limitation of movement that characterizes physical restraint is, in effect, increasing a patient’s disability [37, 48]. For this reason, it is important to provide support and training to nurses when caring for residents with these characteris- tics, especially those aspects of training that could contrib- ute to reducing the use of these procedures. Another aspect that could be related to the high prevalence is that there is very little training in fall pre- vention or the management of behavioral and psychological problems for professionals [34, 36, 40], leading them to see restraint as one of the few alterna- tives available. Previous studies also detected a lack of awareness of the potential complications [34, 36, 40], so that staff may erroneously overestimate the benefits of these devices. Teaching programs addressing these issues, supported by other measures, such as the implementation of technical assistance, changes in institutional culture, or support from a specialist consultant, have been found effective in reducing their usage [16, 41, 42]. The absence of effective national legislation regulating the use of physical restraint in long-term care in Spain [21] could be another factor. In countries where the law signifi- cantly limits the use of these devices, Germany for example, the prevalence is much lower than in Spain, and belts are little used [12, 15]. In the USA, the introduction of the Nursing Home Reform Act (OBRA, 1987), and subsequent regulations, reduced the use of restraints, excluding side rails, from more than 30% to less than 10% [43, 44]. At present, some Spanish regions have begun to develop a legal framework to limit the use of restraint, both physical and chemical [21]. Funding S d d Sociedad Canaria de Geriatría y Gerontología and the Fundación Canaria de Investigación Sanitaria financed this study. The sponsors played no role in the design, methods, participant recruitment, data collection, analysis or preparation of this paper. 9. Saarnio R, Isola A. Use of physical restraint in institutional elderly care in Finland: perspectives of patients and their family members. Res Gerontol Nurs. 2009;2(4):276–86. doi:10.3928/19404921-20090706-02. 10. Lane C, Harrington A. The factors that influence nurses’ use of physical restraint: a thematic literature review. Int J Nurs Pract. 2011;17(2):195–204. doi:10.1111/j.1440-172X.2011.01925.x. Author details 1 1Nursing Department, University of Las Palmas de Gran Canaria, Tahiche, Spain. 2Clinic Sciences Department, University of Las Palmas de Gran Canaria, Tahiche, Spain. 3Nursing Department, University of Santiago de Compostela, Santiago de Compostela, Spain. 4Nursing Department, University of Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, Spain. 5William Randolph Hearst Foundation Chair in Gerontology, Hunter College of the City University of New York, New York, NY, USA. 6Unidad Docente de Enfermería de Lanzarote. Facultad de Ciencias de la Salud. ULPGC, C/Rafael Alberti 50, Tahiche 35507, Spain. Abbreviations f 5. Hofmann H, Hahn S. Characteristics of nursing home residents and physical restraint: a systematic literature review. J Clin Nurs. 2014;23(21–22):3012–24. doi:10.1111/jocn.12384. ADL: Activities of daily living; AOR: Adjusted odds ratio; CI: Confidence interval; DF: Design factor; eg (exempli gratia): for example; ICCC: Intra-cluster correlation coefficients; MMSE: Mini Mental Status Examination; n: Number of characteristic values; US: United States 6. Berzlanovich AM, Schöpfer J, Keil W. Deaths due to physical restraint. Dtsch Arztebl Int. 2012;109:27–32. doi:10.3238/arztebl.2012.0027. 7. Fariña-López E, Camacho-Cáceres A, Estévez-Guerra G, Bros-Serra M. Accidentes asociados al uso de restricciones físicas en ancianos con trastornos cognitivos: estudio de tres casos [Accidents associated with the use of physical restraints in the elderly with cognitive disorders: a study of three cases]. Rev Esp Geriatr Gerontol. 2009;44(5):262–5. doi:10.1016/j.regg.2009.03.010. 7. Fariña-López E, Camacho-Cáceres A, Estévez-Guerra G, Bros-Serra M. Accidentes asociados al uso de restricciones físicas en ancianos con trastornos cognitivos: estudio de tres casos [Accidents associated with the use of physical restraints in the elderly with cognitive disorders: a study of three cases]. Rev Esp Geriatr Gerontol. 2009;44(5):262–5. doi:10.1016/j.regg.2009.03.010. Authors’ contributions GJE-G and EF-L conceived and designed the study with the assistance of the other authors; they were also responsible for the data collection. EN-G per- formed statistical analysis. All authors undertook the analysis and interpretation of the findings. GJE-G, EF-L and EC were responsible for the finalization of this manuscript. All authors commented on drafts and approved the final version. 12. Meyer G, Kopke S, Haastert B, Muhlhauser I. Restraint use among nursing home residents: cross-sectional study and prospective cohort study. J Clin Nurs. 2009;18(7):981–90. doi:10.1111/j.1365-2702.2008.02460.x. 13. Hofmann H, Schorro E, Haastert B, Meyer G. Use of physical restraints in nursing homes: a multicentre cross-sectional study. BMC Geriatr. 2015;15:129. doi:10.1186/s12877-015-0125-x. Availability of data and materials The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request. 11. Feng Z, Hirdes JP, Smith TF, Finne-Soveri H, Chi I, Du Pasquier J-N, et al. Use of physical restraints and antipsychotic medications in nursing homes: a cross-national study. Int J Geriatr Psychiatry. 2009;24(10):1110–8. doi:10.1002/gps.2232. 11. Feng Z, Hirdes JP, Smith TF, Finne-Soveri H, Chi I, Du Pasquier J-N, et al. Use of physical restraints and antipsychotic medications in nursing homes: a cross-national study. Int J Geriatr Psychiatry. 2009;24(10):1110–8. doi:10.1002/gps.2232. Ethics approval and consent to participate The protocol was approved by the Ethics Committee for Human Research of the University of Las Palmas de Gran Canaria (CEIH 2013–10). Prior to data collection, verbal consent of the resident, or in cases of cognitive disorder, authorization of the family was obtained. In order to preserve the confidentiality of the data, the names of centers and residents were converted to numerical codes and only the investigators had access to the list, which was stored in a password-protected file. Finally, the data were collected uniquely in the Canary Islands, but it is probable that the results would be similar throughout Spain, given that these types of public institu- tion have similar characteristics and operate under similar legislation regarding the use of physical restraint. References 1 H 1. Huang H, Huang Y, Lin K, Kuo Y. Risk factors associated with physical restraints in residential aged care facilities: a community-based epidemiological survey in Taiwan. J Adv Nurs. 2014;70(1):130–43. doi:10.1111/jan.12176. 2. Chiba Y, Yamamoto‐Mitani N, Kawasaki M. A national survey of the use of physical restraint in long-term care hospitals in Japan. J Clin Nurs. 2012;21:1314–26. doi:10.1111/j.1365-2702.2011.03971.x. 2. Chiba Y, Yamamoto‐Mitani N, Kawasaki M. A national survey of the use of physical restraint in long-term care hospitals in Japan. J Clin Nurs. 2012;21:1314–26. doi:10.1111/j.1365-2702.2011.03971.x. j 3. Goethals S, Dierckx de Casterlé B, Gastmans C. Nurses’ decision‐making in cases of physical restraint: a synthesis of qualitative evidence. J Adv Nurs. 2012;68(6):1198–210. doi:10.1111/j.1365-2648.2011.05909.x. 4. Shanahan D. Bedrails and vulnerable older adults: how should nurses make ‘safe and sound’ decisions surrounding their use? Int J Older People Nurs. 2012;7(4):272–81. doi:10.1111/j.1748-3743.2011.00285.x. Conclusions The prevalence in the use of physical restraint in the facil- ities studied was higher than those in studies from other countries. This is associated with impairments in functional and cognitive status. The results emphasize the need to provide support and training to staff in the care of these residents. Practices that may favor the routine use of this procedure should be revised to address the underlying issues leading to restraint; nurse consultants and the avail- ability of alternative devices also need to be evaluated in Spain. We believe that these strategies should be supported by specific laws that guide practitioners and institutions to provide care in the least restrictive way possible. The high prevalence of physical restraint use, compared to studies in other countries, will hopefully convince legislators of the need to enact legislation that will restrict usage. Received: 16 August 2016 Accepted: 13 January 2017 Limitations The sample size of this study could be considered a limita- tion as it may have reduced the accuracy of the confidence interval. The data regarding some of the variables was obtained from the clinical records; given that studies in nursing homes have highlighted that not all events are recorded in patient notes [49, 50], it is possible that in some cases information might be missing such as the recording of falls. Other data, such as the cognitive impairment evalu- ation, must be treated with caution as these did not figure in a large number of patient records. In addition to the high prevalence found in this study, the level of restraint use variability among the centers investi- gated was striking. This result has been observed in other studies [12, 13, 25]. Given that many have similar character- istics, we cannot explain the source of this variability. Page 6 of 7 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Page 6 of 7 The decision not to include residents without voluntary movement may also have influenced the results, especially their comparability to other studies that did not follow the same criteria; however, we felt this provided a more represen- tative sample of those in which bilateral full-enclosure side rails, and other devices, would be considered a restraint. Acknowledgements h k h Our sincere thanks to the residents and family members who have participated in this study; to the professionals and managers of the centers for their collaboration; and to Alan Gregory for his invaluable help in proofreading. 8. Strout T. Perspectives on the experience of being physically restrained: an integrative review of the qualitative literature. Int J Ment Health Nurs. 2010;19(6):416–27. doi:10.1111/j.1447-0349.2010.00694.x. 8. Strout T. Perspectives on the experience of being physically restrained: an integrative review of the qualitative literature. Int J Ment Health Nurs. 2010;19(6):416–27. doi:10.1111/j.1447-0349.2010.00694.x. Competing interests The authors declare that they have no competing interests. 14. Kruger C, Meyer G, Hamers J. Mechanische freiheitsentziehende Maßnahmen im Krakenhaus. Ein systematischer Literaturüberblick [Physical restraints in hospital. A systematic overview]. Z Gerontol Geriatr. 2010;43(5):291–6. doi:10.1007/s00391-010-0111-z. 14. Kruger C, Meyer G, Hamers J. Mechanische freiheitsentziehende Maßnahmen im Krakenhaus. Ein systematischer Literaturüberblick [Physical restraints in hospital. A systematic overview]. Z Gerontol Geriatr. 2010;43(5):291–6. doi:10.1007/s00391-010-0111-z. Consent for publication Not applicable. Consent for publication Not applicable. Page 7 of 7 Page 7 of 7 Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Page 7 of 7 34. Fariña-López E, Estévez-Guerra G, Núñez-González E, Pérez-Hernández D, Gandoy-Crego M. Actitudes, conocimientos y práctica de los profesionales de enfermería en el uso de restricciones físicas con personas mayores [Use of physical restraints on the elderly: Attitudes, knowledge and practice among nursing staff]. Rev Esp Geriatr Gerontol. 2013;48(5):209–15. doi:10.1016/j.regg. 2013.01.005. 15. Heinze C, Dassen T, Grittner U. Use of physical restraints in nursing homes and hospitals and related factors: a cross-sectional study. J Clin Nurs. 2012;21(7–8):1033–40. doi:10.1111/j.1365-2702.2011.03931.x. 16. Muñiz R, Gómez S, Curto D, Hernández R, Marco B, García P, et al. Reducing physical restraints in nursing homes: a report from María Wolff and Sanitas. J Am Med Dir Assoc. 2016;17(7):633–9. doi:10.1016/j.jamda.2016.03.011. 35. Möhler R, Meyer G. Attitudes of nurses towards the use of physical restraints in geriatric care: a systematic review of qualitative and quantitative studies. Int J Nurs Stud. 2014;51(2):274–88. doi:10.1016/j.ijnurstu.2013.10.004. 17. Beerens HC, Sutcliffe C, Renom-Guiteras A, Soto ME, Suhonen R, Zabalegui A, et al. Quality of life and quality of care for people with dementia receiving long term institutional care or professional home care: the european righttimeplacecare study. J Am Med Dir Assoc. 2014;15(1):54–61. doi:10.1016/j.jamda.2013.09.010. 36. Kong EH, Choi H, Evans LK. Staff Perceptions of Barriers to Physical Restraint-Reduction in Long-Term Care: A meta-synthesis. J Clin Nurs 2016 Jun 6. doi:10.1111/jocn.13418. 18. Fernández, JN (coordinator). INFORME 2014. Las Personas Mayores en España. Datos Estadísticos Estatales y por Comunidades Autónomas. [REPORT 2014. Older people in Spain. National and regional statistical data]. Madrid: IMSERSO. 2015. Available from: http://www.imserso.es/InterPresent2/groups/imserso/ documents/binario/22029_info2014pm.pdf. Accessed 15 Jun 2016. 37. Castle N, Engberg J. The health consequences of using physical restraints in nursing homes. Med Care. 2009;47(11):1164–73. doi:10.1097/MLR. 0b013e3181b58a69. 38. Fonad E, Wahlin T, Winblad B, Emami A, Sandmark H. Falls and fall risk among nursing home residents. J Clin Nurs. 2008;17(1):126–34. doi:10.1111/j.1365-2702.2007.02005.x. 19. Duque JM, Mateo A (coordinators). Competing interests Los modelos de atención en alojamientos residenciales para personas en situación de dependencia. [The models of care in residences for people in situations of dependence]. Madrid: IMSERSO. 2006. Available from: http://ibdigital.uib.es/greenstone/collect/portal_social/archives/ mtas0016.dir/mtas0016.pdf. Accessed 15 Nov 2016. 39. Evans D, Wood J, Lambert L. Patient injury and physical restraint devices: a systematic review. J Adv Nurs. 2003;41(3):274–82. doi:10.1046/j.1365-2648. 2003.02501.x. 40. Moore K, Haralambous B. Barriers to reducing the use of restraints in residential elder care facilities. J Adv Nurs. 2007;58(6):532–40. doi:10.1111/j.1365-2648.2007.04298.x. 20. IMSERSO. Atención a las personas en situación de dependencia en España. Libro blanco [Caring for people in situations of dependency in Spain. White Paper]. Madrid: IMSERSO. 2005. (updated Jun 24, 2014). Available from: http://www.dependencia.imserso.es/InterPresent1/groups/imserso/ documents/binario/libroblanco.pdf. Accessed 15 Nov 2016. 41. Köpke S, Mühlhauser I, Gerlach A, Haut A, Burkhard H, Möhler R, et al. Effect of a guideline-based multicomponent intervention on use of physical restraints in nursing homes. JAMA. 2012;307(20):2177–84. doi:10.1001/jama.2012.4517. 21. García P, Beltrán J. El uso de contenciones en el ámbito de los servicios sociales de Navarra [The use of restraints in the field of social services in Navarra]. Revista Jurídica de Navarra 2012;53–54:71–112. 42. Gulpers M, Bleijlevens M, Ambergen T, Capezuti E, Rossum E, Hamers J. Belt restraint reduction in nursing homes: effects of a multicomponent intervention program. J Am Geriatr Soc. 2011;59(11):2029–36. doi:10.1111/j.1532-5415.2011.03662.x. 22. Bleijlevens MHC, Wagner LM, Capezuti E, Hamers JPH. Physical restraints: Consensus of a research definition using a modified delphi technique. J Am Geriatr Soc. 2016. doi:10.1111/jgs.14435. 43. Castle N, Mor V. Physical restraints in nursing homes: a review of the literature since the nursing home reform Act of 1987. Med Care Res Rev. 1998;55(2):139–76. doi:10.1177/107755879805500201. 23. Centers for Medicare & Medicaid Services. Revisions to the State Operations Manual-Appendix PP-Guidance to Surveyors for Long Term Care Facilities. Pub. 100–07 State Operations. Transmittal 157. Maryland (U.S.) (updated Jun 10, 2016). Available from: https://www.cms.gov/Regulations-and-Guidance/ Guidance/Transmittals/Downloads/R157SOMA.pdf. Accessed 12 Oct 2016. 44. Luo H, Lin M, Castle N. Physical restraint use and falls in nursing homes: a comparison between residents with and without dementia. Am J Alzheimers Dis Other Demen. 2011;26(1):44–50. doi:10.1177/ 1533317510387585. 24. Hamers J, Gulpers M, Strik W. Use of physical restraints with cognitively impaired nursing home residents. J Adv Nurs. 2004;45(3):246–51. 45. Miles S. Autopsy findings in asphyxia in medical bed rails. Am J Forensic Med Pathol. 2009;30(3):256–61. doi:10.1097/PAF.0b013e31819df5c2. 25. Huizing A, Hamers J, De Jonge J, Candel M, Berger M. Competing interests Organisational determinants of the use of physical restraints: a multilevel approach. Soc Sci Med. 2007;65(5):924–33. doi:10.1016/j.socscimed.2007.04.030. 46. Fariña‐López E, Estévez‐Guerra GJ, Gandoy‐Crego M, Polo‐Luque LM, Gómez‐ Cantorna C, Capezuti EA. Perception of spanish nursing staff on the use of physical restraints. J Nurs Scholarsh. 2014;46(5):322–30. doi:10.1111/jnu.12087. 26. Laurin D, Voyer P, Verreault R, Durand P. Physical restraint use among nursing home residents: a comparison of two data collection methods. BMC Nurs. 2004;3:5. doi:10.1186/1472-6955-3-5. 47. Ludwick R, Meehan A, Zeller R, O’Toole R. Safety work: initiating, maintaining, and terminating restraints. Clin Nurse Spec. 2008;22(2):81–7. doi:10.1097/01.NUR.0000311672.03857.1b. 27. Baztán JJ, Pérez J, Alarcón T, San Cristóbal E, Izquierdo G, Manzarbeitia J. Índice de Barthel: Instrumento válido para la valoración funcional de pacientes con enfermedad cerebrovascular [Barthel Index: functional assessment of patients with cerebrovascular disease]. Rev Esp Geriatr Gerontol. 1993;28:32–40. 48. Freeman S, Spirgiene L, Martin-Khan M, Hirdes JP. Relationship between restraint use, engagement in social activity, and decline in cognitive status among residents newly admitted to long-term care facilities. Geriatr Gerontol Int. 2016 Jan 28. doi:10.1111/ggi.12707. 49. Schnelle JF, Bates-Jensen BM, Chu L, Simmons SF. Accuracy of nursing home medical record information about care-process delivery: Implications for staff management and improvement. J Am Geriatr Soc. 2004;52(8):1378–83. doi:10.1111/j.1532-5415.2004.52372.x. 28. Bergstrom N, Braden B, Laguzza A, Holman V. The Braden Scale for predicting pressure sore risk. Nurs Res. 1987;36(4):205–10. 29. Folstein M, Folstein S, McHugh P. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–98. 50. Hill-Westmoreland EE, Gruber-Baldini AL. Falls documentation in nursing homes: agreement between the minimum data set and chart abstractions of medical and nursing documentation. J Am Geriatr Soc. 2005;53(2):268–73. doi:10.1111/j.1532-5415.2005.53113.x. 30. Lobo A, Saz P, Marcos G, et al. Revalidación y normalización del Mini- Examen Cognoscitivo (primera versión en castellano del Mini-Mental Status Examination) en la población general geriátrica [Revalidation and standardization of the cognition mini-exam (first Spanish version of the Mini-Mental Status Examination) in the general geriatric population]. Med Clin (Barc). 1999;112(20):767–74. 31. Killip S, Mahfoud Z, Pearce K. What is an intracluster correlation coefficient? crucial concepts for primary care researchers. Ann Fam Med. 2004;2(3):204–20. doi:10.1370/afm.141. 32. Gallinagh R, Nevin R, Mc Ilroy D, Mitchell F, Campbell L, Ludwick R, et al. The use of physical restraints as a safety measure in the care of older people in four rehabilitation wards: findings from an exploratory study. Estévez-Guerra et al. BMC Geriatrics (2017) 17:29 Competing interests Int J Nur Stud. 2002;39(2):147–56. 33. Mamun K, Lim J. Use of physical restraints in nursing homes: current practice in Singapore. Ann Acad Med Singapore. 2005;34:158–62. 33. Mamun K, Lim J. Use of physical restraints in nursing homes: current practice in Singapore. Ann Acad Med Singapore. 2005;34:158–62.
https://openalex.org/W3169436730	https://www.researchsquare.com/article/rs-512644/latest.pdf	English	null	Therapeutic potential of an intestinotrophic hormone, glucagon-like peptide 2, for treatment of type 2 short bowel syndrome rats with intestinal bacterial and fungal dysbiosis	BMC infectious diseases	2,021	cc-by	9,906	Therapeutic Potential of an Intestinotrophic Hormone, Glucagon-Like Peptide 2, for Treatment of Type 2 Short Bowel Syndrome Rats with Intestinal Bacterial and Fungal Dysbiosis Page 1/28 Type 2 Short Bowel Syndrome Rats with Intestinal Bacterial and Fungal Dysbiosis Xiuting Hu Nanjing University Wei Cheng Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine Shengxian Fan Nanjing Drum Tower Hospital, Medical School of Nanjing University Yuhua Huang Medical School of Nanjing University Xi Chen Nanjing University Zhiwei Jiang Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine Jian Wang (  jianwangnju@yeah.net ) Nanjing Drum Tower Hospital, Medical School of Nanjing University Research Article Keywords: Intestinal bacterial and fungal, dysbiosis, short bowel syndrome, intestinotrophic hormone, glucagon-like peptide 2 Posted Date: May 26th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-512644/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Bacterial and Fungal Dysbiosis Xiuting Hu Nanjing University Wei Cheng Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine Shengxian Fan Nanjing Drum Tower Hospital, Medical School of Nanjing University Yuhua Huang Medical School of Nanjing University Xi Chen Nanjing University Zhiwei Jiang Jiangsu Province Hospital of Chinese Medicine, Affiliated Hospital of Nanjing University of Chinese Medicine Jian Wang (  jianwangnju@yeah.net ) Nanjing Drum Tower Hospital, Medical School of Nanjing University Research Article Keywords: Intestinal bacterial and fungal, dysbiosis, short bowel syndrome, intestinotrophic hormone, glucagon-like peptide 2 Posted Date: May 26th, 2021 DOI: https://doi.org/10.21203/rs.3.rs-512644/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Abstract Background: Previous studies showed that type 2 short bowel syndrome (SBS) rats were accompanied by severe intestinal bacterial dysbiosis. Limited data are available for intestinal fungal dysbiosis. Moreover, no effective therapeutic drugs are available for these microbiota dysbiosis. The aims of our study were to investigate the therapeutic potential of glucagon-like peptide 2 (GLP-2) for these microbiota dysbiosis in type 2 SBS rats. Methods: 8-week-old male SD rats which underwent 80% small bowel resection, ileocecum resection, partial colon resection and jejunocolostomy, were treated with saline (SBS group, n=5) or GLP-2 (GLP2.SBS group, n=5). The Sham group rats which underwent transection and re-anastomosis were given a saline placebo (Sham group, n=5). 16S rRNA and ITS sequencing were applied to evaluate the colonic bacterial and fungal composition at 22 days after surgery, respectively. Results: The relative abundance of Actinobacteria, Firmicutes and proinflammatory Proteobacteria increased significantly in SBS group rats, while the relative abundance of Bacteroidetes, Verrucomicrobia and Tenericutes decreased remarkably. GLP-2 treatment significantly decreased Proteus and increased Clostridium relative to the saline treated SBS rats. The diversity of intestinal fungi was significantly increased in SBS rats, accompanied with some fungi abnormally increased and some resident fungi (e.g., Penicillium) significantly decreased. GLP-2 treatment significantly decreased Debaryomyces and Meyerozyma, and increased Penicillium. Moreover, GLP-2 partially restored the bacteria-fungi interkingdom interaction network of SBS rats. Conclusion: Our study confirms the bacterial and fungal dysbiosis in type 2 SBS rats, and GLP-2 partially ameliorated these microbiota dysbiosis. Research Article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Page 1/28 Page 1/28 Page 1/28 1. Background Short bowel syndrome (SBS), a main cause of intestinal failure, refers to a kind of disease presented as severe malnutrition, diarrhea, water and electrolyte disorder mainly resulting from massive small bowel resection [1, 2]. The treatment of SBS is facing multiple challenges. Inadequate enteral nutrition may lead to a series of complications, such as malnutrition, parenteral nutrition related liver injury or catheter- related infection, creating huge economic burdens both for families and the whole society. Gut microbiota dysbiosis is another great challenge for SBS patients [3, 4]. In recent years, with the rapid development of high-throughput sequencing technology, a growing number of studies have been focused on intestinal flora dysbiosis in SBS. Previous studies demonstrated that the intestinal flora of SBS was significantly disturbed, manifested as the reduction of intestinal flora diversity, and increase of pathogenic bacteria abundance. Unfortunately, intestinal flora dysbiosis in SBS can lead to intestinal mucosal inflammation, delayed enteral nutrition and prolonged use of parenteral nutrition, impaired intestinal adaptation, and finally resulting in poor prognosis [4–6]. Our Previous studies have also disclosed that the composition of Page 2/28 Page 2/28 intestinal flora was altered in type 2 SBS patients and rats, which have an enterocolonic anastomosis and without ileocecal region [4, 7]. Apart from bacteria, intestinal fungi are an indispensable part of intestinal microbiota. However, due to the restriction of techniques in the past, research on intestinal fungi dysbiosis in SBS patients and rats is still lacking. The development of high-throughput technology provides sufficient technical support for the study of fungal microbiota, which enables researchers to accurately understand the composition of intestinal fungi through internal transcribed spacer (ITS) sequencing [8–11]. Recent studies have revealed that there is an intense disruption in intestinal bacteria-fungi correlation network in inflammatory bowel disease (IBD) [8, 9, 11]. In fact, under circumstances of intestinal bacterial dysbiosis, intestinal fungal dysbiosis is likely to occur. Thus, we hypothesized that the fungal microbiota might be altered and further affect the intestinal adaption procedure in SBS. The intestine is the most developed endocrine organ. There are more than ten kinds of enteric endocrine cells (EEC) in the intestine, which secrete more than twenty kinds of intestinal endocrine hormones [12]. Glucagon like peptide-2 (GLP-2), secreted by distal small intestine and colon, is a well-known intestinal growth factor, which can promote the proliferation of intestinal mucosa and enhance the absorption function of remanent intestine. 1. Background GLP-2 is the most crucial factor in current intestinal rehabilitation therapy of SBS. Our recent study published in Theranostics further demonstrated that GLP-2 exerted its gut mucosal hyperplasia promoting effect through modulating exosomal miRNAs transportation in small intestinal microenvironment [1]. Moreover, Yu Hu et al. [13] declared that GLP-2 significantly alleviated intestinal flora dysbiosis in elderly rats. However, it remains unknown whether GLP-2 could modulate the intestinal bacterial and fungal dysbiosis in SBS. In the current study, we characterized the altered composition of bacteria and fungi in SBS rats and evaluated the effects of GLP-2 on alleviation of bacterial and fungal dysbiosis, and provided insight into the crosstalk between bacteria and fungi in SBS rats. 2.1 Animal experimental protocol Type 2 short bowel syndrome model was created with SD rats (male, 8 weeks old, 250–260 g, Charles River, Beijing, China), according to a previously reported method [1]. Rats were kept at a temperature of 21°C to 25°C, humidity of 52% to 62%, and a 12/12-hour light/dark cycle. Figure 1A displayed the timeline of the experimental procedures. SD rats were fasted for 24 hours before operation with free access to water. All procedures were performed under the anaesthetization of sodium pentobarbital (40 mg/kg, ip) using aseptic technique. Rats were randomly assigned to three experimental groups: Sham, SBS and GLP2.SBS. The Sham group underwent transection and re-anastomosis at approximately 20 cm from the ligament of Treitz without small bowel resection and were given saline as placebo. SBS rats underwent resection of 80% small gut, ileocecum resection, resection of partial colon and jejunocolostomy (shown in Figure 1B). Animals were resuscitated by intraperitoneal injection of 5 ml saline. 250 ul Saline or 100 Page 3/28 Page 3/28 μg/kg body weight degradation-resistant (Gly2) GLP-2 (100 μg/ml dissolved in saline, Creative Peptides, New York, USA) were administered subcutaneously once daily in SBS and GLP2.SBS group, respectively. Rats were fasted within 24 hours after operation, then fed with enteral nutrition solution (full strength) for 24-72 hours, followed by a gradual transition to solid diet 7 days after operation and maintained total solid diet until the rats were sacrificed for sampling. 2.2 Sample collection and genomics DNA isolation After study, all animals were anaesthetized by sodium pentobarbital (40 mg/kg, ip) and then euthanized by cervical dislocation at 22 days after operation. At about 3 cm distal to the anastomotic site of SBS rats or the same colon segment site of Sham rats (Figure 1B), colonic contents were harvested, quickly frozen in liquid nitrogen and stored at -80°C for further analysis. The colonic contents DNA was extracted using the QIAamp DNA Stool Mini Kit (Qiagen, Valencia, CA) following the manufacturer's instructions. DNA was quantified with a Qubit Fluorometer by using Qubit dsDNA BR Assay kit (Invitrogen, USA) and the quality was checked by running aliquot on 1% agarose gel. 2.3 Library construction The 16S rRNA gene was amplified using primers targeting the V4 region (primers in Table 1). The fungal ITS1 region was amplified using ITS-F and ITS-R primers (in Table 1). Both primers (forward and reverse) were labelled with Illumina adapter. After polymerase chain reaction (PCR), Agencourt AMPure XP beads were used to purify the PCR products. Agilent Technologies 2100 bioanalyzer (Agilent, USA) was used to qualify the libraries. Further, the validated libraries were sequenced on Illumina HiSeq 2500 platform. 2.4 Sequencing, bioinformatics and statistical analysis Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and interquartile range (IQR) using GraphPad Prism 7. Only differences with adjusted P-value of less than 0.05 were considered statistically significant. For data passed Shapiro-Wilk normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD or SEM). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and interquartile range (IQR) using GraphPad Prism 7. 3. Results Colonic contents harvested from 15 rats (5 in each group) were collected and a total of 919054 V4 16S tags were generated for final analysis, with 61270 16S tags on average. And 928933 ITS tags were generated in total with 61928 tags per sample on average. 2.4 Sequencing, bioinformatics and statistical analysis Paired-end reads were added to tags by FLASH program [14]. Using UPARSE algorithm [15], tags were clustered into operational taxonomical units (OTUs). OTU representative sequences were then taxonomically classified using Ribosomal Database Project Classifier v.2.2, and trained on the UNITE (V6 20140910) (for ITS sequencing) and Greengenes database v201305 (for 16s rRNA sequencing) by QIIME v1.8.0 [16, 17], respectively. Then we used the USEARCH_global to compare all Tags back to OTU. Alpha and beta diversity were estimated by MOTHUR (v1.31.2) and QIIME (v1.8.0). Sample cluster was conducted by QIIME (v1.8.0). OTU Rank curve was plotted with R package version 3.1.1. The Venn plots in OTUs or in taxa were plotted with R package “Venn Diagram”. Principal Coordinate Analysis (PCoA) was performed by QIIME (v1.8.0). Partial least-squares discrimination analysis (PLS-DA) and Anosim analysis were performed using R package. Heatmap and barplot of different classification levels was plotted with R package “gplots” and R package v3.4.1, respectively. Significant taxa were determined by R package based on Kruskal-Test. LEfSe cluster and LDA analysis were performed by LEfSe (https://huttenhower.sph.harvard.edu/galaxy/). Spearman correlation analysis was applied to build the correlation networks between different bacteria. The correlation networks between top 50 bacteria and top 20 fungi were analyzed using Spearman correlation analysis. Alpha and beta diversity were estimated by MOTHUR (v1.31.2) and QIIME (v1.8.0). Sample cluster was conducted by QIIME (v1.8.0). OTU Rank curve was plotted with R package version 3.1.1. The Venn plots in OTUs or in taxa were plotted with R package “Venn Diagram”. Principal Coordinate Analysis (PCoA) was performed by QIIME (v1.8.0). Partial least-squares discrimination analysis (PLS-DA) and Anosim analysis were performed using R package. Heatmap and barplot of different classification levels was plotted with R package “gplots” and R package v3.4.1, respectively. Significant taxa were determined by R package based on Kruskal-Test. LEfSe cluster and LDA analysis were performed by LEfSe (https://huttenhower.sph.harvard.edu/galaxy/). Spearman correlation analysis was applied to build the correlation networks between different bacteria. The correlation networks between top 50 bacteria and top 20 fungi were analyzed using Spearman correlation analysis. Page 4/28 Only differences with adjusted P-value of less than 0.05 were considered statistically significant. For data passed Shapiro-Wilk normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD or SEM). 3.1 GLP-2 partially reversed intestinal bacterial dysbiosis in SBS rats Rarefaction analysis demonstrated that the estimated OTU richness could approach saturation in each sample (Figure 1C). The OTU counts index in SBS and GLP2.SBS group were dramatically lower than that of Sham group, indicating SBS induced a significant decrease in intestinal flora diversity (Figure 1C-1D). Besides, compared with Sham group, the α diversity of intestinal bacteria in SBS rats was significantly decreased, which was manifested as decreased Chao (P<0.01), ACE (P<0.01), Shannon’s diversity (P<0.01), and increased Simpson’ diversity (P<0.05) and Good’s coverage (P<0.01). However, no significant difference in α-diversity was found between SBS and GLP2.SBS group (Figure 1E-1I). Beta diversity, assessed by the weighted UniFrac Cluster Tree and weighted UniFrac diversity distance, were different in both SBS groups as compared with the Sham group (Figure 2A-2B). We then visualized weighted UniFrac dissimilarity by PCoA analysis (PC1 61.13% and PC2 15.02%), which declared different overall bacterial community among Sham, SBS and GLP2.SBS group. We found the samples of Sham group clustered together and were far away from the fusion clustering of SBS group and GLP2.SBS group (Figure 2C). Additionally, PLS-DA, a supervised analysis method also disclosed a distinctive intestinal bacterial composition in GLP2.SBS and SBS groups when compared with the Sham group (Figure 2D). Furthermore, Anosim similarity analysis based on both weighted_unifrac and Bray_curtis_OTU distance disclosed a statistically different intestinal bacterial composition between Sham group and the two SBS groups (SBS group and GLP2.SBS group) (Figure 2E-2F and Figure S1). The colonic flora composition differed dramatically at phylum level among the three groups (Figure 3A). Analyzed at phylum level, rats underwent massive small bowel resection displayed perturbed microbiota signatures especially in Firmicutes, Proteobacteria, Actinobacteria, Bacteroidetes, Verrucomicrobia, Tenericutes, Deferribacteres, TM7 and Unclassified phyla compared with Sham rats (Figure 3B-L). In SBS and GLP2.SBS group, Firmicutes dramatically increased as compared with Sham group (P=0.0320 and P=0.0204) (Figure 3B). Meanwhile, the relative abundance of Proteobacteria was quite low in Sham group, while it was significantly higher in the colonic contents of SBS and GLP2.SBS group. Furthermore, compared to Sham group, Actinobacteria was significantly increased in SBS group, while GLP-2 treatment decreased the relative abundance of Actinobacteria in SBS rats (Figure 3D). The relative abundance of Bacteroidetes, Verrucomicrobia, Tenerictes, Deferrribates, TM7 and Unclassified phyla were significantly decreased, even Page 5/28 Page 5/28 undetectable in SBS and GLP2.SBS rats, indicating that the intestinal bacterial flora of SBS rats has been seriously disturbed (Figure 3G-3L). 3.1 GLP-2 partially reversed intestinal bacterial dysbiosis in SBS rats We further analyzed the different bacterial strains of the three surgical groups and the subordination of these different bacterial strains by LEfSe. SBS resulted in an obvious reduction of multiple bacterial genera, including Bacteroides, Odoribacter and Prevotella genera from Bacteroidetes phylum, Coprococcus and Dorea genera from Lachnospiraceae family and Firmicutes phylum, Oscillospira, Ruminococcus and Clostridium genera from Ruminococcaceae family and Firmicutes phylum, and Akkermansia genus from Verrucomicrobia phylum. Moreover, SBS caused a significant increase of Proteus genus belonging to the pro-inflammatory Proteobacteria phylum. On the other hand, GLP-2 treatment could partially ameliorate the intestinal bacterial dysbiosis of SBS rats, manifested as significant inhibition of the overgrowth of pro-inflammatory Proteus genus in SBS rats. Moreover, the relative abundance of anti-inflammatory Clostridium genus from Firmicutes phylum was remarkably increased in GLP2.SBS rats when compared with SBS rats (Figure 4 and Table 2). We further built the bacterial abundance correlation network to assess the bacterial structure in SBS rats. An abundant and complicated network of interrelations between bacteria was observed in Sham rats (Figure 5A). In comparison with the Sham rats, the density of bacterial interrelation network was markedly decreased in SBS rats, as attested by a reduced relative connectedness and a decreased number of neighbors. The network in GLP2.SBS rats was larger (nodes n=50, edges n=74) than in SBS rats (nodes n=42, edges n=59), suggesting a potential ameliorate effect of GLP-2 treatment (Figure 5A-5B). 3.2 GLP-2 ameliorated intestinal fungal dysbiosis in SBS rats Intestinal bacterial dysbiosis could result in intestinal fungal dysbiosis, but there is still no research on intestinal fungal dysbiosis in SBS rats and patients [3]. This part aims to investigate the intestinal fungal dysbiosis of SBS and the therapeutic effect of GLP-2 on the intestinal fungal dysbiosis. Rarefaction analysis declared that the estimated OTU richness could approach saturation in each sample (Figure 6A). OTU Venn analysis showed that there were 58 mutual OTUs in the three groups. And there were 17 OTUs, 62 OTUs and 44 OTUs specifically in Sham group, SBS group and GLP2.SBS group (Figure 6B). Moreover, statistics showed that there were significant differences in OTU Counts index between Sham group and SBS group (P<0.05). On the other hand, GLP-2 treatment inhibited the increase of OTU Counts index, manifested by no significant difference in OTU Counts index between Sham group and GLP2.SBS (Figure 6C). Moreover, the α-diversity of intestinal fungi in SBS rats was dramatically increased relative to Sham rats, as evidenced by increased Chao (P<0.05) and ACE index (P<0.05). As expected, no significant difference in Chao and ACE index was found between GLP2.SBS group and Sham group, indicating that GLP-2 treatment might partially reverse the intestinal fungal overgrowth in SBS rats (Figure 6D-E). Weighted UniFrac Cluster Tree and weighted UniFrac diversity distance showed obvious changes in intestinal fungal β-diversity in SBS and GLP2.SBS rats when compared with Sham rats (Figure 7A-B). We then visualized UniFrac dissimilarity by PCoA analysis, which indicated different overall fungal community structures among Sham, SBS and GLP2.SBS group. Moreover, we found samples of SBS Page 6/28 Page 6/28 group clustered together and were far away from the fusion clustering of Sham group and GLP2.SBS group, indicating that GLP-2 treatment could partially reverse the intestinal fungal dysbiosis of SBS rats (Figure 7C). Additionally, the result of PLS-DA was similar with PCoA analysis result (Figure 7D). Furthermore, Anosim similarity analysis based on both weighted_unifrac and Bray_curtis_OTU distance disclosed a statistically different intestinal fungal composition between the three groups (Figure 7E-7F and Figure S2). Figure 8A showed profiling of each sample at the class level and gave an idea of the proportion of different classes in each sample. Furthermore, the differences between groups in the first five classes, including Saccharomycetes, Eurotiomycetes, Sordariomycetes, Dothideomycetes, and Tremellomycetes were investigated. 3.2 GLP-2 ameliorated intestinal fungal dysbiosis in SBS rats The results suggested that the relative abundance of Saccharomycetes class was notably increased in SBS and GLP2.SBS group (P=0.0367 and P=0.0222) (Figure 8B). In comparison with Sham group, the relative abundance of Eurotiomycetes class was remarkably reduced in SBS group (P=0.0067), while GLP-2 treatment partially reversed the decrease of Eurotiomycetes class in SBS rats (Figure 8C). The relative abundance of Sordariomycetes decreased evidently in both SBS and GLP2.SBS group (P= 0.0418 and P= 0.0421) (Figure 8D). The relative abundance of Dothideomycetesand Tremellomycetes in SBS rats was higher than that of rats in Sham group, but there was no significant difference between the three groups (Figure 8E-8F). We then analyzed the different fungal strains in three surgical groups by LEfSe analysis. SBS resulted in a remarkable reduction in the relative abundance of the multiple fungal genera, including Aspergillus and Penicillium genera from Eurotiomycetes class and Wickerhamomyces genus from Saccharomycetes class. Moreover, SBS caused a significant increase in three genera, including Debaryomyces and Meyerozyma genera from Saccharomycetes class and Xerochrysium genus from Eurotiomycetes class. On the other hand, GLP-2 treatment (GLP2.SBS group) can partially ameliorate the intestinal fungal dysbiosis of SBS rats, including significantly inhibited the increase of Meyerozyma and Debaryomyces genera, and significantly inhibited the decrease of Penicillium genera (Figure 9 and Table 3). 4. Discussion In the present study, we comprehensively described the intestinal bacterial and fungal dysbiosis in type 2 SBS rats with ileocecal resection, and found that GLP-2 treatment could partially attenuate the disturbance of intestinal microbiota. Although our previous studies have disclosed severe intestinal bacterial dysbiosis in animal models and patients with ileocecal resection [2, 7], there is still no comprehensive assessment of intestinal fungal dysbiosis accompanied by intestinal bacterial dysbiosis in SBS. Due to the interkingdom interaction between bacteria and fungi, intestinal bacterial dysbiosis is likely to cause intestinal fungal dysbiosis, which may further aggravate the intestinal bacterial dysbiosis, and even exacerbate the intestinal inflammation and thus worsen the prognosis of the disease [19–21]. Limited by previous technology, it was difficult for traditional culture-dependent methods to achieve in- depth understanding of the fungal microbiota. Recently, high-throughput sequencing provides effective technical support for study of intestinal fungi in various diseases [20, 22]. Herein, we used high- throughput ITS sequencing to comprehensively analyze intestinal fungal dysbiosis in type 2 SBS rats. Our previous study found that GLP-2 could promote intestinal mucosal proliferation [1]. However, whether GLP-2 could ameliorate the intestinal microbiota dysbiosis of SBS has not been studied. Yu Hu and colleagues [13] demonstrated that GLP-2 treatment resulted in the increase of several probiotics and the decrease of some pathogenic bacterial genera in intestine of aged rats. Therefore, we performed this study to investigate the colonic microbiota of SBS rats with or without GLP-2 intervention. We first characterized the altered composition of intestinal bacterial dysbiosis in SBS rats and evaluated the therapeutic effects of GLP-2 on intestinal bacteria dysbiosis. Significant decrease of OTU counts and α diversity were observed in the colonic contents of SBS and GLP2.SBS rats which revealed that there was significant intestinal bacterial dysbiosis in type 2 SBS rats. Massive resection of small intestine and ileocecal resection lead to fast pass of food, gastric acid and bile acid, which exerts deleterious effect on the luminal microenvironment. This will inevitably affect the living environment of intestinal microbiota, thus leading to intestinal microbiota dysbiosis, presented as reduction of probiotics and an increase of some pathogenic bacteria [23]. Although GLP-2 possess proliferative and anti-inflammatory effects on intestinal mucosa, no significant improvement in OTU counts and α-diversity was found in GLP-2 treated SBS rats. 3.3 Altered interkingdom network in SBS rats The bacteria–fungi interaction has been proven to be quite important for keeping intestinal microbial homeostasis [11, 18]. To investigate the equilibrium between intestinal bacteria and fungi diversity, we calculated the fungi-to-bacteria diversity by ITS1/16S diversity ratio of Chao index and ACE index. The ratios were significantly higher in SBS and GLP2.SBS rats than those in Sham rats. However, there was no significant difference of fungi-to-bacteria diversity ratio between SBS and GLP2.SBS rats (Figure 10A- 10B). Furthermore, to explore bacterial and fungi interaction network in this study, we selected the top 20 fungal genera and top 50 bacterial genera for further analysis according to the microbial relative abundance. When compared with Sham group, the correlation between microbiota was quite weak (only few lines between fungal and bacterial genera) in SBS group, indicating the interkingdom network was seriously disrupted. The microbial correlation pattern was stronger in GLP2.SBS group relative to SBS group, indicating by increased edges, average number of neighbors, and relative connectedness (Figure Page 7/28 Page 7/28 10C-10D). These data suggested that SBS led to an alteration of bacterial-fungi interactions, whereas GLP-2 treatment could help to partially rebuild these interkingdom interactions. 4. Discussion Both PLS-DA and PCoA strongly indicated distinguished bacterial communities between Sham and SBS group, while clusters of SBS and GLP2.SBS group were close to each other, indicating similar bacterial communities. Moreover, notable differences in β-diversity based on weighted UniFrac analysis and Anosim tests suggested that the variation of microbial community contains not only changes in quantity of bacterial species but also changes in specific bacterial abundance [24]. These results demonstrated that massive resection of small intestine and ileocecal resection could severely disrupt normal intestinal bacterial microbiome, while GLP-2 treatment could not restore the intestinal bacterial diversity of SBS rats. Page 8/28 Detailed analysis of the composition and structure of intestinal bacteria in SBS demonstrated that, in comparison with Sham group, the relative abundance of 9 out of 11 phyla changed significantly, suggesting a thorough dysbiosis of intestinal bacteria in SBS rats. In line with previous studies, there existed a dramatic overabundance of Proteobacteria and its family Enterobacteriaceae in colonic contents of SBS rats, which contains a great number of pathogenic bacteria [2, 25, 26]. Proteus, an opportunistic pathogen within the Enterobacteriaceae family, was remarkably prevalent in SBS group in our study. Previous studies have proved Proteus’ potential ability to produce LPS results in the upregulated secretion of pro-inflammatory cytokines and further leads to the destruction of intestinal mucosal barrier and thus inducing bacterial translocation [2, 27]. A remarkable alteration of the Firmicutes phylum in SBS and GLP2.SBS rats was the decrease of the Lachnospiraceae and Ruminococcaceae families, which are able to produce short-chain fatty acids (SCFAs). SCFAs, important metabolites of bacteria, exerted many important biological functions, such as energy-supplying fuel for intestinal epithelial cells, inducing enterocyte proliferation and differentiation, and promoting the production of antimicrobial peptides by epithelial cells [2, 28, 29]. Therefore, deficiency of Lachnospiraceae and Ruminococcaceae families may have adverse effects on intestinal adaptation of SBS. In line with previous research, the relative abundance of Akkermansia genus from Verrucomicrobia phylum decreased significantly and even disappeared completely in SBS rats [2]. Akkermansia is a newly discovered star probiotics, which has a variety of beneficial biological functions, such as anti-tumor effect, intestinal mucosal barrier protection, and alleviating colonic inflammation [30–32]. Yu Xu and colleagues summarized that Akkermansia could protect the intestinal mucosal barrier and inhibit the translocation of intestinal flora, therefore reducing the level of LPS in circulation and thus inhibiting the inflammatory response. [33]. 4. Discussion The almost complete loss of Akkermansia in the colon caused by SBS might affect the integrity of the intestinal mucosal barrier of SBS, lead to bacterial translocation and aggravate the systemic inflammatory response, which may have adverse effects on the intestinal adaptation of SBS. However, effective therapies for intestinal bacterial dysbiosis are still lacking. The first drug we thought of and used was antibiotics. However, a previous study had found that several days of antibiotics may lead to a reduction of potentially anti-inflammatory Clostridia and an increase of pro-inflammatory Proteobacteria [4]. According to previous studies, GLP-2 may alleviate intestinal microbiota dysbiosis [13]. As we expected, GLP-2 treatment significantly downregulated the relative abundance of pro-inflammatory Proteus genus in SBS rats, and increased the relative abundance of potentially anti-inflammatory inhibitory Clostridium genus from Clostridia class in SBS rats. Notably, some Clostridium spp. could also be unbeneficial, and were even increased in patients with inflammatory bowel disease when compared with healthy people [34]. As we expected, the colonic fungi of SBS rats were abnormal, indicated by increased OTU Counts index, Chao and ACE index. There was obvious overlap between the PCoA cluster of Sham group and the GLP2.SBS group, suggesting that their fungal communities were similar. On the contrary, there was no overlap between the PCoA clusters of Sham group and SBS group, suggesting that their fungal structures Page 9/28 Page 9/28 Page 9/28 were quite different. Therefore, GLP-2 treatment not only remarkably inhibited the abnormal reproduction of colonic fungi, but also potentially improved the SBS colonic fungal communities, making it partially restored to the Sham rats. Kelly B. Flett and Hector Chavez [35, 36] demonstrated a high proportion of blood fungal infections in SBS patients using culture-dependent methods, but did not use high- throughput sequencing to analyze intestinal fungal dysbiosis. To our knowledge, our study comprehensively analyzed the fungal dysbiosis in SBS by high-throughput sequencing for the first time, and the results will provide preliminary data support for relevant studies on SBS fungal dysbiosis. In terms of phyla, we found the relative abundance of Ascomycota phylum in Sham, SBS and GLP2.SBS was over 90%, indicating Ascomycota was the most predominant phylum in the intestine, followed by Basidiomycota phylum. This result is consistent with many previous studies [8, 9, 37]. Furthermore, we comprehensively analyzed the composition of colonic fungi at the class level. 4. Discussion The results showed that 3/5 of the top five abundant fungi classes (including Saccharomycetes, Eurotiomycetes, Sordariomycetes) were significantly different in SBS rats in comparison with the Sham rats. Though the other 2/5 classes (including Tremellomycetes, Dothideomycetes) were increased in SBS rats, they did not reach significant difference (probably because of the small sample size). These results indicated that severe intestinal fungal dysbiosis occurred in SBS rats. LEfSe analysis was used to deeply analyze the different fungi in the 3 groups, and some interesting results were found. For example, the relative abundance of Penicillium genus from Eurotiomycetes class decreased significantly in the SBS rats, but GLP-2 treatment could reverse its decrease in SBS rats. In addition, the abundance of Debaryomyces and Meyerozyma genera from Saccharomycetes class increased significantly in the SBS group, while GLP-2 treatment could inhibit its overgrowth. In fact, at the genus level, SBS group exhibited significant changes in six fungal genera (including Aspergillus, Penicillium, Xerochrysium, Wickerhamomyces, Debaryomyces and Meyerozyma) as compared with Sham group, while GLP-2 treatment could partially reverse the significant changes of 3/6 fungal genera (including Penicillium, Debaryomyces and Meyerozyma) at 22nd days after surgery. Penicillium genus could secrete penicillin G and griseofulvin, which have antibacterial and antifungal effects respectively. These compounds may also inhibit the abnormal growth of some bacteria and fungi in the intestinal tract, thus maintaining intestinal microbiota homeostasis [22, 38]. A decrease of the relative abundance of Penicillium may further worsen intestinal microbiota dysbiosis in SBS rats. GLP-2 treatment could partially reverse the decrease of the relative abundance of Penicillium in SBS rats, which may have beneficial effects on the balance of intestinal microbiota. Coexisting in the gut, intestinal fungi and bacteria interact directly or indirectly with each other [11, 39]. A growing number of studies confirmed that the bacterial-fungal interactions played an important role in the occurrence and exacerbation of many diseases, such as IBD, primary sclerosing cholangitis, neurological disease, and even cancer [8–10, 20, 40]. We herein investigated the bacterial-fungal correlation patterns in SBS rats and found disrupted bacteria–fungi correlations existed in the colonic content of SBS rats. Fortunately, GLP-2 treatment might partially re-build the interkingdom interaction in SBS rats. In terms of phyla, we found the relative abundance of Ascomycota phylum in Sham, SBS and GLP2.SBS was over 90%, indicating Ascomycota was the most predominant phylum in the intestine, followed by Basidiomycota phylum. 4. Discussion This result is consistent with many previous studies [8, 9, 37]. Furthermore, we comprehensively analyzed the composition of colonic fungi at the class level. The results showed that 3/5 of the top five abundant fungi classes (including Saccharomycetes, Eurotiomycetes, Page 10/28 However, our study also has some limitations, such as small sampling size and lack of sampling over time. Our team's previous published results confirmed that the 16S sequencing data were reliable under However, our study also has some limitations, such as small sampling size and lack of sampling over time. Our team's previous published results confirmed that the 16S sequencing data were reliable under Page 10/28 consistent sampling process and the difference was small within the group but significant between groups [2, 7]. From the Anosim consistency analysis in this study, we found that the intra group differences of our samples were small, but the inter group differences were large. Therefore, we still found a lot of bacterial and fungal taxa with significant differences in case of existing small sample size. This also indirectly indicated that the intestinal flora of SBS rats was in severe dysbiosis and GLP-2 might attenuate the intestinal flora dysbiosis caused by SBS to a certain extent. We set the sampling time at the 22nd day after operation, because at this time point the intestinal adaptation of SBS rats had reached the peak while the residual intestinal internal environment and the intestinal flora of SBS rats has reached a relatively stable state so that the results would be more reliable. However, if we could enrich the sampling time point to declare the effects of whole process of intestinal adaptation and GLP-2 treatment on intestinal flora of SBS, this study would be more meaningful. Abbreviations SBS (short bowel syndrome), GLP-2 (glucagon-like peptide 2), ITS (internal transcribed spacer), IBD (inflammatory bowel disease), EEC (enteric endocrine cells), PCoA (principal coordinate analysis), PLS-DA (partial least-squares discrimination analysis), SCFAs (short-chain fatty acids). 5. Conclusions We found that severe colonic bacterial dysbiosis of SBS rats was accompanied by significant fungal dysbiosis, and GLP-2 treatment might partially ameliorate the intestinal bacterial and fungal dysbiosis in SBS rats (Figure 11). Consent for publication Not applicable. Ethics approval and consent to participate The animal protocols were approved by the Institutional Ethics Committee of Life Sciences School of Nanjing University and performed according to guideline for the care and use of laboratory animals published by the National Institutes of Health (8 Edition). And all methods were carried out in compliance with the ARRIVE guidelines. Acknowledgements Not applicable. Funding This study was supported by Lijieshou Intestinal Barrier Academy Foundation (Grant numbers LJS- 201915C) and Jiangsu “Mass Innovation and Entrepreneurship” Talent Programme to Wei Cheng, and National Natural Science Foundation of China (Grant numbers 81800452) to Shengxian Fan, and Social development projects in Jiangsu Province (Grant numbers BE2015687) to Zhiwei Jiang. Availability of data and materials The sequencing data for our study are available in the NCBI Sequence Read Archive under accession no. PRJNA677835 (https://www.ncbi.nlm.nih.gov/sra/PRJNA677835) and PRJNA677844 (//www.ncbi.nlm.nih.gov/sra/PRJNA677844) after the indicated release date. Page 11/28 Page 11/28 Authors' contributions Zhiwei Jiang, Xi Chen, and Jian Wang designed the study. Xiuting Hu and Wei Cheng wrote this manuscript. Xiuting Hu and Wei Cheng finished experiments. Xi Chen gave the structure of this dissertation. Yuhua Huang and Shengxian Fan performed the bioinformatics analysis. All authors have read and approved the final manuscript. Competing interests The authors have declared that no competing interest exists. References 1. Cheng W, Wang K, Zhao Z, Mao Q, Wang G, Li Q, Fu Z, Jiang Z, Wang J, Li J: Exosomes-mediated Transfer of miR-125a/b in Cell-to-cell Communication: A Novel Mechanism of Genetic Exchange in the Intestinal Microenvironment. Theranostics 2020, 10(17):7561-7580. 2. Huang Y, Chen A, Guo F, Wang J, Li Y: Severe Intestinal Dysbiosis in Rat Models of Short Bowel Syndrome with Ileocecal Resection. Digestive diseases and sciences 2020, 65(2):431-441. 3. Piper HG: Intestinal microbiota in short bowel syndrome. Seminars in pediatric surgery 2018, 27(4):223-228. 27(4):223 228. Piper HG, Fan D, Coughlin LA, Ho EX, McDaniel MM, Channabasappa N, Kim J, Kim M, Zhan X, Xie Y et al: Severe Gut Microbiota Dysbiosis Is Associated With Poor Growth in Patients With Short Bowel Syndrome. JPEN Journal of parenteral and enteral nutrition 2017, 41(7):1202-1212. Engstrand Lilja H, Wefer H, Nyström N, Finkel Y, Engstrand L: Intestinal dysbiosis in children with hort bowel syndrome is associated with impaired outcome. Microbiome 2015, 3:18. Goulet O, Joly F: [Intestinal microbiota in short bowel syndrome]. Gastroenterologie clinique et biologique 2010, 34 Suppl 1:S37-43. Huang Y, Guo F, Li Y, Wang J, Li J: Fecal microbiota signatures of adult patients with different types of short bowel syndrome. Journal of gastroenterology and hepatology 2017, 32(12):1949-1957. 4. Piper HG, Fan D, Coughlin LA, Ho EX, McDaniel MM, Channabasappa N, Kim J, Kim M, Zhan X, Xie Y et al: Severe Gut Microbiota Dysbiosis Is Associated With Poor Growth in Patients With Short Bowel Syndrome. JPEN Journal of parenteral and enteral nutrition 2017, 41(7):1202-1212. 5. Engstrand Lilja H, Wefer H, Nyström N, Finkel Y, Engstrand L: Intestinal dysbiosis in children with short bowel syndrome is associated with impaired outcome. Microbiome 2015, 3:18. 6. Goulet O, Joly F: [Intestinal microbiota in short bowel syndrome]. Gastroenterologie clinique et biologique 2010, 34 Suppl 1:S37-43. 7. Huang Y, Guo F, Li Y, Wang J, Li J: Fecal microbiota signatures of adult patients with different types of short bowel syndrome. Journal of gastroenterology and hepatology 2017, 32(12):1949-1957. 7. Huang Y, Guo F, Li Y, Wang J, Li J: Fecal microbiota signatures of adult patients with different types of short bowel syndrome. Journal of gastroenterology and hepatology 2017, 32(12):1949-1957. Page 12/28 8. References Lemoinne S, Kemgang A, Ben Belkacem K, Straube M, Jegou S, Corpechot C, , Chazouillères O, Housset C, Sokol H: Fungi participate in the dysbiosis of gut microbiota in patients with primary sclerosing cholangitis. Gut 2020, 69:92-102. 9. Sokol H, Leducq V, Aschard H, Pham HP, Jegou S, Landman C, Cohen D, Liguori G, Bourrier A, Nion- Larmurier I et al: Fungal microbiota dysbiosis in IBD. Gut 2017, 66(6):1039-1048. 10. Richard ML, Liguori G, Lamas B, Brandi G, da Costa G, Hoffmann TW, Pierluigi Di Simone M, Calabrese C, Poggioli G, Langella P et al: Mucosa-associated microbiota dysbiosis in colitis associated cancer. Gut microbes 2018, 9(2):131-142. 11. Jun X, Ning C, Yang S, Zhe W, Na W, Yifan Z, Xinhua R, Yulan L: Alteration of Fungal Microbiota After 5-ASA Treatment in UC Patients. Inflammatory bowel diseases 2020, 26(3):380-390. 12. Gribble FM, Reimann F: Function and mechanisms of enteroendocrine cells and gut hormones in metabolism. Nature Reviews Endocrinology 2019, 15(4):226-237. 13. Wu J, Ren W, Li L, Luo M, Xu K, Shen J, Wang J, Chang G, Lu Y, Qi Y et al: Effect of Aging and Glucagon-like Peptide 2 on Intestinal Microbiota in SD Rats. Aging and disease 2018, 9(4):566-577. 14. Magoč T, Salzberg SL: FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics (Oxford, England) 2011, 27(21):2957-2963. 15. Edgar RC: UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature methods 2013, 10(10):996-998. 16. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R: UCHIME improves sensitivity and speed of chimera detection. Bioinformatics (Oxford, England) 2011, 27(16):2194-2200. 17. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI et al: QIIME allows analysis of high-throughput community sequencing data. Nature methods 2010, 7(5):335-336. 18. Richard ML, Sokol H: The gut mycobiota: insights into analysis, environmental interactions and role in gastrointestinal diseases. Nature Reviews Gastroenterology & Hepatology 2019, 16(6):331-345. 19. Lang S, Duan Y, Liu J, Torralba MG, Kuelbs C, Ventura-Cots M, Abraldes JG, Bosques-Padilla F, Verna EC, Brown RS, Jr. et al: Intestinal Fungal Dysbiosis and Systemic Immune Response to Fungi in Patients With Alcoholic Hepatitis. Hepatology 2020, 71(2):522-538. 20. Forbes JD, Bernstein CN, Tremlett H, Van Domselaar G, Knox NC: A Fungal World: Could the Gut Mycobiome Be Involved in Neurological Disease? Frontiers in microbiology 2018, 9:3249. 21. References Jiang TT, Shao TY, Ang WXG, Kinder JM, Turner LH, Pham G, Whitt J, Alenghat T, Way SS: Commensal Fungi Recapitulate the Protective Benefits of Intestinal Bacteria. Cell host & microbe 2017, 22(6):809-816.e804. 22. Chin VK, Yong VC, Chong PP, Amin Nordin S, Basir R, Abdullah M: Mycobiome in the Gut: A Multiperspective Review. Mediators of inflammation 2020, 2020:9560684. 23. Lapthorne S, Pereira-Fantini PM, Fouhy F, Wilson G, Thomas SL, Dellios NL, Scurr M, O'Sullivan O, Ross RP, Stanton C et al: Gut microbial diversity is reduced and is associated with colonic Page 13/28 Page 13/28 inflammation in a piglet model of short bowel syndrome. Gut microbes 2013, 4(3):212-221. inflammation in a piglet model of short bowel syndrome. Gut microbes 2013, 4(3):212-221. 24. Levesque CL, Turner J, Li J, Wizzard P, St Pierre B, Lim D, Wales P: In a Neonatal Piglet Model of Intestinal Failure, Administration of Antibiotics and Lack of Enteral Nutrition Have a Greater Impact on Intestinal Microflora Than Surgical Resection Alone. JPEN Journal of parenteral and enteral nutrition 2017, 41(6):938-945. 25. Davidovics ZH, Carter BA, Luna RA, Hollister EB, Shulman RJ, Versalovic J: The Fecal Microbiome in Pediatric Patients With Short Bowel Syndrome. JPEN Journal of parenteral and enteral nutrition 2016, 40(8):1106-1113. 26. Engelstad HJ, Barron L, Moen J, Wylie TN, Wylie K, Rubin DC, Davidson N, Cade WT, Warner BB, Warner BW: Remnant Small Bowel Length in Pediatric Short Bowel Syndrome and the Correlation with Intestinal Dysbiosis and Linear Growth. Journal of the American College of Surgeons 2018, 227(4):439-449. 27. Shin NR, Whon TW, Bae JW: Proteobacteria: microbial signature of dysbiosis in gut microbiota. Trends in biotechnology 2015, 33(9):496-503. 28. Schönfeld P, Wojtczak L: Short- and medium-chain fatty acids in energy metabolism: the cellular perspective. Journal of lipid research 2016, 57(6):943-954. 29. Qing Y, Xie H, Su C, Wang Y, Yu Q, Pang Q, Cui F: Gut Microbiome, Short-Chain Fatty Acids, and Mucosa Injury in Young Adults with Human Immunodeficiency Virus Infection. Digestive diseases and sciences 2019, 64(7):1830-1843. 30. Ma Y, Hu C, Yan W, Jiang H, Liu G: Lactobacillus pentosus Increases the Abundance of Akkermansia and Affects the Serum Metabolome to Alleviate DSS-Induced Colitis in a Murine Model. Frontiers in cell and developmental biology 2020, 8:591408. 31. Wang L, Tang L, Feng Y, Zhao S, Han M, Zhang C, Yuan G, Zhu J, Cao S, Wu Q et al: A purified membrane protein from Akkermansia muciniphila or the pasteurised bacterium blunts colitis associated tumourigenesis by modulation of CD8（+）T cells in mice. Gut 2020, 69:1988-1997. 31. Wang L, Tang L, Feng Y, Zhao S, Han M, Zhang C, Yuan G, Zhu J, Cao S, Wu Q et al: A purified membrane protein from Akkermansia muciniphila or the pasteurised bacterium blunts colitis associated tumourigenesis by modulation of CD8（+）T cells in mice. Gut 2020, 69:1988-1997. 32. Depommier C, Everard A, Druart C, Plovier H, Van Hul M, Vieira-Silva S, Falony G, Raes J, Maiter D, Delzenne NM et al: Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nature medicine 2019, 25(7):1096-1103. 32. 39. Peleg AY, Hogan DA, Mylonakis E: Medically important bacterial-fungal interactions. Nature reviews Microbiology 2010, 8(5):340-349. 39. Peleg AY, Hogan DA, Mylonakis E: Medically important bacterial-fungal interactions. Nature reviews Microbiology 2010, 8(5):340-349. 39. Peleg AY, Hogan DA, Mylonakis E: Medically important bacterial-fungal interactions. Nature reviews Microbiology 2010, 8(5):340-349. 40. Imai T, Inoue R, Kawada Y, Morita Y, Inatomi O, Nishida A, Bamba S, Kawahara M, Andoh A: Characterization of fungal dysbiosis in Japanese patients with inflammatory bowel disease. Journal of gastroenterology 2019, 54(2):149-159. inflammation in a piglet model of short bowel syndrome. Gut microbes 2013, 4(3):212-221. Depommier C, Everard A, Druart C, Plovier H, Van Hul M, Vieira-Silva S, Falony G, Raes J, Maiter D, Delzenne NM et al: Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nature medicine 2019, 25(7):1096-1103. 33. Xu Y, Wang N, Tan HY, Li S, Zhang C, Feng Y: Function of Akkermansia muciniphila in Obesity: Interactions With Lipid Metabolism, Immune Response and Gut Systems. Frontiers in microbiology 2020, 11:219. 34. Morgan XC, Tickle TL, Sokol H, Gevers D, Devaney KL, Ward DV, Reyes JA, Shah SA, LeLeiko N, Snapper SB et al: Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome biology 2012, 13(9):R79. 34. Morgan XC, Tickle TL, Sokol H, Gevers D, Devaney KL, Ward DV, Reyes JA, Shah SA, LeLeiko N, Snapper SB et al: Dysfunction of the intestinal microbiome in inflammatory bowel disease and treatment. Genome biology 2012, 13(9):R79. 35. Fifi AC, Bayes L, Ehrenpreis ED, Chavez H: Prevalence of Bloodstream Infections in Children With Short-Bowel Syndrome With a Central Line Presenting to Emergency Department With Fever. JPEN Journal of parenteral and enteral nutrition 2020, 44(4):655-660. Page 14/28 36. Raphael BP, Fournier G, McLaughlin SR, Puder M, Jones S, Flett KB: Antibiotic Susceptibility and Therapy in Central Line Infections in Pediatric Home Parenteral Nutrition Patients. Journal of pediatric gastroenterology and nutrition 2020, 70(1):59-63. 37. El Mouzan M, Wang F, Al Mofarreh M, Menon R, Al Barrag A, Korolev KS, Al Sarkhy A, Al Asmi M, Hamed Y, Saeed A et al: Fungal Microbiota Profile in Newly Diagnosed Treatment-naïve Children with Crohn's Disease. Journal of Crohn's & colitis 2017, 11(5):586-592. 38. Banani H, Marcet-Houben M, Ballester AR, Abbruscato P, González-Candelas L, Gabaldón T, Spadaro D: Genome sequencing and secondary metabolism of the postharvest pathogen Penicillium griseofulvum. BMC genomics 2016, 17:19. 38. Banani H, Marcet-Houben M, Ballester AR, Abbruscato P, González-Candelas L, Gabaldón T, Spadaro D: Genome sequencing and secondary metabolism of the postharvest pathogen Penicillium griseofulvum. BMC genomics 2016, 17:19. Tables Table 1 PCR primers Table 1 PCR primers Primer name Sequences Bacterial 16S rRNA gene 515-F 5′-GTGCCAGCMGCCGCGGTAA-3′ 806-R 5′-GGACTACHVGGGTWTCTAAT-3′ Fugal ITS1 ITS-F 5’-CTTGGTCATTTAGAGGAAGTAA-3’ ITS-R 5’-GCTGCGTTCTTCATCGATGC-3’ Table 2 Relative abundance of bacteria of colonic samples. Page 15/28 Page 15/28 Page 15/28 Bacterial taxa Sham SBS GLP2.SBS Bacteroidetes.P Bacteroides.G 0.55 (0.42-1.12) 0.00 (0.00-0.01) * 0.00 (0.00-0.00) Odoribacter.G 0.75 (0.48-1.15) 0.00 (0.00-0.00) 0.00 (0.00-0.00) Prevotella.G 6.40 (4.63-8.32) 0.00 (0.00-0.00) 0.00 (0.00-0.02) * Firmicutes.P Lachnospiraceae.F 3.35 (2.49-4.01) 0.00 (0.00-0.21) * 0.02 (0.00-0.20) * Coprococcus.G 0.44 (0.23-0.63) 0.00 (0.00-0.00) 0.00 (0.00-0.00) Dorea.G 0.04 (0.02-0.36) 0.00 (0.00-0.00) 0.0000 (0.00-0.00) Ruminococcaceae.F 5.73 (4.48-7.34) 0.00 (0.00-0.00) * 0.00 (0.00-0.00) * Oscillospira.G 1.55 (1.23-1.98) 0.00 (0.00) ** 0.00 (0.00) * Ruminococcus.G 1.70 (1.29-2.34) 0.00 (0.00-0.00) * 0.00 (0.00-0.00) * Clostridium.G 0.09 (0.05) 15.10 (11.40) 33.55 (12.67) # Verrucomicrobia.P Akkermansia.G 33.90 (23.69-39.75) 0.00 (0.00-0.00) 0.00 (0.00-0.00) * Proteobacteria.P Proteus.G 0.00 (0.00-0.00) 2.01 (1.23-2.37) ** 0.02 (0.02-0.05) # Data were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR P< 0.05, P< 0.01, *P< 0.001 compared with Sham group, #P< 0.05 compared with SBS group. .P= Phylum, .F= Family, .G= Genus Data were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR P< 0.05, P< 0.01, P< 0.001 compared with Sham group, #P< 0.05 compared with SBS group. Data were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR P< 0.05, P< 0.01, P< 0.001 compared with Sham group, #P< 0.05 compared with SBS group. .P= Phylum, .F= Family, .G= Genus .P= Phylum, .F= Family, .G= Genus Table 3 Relative abundance of fungal genera of colonic samples. Table 3 Relative abundance of fungal genera of colonic samples. Tables Page 16/28 Fungal taxa Sham SBS GLP2.SBS Eurotiomycetes.C Aspergillus.G 16.43 (6.48) 5.36 (1.61) * 9.14 (2.99) Penicillium.G 2.46 (1.07) 0.41 (0.16) ** 1.75 (0.65) # Xerochrysium.G 0.00 (0.00-0.03) 0.31 (0.10-0.75) * 0.11 (0.03-0.50) Saccharomycetes.C Wickerhamomyces.G 7.41 (2.27) 3.33 (1.62) * 4.12 (1.63) Debaryomyces.G 0.00 (0.00-0.17) 3.90 (2.43-7.56) 0.30 (0.17-0.37) # Meyerozyma.G 4.66 (1.51) 13.51 (5.27) 1.48 (0.26) ### Fungal taxa Sham SBS GLP2.SBS Eurotiomycetes.C Aspergillus.G 16.43 (6.48) 5.36 (1.61) 9.14 (2.99) Penicillium.G 2.46 (1.07) 0.41 (0.16) 1.75 (0.65) # Xerochrysium.G 0.00 (0.00-0.03) 0.31 (0.10-0.75) * 0.11 (0.03-0.50) Saccharomycetes.C Wickerhamomyces.G 7.41 (2.27) 3.33 (1.62) * 4.12 (1.63) Debaryomyces.G 0.00 (0.00-0.17) 3.90 (2.43-7.56) 0.30 (0.17-0.37) # Meyerozyma.G 4.66 (1.51) 13.51 (5.27) 1.48 (0.26) ### ata were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA d Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, uskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR <0.05, *P< 0.01 compared with Sham group, #P <0.05, ###P< 0.001 compared with SBS group. = Class G= Genus Data were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR # ### Data were expressed as percentages of total sequences. For data passed normality test, one-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean (SD). Otherwise, Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR P <0.05, *P< 0.01 compared with Sham group, #P <0.05, ###P< 0.001 compared with SBS group. P <0.05, *P< 0.01 compared with Sham group, #P <0.05, ###P< 0.001 compared with SBS group. C= Class G= Genus .C= Class, .G= Genus Figures Page 17/28 Figure 1 Bacterial microbiota α-diversity biodiversity analysis. [A] Experimental design and procedure of this study. [B] The establishment of sham and type 2 SBS rat model. [C] Rarefaction curves of each sample. [D] OTU counts analysis among groups. Bacterial α-diversity estimated by Chao [E], ACE [F], Shannon’s diversity [G], Simpson’ diversity [H] and Good’s coverage [I]. D-I Median and IQR were shown. The abnormal value is shown as 'o'. P <0.05, *P <0.01 compared with the Sham group. Fi 1 Figure 1 Figure 1 Bacterial microbiota α-diversity biodiversity analysis. [A] Experimental design and procedure of this study. [B] The establishment of sham and type 2 SBS rat model. [C] Rarefaction curves of each sample. [D] OTU counts analysis among groups. Bacterial α-diversity estimated by Chao [E], ACE [F], Shannon’s diversity [G], Simpson’ diversity [H] and Good’s coverage [I]. D-I Median and IQR were shown. The abnormal value is shown as 'o'. P <0.05, *P <0.01 compared with the Sham group. Page 18/28 Page 19/28 Figure 2 Comparative analyses of the bacterial microbiota communities. [A] Samples’ clustering res (Description, weighted_unifrac). [B] Beta diversity heat map (Description, weighted_unifrac) β-diversity of principal coordinates analysis (PCoA) based on weighted_unifrac distance ra curves of each sample. [D] Partial least squares discriminant analysis (PLS-DA) plots based of OTU. [E] Anosim similarity analysis based on weighted_unifrac distance rarefaction curv Figure 2 Figure 2 Comparative analyses of the bacterial microbiota communities. [A] Samples’ clustering result (Description, weighted_unifrac). [B] Beta diversity heat map (Description, weighted_unifrac). [C] Bacterial β-diversity of principal coordinates analysis (PCoA) based on weighted_unifrac distance rarefaction curves of each sample. [D] Partial least squares discriminant analysis (PLS-DA) plots based on the level of OTU. [E] Anosim similarity analysis based on weighted_unifrac distance rarefaction curves of each Page 19/28 Page 19/28 sample. [F] Anosim similarity analysis based on Bray_curtis_OTU distance rarefaction curves of each sample. [F] Anosim similarity analysis based on Bray_curtis_OTU distance rarefaction curves of each sample. [F] Anosim similarity analysis based on Bray_curtis_OTU distance rarefaction curves of each sample. sample. [F] Anosim similarity analysis based on Bray_curtis_OTU distance rarefaction curves of ea sample. Figure 3 Composition of the bacterial gut microbiota. [A] Proportion of different phyla in each sample. [B-L] Relative abundance of top 11 phylum across groups. [B-D] One-way ANOVA and Tukey's multiple p Figure 3 Figure 3 Composition of the bacterial gut microbiota. [A] Proportion of different phyla in each sample. [B-L] Relative abundance of top 11 phylum across groups. [B-D] One-way ANOVA and Tukey's multiple Page 20/28 Page 20/28 comparisons test were used, and data were shown as mean ± SD；[E-L] Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR. P< 0.05, P<0.01, P<0.001. Q , , p , Q , , Figure 4 Variations of the bacterial gut microbiota. [A] Cladogram depicting the taxonomic hierarchical structure of distinguished phylotype created by LEfSe analysis. Each filled circle indicates a certain type of phylotype. Different colors represent different groups. Colored node consistent with the group color indicated important microbe biomarkers in the group and the name of biomarkers are listed in the upper right corner. The yellow notes suggest biomarkers which do not show any significant differences among groups. [B] Graphics of Linear discriminant analysis (LDA). Horizontal bars demonstrated the effect size. The length of the bar indicates the log10 transformed LDA score, represented by vertical dotted lines. The bacteria with statistically significant changes (p < 0.05) are shown alongside the horizontal lines. Figure 4 Variations of the bacterial gut microbiota. [A] Cladogram depicting the taxonomic hierarchical structure of distinguished phylotype created by LEfSe analysis. Each filled circle indicates a certain type of phylotype. Different colors represent different groups. Colored node consistent with the group color indicated important microbe biomarkers in the group and the name of biomarkers are listed in the upper right corner. The yellow notes suggest biomarkers which do not show any significant differences among groups. [B] Graphics of Linear discriminant analysis (LDA). Horizontal bars demonstrated the effect size. The length of the bar indicates the log10 transformed LDA score, represented by vertical dotted lines. The bacteria with statistically significant changes (p < 0.05) are shown alongside the horizontal lines. Page 21/28 Figure 5 Correlation networks of intestinal bacterial community. Correlations between bacteria were analyzed by Spearman correlation analysis. [A] Correlation networks of abundance are shown, in which each node represents an OTU, its color represents the bacterial phylum and its size indicates the number of direct Figure 5 Correlation networks of intestinal bacterial community. Correlations between bacteria were analyzed by Spearman correlation analysis. [A] Correlation networks of abundance are shown, in which each node represents an OTU, its color represents the bacterial phylum and its size indicates the number of direct edges. Edges suggest the magnitude of distance correlation (positive in green, negative in red). Only OTUs presented in >1/3 of all samples would be taken into account. The table in the middle panel shows Page 22/28 Page 22/28 the parameters of network. The relative connectivity of networks was calculated as the ratio of the number of significant interactions to the number of nodes in the correlation network. [B] Quantification of neighbor amount in Sham, SBS and GLP2.SBS, means and SEM are marked in red. *p<0.001. avg, average. average. Figure 6 Fungal microbiota α-diversity biodiversity analysis. [A] Rarefaction curves of each sample. [B] Core-Pan OTU analysis, circle in the middle indicates the number of shared OTUs in three groups, and the ellipse outside the middle circle suggest the number of OTUs unique to certain group. [C] OTU counts analysis among groups. Fungal α-diversity estimated by Chao [D] and ACE index [E]. [C] One-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean ± SD；[D-E] Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR. The abnormal value is shown as 'o'. P <0.05 compared with the Sham group. Figure 6 Fungal microbiota α-diversity biodiversity analysis. [A] Rarefaction curves of each sample. [B] Core-Pan OTU analysis, circle in the middle indicates the number of shared OTUs in three groups, and the ellipse outside the middle circle suggest the number of OTUs unique to certain group. [C] OTU counts analysis among groups. Fungal α-diversity estimated by Chao [D] and ACE index [E]. [C] One-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean ± SD；[D-E] Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR. The abnormal value is shown as 'o'. P <0.05 compared with the Sham group. Page 23/28 Page 23/28 Figure 7 Comparative analyses of intestinal fungal microbiota communities. [A] Samples’ clu (Description, weighted_unifrac). [B] Beta diversity heat map (Description, weighted_u diversity of PCoA based on weighted_unifrac distance rarefaction curves of each sa plots based on the level of OTU. [E] Anosim similarity analysis based on weighted_u Figure 7 Figure 7 Comparative analyses of intestinal fungal microbiota communities. [A] Samples’ clustering result (Description, weighted_unifrac). [B] Beta diversity heat map (Description, weighted_unifrac). [C] Fungal β- diversity of PCoA based on weighted_unifrac distance rarefaction curves of each sample. [D] PLS-DA plots based on the level of OTU. [E] Anosim similarity analysis based on weighted_unifrac distance Page 24/28 rarefaction curves of each sample. [F] Anosim similarity analysis based on Bray_curtis_OTU distance rarefaction curves of each sample. Figure 8 Composition of intestinal fungal microbiota. [A] Proportion of the different classes in each sample. [B-F] Relative abundance of top 5 classes of gut fungus among three groups. [B-E] One-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean ± SD, [F] Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR. P< 0.05, *P<0.01. Figure 8 Figure 8 Composition of intestinal fungal microbiota. [A] Proportion of the different classes in each sample. [B-F] Relative abundance of top 5 classes of gut fungus among three groups. [B-E] One-way ANOVA and Tukey's multiple comparisons test were used, and data were shown as mean ± SD, [F] Kruskal-Wallis and Dunn's multiple comparisons test were used, and data were shown as median and IQR. P< 0.05, *P<0.01. Page 25/28 Figure 9 Variations of the fungal gut microbiota. [A] Taxonomic Cladogram depicting the hierarchical structure of distinguished phylotype created by LEfSe analysis. [B] LDA score computed by LEfSe analysis. The fungus with statistically significant changes (p < 0.05) is shown alongside the horizontal lines. Figure 9 Variations of the fungal gut microbiota. [A] Taxonomic Cladogram depicting the hierarchical structure of distinguished phylotype created by LEfSe analysis. [B] LDA score computed by LEfSe analysis. The fungus with statistically significant changes (p < 0.05) is shown alongside the horizontal lines. Figure 9 Variations of the fungal gut microbiota. [A] Taxonomic Cladogram depicting the hierarchical structure of distinguished phylotype created by LEfSe analysis. [B] LDA score computed by LEfSe analysis. The fungus with statistically significant changes (p < 0.05) is shown alongside the horizontal lines. Figure 9 Variations of the fungal gut microbiota. [A] Taxonomic Cladogram depicting the hierarchical structure of distinguished phylotype created by LEfSe analysis. [B] LDA score computed by LEfSe analysis. The fungus with statistically significant changes (p < 0.05) is shown alongside the horizontal lines. Page 26/28 Figure 10 Mi bi l i t l ti t k f SBS t i th lid ti t d ITS1/16S di it Page 27/28 Figure 10 Microbial interrelation network of SBS rats in the validation study. ITS1/16S diversity ratio by the Chao [A] and ACE index [B] in the three surgical groups, One-way ANOVA test follow multiple comparisons test were applied to determine differences among groups. [C] The a top 50 bacteria and top 20 fungi was analyzed using Spearman correlation analysis. Only correlations (P ≤ 0.05) are displayed with an edge. The colors of edge denote positive cor Figure 10 Figure 10 Page 27/28 Microbial interrelation network of SBS rats in the validation study. ITS1/16S diversity ratio was computed by the Chao [A] and ACE index [B] in the three surgical groups, One-way ANOVA test followed by Tukey's multiple comparisons test were applied to determine differences among groups. [C] The abundance of top 50 bacteria and top 20 fungi was analyzed using Spearman correlation analysis. Only significant correlations (P ≤ 0.05) are displayed with an edge. The colors of edge denote positive correlation in green Page 27/28 Page 27/28 and negative correlation in red. Each node represents certain microbial genera and each color indicates specific microbial phyla. The table shows the parameters of network. The relative connectivity of networks was calculated as the ratio of number of significant interactions (edges) to the number of taxa (nodes) in the network. [D] Average number of neighbors. Means and SEM are indicated in figure [A], [B] and [D]. p<0.05, **p<0.001. avg, average. and negative correlation in red. Each node represents certain microbial genera and each color indicates specific microbial phyla. The table shows the parameters of network. The relative connectivity of networks was calculated as the ratio of number of significant interactions (edges) to the number of taxa (nodes) in the network. [D] Average number of neighbors. Means and SEM are indicated in figure [A], [B] and [D]. p<0.05, ***p<0.001. avg, average. Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Figure11jpeg.jpg FigureS1jpeg.jpg FigureS2jpeg.jpg Figure11jpeg.jpg FigureS1jpeg.jpg Page 28/28
https://openalex.org/W2791148848	https://digibug.ugr.es/bitstream/10481/55900/1/Rabello%20Mendes_Infant%20Stools.pdf	English	null	TECHNOLOGY PROSPECTING OF INFANT STOOL AS A SOURCE OF NOVEL PROBIOTIC PRODUCTS	Revista GEINTEC	2,018	cc-by	3,244	Hermínio Benítez Rabello Mendes1,2, Julio Plaza-Díaz4, Maria Rosa Quaresma Bomfim2, Angel Gil Hernandez4, and Valério Monteiro-Neto 2, 3 1 Programa de Pós-graduação em Biodiversidade e Biotecnologia da Amazônia Legal-Rede BIONORTE, Universidade Federal do Maranhão-UFMA, São Luís-MA, Brasil herminio.mendes@ceuma.br 2 Universidade CEUMA, São Luís-MA, Brasil marqbomfim@gmail.com 3 Universidade Federal do Maranhão-UFMA, São Luís-MA, Brasil vmonteironeto@ceuma.br 4Universidad de Granada, Granada, Andalucia, España jplaza@ugr.es; a.gil@ugr.es Hermínio Benítez Rabello Mendes1,2, Julio Plaza-Díaz4, Maria Rosa Quaresma Bomfim2, Angel Gil Hernandez4, and Valério Monteiro-Neto 2, 3 1 Programa de Pós-graduação em Biodiversidade e Biotecnologia da Amazônia Legal-Rede BIONORTE, Universidade Federal do Maranhão-UFMA, São Luís-MA, Brasil herminio.mendes@ceuma.br 2 Universidade CEUMA, São Luís-MA, Brasil marqbomfim@gmail.com 3 Universidade Federal do Maranhão-UFMA, São Luís-MA, Brasil vmonteironeto@ceuma.br 4Universidad de Granada, Granada, Andalucia, España jplaza@ugr.es; a.gil@ugr.es Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 1. Introduction According to the Food and Agriculture Organization of the United Nations and the World Health Organization (WHO), probiotics are live microorganisms that confer a health benefit to the host when they are administered in adequate amounts (FAO/WHO, 2006). Specially, bacteria belonging to Bifidobacterium and Lactobacillus, which are the predominant and subdominant groups of the gastrointestinal microbiota, respectively, are the most widely used probiotic bacteria and are included in many functional foods and dietary supplements (GOURBEYRE et al., 2011). Generally, the selection of new probiotic strains for human use is sourced from human intestinal flora and breast milk (MARTIN et al., 2009). With a long history of the safe use of probiotics in fermented dairy products and growing recognition of their beneficial effects on human health, the food industry has become increasingly interested in these types of microorganisms (SHERIDAN et al., 2014). Often, the criteria for the selection of probiotics include tolerance to gastrointestinal conditions (gastric acid and bile), the ability to adhere to the gastrointestinal mucosa and competitive exclusion of pathogens (SANDERS, 2013). The mechanisms underlying the beneficial effects of probiotics are largely unknown but are likely to be multifactorial. Several mechanisms related to the antagonistic effects of probiotics on various microorganisms include the following: secretion of antimicrobial substances, competitive adherence to the mucosa and epithelium, strengthening of the gut epithelial barrier, and modulation of the immune system (NEAL- MCKINNEY et al., 2012; GIORGETTI et al., 2015). Probiotic bacteria, both native and those introduced by feeding the host organism, can control the initiation and/or development of many pathogenic microorganisms, including Salmonella typhymurium, Shigella spp., Clostridium difficile, Campylobacter jejuni and Escherichia coli. Probiotics also provide important protection against urogenital pathogens such as Gardnerella vaginalis, Candida albicans, and Chlamydia trachomatis (MENARD, 2011). Thus, these probiotic strains have been the subject of numerous studies because of an increase in the variability of microbial species that have been discovered and tested with potential probiotic capabilities. Abstract Probiotics are live microorganisms that provide health benefits to the host when they are ingested in adequate amounts. The strains most frequently used as probiotics include lactic acid bacteria and bifidobacteria. These microorganisms have significant potential as therapeutic options for a variety of diseases, but the mechanisms responsible for these effects have not yet been fully elucidated. In general, new probiotic strains for human use are sourced from human intestinal flora and breast milk. This study sought to survey the existing research regarding the potential for and development of technologies regarding infant stool-associated probiotics. The study was performed using information from the European Patent Bank, the Bank of the World Intellectual Property Organization and the database of the National Institute of Industrial Property of Brazil. The most abundant international classification prospected was Subsection C12, particularly class A61N. The Republic of Korea has the most patents filed. The results indicate a promising area with significant growth in the number of patents. Keywords: infant stools; probiotics products; technology prospecting; patents Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4240 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 2. Methods We used the following terms in conjunction with probiotics in the database search: infant stools, breast-fed-infant stools, feces of infants, and baby’s stool. The databases used for patent research were: INPI - National Institute of Industrial Property; Espacenet - European Patent Office (EPO); USPTO - United States Patents and Trademarks Office; and WIPO - Intellectual Property Digital Library. The databases used for research papers, theses and dissertations were CAPES Periodicals; PubMed; and SCIELO. Initially, the search in the patent databases used the search term "probiotic and infant stool”, which resulted in a much lower number of results in the EPO and WIPO databases compared with the USPTO database. 1. Introduction This change has meant that the previous target audience, consisting of elderly people, children and adults with poor health, has since expanded to also cover healthy individuals and even animals (KORTERINK et al., 2014) Therefore, the accelerated development of scientific knowledge regarding the role of probiotics conveyed by dairy products and certain non-dairy products on the health of the host will almost certainly result in an expansion of the range of probiotic product options available to consumers (ROWLAND et al., 2010). Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4241 Furthermore, consumers are now more aware of probiotics and are increasingly opting for products that support a balanced lifestyle, provide health benefits and are attractive from a sensory standpoint. Consequently, the market for these products is increasingly competitive. The objective of this research is to conduct a survey on patent bases of documents related to probiotic products from infants stools, in order to evaluate the panorama of the use of this raw material in the elaboration of these products for human health. 3. Results and discussion The searches of the patent databases (INPI, EPO, USPTO and WIPO) identified a total of 7 relevant patents in the WIPO database, 2 patents in the EPO database and 2 patents in the USPTO database, as shown in the data presented in Figure 1. No patents were found in the INPI database. Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 4242 D.O.I.: 10.7198/geintec.v8i1.942 0 1 2 3 4 5 6 7 8 INPI ESPACENET USPTO WIPO Number of Patents Database PATENTS Figure 1. Patents filed by database Source: Author's own figure 0 1 2 3 4 5 6 7 8 INPI ESPACENET USPTO WIPO Number of Patents Database PATENTS Source: Author's own figure 4242 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Recently, the INPI has concentrated on using the system of industrial property for more than its function of protecting intellectual property. The restructuring work, undertaken since 2004, has intended to allow use of the system as a tool of empowerment and competitiveness, which are essential to improving countries’ economic and technological development (BARROSO et al., 2009) The documents identified by our searches provided us with the number of relevant patents deposited each year. Using the WIPO, which is the most specific patent database (because the necessary information to register a patent in the WIPO includes a set of legal references related to the patent’s theme of development), we noted that the deposition of patents fluctuated in the past decade but remained constant in recent years, although the last year of record is 2014. Figure 2. Number of patent filed per year, 2016 0 1 2 3 Number of Patents Year of Filing The analysis of the country of origin of each depositor can be observed in Figure 3. The epublic of Korea is the largest patent holder, with 5 deposited patents. The United States and apan each have only one registered patent. Source: Author's own figure 0 1 2 3 Number of Patents Year of Filing Source: Author's own figure 0 1 2 3 Number of Patents Year of Filing Source: Author's own figure Source: Author's own figure The analysis of the country of origin of each depositor can be observed in Figure 3. The Republic of Korea is the largest patent holder, with 5 deposited patents. The United States and Japan each have only one registered patent. Figure 2. Number of patent filed per year, 2016 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 3. Results and discussion 4243 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Figure 3. Patents filed by country of origin, 2016 Source: Author's own figure 0 1 2 3 4 5 6 Republic of Korea United States Japan Number of Patents Source: Author's own figure Brazil does not have any patents in the area of probiotics associated with infant stools. perhaps this fact maybe associated with the provisions of Brazilian Law n° 13.123/15 which provides for access to genetic heritage, protection and access to associated traditional knowledge, and sharing of benefits for the conservation and sustainable use of biodiversity. However it is possible to patent genetically modified microorganisms, except for all plants or animals or parts thereof (MATTOS; CARDOZO, 2013). In Brazil, the ANVISA, National Sanitary Surveillance Agency, which is responsible for certifying the production of food products, decided to exclude the list of approved probiotics and defines that the company should propose a functional or health claim and it will be evaluated case by case, based on the definitions and principles set out in Resolution no. 18/1999. At the same time, functional properties and/or health claims on a food label are tools that allow the consumer to know the possible benefits the food can offer. However, it is necessary to clearer rules for the industry so that the consumer is not deceived by false cure promises (SILVEIRA et al., 2009) Among the documents identified in the survey, 55% are related to the field of microorganisms or enzymes and their compositions (C12N); 18% are in the area of food, food products and non-alcoholic beverages and preparations for medical purposes (A61K and A23C, respectively); and 9% are in the area of dairy products (A61K) (Figure 4). Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4244 Figure 4. Patents filed by application area, 2016. 18% 9% 55% 18% A61K A23L C12N A23C Source: Author's own figure Source: Author's own figure Source: Author's own figure As proposed by Vasconcellos et al. (2001), the subclass C12N was used as most representative of modern biotechnology, as it encompasses the technologies related to "Microorganisms or enzymes; their compositions; propagation, preservation or maintenance of micro-organisms or tissues; genetic engineering or mutation; culture mediums". 3. Results and discussion Significant advances in the area of biotechnology in Brazil came mainly with the approval of Law No. 9279/96, which controls industrial property rights and obligations. A61K (preparations for medical, dental or hygienic purposes) and A23C which is part of Section A - for human needs, linked to class 23 related to food or food products; not covered by other classes and subclass C dairy products, ex. milk, butter, cheese; substitutes for milk or cheese; important feature associated mainly with products with probiotic purposes. Among the probiotics, the genera Bifidobacterium and Lactobacillus stand out (VARAVALLO et al., 2008), in particular, the species Lactobacillus acidophilus (THAMER; PENNA, 2005). These genera are present in yogurts, fermented dairy products and dietary supplements (RAIZEL et al., 2011). Enterococcus faecium are also employed, but on a smaller scale, in addition to other lactic acid bacteria, non-lactic acid bacteria and yeasts (PROBIOTICS, [2010?]). The search on patent databases and scientific production identified 17,166 documents in PubMed (US National Library of Medicine), 596 scientific papers in the Scielo (Scientific Electronic Library Online) and 15,672 scientific papers in the CAPES database, as shown in Table 1. Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4245 Table 1. Number of research papers, theses and dissertations from keyword searches in different databases Keyword CAPES PubMed SCIELO Probiotic* 15,478 16,682 573 Probiotic* and (breast-fed-infant stools OR infant feces OR baby's stool) 130 324 21 Food* and probiotic* and (breast-fed-infant stools OR infant feces OR baby’s stool) 37 147 1 Probiotic* and Product* and (breast-fed-infant stools OR infant feces OR baby's stool) 27 13 1 Total 15,672 17,166 596 With the search keywords and associations using Boolean operators outlined in Table 1, the largest volume of documents was found with the PubMed search. This is because PubMed is the search engine for the MEDLINE database of references and abstracts, and MEDLINE is the largest medical and natural sciences database in the world. According to the analysis of the articles extracted from MEDLINE by year, the annual production of work related to probiotics and associated with the word "infant stool” was scarce until the end of the 1990s and fluctuated greatly in the 2000s, with the highest number of publications in 2014. 3. Results and discussion The number of scientific papers found was much higher than the number of patents in the WIPO database (Figure 5). Figure 5. Annual number of articles deposited into the PubMed database (1992-2016). Figure 5. Annual number of articles deposited into the PubMed database (1992-2016). Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 4246 D.O.I.: 10.7198/geintec.v8i1.942 0 5 10 15 20 25 30 35 40 1993 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Source: Author's own figure 0 5 10 15 20 25 30 35 40 1993 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4246 4. Conclusions The Republic of Korea has largest number of patent applications in this area, followed by the United States and Japan. Brazilian legislation is not clear as to the use of probiotics, since in the market are found foods considered probiotics, and often used as medicines. Although, several scientific projects specifically in the area of probiotics has grown tremendously in recent years. The public expects future innovations to produce more diverse probiotic products aimed at contributing to the improvement of consumer health. References BARROSO, W.; QUONIAM, L.; PACHECO, E. Patents as technological information in Latin America. World Patent Information, vol. 31, n 3, p 207-215, 2009. CAPES, 2016. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior. Disponível em: Acesso em 2016 >http://www-periodicos-capes-gov-br.ez14.periodicos.capes.gov.br/< BARROSO, W.; QUONIAM, L.; PACHECO, E. Patents as technological information in Latin America. World Patent Information, vol. 31, n 3, p 207-215, 2009. CAPES, 2016. Coordenação de Aperfeiçoamento de Pessoal de Nível Superior. Disponível em: Acesso em 2016 >http://www-periodicos-capes-gov-br.ez14.periodicos.capes.gov.br/< EPO, 2016. Europen Patent Office. Disponível em:. >https://www.epo.org/index.html< Acesso em: jun.2016 FAO/WHO. Probiotics in Food: Health and Nutritional Properties and Guidelines for Evaluation. FAO Food and Nutrition Paper 85. Food and Agriculture Organization of the United Nations and World Health Organization, Rome. 2006 GIORGETTI, G.; BRANDIMARTE, G.; FABIOCCHI, F., et al. Interactions between Innate Immunity, Microbiota, and Probiotics. Journal Immunology Res, v. 2015, n., p. 501361, 2015. GOURBEYRE P, DENERY S, BODINIER M: Probiotics, prebiotics, and synbiotics: impact on the gut immune system and allergic reactions. Journal Leukocyte Biol 2011; 89: 685–695 INPI, 2016. Instituto Nacional de Propriedade Industrial. Disponível em: >http://www.inpi.gov.br/.< Acesso em jun.2016. KORTERINK, J. J.; OCKELOEN, L.; BENNINGA, M. A., et al. Probiotics for childhood functional gastrointestinal disorders: a systematic review and meta-analysis. Acta Paediatrica, v. 103, n. 4, p. 365-372, 2014. LIONG MT: Probiotics: Biology, Genetics and Health Aspects. Microbiology Monographs. Heidelberg, Springer, 2011 MENARD, J. P. Antibacterial treatment of bacterial vaginosis: current and emerging therapies. International Journal Womens Health, v. 3, n., p. 295-305, 2011. MATOS, C.; CARDOZO; CARDOZO, V.S. Brazil: General overview on patenting of microrganisms in Brazil. Mondaq Connecting Knowledge & Peaple. Disponivel em: < http://www.mondaq.com >.Acesso em: 29 de out. 2017 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4247 – 2018 MARTIN, R.; JIMENEZ, E.; HEILIG, H., et al. Isolation of bifidobacteria from breast milk and assessment of the bifidobacterial population by PCR-denaturing gradient gel electrophoresis and quantitative real-time PCR. Applied Environ Microbiol, v. 75, n. 4, p. 965-969, 2009 NEAL-MCKINNEY, J. M.; LU, X.; DUONG, T., et al. Production of organic acids by probiotic lactobacilli can be used to reduce pathogen load in poultry. PLoS One, v. 7, n. 9, p. e43928, 2012. OLIVEIRA, L. G; PINTO, C. A.; RIBEIRO, N. M. et. al. Informação de patentes: ferramenta indispensável para a pesquisa e o desenvolvimento tecnológico. Quimica Nova, v. 28, supl., p. S36-S40, 2005. OLIVEIRA, A. Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 References F.; ROMAN, J. A. Alimentos Funcionais. In:____. Nutrição para Tecnologia e Engenharia de Alimentos, Curitiba: CRV, 2013. p. 151-164. PINHEIRO, L. V. R. Inteligência competitiva como disciplina da Ciência da Informação e sua trajetória e evolução no Brasil. In: STAREC, C.; GOMES, E. B. P.; CHAVES, J. B. L. Gestão estratégica da informação e inteligência competitiva. São Paulo: Saraiva, 2005. p. 17-32 PROBIÓTICOS, prebióticos e simbióticos. Funcionais Nutracêuticos, São Paulo, n. 2, p. 55- 62, [2010?] RAIZEL, R. et al. Efeitos do consumo de probióticos, prebióticos e simbióticos para o organismo humano. Revista Ciência & Saúde, Porto Alegre, v. 4, n. 2, p. 66-74, 2011 ROWLAND, I; CAPURSO L; COLLINS, K; et al. Current level of consensus on probiotic science. Gut Microbes. v. 1, n. 6, p. 436-439, 2010 SANDERS, M. E.; GUARNER, F.; GUERRANT, R., et al. An update on the use and investigation of probiotics in health and disease. Gut, v. 62, n. 5, p. 787-796, 2013. SCIELO, 2016. Scientific Electronic Library Online. Disponível em: Acesso em jun. 2016 SHERIDAN, P. O.; BINDELS, L. B.; SAULNIER, D. M., et al. Can prebiotics and probiotics improve therapeutic outcomes for undernourished individuals? Gut Microbes, v. 5, n. 1, p. 74-82, 2014. SILVEIRA, T.F.V.; VIANA, C.M.M.; MOSEGUI, G.B.G. Brazilian legislation for functional 1189 foods and the interface with the legislation for other food and medicine classes: contradictions and omissions. Revista de Saúde Coletiva, Rio de Janeiro, 19 [ 4 ]: 1189-1202, 2009. THAMER, K. G.; PENNA, A. L. B. Efeito do teor de soro, açúcar e de fruto-oligossacarídeos sobre a população de bactérias lácticas probióticas em bebidas fermentadas. Revista Brasileira de Ciências Farmacêuticas, São Paulo, v. 41, n. 03, 2005 USPTO, 2016. United States Patent and Trademark Office. Acesso em: 6 jun. 2016. VARAVALLO, M. A. et al. Aplicação de bactérias probióticas para profilaxia e tratamento de doenças gastrointestinais. Semina: Ciências Biológicas e da Saúde, Londrina, v. 29, n. 1, p. 83- 104, 2008 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4248 VASCONCELLOS, A. G.. Propriedade intelectual dos conhecimentos associados à biodiversidade, com ênfase nos derivados de plantas medicinais – desafio para inovação biotecnológica no Brasil. Tese de Doutorado, Universidade Federal do Rio de Janeiro, Brasil, 170 pp, 2003 WIPO, 2016. World Intellectual Property Organization. Acesso em: 6 jun. 2016 Recebido: 21/06/2016 Aprovado: 07/03/2018 Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Revista GEINTEC– ISSN: 2237-0722. Aracaju/SE. Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 References Vol.8, n.1, p.4240-4249, jan/fev/mar – 2018 D.O.I.: 10.7198/geintec.v8i1.942 4249
https://openalex.org/W4383989125	https://journals.plos.org/climate/article/file?id=10.1371/journal.pclm.0000241&type=printable	English	null	Climate obstruction in the Global South: Future research trajectories	PLOS climate	2,023	cc-by	2,755	Guy Edwards1☯, Paul K. GellertID2☯, Omar FaruqueID3☯, Kathryn Hochstetler4☯, Pamela D. McElwee5☯, Prakash Kaswhan6☯, Ruth E. McKieID7☯, Carlos MilaniID8☯, Timmons Roberts9☯, Jonathan WalzID10☯ 1 School of Global Studies, University of Sussex, Sussex, United Kingdom, 2 Department of Sociology, University of Tennessee, Knoxville, Tennessee, United States of America, 3 Department of Global Development Studies, Queen’s University, Ontario, Canada, 4 Department of International Development, London School of Economics and Political Science, London, United Kingdom, 5 Department of Human Ecology, Rutgers University, New Brunswick, New Jersey, United States of America, 6 Heller School for Social Policy & Management, Brandeis University, Waltham, Massachusetts, United States of America, 7 School of Criminology and Community Justice, De Montfort University, Leicester, United Kingdom, 8 Institute for Social and Political Studies, Rio De Janeiro State University, Rio De Janeiro, Brazil, 9 Institute at Brown for Environment and Society, Brown University, Providence, Rhode Island, United States of America, 10 SIT Graduate Institute, Brattleboro, Vermont, United States of America ☯These authors contributed equally to this work. @ a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 PLOS CLIMATE PLOS CLIMATE 1. Introduction Copyright: © 2023 Edwards 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. “Climate Obstruction” broadly refers to campaigns and other policy actions led by well-orga- nized and financed networks of corporate and other actors who have actively sought to prevent global and/or national action on climate change over the past four decades [1]. Their efforts have delayed ambitious climate action using organized mainstream and social media cam- paigns, lobbying, funding politicians and political campaigns, and disseminating climate- delaying discourses and practices [2]. In turn, these campaigns often shape public debates, which can affect political support and collective mobilization to mitigate climate change. How- ever, to date, most of the research on climate obstruction has focused on countries in the Global North, especially the United States. Given considerable gaps in research and knowl- edge, a future research agenda needs to shine greater light on if and in what form climate obstruction in the Global South appears. Editor: Jamie Males, PLOS Climate, UNITED KINGDOM Editor: Jamie Males, PLOS Climate, UNITED KINGDOM OPINION Climate obstruction in the Global South: Future research trajectories Guy Edwards1☯, Paul K. GellertID2☯, Omar FaruqueID3☯, Kathryn Hochstetler4☯, Pamela D. McElwee5☯, Prakash Kaswhan6☯, Ruth E. McKieID7☯, Carlos MilaniID8☯, Timmons Roberts9☯, Jonathan WalzID10☯ Guy Edwards1☯, Paul K. GellertID2☯, Omar FaruqueID3☯, Kathryn Hochstetler4☯, Pamela D. McElwee5☯, Prakash Kaswhan6☯, Ruth E. McKieID7☯, Carlos MilaniID8☯, Timmons Roberts9☯, Jonathan WalzID10☯ OPEN ACCESS Citation: Edwards G, Gellert PK, Faruque O, Hochstetler K, McElwee PD, Kaswhan P, et al. (2023) Climate obstruction in the Global South: Future research trajectories. PLOS Clim 2(7): e0000241. https://doi.org/10.1371/journal. pclm.0000241 ☯These authors contributed equally to this work. ruth.mckie@dmu.ac.uk Abstract “Climate Obstruction” broadly refers to campaigns and other policy actions led by well-orga- nized and financed networks of corporate and other actors who have actively sought to pre- vent global and/or national action on climate change over the past four decades. In turn, these campaigns often shape public debates, which can affect political support and collec- tive mobilization to mitigate climate change. However, to date, most of the research on cli- mate obstruction has focused on countries in the Global North, especially the United States. Given considerable gaps in research and knowledge, this opinion paper presents a future research agenda needs to shine greater light on if and in what form climate obstruction in the Global South appears. Funding: The authors received no specific funding for this work. Funding: The authors received no specific funding for this work. Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: We note that author Pamela McElwee is a member of the PLOS Climate editorial board. 1 / 5 PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 PLOS CLIMATE 2. Locating climate obstruction in domestic policy and politics Domestic policy options and politics in the Global South are often distinct from those in the Global North. In the Global South, climate delay is a frequently deployed tactic of climate obstruction and is often linked to a justification for expanded economic growth or the need to address energy poverty. Underpinned by claims of improving well-being and economic growth, approaches to domestic energy and environmental policy reinforce trajectories that include and, in some cases expand, the use of fossil fuel-based resources, particularly where alternative energy sources are limited, and cheaper (often subsidized) fossil fuels are promoted [3]. In turn, these resources are often supplied by powerful state and non-state actors that can dictate the terms of debate. As an example, in Vietnam, strong ties between Vinacomin, a state-owned mining company, Electricity of Vietnam, the state-owned power company, and the Ministry of Industry and Transportation have allowed these three entities to slow progress towards renewable energy development [4]. Claims of improving well-being through fossil-fuel-powered economic growth often dis- guise the fact that benefits are often captured exclusively by economic and political elites, both nationally and transnationally [5]; a similar feature to climate obstruction in the Global North. For instance, recent research shows that despite its progress in generating electricity from renewable energy, India has also increased coal use in its overall energy mix to enhance its eco- nomic growth. Meanwhile, India’s delegation at the United Nations Climate Conference in Paris in 2015 argued that poor countries like India should be allowed to continue with carbon emissions to grow their economies, characterizing its position as championing ‘climate justice’ for poor countries [6]. Yet in-depth research on Indian government’s actions suggests a dismal record of prioritizing the needs of economically poor and socially marginalized communities, whose lives, in over a quarter century of steady economic growth, has increased domestic eco- nomic inequalities [7]. Similar impacts can be found in Vietnam, where the state focus on expanding coal-fired power plants with little consultation has resulted in environmental justice harms to local communities from pollution and loss of land [8]. PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 3. Locating climate obstruction in global and transnational networks The international political economy and the nation-state are critical in creating the conditions and the arenas, which have shaped the emergence of climate obstruction in the Global North. Therefore, research is needed on how in-country or domestic actors (including the state, and corporations) respond to the global and transnational networks, are funded by them, and begin implementing activities related to lobbying, disinformation, and dissemination of non- scientific views on climate change. For example, the fossil fuel industry is the leading purveyor of climate obstruction in the US, the agribusiness sector, one of the fastest-growing contribu- tors to emissions (now recognized in the Global North too), is more prominent in climate obstruction activities in Latin America [9]. Identifying and examining actors who have benefited from the current economic status quo and have heavily contributed to the rise in greenhouse gas emissions in the Global South is thus a complex task. Within the complex relations of the international political economy, transnational links may then impact climate obstruction in the Global South. For instance, there is growing evi- dence of the documented and considerable ties between, for example, USA, Japanese, and European financial and private sector institutions, with high-carbon/dirty sectors in the Global South [10]. China and other (re)emerging powers also have a role in framing and practicing development, through South-South cooperation programs, with China rising as the favored source of development financing. While some climate activist NGOs in the Global South have PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 2 / 5 PLOS CLIMATE been able to network and connect to counter these powerful industry and other lobby groups, the constricted space for civil society groups in many global South countries and lack of financing for their activities has prevented them from having stronger impacts [11], and in some cases they have been specifically targeted for repression. For example, in Vietnam, sev- eral prominent anti-coal activists were recently prosecuted on trumped-up tax evasion charges and jailed [12]. 4. Theorizing climate obstruction in the global South The Global South is not a homogenous group, with variations in structural position, geograph- ical size and latitude, political systems, economies, industries, demography, and exposure to climate changes, among others, calling for a look at these similarities and differences among these countries as well. A particular complication arises because considerations of theory to understand climate obstruction are generally situated in knowledge based on US experiences and scholarly articles. As such, conceptual and theoretical understandings of climate obstruc- tion in the Global South are under development. The conceptual and theoretical orientation adopted must also reflect on how this is related to historic emissions. We argue that the idea of common but differentiated responsibility must actively shape our understanding and exami- nation of climate obstruction in Global South nations. Notably, one argument that has been advanced is that many climate-vulnerable countries in the Global South bear little or no responsibility to address carbon emissions because of their historically lower contribution to global warming than advanced industrial countries. Contradictions between climate justice and obstruction must be at the centre of our analyses. The response to the structural condi- tions imposed by the Global North, states and corporations, can often be understood theoreti- cally as ecologically unequal exchange and climate injustice [13]. Hence, in response to the systematic inequalities rooted in and shaped by the damage of colonial and imperialist histo- ries, some countries in the Global South have prioritized highly centralized models of eco- nomic development often leading to the sidelining of climate and other environmental policies. In other instances, domestic political and economic elites have centralized policymak- ing on both environment and development questions, which allows these domestic elites to cynically exploit domestic policymaking processes for maintaining the status quo [14]. More- over, exacerbated by the Global North’s lack of confirmed financial and technological support for mitigation, adaptation and compensation for loss and damages, understanding climate obstruction in Global South nations must be understood with this historic, political, and eco- nomic context in mind while also acknowledging if and how these arguments are used as legit- imations for further higher unequal development. PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 Author Contributions Writing – original draft: Guy Edwards, Paul K. Gellert, Omar Faruque, Kathryn Hochstetler, Pamela D. McElwee, Prakash Kaswhan, Ruth E. McKie, Carlos Milani, Timmons Roberts, Jonathan Walz. Writing – original draft: Guy Edwards, Paul K. Gellert, Omar Faruque, Kathryn Hochstetler, Pamela D. McElwee, Prakash Kaswhan, Ruth E. McKie, Carlos Milani, Timmons Roberts, Jonathan Walz. Writing – review & editing: Guy Edwards, Paul K. Gellert, Omar Faruque, Kathryn Hochste- tler, Pamela D. McElwee, Prakash Kaswhan, Ruth E. McKie, Carlos Milani, Timmons Rob- erts, Jonathan Walz. Writing – review & editing: Guy Edwards, Paul K. Gellert, Omar Faruque, Kathryn Hochste- tler, Pamela D. McElwee, Prakash Kaswhan, Ruth E. McKie, Carlos Milani, Timmons Rob- erts, Jonathan Walz. PLOS CLIMATE This is not an exhaustive list with research gaps that merit investigation. sovereignty claims, climate nationalism, and domestic energy industries and their influence as part of obstruction efforts that appear to emerge in the Global South. This is not an exhaustive list with research gaps that merit investigation. PLOS CLIMATE PLOS CLIMATE Table 1. Research topics and questions. Topic Potential Question(s) Developmentalism discourses What role does developmental discourse play in justifying the expansion of unsustainable developments? What role does developmental discourse play in justifying the delay of low carbon development path? Negotiating blocs What role do negotiating blocs play in stalling in-country efforts to reduce GHG emissions in countries in the Global South? Political leaders and their coalitions What political regimes and leadership types lead to different climate obstruction modalities and trajectories? Transnational links What role do different types of economies such as extractive, industrial and agro-industrial, service, tourism-dependent etcetera, play in different types of climate obstruction strategies, discourses, influence over governments and legislative bodies, and organizational structures? Sovereignty claims Does colonial history, land/resource ownership, and sovereignty help us understand energy transitions in the Global South? How do discourses of sovereignty and energy independence impact the construction of climate related legislation and policy? Domestic energy industry How does obstruction relate to the state’s role (or state-market relations) and other actors like civil society organizations (or state- society relations) in the Global South? How are domestic policies in the Global South influenced by technical assistance projects implemented by their development partners (donors)? How do international partnerships affect climate policies in the Global South? What tensions exist between the requirements of current domestic energy consumers and the impacts of delayed energy transitions and how do these manifest in political discourse? Religion and ethics What role do religions, religious beliefs and religious organizations play in influencing the nature of civil society and state and legal structures related to climate change? Media How are media outlets connected to obstruction actors in Global South nations? If and how do states and private media companies influence broadcast, print, and social media content in-country on the issue of climate change? Impacts of green technology What role does technical assistance and development partnerships play in transitioning from fossil fuels to renewable energy and other mitigation actions in expanding or reducing climate obstruction? Cooptation of scientific expertise To what degree does the cooptation of educational institutions and their associates and expertise enable climate obstruction in the Global South? https://doi.org/10.1371/journal.pclm.0000241.t001 Table 1. Research topics and questions. sovereignty claims, climate nationalism, and domestic energy industries and their influence as part of obstruction efforts that appear to emerge in the Global South. 5. Conclusion COP26 and COP27 revealed an apparent resurgence and series of commitments to tackling the climate emergency, where world leaders verbally agreed to implement strategies to mitigate and adapt to climate changes. However, climate obstruction movements, including institu- tional and non-state actors, corporations and conservative think tanks who consistently delayed commitments in the past, appear to have increased their influence on climate dis- course, especially in the context of the Russian invasion of Ukraine, and pro-growth post-pan- demic packages wherein gas and coal have reemerged as important energy sources. In the Global South, climate obstruction is especially fraught due to historical and contemporary structures of inequality bolstering justifications for various forms of delay. As such, Table 1 incorporates urgent future research directions include the role of developmentalism discourses, negotiating blocs at the negotiations, political leaders and their coalitions, transnational links, PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 3 / 5 PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 References 1. Brulle RJ., Dunlap RE. (2021) A Sociological View of the Effort to Obstruct Action on Climate Change. Footnotes, American Sociological Association Magazine. Available from: https://www.asanet.org/ sociological-view-effort-obstruct-action-climate-change. 2. Lamb WF, Mattioli G, Levi S, Roberts JT, Capstick S, Creutzig F, et al. Discourses of climate delay. Global Sustainability. Cambridge University Press; 2022; 3;e17. 3. Gonza´lez-Eguino M. (2015). Energy poverty: An overview. Renewable and sustainable energy reviews, 47, 377–385. PLOS Climate \| https://doi.org/10.1371/journal.pclm.0000241 July 12, 2023 4 / 5 PLOS CLIMATE 4. Brown D. 2020. Analysis: Vietnam’s leadership flex shows how to drive electricity reform. Mongabay, Aug 6 https://news.mongabay.com/2020/08/analysis-vietnams-leadership-flex-shows-how-to-drive- electricity-reform/ 5. Chakrabarty D. The climate of history: Four theses. Critical inquiry. 2009; 35(2):197–222. 6. Roy B, Schaffartzik A. Talk renewables, walk coal: The paradox of India’s energy transition. Ecological Economics. 2021 Feb 1; 180:106871. https://doi.org/10.1016/j.ecolecon.2020.106871 PMID: 33071457 7. Kashwan P, editor. Climate Justice in India: Volume 1. Cambridge University Press; 2022 Dec 31. 8. Dao Nga. 2022. Environmental justice and the politics of coal-fired thermal power in Vietnam’s Mekong Delta. Asia Pacific Viewpoint 63(2): 262–277. 9. Milani CR, Chaves LN. How and why European and Chinese pro-climate leadership may be challenged by their strategic economic interests in Brazil. Asia Europe Journal. 2022 Dec 1:1–20. 10. Urban MA, Wo´jcik D. Dirty banking: Probing the gap in sustainable finance. Sustainability. 2019 Mar 22; 11(6):1745 11. Gereke Marika & Bru¨hl Tanja (2019) Unpacking the unequal representation of Northern and Southern NGOs in international climate change politics, Third World Quarterly, 40:5, 870–889, https://doi.org/10. 1080/01436597.2019.1596023 12. Swanton B. 2023. Weaponizing the law to prosecute the Vietnam Four. Project 88. 13. Frey RS, Gellert PK, Dahms HF. Introduction: Ecologically unequal exchange in comparative and his- torical perspective. Ecologically unequal exchange: Environmental injustice in comparative and histori- cal perspective. 2019:1–0. 14. Kashwan P. Inequality, democracy, and the environment: A cross-national analysis. Ecological Eco- nomics. 2017 Jan 1; 131:139–51. 5 / 5
https://openalex.org/W2151582013	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0104037&type=printable	English	null	Field Metabolic Rate and PCB Adipose Tissue Deposition Efficiency in East Greenland Polar Bears Derived from Contaminant Monitoring Data	PloS one	2,014	cc-by	10,563	Abstract Climate change will increasingly affect the natural habitat and diet of polar bears (Ursus maritimus). Understanding the energetic needs of polar bears is therefore important. We developed a theoretical method for estimating polar bear food consumption based on using the highly recalcitrant polychlorinated biphenyl (PCB) congener, 2,29,4,49,55-hexaCB (CB153) in bear adipose tissue as an indicator of food intake. By comparing the CB153 tissue concentrations in wild polar bears with estimates from a purposely designed individual-based model, we identified the possible combinations of field metabolic rates (FMR) and CB153 deposition efficiencies in East Greenland polar bears. Our simulations indicate that if 30% of the CB153 consumed by polar bear individuals were deposited into their adipose tissue, the corresponding FMR would be only two times the basal metabolic rate. In contrast, if the modelled CB153 deposition efficiency were 10%, adult polar bears would require six times more energy than that needed to cover basal metabolism. This is considerably higher than what has been assumed for polar bears in previous studies though it is similar to FMRs found in other marine mammals. An implication of this result is that even relatively small reductions in future feeding opportunities could impact the survival of East Greenland polar bears. Citation: Pavlova V, Nabe-Nielsen J, Dietz R, Svenning J-C, Vorkamp K, et al. (2014) Field Metabolic Rate and PCB Adipose Tissue Deposition Efficiency in East Greenland Polar Bears Derived from Contaminant Monitoring Data. PLoS ONE 9(8): e104037. doi:10.1371/journal.pone.0104037 Editor: James P. Meador, Northwest Fisheries Science Center, NOAA Fisheries, United States of America Received March 26, 2014; Accepted July 9, 2014; Published August 7, 2014 Received March 26, 2014; Accepted July 9, 2014; Published August 7, 2014 Received March 26, 2014; Accepted July 9, 2014; Published August 7, 2014 Copyright: 2014 Pavlova 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: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and its Supporting Information files. Funding: This study was financially supported by the European Union under the 7th Framework Programme (project acronym CREAM, contract number PITN- GA-2009-238148) http://ec.europa.eu/research/fp7/index_en.cfm, VP JNN VG. Abstract The research on contamination of ringed seals was funded by the Danish Environmental Protection Agency within the Danish Cooperation for Environment in the Arctic (DANCEA). The early sampling of polar bears in Greenland was funded by the Greenland Institute of Natural Resources (Nuuk), Aage V. Jensens Foundation. Samplings since 1999 and all chemical analyses were funded by a number of projects under the DANCEA (Danish Cooperation for Environment in the Arctic) programme including the CORE programme and the large scale IPY programme ‘‘BearHealth’’, which was also supported by KVUG (The Commission for Scientific Research in Greenland) and The Prince Albert II Foundation for the POP time trend analysis; KV CS RD RJL FF. The Center for Informatics Research on Complexity in Ecology (CIRCE, Aarhus University and Aarhus University Research Foundation for funding CIRCE under the AU IDEAS program, JCS. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. Competing Interests: The authors have declared that no competing interests exist. * Email: vpa@dmu.dk * Email: vpa@dmu.dk understand polar bear feeding habits and energetic needs. The diet composition of polar bears has already been studied extensively. Polar bears predominately eat the blubber and meat of ringed seals (Phoca hispida) or other seal species like the bearded (Erighnatus barbatus), harp (Pagophilus groenlandica) and hooded seal (Cystophora cristata) as well as other marine mammals [2,10,11,12,13,14,15]. Viola Pavlova1, Jacob Nabe-Nielsen1,2, Rune Dietz1,2, Jens-Christian Svenning3, Katrin Vorkamp2,4, Frank Farsø Rige´t1, Christian Sonne1,2, Robert J. Letcher7, Volker Grimm5,6 1 Aarhus University, Department of Bioscience, Section for Marine Mammal Research, Roskilde, Denmark, 2 Aarhus University, Arctic Research Centre, Aarhus, Denmark, 3 Aarhus University, Department of Bioscience, Section for Ecoinformatics and Biodiversity, Aarhus, Denmark, 4 Aarhus University, Department of Environmental Science, Section for Environmental Chemistry and Toxicology, Roskilde, Denmark, 5 Helmholtz Centre for Environmental Research - UFZ, Department of Ecological Modelling, Leipzig, Germany, 6 University of Potsdam, Institute for Biochemistry and Biology, Potsdam, Germany, 7 Ecotoxicology and Wildlife Division, Science and Technology Branch, Environment Canada, National Wildlife Research Centre, Carleton University, Ottawa, Ontario, Canada Field Metabolic Rate and PCB Adipose Tissue Deposition Efficiency in East Greenland Polar Bears Derived from Contaminant Monitoring Data Viola Pavlova1, Jacob Nabe-Nielsen1,2, Rune Dietz1,2, Jens-Christian Svenning3, Katrin Vorkamp2,4, Frank Farsø Rige´t1, Christian Sonne1,2, Robert J. Letcher7, Volker Grimm5,6 August 2014 \| Volume 9 \| Issue 8 \| e104037 Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears pharmacokinetic model, but accounts for direct transfer from prey to bears, and to offspring through lactation. The model predicted the CB153 adipose tissue concentrations between 1986 and 2009 (section 2.2 Model description). With this model we ran numerous simulations with varying combinations of food intake level and CB153 deposition efficiency. We then compared the model predictions of CB153 concentration in the population over multiple years to actual field data on CB153 concentrations in East Greenland polar bears to determine which of the parameter combinations produced CB153 loads similar to those observed over time in East Greenland polar bears (section 2.3 Analysis of model predictions). A test of model sensitivity was also performed (section 2.4 Local sensitivity analysis). For all statistical analyses in this study we used the R language, version 3.0.0 (R Core Team 2012). The values and references to all model parameters are presented in Table S1. Parameter values and references in Supporting Information (SI). doubly tritium labelled water method or heart rate measurements [16,17], however, to our knowledge, such measurements have not been done for wild polar bears. Studies undertaken during the summer in the high Arctic reported the field metabolic rate of polar bears to range between 2.0–2.6 times the basal metabolic rate (BMR) [18,19], or between ‘‘12 to 16,000 kcal/day’’ using captive bear feeding trials. This represents 50,400 to 67,200 kJ/ day. Further estimates of biomass consumption have been inferred by extrapolating from ringed seal population data and from polar bear behaviour studies [10,20,21]. Metabolic costs related to locomotion have also been investigated in captive polar bears and it has been shown that movement increases the metabolic rates considerably [22,23]. It remains unclear how these estimates reflect the annual cycle of wild polar bears. To our knowledge, no study on polar bear energetics has been undertaken in East Greenland and it is thus unknown how closely the estimates of energetic demands from other areas or captive bears would match the situation there. Our study was designed to provide independent insight into polar bear energetics using a new approach and with a focus on the subpopulation in East Greenland. 2.1 Ringed seal contamination & growth g g The field data on CB153 concentrations in blubber of ringed seals in East Greenland were collected in the years 1986, 1994 and every year in the period 1999–2004 and in 2006, 2008 and 2010 (collection methods are described in [31,32,37]. We used this data set to predict the age- and time-dependent concentrations of CB153 for every year and each age class of seals during the period between 1986 and 2009. The data consisted of information on sex, age, weight, length, blubber depth, and concentration of CB153 of each sampled seal. We divided the seals into 12 age classes: 0, 1…10 years old, and 11+ for seals 11 years old and older, because there were only few individuals of older age. The log-transformed CB153 concentrations in seal blubber and seal age classes were then fitted with a linear model: log10 [CB153] , year * age class (F12,232 = 7.518, p,0.0001. R2 = 0.28). The concentra- tions predicted by this model decreased in all age classes throughout the study period (details available in Supporting Information (SI): Figure S1 and SI Text S1. Ringed seal contamination & growth). The same dataset was also used to find the age – weight growth parameters of seals in East Greenland assuming a von Bertalanffy growth curve (Equation [S1], SI Text S1, values of parameters shown in Table S1), which has previously been used for ringed seals in Svalbard [38,39]. We also estimated the amount of blubber present on each seal of certain age class (Text S1). The lists of the model predicted CB153 concentrations for seal blubber are available within the model code (available as a text file (File S1) or as a NetLogo file (File S2)). 2.2 Model description A detailed model description, following the ODD (Overview, Design concepts, Details) protocol for describing individual-based and agent-based models [40,41] is provided in SI Text S2. Overview, Design & Details (ODD) protocol. The main purpose of the model was to produce predictions of CB153 concentrations in adipose tissue of a whole subpopulation of polar bears in East Greenland between 1986 and 2009. The model did not account for any competition for food or any behavioural or other interactions among polar bears (such as territoriality, mating etc.), except for the transfer of contaminants between female bears and their offspring. For implementation we used the Net Logo 4.1 platform [42]; the code is available in the Supporting Information (File S1, File S2). The objective of the study was therefore to estimate the possible FMR in relationship to the deposition efficiency of CB153 in East Greenland polar bears. The basic idea of our approach was to infer possible combinations of field metabolic rates and deposition efficiency based on data on CB153 in ringed seal blubber taking into account lactation contaminant transfer. Estimating the Field Metabolic Rate of East Greenland Polar Bears We developed a theoretical method that is based upon using the concentration of an accumulated and recalcitrant contaminant (polychlorinated biphenyl (PCB) congener, 2,29,4,49,5,59-hexaCB (CB153)) in polar bear adipose tissue as a chemical marker of polar bear food intake rate. Many studies have documented high levels of PCBs in polar bear tissues in the East Greenland subpopulation [24,25,26,27,28]. The PCBs are pre- dominantly acquired through the diet [29,30]. Over the past decades the PCB concentrations in the East Greenland ringed seals, the dominant food source of polar bears [15], have been decreasing [31,32]. The concentrations in adipose tissues of the polar bears in East Greenland follow the trend accordingly [27]. The diet composition may influence the contaminant burden of the bears, but most types of food are likely to contain some level of PCBs [28,33]. The final adipose tissue residues of the highly recalcitrant PCB congeners depend on the amount of food consumed and its contamination level, as well as on the efficiency of deposition of the contaminant into the adipose tissues (hereafter referred to as deposition efficiency), which is an outcome of internal physiological pharmacokinetic processes [34]. In females the total body burdens are further affected by the lactation transfer of contaminant to the offspring [35]. The PCB deposition efficiency has not been measured in polar bears, but the closely related grizzly bears (Ursus arctos horribilis) deposit only about 10% of the ingested PCBs [36] while the rest is bio-transformed or excreted. Given the relatedness of grizzly and polar bears the PCB deposition efficiency could be similar in the two species, but no conclusions could be made based on available literature. Introduction Polar bears (Ursus maritimus) are currently listed as vulnerable on the International Union for Conservation of Nature and Natural Resources (IUCN) red list, with the main threats identified as global climate change, pollutants, oil exploration, and hunting [1]. As an apex predator that predominantly feeds on ice dwelling seals, the polar bear is highly dependent on the presence of sea ice, which forms their platform for hunting [2]. In the near future, the potentially longer sea-ice free periods in the Arctic present a risk of food shortage that could directly influence the viability of bear subpopulations [3,4]. The health of bears, litter mass, litter size, survival, and overall subpopulation size could be affected [5,6,7,8,9]. For effective conservation it is therefore essential to While it is important to know the composition of the polar bear diet, it is also crucial to know their overall energy needs. The energetic requirement is dependent on the field metabolic rate (FMR) that polar bears exhibit. The FMR accounts for basal metabolism and any additional energy requirements in connection to movement and other activities in their natural habitat. Measuring FMR in free-ranging animals is possible using the PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e104037 1 Estimating the Field Metabolic Rate of East Greenland Polar Bears As a next step, each bear satisfied its annual energy requirements by feeding exclusively on ringed seal blubber, which was assumed to consist only of lipids and to have been totally digestible. The blubber was acquired from individual seals caught in an iterative process until the bears satisfied their energy needs. The caught seals were of random sex, but their age was determined according to the documented hunting preferences of polar bears for seal pups, one to two year old seals and three to nineteen year old seals [20] (Table S1). With each seal consumed, the bears correspondingly increased their CB153 burden by the amount that was present in the seal blubber multiplied by the deposition efficiency (AA for all bears two years old and older, hereafter referred to as ‘‘older bears’’). To explore the whole range of possible values of deposition efficiency we tested ten values [0.1, 0.2–1] representing the storage of [10, 20, -100]% of ingested contaminant into adipose tissue. We assumed that all bears satisfied their energy needs and that there was no competition for food (Submodel Feed, SI Text S2). Below we describe processes simulated in the model as well as some algorithms and equations essential for the understanding of the main drivers determining CB153 contamination in the model bears. All further equations and their rationale are included in the ODD protocol in the Supporting Information (Text S2). For parameter values and references, see Table S1. During each time step, the following processes were simulated in the given order: Firstly, offspring survival and pregnancy of mature females during the current year was determined according to survival and breeding probabilities originating from [43] (Table S1). (Submodel Offspring survival and pregnancy, SI Text S2). The CB153 body burdens of lactating females were reduced by the amount of contaminant transferred to the milk (Text S2, eq. S12, S14). This amount was partially retained by the offspring, depending on the deposition efficiency AC for cubs and AY for yearlings. Because the Ac value was unknown and could not be determined via calibration, we used estimations published by [46] on the amount of sum PCBs ingested by cubs and the resulting residues in their tissues, and estimated the value to be 0.23 (Table S1). Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears coexisted in the model, but male agents engaged only in selected processes: they grew and fed according to their weight, accumu- lated the contaminant and died. Because females produced offspring their energy needs and contamination burdens were influenced by fluctuation in weight related to reproduction, by milk production and resulting lactation transfer. Females therefore had additional variables representing pregnancy status (i.e. pregnant or not), reproductive status (single, with cubs, with yearlings) and the number of offspring in current litter. The offspring were either cubs or yearlings dependent on the reproductive status of their mother. Information on the CB153 body burdens of the offspring, energy requirements per yearling per annum, and on the number of days of survival for each offspring was also stored as a variable of each female. Cubs and yearlings were not represented as individual agents because their contamination levels were dependent on the contamination levels of their mothers. The model did not incorporate any spatial movement of individuals or any variation in food or bear properties that would result from their spatial distribution. The time step was one year. We simulated the period between 1986 and 2009. The only environmental variable was the content of CB153 in ringed seals. contained in mother’s milk produced during one year (Text S2, p. 6–7), while for yearlings the above presented formula was applied, but using a separate field metabolic factor fY. In this case, after running initial trials we decided to test a slightly wider span of values [1,2–13]. The energy requirements of yearlings were partly covered by nursing and partly by eating the seal blubber. Further on, the yearly energy demands of females with cubs or with yearlings were equal to EA plus the total amount of energy contained in the milk produced by the female during the year, which depended on the number and age of their offspring (Text S2, eq. S5, S6). The above described calculation was solely for the purpose of determining the amount of food and hence the amount of contaminant ingested annually by the bears and was not related to survival or breeding probability. (Submodel Set energy requirements, Text S2). Estimating the Field Metabolic Rate of East Greenland Polar Bears In yearlings we tested values: [0.1, 0.2–1], It should be noted that because no relevant data was available we neglected any trans-placental transfer of the contaminant in pregnant females. (Submodel Lactation transfer, SI Text S2). Secondly, weight and blubber content of each individual was updated according to sex, age and reproductive status. The bear weight was calculated according to the von Bertalanffy growth function with parameters for polar bears following [44] (Table S1). (Submodels Update weight & Update blubber, SI Text S2). Third, the energy requirements of each bear were calculated according to its reproductive status, weight and, in female bears, number of days of lactation. The annual energy requirement of older bears (older than a yearling), EA [kJ], was assumed to equal to 365 times the field metabolic rate [kJ/day]. The latter was calculated using Kleiber’s rule [45] for basal metabolic rate [kJ/ day] multiplied by a factor fA that accounted for excess energy needed for movement and activity: Finally, population dynamics were addressed as follows: all individuals .30 years died, in addition to a random selection of the remaining bears (according to survival probabilities from [43], Table S1). New individuals were created for weaned yearlings that were approaching two years of age. The reproductive status of females was updated (Submodel Population dynamics, SI Text S2). FMR~k fA BMR~k fA 70 W 0:75 ð1Þ August 2014 \| Volume 9 \| Issue 8 \| e104037 Methods The present study had the following overall methodology. First, in order to obtain the input for our model, we analysed time series data on CB153 concentration in blubber of East Greenland seals and calculated their growth parameters (section 2.1 Ringed seal contamination & growth). Second, we implemented a simple individual-based model representing bioaccumulation of polar bears preying on seals. The model is not a physiologically based The model simulated polar bear individuals from 2 to 30 years of age. Bears were characterized by their age, sex, yearly energy requirement, weight, the proportion of storage blubber and the total body burden of CB153. Polar bear females and males PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e104037 2 Results and Discussion Only 24 of the 13,000 tested combinations of field metabolic factor fA and fY and deposition efficiencies AA and AY produced CB153 concentrations in the bear population that were not significantly different from the East Greenland observations (Table 1). The CB153 concentrations in East Greenland polar bears decreased during the study period and the 24 parameter sets described above reproduced this pattern accordingly (Fig. 1). Among the ANCOVA selected parameter combinations (Ta- ble 1) the deposition efficiencies in older bears and yearlings were inversely proportional to field metabolic factors: AA = vA/fA and AY = vY/fY, where vA and vY are the proportionality constants (equal to the products of the studied parameters for adults: vA = AA6fA and yearlings: vY = AY6fY) with values of 0.6626 and 2.42 respectively (Figure 2). AA = vA/fA and AY = vY/fY, where vA and vY are the proportionality constants (equal to the products of the studied parameters for adults: vA = AA6fA and yearlings: vY = AY6fY) with values of 0.6626 and 2.42 respectively (Figure 2). Visual representation of all values of proportionality constants vA and vY allowed us to gain a better insight into the influence of the combined age group specific parameters on the significance (or non – significance) of the results (Fig. 3a). The contamination levels in yearlings never prevented a realistic contamination time trend to arise for adult bears: with almost any value of vY non- significantly different regression lines in both slope and intercept could be obtained for adults. On the other hand, for yearlings neither vA nor vY could be too high. This means that the contaminant levels in older bears determined the contaminant levels in yearlings to a large degree, but not vice versa. The reason is that the body burdens in yearlings formed only a small part of the total body burden of the older bears. We tested all combinations of fA (values 1, 2 … 10), fY (values 1,2,…13) and AA and AY (both 0.1, 0.2… 1) (total number of simulations: 13,000). Each simulation (s) represented a unique combination of the four parameters. The individual CB153 concentrations were log transformed. 2.4 Local sensitivity analysis We analyzed the influence of changing each parameter in the model by 610% relative to its reference value (fA = 6, fY = 11, AA = 0.1, AY = 0.3 and other parameters as in Table S1) on the mean body burden of yearlings and older bears. Only in case of the survival probability sy a change of 69% had to be used as the original value was too close to 1. The resulting change (in%) was determined for the year 1999. Five simulations were run for each parameter combination and their results averaged. ð1Þ ð1Þ where W is the mass [kg] of the bear and k was set to 4.2 and converts kilocalories to kilojoules. The factor fA is henceforth referred to as the ‘field metabolic factor’. We thus assumed that older bears required the following amount of energy per annum [kJ]: As the last step age was updated as well as was weight and proportion of storage blubber (Submodels Update age, Update weight, Update blubber, SI Text S2, Table S1). To initialize simulations, we started with 2000 individuals with random age between 2 and 20 years and randomly assigned sex and in females the initial pregnancy status (pregnant or not) and the number of offspring. The initial body burdens of all bears were set to zero and therefore (and in order to reach a stable age structure) the simulations were started already in 1950. This allowed contamination loads to build up to the 1986 level. Only the output for the period between 1986 and 2009 was used because no data were available for either seals or bears for the EA~365 k fA 70 W 0:75 ð2Þ ð2Þ where W is the mass [kg] of the bear and fA is the field metabolic factor bear and k was set to 4.2 and converted kilocalories to kilojoules. Ten different values [1,2–10] of the field metabolic factor (chosen with respect to values published for other marine mammals [17,53,54]) were tested. The annual energy requirement of cubs was assumed to be equal to the amount of energy PLOS ONE \| www.plosone.org August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 3 Estimating the Field Metabolic Rate of East Greenland Polar Bears previous period. The contamination load of seals from 1986 was used as an approximation for the period between 1950 and 1986 as input due to lack of monitored data from this period. 2.3 Analysis of the model output y p Time trend analysis. We investigated the influence of various combinations of the field metabolic factor (for both yearlings and older bears) and of the deposition efficiencies of the two groups on the individual adipose tissue concentrations of CB153 in the bear subpopulation. The best parameter combina- tions were sought by comparing the linear regression lines of a random subsample of individual CB153 concentrations for periods between 1986 and 2009 with the real observations from East Greenland polar bears collected during the same time period (with gaps). The methods of data collection and a thorough time trend analysis of the collected data were provided by Dietz et al. [27]. For the purpose of our analysis, the original data was subsampled to 13 yearlings and 295 individuals two years old and older; one outlier with extremely low CB153 concentration was excluded due to suspected measurement errors. Although the maximum age of bears in the model was 30 years, we included only individuals younger than 26 years in the analysis from both the empirical data and model predictions. This was necessary, because the oldest bears formed a very small proportion of the East Greenland sample compared to model output. Cubs also had to be excluded from the analysis, because there were too few in the empirical data. Data for yearlings were not available for the whole period and were thus compared only for the period between 1987 and 2000. Yearlings and older bears formed two separate groups in this analysis. Results and Discussion To find the combinations of parameters producing time series data with regression lines best matching those of the data collected in Greenland for both the yearlings and the older bears, we used ANCOVA and tested the model ln (CB153 concentration) , year * simulation for differences in slope and intercept. The predictor variable (year) was centred in the linear regression so that the intercept could be interpreted as the value of the dependent variable for the centre of the time interval, rather than for year 0 (which would not be useful). We tested the model separately for yearlings and older bears, and only if the results were not significantly different for both categories (using significance levels p.0.1) was the simulation accepted as producing contaminant levels similar to those observed in real polar bears. The parameter combinations that produced these results were further assessed for biological relevance in order to identify the most probable parameter combinations. Most importantly, there existed only narrow ranges of vA and vY values that resulted in regression lines that were not significantly different from the data for both yearlings and older bears: from 0.6 to 0.8 and from 2 to 3.2, respectively. Within these narrow intervals, vA and vY exhibited a negative relationship (Fig 3b), which means that higher levels of contamination in yearlings corresponded to lower levels of contamination in adults. According to the presented results the deposition efficiency in yearlings or their field metabolic factor (or both at the same time) must be considerably higher than that of adults. Thus, even if the yearlings retained 100% of the contaminants they acquired from food and milk (AY = 1) their field metabolic rate would still have to be 2.0 to 3.2 higher than their basal metabolic rate. On the other hand if the older bears exhibited 100% deposition efficiency, their field metabolic rate should only be about 66% of their basal metabolic rate. This indicates that there must be a considerable loss of the contaminant between its ingestion and incorporation into the adipose tissue in older bears. Model validity. In order to verify that the matching time- trend predictions as identified in previous steps were not an artefact of data pooling, we divided the bears into subgroups according to their age category and sex: yearling, subadult females, subadult males, adult females and adult males. August 2014 \| Volume 9 \| Issue 8 \| e104037 Results and Discussion metabolism can be very efficient. Although the bears often hunt by still-hunting [49], which can be potentially energetically quite inexpensive, they are known to migrate large distances [50] and engage in swimming [51]. Swimming in particular may be energy costly as the insulation qualities of the bear fur strongly decrease in water and the body is then primarily insulated by adipose tissue [52]. The maintenance of sufficient insulation is again likely to require a large amount of food. In other marine mammals rather high field metabolic rates have also been found. Common dolphins (Tursiops truncatus) can use up to six times more energy than required for their basal metabolism [53], while FMR values between 5 to 7.4x BMR were estimated for killer whales (Orcinus orca) [54]. In walrus (Odobenus rosmarus) direct measurements have indicated FMR to be up to 5.5 to 6.5x BMR [17]. This finding was discussed by the authors as conflicting with the common practice of using a field metabolic factor of three for pinnipeds, since it may lead to underestimation in calculations of their food intake [55,56]. Although polar bears are not close relatives of any of these species, they live in a similarly extreme environment. Using the field metabolic factor of 6 (FMR thus being equal to six times BMR), a bear weighing 200 kg would spend roughly 93,800 kJ/day. This is close to the 95,827 kJ/day predicted for bears based on an allometric relationship developed y g In theory, these results could support the field metabolic rate of 2–2.6 x BMR as suggested for polar bears previously [18,19]. The older East Greenland polar bears should then have incorporated about 30% of the ingested CB153 into their adipose tissues. However, if the deposition efficiency were as low as in grizzly bears [36], with the bears retaining only about 10% of the contaminant, the field metabolic rate of older bears would reach values of 6 to 7 times the basal metabolic rate. This finding is particularly interesting, because it is considerably higher than what has previously been reported for polar bears [18,19,20]. We have no data to indicate whether polar bears from different geographical areas might differ in their metabolic rates to such an extent. Results and Discussion Then we compared the medians of predicted CB153 concentrations within these subgroups with the East Greenland data. This allowed us to check whether the results produced by the model also corresponded with the real observations with respect to the sex and age of bears. The identification of independent and realistic properties emerging from the model was used to support the validity of principles and assumptions used in our model [47]. While 24 parameter sets were identified as producing realisti- cally decreasing contaminant loads for the period 1986–2009, some of these are not biologically justifiable, as often is the case for calibrated parameter sets [48]. Thus, all sets with fA equal one, where the field metabolic rate would equal to the basal metabolic rate, can be dismissed as unlikely as well as the few combinations where the per unit body weight energy requirement of yearlings August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 4 Estimating the Field Metabolic Rate of East Greenland Polar Bears Figure 1. Time trend in polar bear CB153 concentrations. Red dots show the CB153 concentrations in individual East Greenland bears across the study period. Red curve shows log – linear trend in the data. Grey and black curves show simulated trends in CB153 concentrations produced using parameter combinations that produced predictions that did not differ from the observed trends in East Greenland for yearlings and older bears, respectively. doi:10.1371/journal.pone.0104037.g001 Figure 1. Time trend in polar bear CB153 concentrations. Red dots show the CB153 concentrations in individual East Greenland bears across the study period. Red curve shows log – linear trend in the data. Grey and black curves show simulated trends in CB153 concentrations produced using parameter combinations that produced predictions that did not differ from the observed trends in East Greenland for yearlings and older bears, respectively. doi:10.1371/journal.pone.0104037.g001 would be smaller than that of older bears (Table 1, combinations in italics). In the remaining solutions the parameters of older bears typically occur in two combinations fA = 2 or 3 with AA = 0.3 or fA = 6 or 7 with AA = 0.1, while the field metabolic factor for yearlings, fY, ranges between 9 and 11 in most cases, with a corresponding deposition efficiency AY of 0.3. In two occasions fY is lower, reaching only 5 and 7 and the corresponding deposition efficiency is thus higher: 0.6 or 0.4. Results and Discussion Stirling and Øritsland [20], used a conservative field metabolic factor of 2 in their polar bear study based on the assumption that polar bears are well adapted to the life in the Arctic and their August 2014 \| Volume 9 \| Issue 8 \| e104037 5 PLOS ONE \| www.plosone.org Estimating the Field Metabolic Rate of East Greenland Polar Bears Table 1. ANCOVA selected parameter combinations. Nr. AY AA fY fA 1 0.3 0.7 8 1 2 0.3 0.7 9 1 3 0.3 0.3 9 2 4 0.3 0.3 10 2 5 0.3 0.2 10 3 6 0.3 0.1 10 6 7 0.3 0.1 10 7 8 0.4 0.6 7 1 9 0.4 0.7 6 1 10 0.4 0.3 7 2 11 0.5 0.6 5 1 12 0.5 0.7 5 1 13 0.6 0.6 5 1 14 0.6 0.3 5 2 15 0.6 0.1 4 7 16 0.7 0.6 4 1 17 0.7 0.8 3 1 18 0.8 0.7 3 1 19 1 0.1 2 7 20 0.3 0.6 11 1 21 0.3 0.3 11 2 22 0.3 0.1 11 6 23 0.2 0.7 12 1 24 0.2 0.7 13 1 Parameter combinations that produced CB153 contamination patterns not significantly different from those of East Greenland polar bears (biologically relevant combinations where fA.1 or fY$fA are in bold font, non-relevant combinations in italics). doi:10.1371/journal.pone.0104037.t001 produced CB153 contamination patterns not significantly different from those of East Greenland polar bears (biologically relevant Y$fA are in bold font, non-relevant combinations in italics). Parameter combinations that produced CB153 contamination patterns not significantly different from those of East Greenland polar bears (biologically relevant combinations where fA.1 or fY$fA are in bold font, non-relevant combinations in italics). doi:10.1371/journal.pone.0104037.t001 Parameter combinations that produced CB153 contamination patterns not significantly different from those of East Greenland polar bears (biologically relevant combinations where fA.1 or fY$fA are in bold font, non-relevant combinations in italics). doi:10.1371/journal.pone.0104037.t001 Figure 2. Relationship between deposition efficiency and field metabolic factor. Within the parameter combinations that produced matching predictions the deposition efficiency was inversely propor- tional to the filed metabolic factor in both groups; the yearlings and older bears. Solid curve: AA = 0.6626/fA, dashed curve: AY = 2.42/fY. doi:10.1371/journal.pone.0104037.g002 for non-herbivorous large mammals [57]. With respect to this and the arguments above, a field metabolic factor of 6 seems possible for older polar bears. August 2014 \| Volume 9 \| Issue 8 \| e104037 Results and Discussion Because of lack of data we assumed the daily milk consumption during the whole year to equal to that measured during the summer ice-free period [64]. Furthermore, we have no information on whether the contaminant is absorbed from the milk at a similar rate as from the seal blubber. In the case of cubs totally dependent on milk, we were unable to use our data to calibrate the deposition efficiency because the sample size was too small. Instead we had to use a value calculated from data by [46] (Table S1). The calculated value could have been influenced by the sample size used in [46] or other factors and therefore had associated uncertainty. Nevertheless the calculated value was not widely different from our findings for yearlings and older bears. While these results were expected, the analysis pointed out the importance of the energy content of food. We have set the energy content of seal blubber to 37.8 kJ/g (or 9 kcal/g) based on the assumption that it consists of 100% blubber and is entirely digestible. We did not have sufficient data to include detailed feeding preferences (including various degree of contamination, energy content and digestibility of all types of food that East Greenland bears regularly feed on), but adding such details in future would help to improve the estimations of energy require- ment. Finally, the sensitivity analysis showed that the body burdens of yearlings were also strongly dependent on their weight, because their energy requirement is calculated on the basis of their body mass. The East Greenland bear data set did not contain data on body mass and therefore we based our calculations on a value obtained from the literature. This may have had an impact on the estimation of fY as discussed above. Model validation. The medians of CB153 concentration of each group of polar bears: yearlings, subadult females, subadult males, adult females and adult males matched fairly well with the East Greenland data when using the ANCOVA selected parameter sets (Table 1), but not for some of the other (randomly chosen) combinations (Fig. 4). The existence of parameter sets that produce not only matching time trend of pooled data but also of the sex and age dependent patterns supports the validity of the model structure and parameterization. Results and Discussion The PCB deposition efficiency in yearlings was mostly 30% and the field metabolic factor ranged from 9 to 11 (Table 1). The deposition efficiency is thus the same or somewhat higher than in older bears. This seems plausible, considering that in the PCB deposition efficiency can range from 100% in infants [58] to 10% in adults [59]. We do not know the reasons for this phenomenon in polar bears, but in children immaturity of enzymatic systems or of hepatic and renal functions can cause longer half-lives of xenobiotic [60]. When designing the study, keeping in mind the proposed field metabolic rate of 2xBMR from previous studies, we restricted the range of tested field metabolic factor values to 1 to 13, which we assumed covered what seems biologically possible. If we had tested even higher values, we could have possibly found some parameter combination with deposition efficiency as low as 0.1. Nevertheless, according to the current results, juvenile polar bears seem to have an even higher field metabolic rate than older bears. In dogs (Canis familiaris), growing puppies need about 1.8 times more energy per unit body weight compared to adults until they reach 50% of adult weight, and their energy requirement continues to be higher than that of adults until reaching the full body size or even longer [61]. According to published polar bear nutrition guidelines a captive 3-year old polar bear was fed 2 to 3 Figure 2. Relationship between deposition efficiency and field metabolic factor. Within the parameter combinations that produced matching predictions the deposition efficiency was inversely propor- tional to the filed metabolic factor in both groups; the yearlings and older bears. Solid curve: AA = 0.6626/fA, dashed curve: AY = 2.42/fY. doi:10.1371/journal.pone.0104037.g002 August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 6 Estimating the Field Metabolic Rate of East Greenland Polar Bears B Figure 3. Tested combinations of deposition efficiencies and field metabolic factors. a) Only few combinations of proportionality constants: vA ( = AA6fA) and vY( = AY6fY) produced predictions not significantly different from actual data: in older bears (blue), yearlings (yellow) or both groups (green). All tested combinations are coloured grey. b) Combinations of vA and vY that produced predictions not significantly different from actual data in both studied groups exhibited a negative relationship. doi:10.1371/journal.pone.0104037.g003 B B Figure 3. Tested combinations of deposition efficiencies and field metabolic factors. Results and Discussion a) Only few combinations of proportionality constants: vA ( = AA6fA) and vY( = AY6fY) produced predictions not significantly different from actual data: in older bears (blue), yearlings (yellow) or both groups (green). All tested combinations are coloured grey. b) Combinations of vA and vY that produced predictions not significantly different from actual data in both studied groups exhibited a negative relationship. doi:10.1371/journal.pone.0104037.g003 body burdens of older bears depended mostly on three parame- ters: the energy contained in a gram of seal blubber (Es), the field metabolic factor fA and the deposition efficiency, AA. times more food in relation to its body size than a nine years old bear [62]. Some of the combinations of the parameters (parameter sets 6, 7, 14 and 22 in Table 1) do comply with these ratios. It should be noted though, that over-estimation of the blubber consumption could have occurred due to underestimation of milk consumption rates in cubs and yearlings. It is known that milk production/consumption varies during the year in grizzly bears [63]. Because of lack of data we assumed the daily milk consumption during the whole year to equal to that measured during the summer ice-free period [64]. Furthermore, we have no information on whether the contaminant is absorbed from the milk at a similar rate as from the seal blubber. In the case of cubs totally dependent on milk, we were unable to use our data to calibrate the deposition efficiency because the sample size was too small. Instead we had to use a value calculated from data by [46] (Table S1). The calculated value could have been influenced by the sample size used in [46] or other factors and therefore had associated uncertainty. Nevertheless the calculated value was not widely different from our findings for yearlings and older bears. times more food in relation to its body size than a nine years old bear [62]. Some of the combinations of the parameters (parameter sets 6, 7, 14 and 22 in Table 1) do comply with these ratios. It should be noted though, that over-estimation of the blubber consumption could have occurred due to underestimation of milk consumption rates in cubs and yearlings. It is known that milk production/consumption varies during the year in grizzly bears [63]. August 2014 \| Volume 9 \| Issue 8 \| e104037 Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears Figure 4. Medians of CB153 concentration of bear sub groups. Comparison between median observed CB153 concentrations in East Greenland polar bears (solid red lines) and CB153 concentration simulated using the best-fitting parameter combinations (black) or randomly selected parameter combinations (blue). Quartiles are shown as bars for model predictions and as dashed red lines for East Greenland bears. doi:10.1371/journal.pone.0104037.g004 Figure 4. Medians of CB153 concentration of bear sub groups. Comparison between median observed CB153 concentrations in East Greenland polar bears (solid red lines) and CB153 concentration simulated using the best-fitting parameter combinations (black) or randomly selected parameter combinations (blue). Quartiles are shown as bars for model predictions and as dashed red lines for East Greenland bears. doi:10.1371/journal.pone.0104037.g004 to model the bioaccumulation - a parameter that is measurable in large mammals such as bears as Christensen et al. [36] demonstrated - we avoided the necessity of using a number of physiological and pharmacokinetic parameters derived from laboratory animal studies. Such an approach helped us to avoid the uncertainty associated with cross-species extrapolation. Using a single measurable parameter also increases the likelihood of such measurement being undertaken in relation to polar bears in future. In view of our having quantified the relationship between deposition efficiency and field metabolic rate this would then allow further refinement of our results regarding the field metabolic rate. However a physiologically based pharmacokinetic (PBPK) model has been developed for polar bears in the past [66] and using such a model would be advisable in order to understand the physiological processes that lead to the relatively low deposition of the highly persistent congener CB153 into polar bear adipose tissue. to model the bioaccumulation - a parameter that is measurable in large mammals such as bears as Christensen et al. [36] demonstrated - we avoided the necessity of using a number of physiological and pharmacokinetic parameters derived from laboratory animal studies. Such an approach helped us to avoid the uncertainty associated with cross-species extrapolation. Using a single measurable parameter also increases the likelihood of such measurement being undertaken in relation to polar bears in future. In view of our having quantified the relationship between deposition efficiency and field metabolic rate this would then Using a contaminant as a chemical marker of food intake has proven to be feasible. Results and Discussion For our purpose of predicting the long-term time trend in CB153 adipose tissue concentrations across the population, it was necessary to couple the bioaccumulation model with population dynamics. At this stage we have not related the vital rates to the food supply of the bears and have instead assumed that bears forage according to their requirements. However it has been shown that longer ice free periods may result in decreasing litter sizes with direct effects on the population viability of polar bears [9]. Decreased survival in connection to prolonged fasting has also been demonstrated in male polar bears using a modelling platform Niche Mapper [65]. Our individual based model provides a platform that would allow for researching these issues in the context of a whole population in the future. Sensitivity analysis. The sensitivity analysis revealed that the mean CB153 body burdens were indeed particularly sensitive to variations in the four focal parameters (Figure S2); the body burdens of yearlings were most sensitive to the field metabolic factor of yearlings, fY, the deposition efficiency of yearlings AY and further on by the energy contained in a gram of seal blubber (Es), and the weight of the yearlings (WY). Similarly, the mean CB153 We aimed to construct a model of low complexity and a lowest possible number of parameters. By using only a single parameter August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 7 Estimating the Field Metabolic Rate of East Greenland Polar Bears Acknowledgments We thank Erik Born, Aqqaluq Rosing Asvid, Hanne Tuborg, Birger Sandell, and Sigga Joensen helped to obtain the polar bears and ringed seal together with the local hunters in East Greenland. Lene Bruun, Annegrete Ljungqvist, Birgit Groth, and Sigga Joensen are acknowledged for their work in the AU laboratory together with Eric Pelletier, David Blair and Wouter Gebbink in the OCRL/Letcher Labs at EC-NWRC where the polar bear OHC analysis of the majority of polar bear fat/adipose tissue samples were conducted. Derek Muir and his team at NLET, Environment Canada, performed the PCB analyses in ringed seals from 1999. The aging of individual polar bears was assisted by Erik W. Born, M. Kierkegaard, S. Joensen, and L. Bruun. References 12. Derocher AE, Lunn N, Stirling I (2004) Polar Bears in a Warming Climate. Integrative and Comparative Biology 44: 163–76. 1. Schliebe S, Wiig Ø, Derocher A, Lunn N (IUCN SSC Polar Bear Specialist Group) (2008) Ursus maritimus. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. Available: www.iucnredlist.org. Accessed 27 February 2013. g p gy 13. Thiemann GW, Iverson SJ, Stirling I (2008) Polar bear diets and arctic marine food webs: insights from fatty acids analysis. Ecological Monographs 78(4): 591– 613. 2. Smith TG (1980) Polar bear predation of ringed and bearded seals in the land- fast sea ice habitat. Canadian Journal of Zoology 58: 2201–09. 14. Iversen M, Aars J, Haug T, Alsos IG, Lydersen C, et al. (2013) The diet of polar bears (Ursus maritimus) from Svalbard, Norway, inferred from scat analysis. Polar Biology 36: 561–571 3. Stirling I (1993) Possible impacts of climatic warming on polar bears. Arctic 46(3): 240–245. 4. Derocher AE, Aars J, Amstrup SC, Cutting A, Lunn NJ, et al. (2013) Rapid ecosystem change and polar bear conservation. Conservation Letters 6: 368– 375. 15. McKinney M, Iverson S, Fisk A, Sonne C, Riget F, et al. (2013) Causes and consequences of long-term change in East Greenland polar bears diets: Investigation using quantitative fatty acid estimates and fatty acid carbon isotope patterns. Global Change Biology 19: 2360–2372. doi:10.1111/ gcb.12241 5. Derocher AE, Stirling I (1994) Age-specific reproductive performance of female polar bears (Ursus maritimus). Journal of Zoology 234: 527–536. 6. Derocher AE, Stirling I (1996) Aspects of survival in juvenile polar bears. Canadian Journal of Zoology 74: 1246–1252. 16. Butler PJ, Green JA, Boyd IL, Speakman JR (2004) Measuring metabolic rate in the field: the pros and cons of the doubly labelled water and heart rate methods. Functional Ecology 18: 168–183. 7. Regehr EV, Lunn NJ, Amstrup SC, Stirling I (2007) Effects of earlier sea ice breakup on survival and population size of polar bears in western Hudson Bay. Journal of Wildlife Management 71: 2673–2683. 17. Acquarone M, Born EW, Speakman JR (2006) Field metabolic rates of walrus (Odobenus rosmarus) measured by the doubly labeled water method. Aquatic Mammals 32(3): 363–369. 8. McKinney MA, Atwood T, Dietz R, Sonne C, Iverson SJ, et al. (2014) Validation of adipose lipid content as a body condition index for polar bears. Ecology and Evolution 4(4) 516–527. doi:10.1002/ece3.956 18. Table S1 Parameter values and references. (DOC) Table S1 Parameter values and references. (DOC) In conclusion, the field metabolic rate of older East Greenland polar bears may be as low as 2xBMR under the assumption that polar bears deposit around 30% of the consumed CB153 into their adipose tissue. However, if they deposited only 10% of consumed CB153 into their adipose tissue their indicated field metabolic rate would be three times higher than previously assumed. This finding is of concern especially in view of the on-going climate changes, where polar bears will have less access to feeding opportunities. Polar bear yearlings could be in particular vulnerable to starvation and reduced food intake since their per unit body weight energy needs appear to be even higher than the energy needs of older bears, possibly reaching 11x BMR. While our model is too simplistic to provide precise estimates we would like to point out the necessity of further research into this area. It is essential to know how much food polar bears consume in order to assess the impact of potential loss of foraging opportunities. A direct measurement of metabolic rates using the doubly tritium labelled water method at different seasons and for different age groups and sexes would be extremely valuable. Likewise, researching activity budgets together with data from GPS equipped polar bears on movements (which were disregarded in our model) coupled with spatial modelling or using the method by [65] for all seasons of the year could provide further insight into the amount of energy polar bears spend when roaming their natural habitat. File S1 Model code – text file. (TXT) File S1 Model code – text file. (TXT) File S2 Model code – NetLogo file. (NLOGO) File S2 Model code – NetLogo file. (NLOGO) Text S1 Ringed seal contamination & growth. (DOC) Text S1 Ringed seal contamination & growth. (DOC) Text S1 Ringed seal contamination & growth. (DOC) Text S2 Overview, Design & Details (ODD) protocol. (DOC) Text S2 Overview, Design & Details (ODD) protocol. (DOC) Author Contributions Figure S1 CB153 concentration in blubber of East Greenland seals. Black dots: field measurements, lines: linear model predictions for various age classes of seals. (EPS) Figure S1 CB153 concentration in blubber of East Greenland seals. Black dots: field measurements, lines: linear model predictions for various age classes of seals. (EPS) Conceived and designed the experiments: VP. Performed the experiments: VP. Analyzed the data: VP. Contributed to the writing of the manuscript: VP JNN RD JCS KV FFR CS RJL VG. Provided field data and/or permissions to use them (on seal and bear contamination): RD KV FFR CS RJL. Contributed to statistical analysis: JNN. Contributed to model development: VG. Conceived and designed the experiments: VP. Performed the experiments: VP. Analyzed the data: VP. Contributed to the writing of the manuscript: VP JNN RD JCS KV FFR CS RJL VG. Provided field data and/or permissions to use them (on seal and bear contamination): RD KV FFR CS RJL. Contributed to statistical analysis: JNN. Contributed to model development: VG. Figure S2 Sensitivity analysis. Change (%) in mean CB 153 body burdens in yearlings and older bears following the change of Figure S2 Sensitivity analysis. Change (%) in mean CB 153 body burdens in yearlings and older bears following the change of Estimating the Field Metabolic Rate of East Greenland Polar Bears While we used CB153 in polar bears, we believe that our method, if adapted, could be used in other species August 2014 \| Volume 9 \| Issue 8 \| e104037 August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 8 Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears parameters by +10% (red plus symbol) and - 10% (black minus symbol) of the reference value. Dashed red lines indicate standard deviation (in%) in the body burdens in the population when using original parameter values. (EPS) using similar or other suitable contaminants. Although our model is based on many assumptions which carry varying degrees of certainty, we have pointed out an alternative, non-invasive way of estimating the energy requirements using contaminant monitoring data in relation to a species that is otherwise very difficult to study in the field. Table S1 Parameter values and references. (DOC) Estimating the Field Metabolic Rate of East Greenland Polar Bears Estimating the Field Metabolic Rate of East Greenland Polar Bears 22. Hurst RJ, Oritsland NA, Watts PD (1982) Body mass, temperature and cost of walking in polar bears. Acta Physiologica Scandinavica 115(4): 391–5. 45. Kleiber M (1975) The fire of life: An introduction to animal energetics. 2nd edition. Robert E. Kreiger Publishing, Huntington, NY. g g g uk SC (1999) Organochlorine dynamics in free-ranging polar bears an 23. Watts PD, Ferguson KL, Draper BA (1991) Energetic output of subadult polar bears (Ursus maritimus): resting, disturbance and locomotion. Comparative Biochemistry and Physiology A 98(2): 191–193. 46. Polischuk SC (1999) Organochlorine dynamics in free-ranging polar bears and their cubs. PhD Thesis, University of Saskatchewan, Saskatoon, Canada, 188 p. 47. Wiegand T, Jeltsch F, Hanski I, Grimm V (2003) Using pattern-oriented modeling for revealing hidden information: a key for reconciling ecological theory and application. Oikos, 100(2), 209–222. doi:10.1034/j.1600- 0706.2003.12027.x y y gy ( ) 24. Henriksen EO, Wiig Ø, Skaare JU, Gabrielsen GW, Derocher AE (2001) Monitoring PCBs in polar bears: lessons learned from Svalbard. Journal of Environmental Monitoring 3: 493–498. g 25. Bentzen TW, Muir DCG, Amstrup SC, O9Hara TM (2008) Organohalogen concentrations in blood and adipose tissue of Southern Beaufort Sea polar bears. Science of the Total Environment 406: 352–367. 48. Jakoby O, Grimm V, Frank K (2014) Pattern-oriented parameterization of general models for ecological application: towards realistic evaluations of management approaches. Ecological Modelling 275: 78–88. 26. Dietz R, Rige´t FF, Sonne C, Letcher RJ, Born EW, et al. (2004) Seasonal and temporal trends in polychlorinated biphenyls and organochlorine pesticides in East Greenland polar bears (Ursus maritimus), 1990–2001. Science of the Total Environment 331(1–3): 107–124. 49. Stirling I (1974) Midsummer observations on the behavior of wild polar bears (Ursus maritimus). Canadian Journal of Zoology 52 (9): 1191–1198. doi:10.1139/z74-157 50. Mauritzen M, Derocher AE, Pavlova O, Wiig O (2003) Female polar bears (Ursus maritimus) on the Barents Sea drift ice: walking the trade mill. Animal Behaviour 66: 107–113. 27. Dietz R, Rige´t FF, Sonne C, Born EW, Bechshøft T, et al. (2013) Part 1: Three decades (1984–2010) of legacy contaminant trends in East Greenland polar bears (Ursus maritimus). Environment International 59: 485–493. Available: http://dx.doi.org/10.1016/j.envint.2012.09.004. 51. Durner GM, Whiteman JP, Harlow HJ, Amstrup SC, Regehr EV, et al. (2011) Consequences of long-distance swimming and travel over deep-water pack ice for a female polar bear during a year of extreme sea ice retreat. Estimating the Field Metabolic Rate of East Greenland Polar Bears Polar Biology 34: 975–984. p g j 28. McKinney MA, Letcher RJ, Born EW, Branigan M, Dietz R, et al. (2011) Flame retardants and legacy contaminants in polar bears from Alaska, Canada, East Greenland and Svalbard, 2005–2008 I. Environment International 37: 365–374. 52. Schmidt-Nielsen K (1997) Animal Physiology: Adaptation and environment, Cambridge University Press. 607 p. 29. Letcher RJ, Gebbink WA, Sonne C, Born EW, McKinney MA, et al. (2009) Bioaccumulation and biotransformation of brominated and chlorinated contaminants and their metabolites in ringed seals (Pusa hispida) and polar bears (Ursus maritimus) from Eastern Greenland. Environment International 35: 1180–1124. 53. Costa DP, Williams TM (2000) Marine mammal energetics. In: Reynolds J, and Twiss J, editors. The Biology of Marine Mammals. Washington, DC: Smithsonian Institution Press, pp. 176–217. 54. Noren DP (2011) Estimated field metabolic rates and prey requirements of resident killer whales. Marine Mammal Science 27(1): 60–77. 30. Fisk AT, Hobson KA, Norstrom RJ (2001) Influence of chemical and biological factors on trophic transfer of persistent organic pollutants in the Northwater Polynya marine food web. Environmental Science and Technology 35(4): 732– 738. 55. Trites AW, Chrisensen V, Pauly D (1997) Competition between fisheries and marine mammals for prey and primary production in the Pacific Ocean. Journal of North Atlantic Fisheries Science 22: 173–187. 31. Vorkamp K, Bossi R, Rige´t FF, Dietz R (2011) Time trends of hexabromo- cyclododecane, polybrominated diphenyl ethers and polychlorinated biphenyls in ringed seals from East Greenland. Environmental Science and Technology 45: 1243–1249. 56. Winship AJ, Trites AW, Rosen DAS (2002) A bioenergetic model for estimating the food requirements of Steller sea lions (Eumetopias jubatus) in Alaska, USA. Marine Ecology – Progress Series 229: 291–312 gy g 57. Nagy KA (1987) Field metabolic rate and food requirement scaling in mammals and birds. Ecological Monographs 57: 111–128. gy g 57. Nagy KA (1987) Field metabolic rate and food require and birds. Ecological Monographs 57: 111–128. 32. Rige´t F, Vorkamp K, Dietz R, Rastogi SC (2006) Temporal trend studies on polybrominated diphenyl ethers (PBDEs) and polychlorinated biphenyls (PCBs) in ringed seals from East Greenland. Journal of Environmental Monitoring 8: 1000–1005. gy g 57. Nagy KA (1987) Field metabolic rate and food requirement s and birds. Ecological Monographs 57: 111–128. 58. McLachlan MS (1993) Digestive Tract Absorption of Polychlorinated Dibenzo- p-dioxins, Dibenzofurans, and Biphenyls in a Nursing Infant. Toxicology and Applied Pharmacology 123 (1): 68–72. 33. Estimating the Field Metabolic Rate of East Greenland Polar Bears McKinney MA, Peacock E, Letcher RJ (2009) Sea ice-associated diet change increases the levels of chlorinated and brominated contaminants in polar bears. Environmental Science and Technology 43: 4334–4339. 59. Schlummer M, Moser GA, McLachlan MS (1998) Digestive tract absorption of PCDD/Fs, PCBs, and HCB in humans: mass balances and mechanistic considerations. Toxicology and Applied Pharmacology 152: 128–137. 34. Hickie BE, Mackay D, De Koning J (1999) Lifetime pharmacokinetic model for hydrophobic contaminants in marine mammals. Environmental Toxicology and Chemistry 18: 2622–2633 60. Ginsberg G, Hattis D, Sonawane B, Russ A, Banati P, et al. (2002) Evaluation of Child/Adult Pharmacokinetic Differences from a Database Derived from the Therapeutic Drug Literature Toxicological Science 66 (2): 185–200. doi:10.1093/toxsci/66.2.185 35. Knott KK, Boyd D, Ylitalo GM, Todd M, O9Hara TM (2012) Lactational transfer of mercury and polychlorinated biphenyls in polar bears. Chemosphere 88(4): 395–402. 61. FEDIAF (2011) Nutritional guidelines for complete and complementary pet food for cats and dogs. FEDIAF - European pet food industry federation, Av. Louise 89/B-1050 Bruxelles. Available: http://www.nutricao.vet.br/pdfs/FEDIAF_ Nutritional_Guidelines_-_final_version_6-09-11.pdf. Accessed 27 September 2013. 36. Christensen JR, Letcher RJ, Ross PS (2009) Persistent or not persistent? Polychlorinated biphenyls are readily depurated by grizzly bears (Ursus arctos horribilis) Environmental Toxicology and Chemistry 28(9): 2206–2215. 37. Rige´t F, Vorkamp K, Dietz R, Muir DCG (2008) Levels and temporal trends of HCH isomers in ringed seals from West and East Greenland. Journal of Environmental Monitoring 10: 935–940. 62. Lintzenich BA, Ward AM, Edwards MS, Griffin ME, Robbins CT (2006) Polar bear nutrition guidelines, AZA Bear Taxon Advisory Group (TAG) & AZA Polar Bear Species Survival Plan (SSP) & Polar Bears International, 65 p. Available: http://www.polarbearsinternational.org/sites/default/files/ pbnutritionguidelines.pdf. Accessed 15 January 2014. g 38. McLaren LA (1993) Growth in pinnipeds. Biological Reviews 68: 1–79. 39. Krafft BA, Kovacs KM, Frie AK, Haug T, Lydersen C (2006) Growth and population parameters of ringed seals (Pusa hispida) from Svalbard, Norway, 2002-04. ICES Journal of Marine Science 63: 1136–1144. 63. Farley SD, Robbins CT (1995) Lactation, hibernation and mass dynamics of American black bears and grizzly bears. Canadian Journal of Zoology 73: 2216– 22. 40. Grimm V, Berger U, Bastiansen F, Eliassen S, Ginot V, et al. (2006) A standard protocol for describing individual-based and agent-based models. Ecological Modelling 198: 115–126. 64. Arnould JPY, Ramsay MA (1994) Milk production and milk consumption in polar bears during the ice-free period in western Hudson Bay. Canadian Journal of Zoology 72: 1365–1370. References Best RC (1976) Ecological energetics of the polar bear. M.S. thesis, Univ. of Guelph, Guelph, Ontario, canada, 136 pp. 19. Best RB (1985) Digestibility of ringed seals by the polar bear. Canadian Journal of Zoology 63: 1033–1036. 9. Molna´r PK, Derocher AE, Klanjscek T, Lewis MA (2011) Predicting climate change impacts on polar bear litter size. Nature Communications 2: 186. doi:10.1038/ncomms1183 20. Stirling I, Øritsland NA (1995) Relationships between estimates of ringed seal (Phoca hispida) and polar bear (Ursus maritimus) populations in the Canadian Arctic. Canadian Journal of Fisheries Aquatic Science 52: 2594–2612. 10. Stirling I, McEwan EH (1975) The caloric value of whole ringed seals (Phoca hispida) in relation to polar bear (Ursus maritimus) ecology and hunting behavior. Canadian Journal of Zoology 53(8): 1021–1027. 21. Pilfold NW, Derocher AE, Stirling I, Richardson E, Andriashek D (2012) Age and sex composition of seals killed by polar bears in the Eastern Beaufort Sea. Plos One. doi:10.1371/journal.pone.0041429 J gy ( ) 11. Stirling I, Archibald WR (1977) Aspects of predation of seals by polar bears. Journal of the Fisheries Research Board of Canada 34: 1126–29. Journal of the Fisheries Research Board of Canada 34: 1126–29. PLOS ONE \| www.plosone.org 9 August 2014 \| Volume 9 \| Issue 8 \| e104037 August 2014 \| Volume 9 \| Issue 8 \| e104037 Estimating the Field Metabolic Rate of East Greenland Polar Bears g 41. Grimm V, Berger B, DeAngelis DL, Polhill JG, Giskee J, et al. (2010) The ODD protocol: A review and first update. Ecological Modelling 221: 2760–2768. 42. Wilensky U (1999) NetLogo center for connected learning and computer-based modelling. Northwestern University, Evanston, IL. Available: http:// cclnorthwesternedu/netlogo/. 65. Mathewson PD, Porter WP (2013) Simulating polar bear energetics during a seasonal fast using a mechanistic model. PLOS One 8(9): e72863. doi:10.1371/ journal.pone.0072863 43. Regehr EV, Hunter CM, Caswell H, Amstrup SC, Stirling I (2010) Survival and breeding of polar bears in the southern Beaufort Sea in relation to sea ice. Journal of Animal Ecology 79(1): 117–127. 66. Sonne C, Gustavson K, Rige´t FF, Dietz R, Birkved M, et al. (2009) Reproductive performance in East Greenland polar bears (Ursus maritimus) may be affected by organohalogen contaminants as shown by physiologically- based pharmacokinetic (PBPK) modelling. Chemosphere 77(11): 1558–68. 44. Derocher AE, Wiig O (2002) Postnatal growth in body length and mass of polar bears (Ursus maritimus) at Svalbard. Journal of Zoology 256: 343–349. August 2014 \| Volume 9 \| Issue 8 \| e104037 PLOS ONE \| www.plosone.org 10
https://openalex.org/W3163849916	https://www.nature.com/articles/s41467-021-23204-x.pdf	English	null	Biomimetic high performance artificial muscle built on sacrificial coordination network and mechanical training process	Nature communications	2,021	cc-by	9,405	ARTICLE Results Preparation and mechanical performance of the EPDM-based artiﬁcial muscle material. A three-step compounding process in one internal mixer was conducted to incorporate Zn-based coordination bonds into the EPDM/lignin composite system, as shown in Fig. 1a. EPDM was ﬁrst blended with lignin, then mixed with zinc dimethacrylate (ZDMA), and ﬁnally with vulcanizing agents. The sample was named as LxZy@m%, with “L” standing for lignin, “Z” for ZDMA, and the numbers “x” and “y” for the parts per hundred of EPDM (100 phr), “@” for mechanical training, and the number “m” for the mechanical training strain. The biomass lignin is the largest aromatic biopolymer in nature and rich in oxygen-containing polar groups (Supplementary Fig. 2), which can efﬁciently function as natural ligands for the construction of interfacial coordination bonds. During the vul- canization process, the unsaturated double bonds in EPDM reacted with the methacrylate groups in ZDMA, generating the ZDMA-modiﬁed EPDM with functional carboxylate groups (Fig. 1b and Supplementary Fig. 3)23. The Zn-based coordination bonds could form between the carboxylate groups grafted on EPDM backbones and the polar functional groups in lignin such as the phenoxy and carboxylate groups24, which was conﬁrmed by FTIR spectra in Fig. 2a. The absorption peaks at 1276 cm−1 were attributed to the C−O stretching vibration peaks of aryl groups in lignin, and the signals at 1110 cm−1 and 1023 cm−1 were assigned to the C–O stretching vibration of alkyl groups in the composites, respectively25. All of these peaks had evident increase in intensity, accompanied by the increasing content of ZDMA, indicating the coordination effect between Zn2+ and oxygen-containing groups. Additionally, an obvious redshift from 1110 cm−1 to 1097 cm−1 was also observed as the ZDMA content increased, which further demonstrated the complexation of coordination bonds in composites. The carboxylate groups in ZDMA could also form hydrogen bonds with the polar groups in lignin. Typical engineering stress–strain behaviors of the com- posites with different content of ZDMA also veriﬁed the forma- tion of coordination bonds. As shown in Fig. 2b, the tensile strength and toughness of composites increased gradually with the ZDMA content, due to the dynamic destruction and recon- struction process of coordination bonds, through which they could not only dissipate energy on molecular level, but also eliminate stress concentration, constrain the chain segments, and promote the chain orientation, thus endowing materials with improved strength and toughness18,20. Biomimetic high performance artiﬁcial muscle built on sacriﬁcial coordination network and mechanical training process Zhikai Tu1,4, Weifeng Liu 1,4✉, Jin Wang2, Xueqing Qiu3✉, Jinhao Huang1, Jinxing L Liu 1,4✉, Jin Wang2, Xueqing Qiu3✉, Jinhao Huang1, Jinxing Li1 & Hongming Lou1 Zhikai Tu1,4, Weifeng Liu 1,4✉, Jin Wang2, Xueqing Qiu3✉, Jinhao Huang1, Jinxing Li1 & Hongming Lou1 Zhikai Tu1,4, Weifeng Liu 1,4✉, Jin Wang2, Xueqing Qiu3✉, Jinhao Huang1, Jinxing Li1 & Hongming Lou1 Zhikai Tu1,4, Weifeng Liu 1,4✉, Jin Wang2, Xueqing Qiu3✉, Jinhao Huang1, Jinxing Li1 & Hongming Lou1 Artiﬁcial muscle materials promise incredible applications in actuators, robotics and medical apparatus, yet the ability to mimic the full characteristics of skeletal muscles into synthetic materials remains a huge challenge. Herein, inspired by the dynamic sacriﬁcial bonds in biomaterials and the self-strengthening of skeletal muscles by physical exercise, high per- formance artiﬁcial muscle material is prepared by rearrangement of sacriﬁcial coordination bonds in the polyoleﬁn elastomer via a repetitive mechanical training process. Biomass lignin is incorporated as a green reinforcer for the construction of interfacial coordination bonds. The prepared artiﬁcial muscle material exhibits high actuation strain (>40%), high actuation stress (1.5 MPa) which can lift more than 10,000 times its own weight with 30% strain, characteristics of excellent self-strengthening by mechanical training, strain-adaptive stif- fening, and heat/electric programmable actuation performance. In this work, we show a facile strategy for the fabrication of intelligent materials using easily available raw materials. 1 School of Chemistry and Chemical Engineering, Guangdong Engineering Research Center for Green Fine Chemicals, South China University of Technology, Guangzhou, P. R. China. 2 The National Engineering Research Center of Novel Equipment for Polymer Processing, School of Mechanical & Automotive Engineering, South China University of Technology, Guangzhou, P. R. China. 3 School of Chemical Engineering and Light Industry, Guangdong University of Technology, Guangzhou, P. R. China. 4These authors contributed equally: Zhikai Tu, Weifeng Liu. ✉email: weifengliu@scut.edu.cn; qxq@gdut.edu.cn 1 NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications TURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/ ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x T he majority of biological load-bearing tissues including skeletal muscles present typical J-shaped stress–strain curves with their initially soft elastic modulus rapidly increased by orders of magnitude during deformation (Supple- mentary Fig. 1), which is also known as the strain-adaptive stif- fening performance1–3. Speciﬁcally, skeletal muscles can self- strengthen after physical exercise, because of the destruction and reconstruction process of the muscle ﬁbrils upon mechanical training4. Biomimetic high performance artiﬁcial muscle built on sacriﬁcial coordination network and mechanical training process Moreover, they can intelligently convert adenosine triphosphate (ATP) as chemical energy source into macroscopic motions to adapt the surrounding environment according to the electrical signals sent by the nervous system5. These combina- tional characteristics of skeletal muscles are indispensable for the survival of living organisms in complex environment, which has triggered tremendous interest for the development of artiﬁcial muscle materials. Similar to the destruction and reconstruction process of the muscle ﬁbrils upon physical exercise, through repetitive pre- stretching of mechanical training process, the rupture and reconstruction of randomly distributed coordination bonds prompt the orientation of chain segments along the pre-stretching direction. After mechanical training, the newly regenerated coor- dination bonds along the pre-stretching direction can not only prompt strain-induced crystallization (SIC) more efﬁciently, but also break at a relatively concentrated strain range, which imparts a rapidly increasing modulus. The prepared EPDM-based artiﬁcial muscle material exhibits excellent self-strengthening effect by mechanical training and strong strain-adaptive stiffening perfor- mance. Meanwhile, it can actuate programmable reciprocating motions under the stimulation of heat and electricity. The EPDM- based artiﬁcial muscle material in this work successfully mimics the comprehensive mechanical characteristics of skeletal muscles via a facile approach built on sacriﬁcial coordination bonds and mechanical training process, laying a foundation for large-scale fabrication of artiﬁcial muscle materials. T The bio-inspired artiﬁcial muscle materials are supposed to be able to convert external stimulus energy into macroscopic motions for a wide scope of potential applications such as actuators, robotics, and medical apparatus6. Under the stimulation of external heat, light, electricity etc., the difference in thermal expansion ratios7, the liquid crystalline-isotropic phase transition8 or the change of elec- tric ﬁeld9 inside the artiﬁcial muscle materials would result in macroscopic reversible deformation. For example, a tendril-like ﬁber-based artiﬁcial muscle material was recently fabricated by thermal drawing from the hybrid of cyclic oleﬁn copolymer elas- tomer and high-density polyethylene7. The macroscopic reversible deformation originating from the difference of two polymers in thermal expansion coefﬁcients endowed the material with rapid thermal actuation7. Although these previously reported artiﬁcial muscle materials could convert external stimulus energy into macroscopic motions, the basic mechanical performance did not ﬁt the features of skeletal muscles such as self-strengthening by mechanical training and strain-adaptive stiffening. Some recent studies have reported self-strengthening hydrogels by repetitive mechanical training strategy10–12. Biomimetic high performance artiﬁcial muscle built on sacriﬁcial coordination network and mechanical training process However, these materials with self-strengthening characteristics lack the intelligent features such as precise modulation and programmable control of actuation in response to external stimulus. Until now, integrating the full characteristics of skeletal muscles such as self-strengthening by mechanical training, strain-adaptive stiffening and intelligent actuation upon external stimulus, into a single synthetic material remains to be a huge challenge. Results Remarkly, L0Z12@0% (14.7 MPa) without lignin exhibited much poorer tensile strength Recently, the excellent mechanical performance of some bio- logical materials such as mussel byssus has been partly attributed to the reinforcing mechanism of dynamic sacriﬁcial bonds including the sacriﬁcial coordination bonds and hydrogen bonds13,14. Many efforts have been devoted to incorporate dynamic sacriﬁcial bonds into polymeric materials to enhance the mechanical properties15–21. For instance, Filippidi et al. found that via introducing Fe3+-catechol coordination bonds into the lightly cross-linked elastic epoxy network, the tensile strength and toughness of the elastomer was increased by 58 times and 92 times, respectively22. Nevertheless, the coordination bonds in these systems were randomly distributed in the polymer matrix, followed by irregular rupture during the deformation course, giving the material limited strain-adaptive stiffening performance. In this work, inspired by the dynamic sacriﬁcial bonds in bio- materials and the self-strengthening mechanism of skeletal mus- cles by physical exercise, we propose a strategy to combine the superiority of sacriﬁcial coordination bonds with mechanical training process for construction of high performance artiﬁcial muscle materials. The sacriﬁcial coordination bonds are ﬁrst introduced into a poly(ethylene-propylene-diene monomer) (EPDM) elastomer via incorporating biomass lignin as natural green reinforcer, followed by a mechanical training process. Recently, the excellent mechanical performance of some bio- logical materials such as mussel byssus has been partly attributed to the reinforcing mechanism of dynamic sacriﬁcial bonds including the sacriﬁcial coordination bonds and hydrogen bonds13,14. Many efforts have been devoted to incorporate dynamic sacriﬁcial bonds into polymeric materials to enhance the mechanical properties15–21. For instance, Filippidi et al. found that via introducing Fe3+-catechol coordination bonds into the lightly cross-linked elastic epoxy network, the tensile strength and toughness of the elastomer was increased by 58 times and 92 times, respectively22. Nevertheless, the coordination bonds in these systems were randomly distributed in the polymer matrix, followed by irregular rupture during the deformation course, giving the material limited strain-adaptive stiffening performance. In this work, inspired by the dynamic sacriﬁcial bonds in bio- materials and the self-strengthening mechanism of skeletal mus- cles by physical exercise, we propose a strategy to combine the superiority of sacriﬁcial coordination bonds with mechanical training process for construction of high performance artiﬁcial muscle materials. The sacriﬁcial coordination bonds are ﬁrst introduced into a poly(ethylene-propylene-diene monomer) (EPDM) elastomer via incorporating biomass lignin as natural green reinforcer, followed by a mechanical training process. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x 1. EPDM, Lignin, 8 mins; 2. ZDMA, 10 mins; 3. Curing ingredients, 5 mins. Blending Vulcanization Mechanical training LxZy@m% 170 oC LxZy@0% 80 oC a Covalent bonds EPDM chains Dynamic coordination bonds Lignin b Polyolefin fringed micelles 1 250 N ...... Strain Cyclic Prestretching Fig. 1 The preparation schematic diagram and mechanical perfomance of artiﬁcial muscle materials. a Preparation diagram of EPDM-based artiﬁcial muscle materials. b Schematic illustration for the interfacial coordination bonds in the EPDM composite. Mechanical training LxZy@m% 1 250 N ...... in Cyclic Prestretching LxZy@0% Mechanical training Mechanical training Covalent bonds on Polyolefin fringed micelles Polyolefin fringed micelles Dynamic coordination bonds Covalent bonds Fig. 1 The preparation schematic diagram and mechanical perfomance of artiﬁcial muscle materials. a Preparation diagram of EPDM-based artiﬁcial muscle materials. b Schematic illustration for the interfacial coordination bonds in the EPDM composite. 0 100 200 300 400 500 600 700 0 5 10 15 20 25 Stress (MPa) Strain (%) L40Z0@0% L40Z4@0% L40Z8@0% L40Z12@0% 0 100 200 300 400 500 600 0 5 10 15 20 25 30 35 Stress (MPa) Strain (%) L40Z12@0% L40Z12@200% L40Z12@300% L40Z12@400% L40Z12@500% L40Z12@600% 0 50 100 150 200 250 0 5 10 15 20 25 30 35 Stress (MPa) Strain (%) E0=2.0 MPa E1=33.2 MPa E1/E0=16.5 L40Z12@600% a b c d 1350 1300 1250 1200 1150 1100 1050 1000 1023 1110 1107 1097 1276 Absorbance (a.u.) Wavenumber (cm-1) L40Z0@0% L40Z8@0% L40Z12@0% Fig. 2 Mechanical properties of artiﬁcal muscle material. a The FTIR spectra of EPDM/lignin composites with different content of ZDMA (a.u. means absorbance unit). b The engineering stress–strain curves of EPDM/lignin composites with different content of ZDMA. c The engineering stress–strain curves of EPDM composites after mechanical training for 250 cycles at the stretching speed of 200 mm/min under different training strain. d The tensile stress–strain curve of L40Z12@600% after mechanical training for 250 cycles at 600% strain. Source Data for Fig. 2a–d are available as an associated article ﬁle. 25 b Fig. 2 Mechanical properties of artiﬁcal muscle material. a The FTIR spectra of EPDM/lignin composites with different content of ZDMA (a.u. means absorbance unit). b The engineering stress–strain curves of EPDM/lignin composites with different content of ZDMA. c The engineering stress–strain curves of EPDM composites after mechanical training for 250 cycles at the stretching speed of 200 mm/min under different training strain. Results NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications 2 ARTICLE ARTICLE 7c, d), indicating that a certain part of chain alignment was stabilized by the coordination bonds after mechanical training, which was veriﬁed by the relative higher crystallinity of L40Z12@300% than L40Z12@0% shown in Supplementary Fig. 8. The stress relaxation cure also conﬁrmed that only part of rubber chain strands relaxed after mechanical training (Supplementary Fig. 9). To further reveal the deformation mechanism for the dual- stage enhancement phenomenon, the hysteresis tensile tests at various strains from 50 to 300% were conducted to study the energy dissipation of L40Z12@300% during stretching, as shown in Fig. 3b. It is obvious that the hysteresis loss of L40Z12@300% increased signiﬁcantly with the applied strain and was larger than that of L40Z12@0% (Supplementary Fig. 11a, b). The hysteresis ratio of L40Z12@300% increased rapidly when the applied strain was between 75 and 150%, but leveled off at strains larger than 150% (Fig. 3b). This suggested that, for L40Z12@300% after mechanical training, the dynamic fracture of coordination bonds might concentrate in the early stage of strain range from 75 to 150%, which led to a rapid growth in the energy dissipation ratio and elastic modulus of the ﬁrst-stage enhancement. The dynamic fracture of coordination bonds in the strain from 75 to 150% was also demonstrated by the signal variations in the FTIR analysis (shown in Supplementary Fig. 12). Remarkably, the materials presented typical J-shaped stress–strain curves after mechanical training, conforming to the strain-adaptive stiffening characteristic of soft biological tissues1–3. For example, the elastic modulus of the sample L40Z12@600% stiffened 16.5 times from 2.0 MPa at small strain to 33.2 MPa when the strain deformed from 75 to 150%, as shown in Fig. 2d. The self-strengthening effect by mechanical training and the typical strain-adaptive stiffening of the EPDM-based elastomer composites matched well with the mechanical performance of skeletal muscles, demonstrating great potential as artiﬁcial muscle materials. The reason for the second-stage enhancement in the stress–strain curve of L40Z12@300% was attributed to the strain-induced-crystallization (SIC) of the elastomer chain segments after dynamic fracture of coordination bonds. The wide-angle X-ray diffractograms (XRD) of L40Z12@300% at various stretching strains were depicted in Fig. 3c. The sample L40Z12@300% exhibited wide amorphous diffraction peak at 0% strain, with weak reﬂection of fringed micellar crystals at 2θ of 20.5° hidden in the amorphous diffraction peak. ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x were constituted of ZDMA providing metal ions and lignin acting as ligands in this EPDM composite system. Without ZDMA, the increase in modulus versus strain (slope of the stress–strain curve) was much weaker in L40Z0@300% compared with L40Z12@300%. While without lignin, the phenomenon of dual- stage enhancement disappeared in the tensile curve of L0Z12@300%. These results demonstrated that the coordination bonds and the green reinforcer lignin played a vital role on the self-strengthening and the strain-adaptive stiffening of the EPDM composite after mechanical training. As the dynamic coordina- tion bonds could increase the interfacial interactions between lignin and polymer matrix21, the interfacial coordination bonds promoted the internal stress transfer from EPDM matrix to lignin particles, leading to repeated shearing on lignin particles during the mechanical training process, and thus resulting in smaller particle size and better dispersion of lignin in polymer matrix after mechanical training (shown in Supplementary Fig. 10), which also contributed to the self-strengthening of the EPDM composite. compared with L40Z12@0% (24.8 MPa) (Supplementary Fig. 4). The above results proved that reactive coordination bonds and coordinative reinforcer lignin were both necessary to prepare the high-performance EPDM elatomer composite, providing a direction for the utilization of lignin in high-performance elas- tomer composite, which was a huge challenge in the past. p g g p The EPDM/lignin composite L40Z12@0% was further sub- mitted for mechanical training to mimic the self-strengthening performance of skeletal muscles. The training process was conducted via repeated stretching and unloading (Supplementary Fig. 5), and the corresponding loading–unloading stress–strain curves were illustrated in Supplementary Fig. 6. The engineering stress–strain curves in Fig. 2c demonstrated that the ﬁnal EPDM/ lignin composites exhibited signiﬁcantly improved strength after mechanical training, just like the self-strengthening effect of skeletal muscles by physical exercise. As the training strain increased from 0 to 600%, the tensile strength at failure of the elastomer composite increased gradually from 24.8 to 30.7 MPa. The stress at 200% strain increased nearly 2.5 times for L40Z12@600% compared with L40Z12@0% (Supplementary Fig. 7a). Increasing the training frequency from 100 to 1000 cycles led to a similar variation tendency in the tensile stress–strain curves (Supplementary Fig. 7b). Under the repetitive pre-stretching of mechanical training, the residual strain of the elastomer also gradually increased with the training strain and training frequency (Supplementary Fig. NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x d The tensile stress–strain curve of L40Z12@600% after mechanical training for 250 cycles at 600% strain. Source Data for Fig. 2a–d are available as an associated article ﬁle. 3 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x Fig. 3 Characterizations for the self-strengthening mechanism. a The engineering stress–strain curves of L40Z12@300%, L0Z12@300%, and L40Z0@300%. b The hysteresis tensile curves and the corresponding hysteresis ratio of L40Z12@300% at various strains from 50 to 300%. c The XRD patterns of L40Z12@300% at various ﬁxed strains (a.u. means absorbance unit). d The relative crystallinity calculated from the XRD patterns of L40Z12@300%. e The XRD patterns of L40Z12@0% and L40Z12@300% at various ﬁxed strains. f Normalized 1D correlation function curves for L40Z12@300% at various strains. g The core crystalline layer length (d0) at different strains for L40Z12@0% and L40Z12@300%. Source Data for Fig. 3a–g are available as an associated article ﬁle. Fig. 3 Characterizations for the self-strengthening mechanism. a The engineering stress–strain curves of L40Z12@300%, L0Z12@300%, and L40Z0@300%. b The hysteresis tensile curves and the corresponding hysteresis ratio of L40Z12@300% at various strains from 50 to 300%. c The XRD patterns of L40Z12@300% at various ﬁxed strains (a.u. means absorbance unit). d The relative crystallinity calculated from the XRD patterns of L40Z12@300%. e The XRD patterns of L40Z12@0% and L40Z12@300% at various ﬁxed strains. f Normalized 1D correlation function curves for L40Z12@300% at various strains. g The core crystalline layer length (d0) at different strains for L40Z12@0% and L40Z12@300%. Source Data for Fig. 3a–g are available as an associated article ﬁle. was weaker than that of L40Z12@300%, and this difference was more obvious when the stretching strain increased. These results indicated that, in the presence of coordination bonds, the mechanical training process facilitated chain orientation in the elastomer composite and promote more obvious SIC phenomena during stretching, thus causing the strain-induced stiffening in the second-stage enhancement of L40Z12@300%. strains27, the structure parameters such as the long period length (L), the transition layer length (dtr), the length of the crystalline layer plus transition layer (lc), were achieved in Supplementary Table 1. The core-crystalline layer length (d0 = lc −dtr) was then calculated accordingly. The evolution of these structure para- meters reﬂected the deformation mechanism of L40Z12@300%. The core-crystalline layer length (d0) increased as the strain enlarged, implying that new thicker micellar crystals formed upon stretching (Fig. 3g). Speciﬁcally, for L40Z12@300%, d0 main- tained constant value between 75–100% strain, consistent with the plateau revealed in the relative crystallinity analysis in Fig. 3d. ARTICLE As the strain increased from 50 to 320%, the obvious crystalline peaks at 2θ of 20.5° and 22.8° assigning to the (110) and (200) lattice planes of the polyethylene orthorhombic crystal cell26, respectively, were observed in L40Z12@300%, and the two peaks both intensiﬁed with the strain, which was the typical phenomenon of SIC. As shown in Fig. 3d, the relative crystallinity of L40Z12@300% increased with the strain, and meanwhile, a clear plateau was disclosed in the strain range from 75 to 150%, in which the dynamic coordination bonds underwent the fracture process as explained earlier. This veriﬁed that, in the ﬁrst reinforcing stage of L40Z12@300% (strain range from 75 to 150%), the relative crystallinity had little change, the rapid growth in modulus was mainly resulted from the dynamic fracture of coordination bonds. The increase trend in the relative crystallinity of L40Z12@300% was obviously accelerated as the strain reached over 150%, verifying that the SIC course followed the fracture of coordination bonds in the second-stage enhancement (strain >150%). Interestingly, the stress–strain curves of the elastomers after mechanical training revealed a phenomenon of obvious dual- stage enhancement in modulus, which has been rarely reported before. Take the sample L40Z12@300% as an example (Fig. 2c), the ﬁrst-stage enhancement in modulus occurred in the strain range of 75–150%, followed by the second-stage enhancement. A closer inspection could ﬁnd that, as the training strain increased from 200 to 600%, the elastic modulus for the both enhancing stages increased gradually, the ﬁrst-stage enhancement was postponed to a larger strain and the second-stage enhancement was pulled up to a smaller strain, as shown in Fig. 2c, suggesting a special reinforcing mechanism for this EPDM-based artiﬁcial muscle material. Self-strengthening mechanism by mechanical training. To investigate the effect of coordination bonds on the mechanical training performance, the control sample L40Z0@300% without ZDMA and the sample L0Z12@300% without lignin were com- pared by mechanical training at 300% strain for 250 cycles. As shown in Fig. 3a, the tensile strength of L40Z0@300% and L0Z12@300% was much lower than that of the sample L40Z12@300%. Apparently, the dynamic coordination bonds For comparison, the XRD patterns were also examined for the control sample L40Z12@0% without mechanical training. As shown in Fig. 3e, either before stretching (0% strain) or under the same strain, the crystalline diffraction intensity of L40Z12@0% 4 ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x faster, especially when the strain was larger than 150%. This reconﬁrmed that, the rapid modulus growth in the ﬁrst-stage enhancement of L40Z12@300% was mainly resulted from the dynamic fracture of coordination bonds, after which SIC occurred and contributed to the second-stage enhancement. However, the d0 value of L40Z12@0% and L40Z0@300% increased steadily during the whole stretching course without the similar plateau (Fig. 3g and Supplementary Fig. 13d), showing typically uniform deformation of elastomers. As shown in Supplementary Table 1, comparison on the d0 and Δd0 values of L40Z12@0% and L40Z0@300% with L40Z12@300% further demonstrated that, after mechanical training, the sacriﬁcial coordination bonds in the elastomer composite could not only stabilize chain orientation, but also promote the orientation of chain segments during the stretching course. faster, especially when the strain was larger than 150%. This reconﬁrmed that, the rapid modulus growth in the ﬁrst-stage enhancement of L40Z12@300% was mainly resulted from the dynamic fracture of coordination bonds, after which SIC occurred and contributed to the second-stage enhancement. However, the d0 value of L40Z12@0% and L40Z0@300% increased steadily during the whole stretching course without the similar plateau (Fig. 3g and Supplementary Fig. 13d), showing typically uniform deformation of elastomers. As shown in Supplementary Table 1, comparison on the d0 and Δd0 values of L40Z12@0% and L40Z0@300% with L40Z12@300% further demonstrated that, after mechanical training, the sacriﬁcial coordination bonds in the elastomer composite could not only stabilize chain orientation, but also promote the orientation of chain segments during the stretching course. surface of lignin particles (State 3 in Fig. 4), which led to the decrease in the modulus (110–150%) and resulted in the S-shape turning of the stress–strain curve in Level 2. Following Level 2, further stretching to a large strain in Level 3 led to the ordered chain alignment and SIC, which offered another dramatic increment in the modulus in Level 3. Specially, the sacriﬁcial coordination bonds after mechanical training promoted SIC more efﬁciently as they dynamically ruptured in a focused strain region in Level 2 and facilitated the chain alignment in Level 3. It should be noted that, without lignin, there would be no slippage of elastomer chain segments on the particle surface and the phenomenon of dual-stage enhancement would disappear, while without coordination bonds, the self-strengthening effect and the strain-adaptive stiffening would disappear, which both would result in poor mechanical performance of the elastomer composite. ARTICLE Based on the analysis above, the self-strengthening mechanism by mechanical training for the EPDM-based elastomer composite was proposed in Fig. 4. Firstly, through repetitive pre-stretching of mechanical training process, a certain part of chain orientation was stabilized by the coordination bonds along the pre-stretching direction (State 1 in Fig. 4), forming a more well-deﬁned strand conﬁguration, which would beneﬁt for the stress transfer from elastomer matrix to the coordination bonds. Then, as the elastomer sample after mechanical training was stretched, the tensile stress–strain curve could be divided into three levels (Fig. 4). Take the sample L40Z12@300% as an example, in Level 1, the physically entangled elastic network was deformed ﬁrst at small strain, during which the elastomer chain segments were tensioned (State 2 in Fig. 4). Upon further stretching to the medium strain (75–150% for L40Z12@300%), the stress was transferred to the coordination bonds, the sacriﬁcial coordination bonds started to resist the applied stress and dynamically ruptured to dissipate the energy, which led to the substantial increase in the modulus in Level 2. Once the coordination bonds were sacriﬁced, the elastomer chain segments slipped from the In short, the sacriﬁcial coordination bonds underwent destruction and reconstruction process in the elastomer compo- site, which not only stabilized the chain orientation along the pre- stretching direction after mechanical training, but also efﬁciently dissipated energy and promoted the oriented crystallization, thus endowing the material with strong self-strengthening and strain- adaptive stiffening performance. In order to demonstrate the variation of related network inside the elastomer composite during the deformation course, a one- dimensional constitutive model was employed for describing the stress–strain relationship of materials. According to the char- acteristics of three-level stress response illustrated in Fig. 4, we assume that the molecular chain network in this elastomer composite was composed of three sub-networks, including the covalent-bond network, physically entangled network, and coordination-bond network (Fig. 5a). The detailed explanation for the one-dimensional constitutive model was provided in the Supporting Information. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x After the plateau, the d0 value of L40Z12@300% increased much g The evolution of polymer matrix during deformation could be further revealed by in-situ tensile small-angle X-ray scattering (SAXS) analysis. As shown in Fig. 3f and Supplementary Fig. 13a–c, through analysis of the normalized one-dimensional correlation function curves of the SAXS proﬁles at different 5 NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications 5 ARTICLE The value of σ0, σ1, and σ2 corresponds to the stress of covalent-bond network, physically entangled Level 1 Stretch small strain State 0 State 4 State 3 Level 2 Stretch medium strain Mechanical training State 1 State 2 Level 3 Stretch large strain Covalent bonds EPDM chains Dynamic coordination bonds Lignin Polyolefin fringed micelles Strain induced crystallization 0 50 100 150 200 250 300 350 0 5 10 15 20 25 State 1 State 2 State 3 Level 3 Level 2 Stress (MPa) Strain (%) Dynamic rupture of coordination bonds Level 1 Strain-induced crystallization Deformation of elastic network State 4 Fig. 4 Schematic illustration. Self-strengthening mechanism by mechanical training for the EPDM-based elastomer composite. State 4 Stretch large strain Level 3 State 3 Stretch medium strain Level 2 Level 1 Stretch small strain State 2 State 1 Strain induced crystallization Polyolefin fringed micelles Dynamic coordination bonds Covalent bonds ic illustration. Self-strengthening mechanism by mechanical training for the EPDM-based elastomer composite. Fig. 4 Schematic illustration. Self-strengthening mechanism by mechanical training for the EPDM-based elastom NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunicatio 6 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x Fig. 5 Theoretical simulation for the artiﬁcial muscle material. a Schematic illustration of the variation of the covalent-bond network (σ0), physical entangled network (σ1), and coordination-bond network (σ2) when the training strain increased. b The simulation of engineering stress–strain curve for L40Z12@300%, including covalent-bond network (σ0), elastic network (σ1), and coordination-bond network (σ2). c The simulation results of engineering stress–strain curves for the samples trained at different strains. Source Data for Fig. 5b–c are available as an associated article ﬁle. Fig. 5 Theoretical simulation for the artiﬁcial muscle material. a Schematic illustration of the variation of the covalent-bond network (σ0), physical entangled network (σ1), and coordination-bond network (σ2) when the training strain increased. b The simulation of engineering stress–strain curve for L40Z12@300%, including covalent-bond network (σ0), elastic network (σ1), and coordination-bond network (σ2). c The simulation results of engineering stress–strain curves for the samples trained at different strains. Source Data for Fig. 5b–c are available as an associated article ﬁle. materials. As more chain alignments were stabilized along the prestretching direction under larger training strain (Supple- mentary Fig. 7c), L40Z12@600% was selected to study the thermal-induced actuation performance by the dynamic mechanical analyzer (DMA) under the isoforce mode. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x Fig. 6 Thermal-induced actuation performance. a Strain variation in isoforce mode and the corresponding temperature of the sample L40Z12@600% plotted against time. b Schematic graph of reversible actuation effect of EPDM composites after mechanical training. c Stress variation in isostrain mode of L40Z12@600% plotted against temperature. d Photographs of the angle change and e plot of the angle change vs cycles for the thermal actuation of L40Z12@600% lifting up a 200 g load (ﬁre represents heating and ice represents cooling). f Photographs of the length change and g plot of the strain change vs cycles for the thermal actuation of L40Z12@600% (20 mg) bearing a load of 205 g which was 10,000 times higher than its own weight (ﬁre represents heating and ice represents cooling). Source Data for Fig. 6a, c e, g. are available as an associated article ﬁle. Fig. 6 Thermal-induced actuation performance. a Strain variation in isoforce mode and the corresponding temperature of the sample L40Z12@600% plotted against time. b Schematic graph of reversible actuation effect of EPDM composites after mechanical training. c Stress variation in isostrain mode of L40Z12@600% plotted against temperature. d Photographs of the angle change and e plot of the angle change vs cycles for the thermal actuation of L40Z12@600% lifting up a 200 g load (ﬁre represents heating and ice represents cooling). f Photographs of the length change and g plot of the strain change vs cycles for the thermal actuation of L40Z12@600% (20 mg) bearing a load of 205 g which was 10,000 times higher than its own weight (ﬁre represents heating and ice represents cooling). Source Data for Fig. 6a, c e, g. are available as an associated article ﬁle. L40Z12@0% without mechanical training process exhibited only less than 25% unstable reversible strain variations upon the heating–cooling cycles (Supplementary Fig. 15a), demonstrating that the mechanical training process was an essential and effective method for improving the actuation strain of the arti- ﬁcial muscle material. to 90 °C, the maximum actuation stress of L40Z12@600% reached 1.5 MPa at 90 °C, demonstrating excellent ability to do work. Remarkably, the actuation stress of L40Z12@600% (1.5 MPa, 0.1% isostrain) was larger than most of previous LCE-based actuators34–36, and was more than 4 times higher than that of the human skeletal muscle (0.35 MPa). ARTICLE Using 1.3 MPa as a constant stress of preloading, the strain variation of L40Z12@600% against temperature was plotted in Fig. 6a. Upon heating, a substantial strain change from 21 to −20% was pre- sented as the temperature increased from −30 oC to 90 oC, while the strain value recovered to original length again along with the temperature decreased to −30 oC on the cooling process. As the heating–cooling cycle repeated, the reversible strain variation exhibited good repeatability. The mechanism for the thermal- induced reversible strain variation was attributed to the oriented crystallization and crystallization-induced elongation (CIE) upon cooling and melting-induced contraction (MIC) upon heating of the elastomer matrix28,29, as illustrated in Fig. 6b. The reversible actuation strain reached 41% in this work, which was the largest value ever achieved in the polyoleﬁn-based actuation materials30–33. The value of actuation strain (41%) performed by the EPDM-based artiﬁcial muscle material also met the requirement of human skeletal muscle (40%). In contrast, network, and dynamic coordination-bond network, respectively. As shown in Fig. 5b, c, the engineering stress–strain curves of the elastomer composite after mechanical training at different training strain were perfectly simulated by the one-dimensional constitutive model, demonstrating that the three sub-network model could accurately describe the stress–strain behaviors of the elastomer composite after mechanical training. As summarized in Supplementary Table 2, through analysis on the related structure parameters ﬁtted from the model simulation (Supplementary Fig. 14), the mechanical property of the artiﬁcial muscle material could be facilely customized by precisely programming the input mechanical training strain, the density of coordination-bond network and covalent-bond network. Note that one merit of this strategy is that it does not require complex molecular design or sophisticated synthetic process comparing with those previously reported artiﬁcial muscle materials. Programmable actuation performance of the EPDM-based artiﬁcial muscle material. The key function of skeletal muscles is the intelligent actuation performance in response to external stimulus, which is the long-term pursuit of artiﬁcial muscle TURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications 7 7 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x On the contrary, the typical “thermal expansion and cold shrinkage” effect occurred in the sample L40Z12@0% during the heating–cooling cycle at isostrain mode (Supplementary Fig. 15c), verifying that the elastomer composite without the mechanical training process did not exhibit the ability to do work. The control samples L40Z0@300% without ZDMA and L0Z12@300% without lignin only generated the actuation stress of 1.0 MPa (Supplementary Fig. 15d, e). Apparently, incorporating lignin and sacriﬁcial coordination The actuation stress of L40Z12@600% was then measured by DMA under the isostrain mode at 0.01% strain in tension. As shown in Fig. 6c, the actuation stress generated from the contraction of the sample upon heating slowly increased as the temperature grew from −10 °C to 50 °C. Then the incremental process was accelerated when the sample was mostly melted after 50 °C (the DSC curves of the elastomer composites were provided in Supplementary Fig. 15b. In a heating–cooling cycle of −10 °C NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications 8 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x material triggered by electrical signal. For this purpose, an electric-programmable EPDM-based artiﬁcial muscle material was fabricated by substituting half mass of lignin with conductive carbon black. The prepared sample L20C20Z12@300% also exhibited the self-strengthening effect and the strain-adaptive stiffening behavior, which were similar as L40Z12@600% and could mimic the mechanical performance of skeletal muscles (Supplementary Fig. 16). bonds into the elastomer network not only enhanced the mechanical properties, but also improved the actuation stress. In order to examine the potential application of the EPDM- based artiﬁcial muscle material, the sample L40Z12@600% was attached to the arm of a puppet lifting a 200 g load as a simulation of human skeletal muscle, as shown in Fig. 6d. During the multiple heating−cooling cycles, a reversible switching angle of about 30o between the forearm and the humerus was observed (Fig. 6e and Supplementary Movie 1). When the sample of 20 mg was directly attached to the total weight of 205 g (Fig. 6f), it could repetitively lift it up and down with the actuation strain larger than 30%, which was more than 10,000 times of its own weight, as illustrated in Fig. 6f, g and Supplementary Movie 2. The sample L20C20Z12@300% was assembled into a program- controlled circuit, as shown in Fig. 7a. Discussion l In conclusion, inspired by the dynamic sacriﬁcial bonds in bio- materials and the self-strengthening mechanism of skeletal muscles by physical exercise, we have established a facile strategy for preparing the programmable artiﬁcial muscle material by integrating the superiority of sacriﬁcial coordination bonds and mechanical training process. The biomass lignin was introduced into the commercially available EPDM as natural green reinforcer to serve as ligands for the construction of coordination bonds, followed by a repetitive mechanical training process. An obvious dual-stage enhancement in modulus was obtained for the elas- tomer composite after the mechanical training process. It was demonstrated that the destruction and reconstruction process of the sacriﬁcial coordination bonds along the pre-stretching direction of repetitive training helped to form a more well- deﬁned strand conﬁguration, which not only contributed efﬁcient energy dissipation in the ﬁrst-stage enhancement, but also pro- moted the strain-induced crystallization for the second-stage enhancement, thus endowing the material with strong self- strengthening by mechanical training and strain-adaptive stif- fening performance, in which the elastic modulus could increase 16.5 times from 2 MPa to 33.2 MPa as the strain increased. The simulation results of one-dimensional constitutive model demonstrated that the mechanical property of the EPDM-based artiﬁcial muscle material could be customized by precisely pro- gramming the mechanical training strain, the density of coordination-bond network and covalent-bond network. The EPDM-based artiﬁcial muscle material could actuate program- mable reciprocating motions in accordance to the stimulation of heat and electricity, with the actuation stress and actuation strain reached 1.5 MPa and 41%, respectively. Most importantly, this EPDM-based elastomer composite is the ﬁrst commercial polyoleﬁn-derived artiﬁcial muscle material that fulﬁlls all basic requirements for macroscopic mechanical performance of skeletal muscles including the actuation strain (>40%), actuation stress (>0.35 MPa), characteristics of self-strengthening by mechanical training, strain-adaptive stiffening with more than 10 times increase in modulus, and heat/electric programmable actuation. We hope this facile strategy via combing the sacriﬁcial coordi- nation bonds and mechanical training process to be expanded in the fabrication of sophisticated intelligent materials using easily available green raw materials by means of easily industrial scale- up. It not only provides a method for the intellectualization of traditional elastomer materials, but also provides a direction for the high-value utilization of industrial biomass resources such as lignin. Characterization. Bruker Vertex 70 FTIR spectrometer (Bruker, Germany) was used to collect Fourier transform infrared (FTIR) spectra at the attenuated total reﬂection (ATR) mode. Discussion l In the isostrain experiments, constant strain was ﬁrst ﬁxed at 0.01%, the sample was cooled to −30 °C, then the actuation stress was recorded along with the temperature ranged from −30 °C to 90 °C at a rate of 3.0 °C/min. The range of actuation stress in the isostrain tests was between 0.9 MPa and 1.6 MPa. X-ray diffraction (XRD) test was performed on a X-ray diffractometer (PANalytical, X’pert3 Powder) using Cu radiation with incident wavelength of 0.1541 nm. The sample was stretched to the ﬁxed strain before testing, and was then scanned at room temperature with the scanning step of 0.013° and the scanning speed of 15 s/step. The resulting proﬁles were transformed to the relative crystallinity by the software (Jade 6). Small-angle X-ray scattering (SAXS) experiments were conducted by a Bruker Nanostar U instrument. The wavelength of X-ray source was 1.54 Å and the detector was Vantec 2000 with the sample-to-detector distance of 1078 mm. The exposure time was set 600 s for each scan at each strain. In the thermo-stimulus actuation test, the sample was heated by a hair drier and cooled down, the weights were lifted up and down, a switching angle and a reversible strain were observed and recorded. In the electric-programmable actuation test, the sample L20C20Z12@300% was assembled into a program-controlled circuit, as shown in Fig. 7a. According to Joule’s law, under a constant voltage of 100 V, the current passing through the sample generated heat, thus increased the temperature inside the sample, leading to the sample contraction. When the current was turned down or decreased, the sample cooled down and relaxed to its original length. The temperature inside the sample was detected by the infrared camera (FLIR T530, 30 mK thermal sensitivity). The current was output from a DC power supply (MAISHENG, MS-1001D). ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x thermal energy into macroscopic actuations in accordance to the current signals, which is a good bionic duplicate of skeletal muscles converting the chemical energy source (ATP) into macroscopic motions according to the current signals sent by the nervous system. Note that the current values (5–30 mA) adopted for the actuation control in this work was quite close to ﬁt for the demand of industrial standard current-control signal (4–20 mA). The EPDM-based artiﬁcial muscle material with electric- programmable actuation performance has demonstrated a great potential for the practical industrial application. industrial products and used as received. Poly-switch (PPTC) was provided by Shenzhen Tairuiyuan Technology Co., Ltd. industrial products and used as received. Poly-switch (PPTC) was provided by Shenzhen Tairuiyuan Technology Co., Ltd. industrial products and used as received. Poly-switch (PPTC) was provided by Shenzhen Tairuiyuan Technology Co., Ltd. Fabrication methods. Firstly, EPDM (100 phr) was mixed with lignin (40 phr) at 50 rpm and 80 °C for 8 min in an internal mixer (Guangdong Lina Co., Ltd., Dongguan, China). Then, a certain amount of ZDMA was added and mixed for 10 min. Next, all other ingredients such as ZnO (5 phr), SA (1 phr), BIPB (1 phr), TAIC (0.5 phr), were added and mixed for another 5 min, followed by hot-pressing at 170 °C for 20 min. Finally, the vulcanized elastomer was repetitively stretched for 250–1000 times at a ﬁxed strain by electronic universal testing machine (MTS, China) to get the artiﬁcial muscle materials. All the artiﬁcial muscle materials were rested for half an hour prior to use, and were then cut into the needed shape (dumbbell or ribbon) for further study. The sample was named as LxZy@m%, with “L” standing for lignin, “Z” for ZDMA, and the numbers x and y for the parts per hundred of EPDM (100 phr), “@” for mechanical training, and the number m for the mechanical training strain. For instance, L40Z12@300% means lignin 40 phr, ZDMA 12 phr, and the ﬁxed 300% training strain. L40Z12@0% means the sample was not submitted for mechanical training. The “C” in L20C20Z12@300% repre- sents for conductive carbon black. The sample L40Z0@300% without ZDMA and the sample L0Z12@300% without lignin were compared by mechanical training at 300% strain for 250 cycles. Data availability The source data that support the ﬁndings of this study are available in materialscloud with the identiﬁer https://doi.org/10.24435/materialscloud:9a-7y37. Source data are provided with this paper. Methods M t i l Materials. EPDM 3745 P (ethylene 70 wt%, propylene 29.5 wt%, ethylidene nor- bornene 0.5 wt%) with antioxidant 1010 was obtained from Dow Chemical. Lignin was provided by Shanghai Dongsheng Co., Ltd. Zinc dimethacrylate (ZDMA) was purchased from Sigma-Aldrich. Conductive carbon black (Ketjen Black, ECP600JD, Japanese LION) was purchased from Suzhou Yilongsheng Energy Technology Co., Ltd. Other ingredients such as zinc oxide (ZnO), stearic acid (SA), bis(tert-butyldioxyisopropyl) benzene (BIPB), and triallyl isocyanurate (TAIC) are Materials. EPDM 3745 P (ethylene 70 wt%, propylene 29.5 wt%, ethylidene nor- bornene 0.5 wt%) with antioxidant 1010 was obtained from Dow Chemical. Lignin was provided by Shanghai Dongsheng Co., Ltd. Zinc dimethacrylate (ZDMA) was purchased from Sigma-Aldrich. Conductive carbon black (Ketjen Black, ECP600JD, Japanese LION) was purchased from Suzhou Yilongsheng Energy Technology Co., Ltd. Other ingredients such as zinc oxide (ZnO), stearic acid (SA), bis(tert-butyldioxyisopropyl) benzene (BIPB), and triallyl isocyanurate (TAIC) are Discussion l The sample was stretched to a ﬁxed strain (0%, 75%, 150%) before testing, and the results were normalized at 2918 cm−1 to evaluate the var- iation of coordination bonds. The scanning electron microscopy (SEM) was conducted on a Hitachi UHR FE- SEM SU8220 instrument (Hitachi, Tokyo, Japan) with an accelerating voltage of 5 kV, and the samples were sputtered with a thin gold ﬁlm before test to enhance the conductivity. Mechanical training and tensile tests were carried out on a CMT electronic universal testing machine (MTS, China) with the speed of 200 mm/min at room temperature. The loss energy/tensile toughness in the hysteresis tensile curves were calculated from the integral area under the loading and unloading curves in the cyclic tensile tests. y Dynamic mechanical analyzer (DMA) was used to measure the actuation stress and actuation strain of the samples. In the isoforce tests, constant stress (ranging from 1.0 MPa to 1.3 MPa) was ﬁrst applied to the sample, the sample was cooled to −30 °C, then the actuation strain was recorded along with the temperature varied from −30 °C to 90 °C at a rate of 3.0 °C/min. The actuation strain was deﬁned as (L −L0)/L0, where L is the real-time length of the sample measured at any Dynamic mechanical analyzer (DMA) was used to measure the actuation stress and actuation strain of the samples. In the isoforce tests, constant stress (ranging from 1.0 MPa to 1.3 MPa) was ﬁrst applied to the sample, the sample was cooled to −30 °C, then the actuation strain was recorded along with the temperature varied from −30 °C to 90 °C at a rate of 3.0 °C/min. The actuation strain was deﬁned as (L −L0)/L0, where L is the real-time length of the sample measured at any temperature and L0 is the initial length before testing. In the isostrain experiments, constant strain was ﬁrst ﬁxed at 0.01%, the sample was cooled to −30 °C, then the actuation stress was recorded along with the temperature ranged from −30 °C to 90 °C at a rate of 3.0 °C/min. The range of actuation stress in the isostrain tests was between 0.9 MPa and 1.6 MPa. ( 0)/ 0, g p y temperature and L0 is the initial length before testing. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x By adjusting the current value from 5 mA to 30 mA, the EPDM-based artiﬁcial muscle material was able to lift up a 505 g load (2000 times its own weight), with the reversible motion distance (\|ΔL\|) of 7 mm (13% strain), and this behavior was repeatable (Fig. 7b and Supple- mentary Movie 3). When the current value was adjusted between 0 mA and 25 mA, the reversible actuation strain could be further increased to 20% (Supplementary Fig. 17). As shown in Fig. 7c, both the actuation strain and the sample temperature increased with the current value. The elastomer composite converted the As is well known, the nervous system programs the movements based on the sensory information, and then sends these instructions to the muscles in the form of electrical signals, followed by the muscles’ contraction or relaxation according to the variation of potential/electricity5. This interesting behavior of real muscles inspired us to develop verisimilar artiﬁcial muscle Fig. 7 Electric-programmable actuation performance. a Schematic diagram of the circuit for the electric-programmable actuation tests. b Photographs of the length change of L20C20Z12@300% under varied current signals (the color change of bulbs represent the variation of the current). c Quantitative evaluation of the variation of actuation strain in response to the current signal in the electric-triggered actuation tests. Source Data for Fig. 7c is available as an associated article ﬁle. NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications 9 Author contributions 13. Harrington, M. J., Masic, A., Holten-Andersen, N., Waite, J. H. & Fratzl, P. Iron-clad ﬁbers: a metal-based biological strategy for hard ﬂexible coatings. Science 328, 216–220 (2010). Z.T. and W.L. designed and performed the experiments and wrote the manuscript. J.W. performed the theoretical simulations. J.H. participated in data characterization for SAXS. J.L. took part in data recording for actuation experiment. And H.L. participated in the data discussion for XRD. W.L. and X.Q. revised the manuscript and supervised the research. 14. Lee, S. M. et al. Greatly increased toughness of inﬁltrated spider silk. Science 324, 488–492 (2009). 15. Ducrot, E., Chen, Y., Bulters, M., Sijbesma, R. P. & Creton, C. Toughening elastomers with sacriﬁcial bonds and watching them break. Science 344, 186–189 (2014). Acknowledgements The authors gratefully thank the National Natural Science Foundation of China (22038004, 22078116, 21706082), Guangdong Provincial Key Research and Development Program (2020B1111380002), Natural Science Foundation of Guangdong Province ( ) f h ﬁ l The authors gratefully thank the National Natural Science Foundation of China (22038004, 22078116, 21706082), Guangdong Provincial Key Research and Development Program (2020B1111380002), Natural Science Foundation of Guangdong Province (2019A1515012154, 2018B030311052) for the ﬁnancial supports. 11. Song, L. et al. Ultra-strong long-chain polyamide elastomers with programmable supramolecular interactions and oriented crystalline microstructures. Nat. Commun. 10, 1–8 (2019). 12. Matsuda, T., Kawakami, R., Namba, R., Nakajima, T. & Gong, J. P. Mechanoresponsive self-growing hydrogels inspired by muscle training. Science 363, 504–548 (2019). References 1. Vatankhah-Varnosfaderani, M. et al. Chameleon-like elastomers with molecularly encoded strain-adaptive stiffening and coloration. Science 359, 1509–1513 (2018). polymer networks. Macromolecules 51, 8956–8963 (2018). 32. Gao, Y., Liu, W. & Zhu, S. Thermoplastic polyoleﬁn elastomer blends for multiple and reversible shape memory polymers. Ind. Eng. Chem. Res. 58, 19495–19502 (2019). 2. Sheiko, S. S. & Dobrynin, A. V. Architectural code for rubber elasticity: from supersoft to superﬁrm materials. Macromolecules 52, 7531–7546 (2019). 33. Fan, L. F., Huang, Y. N., Rong, M. Z., Zhang, M. Q. & Chen, X. Imparting external stress-free two-way shape memory effect to commodity polyoleﬁns by manipulation of their hierarchical structures. ACS Macro Lett. 8, 1141–1146 (2019). 3. Vatankhah-Varnosfaderani, M. et al. Mimicking biological stress–strain behaviour with synthetic elastomers. Nature 549, 497–501 (2017). 4. Schoenfeld, B. J. The mechanisms of muscle hypertrophy and their application to resistance training. J. Strength. Cond. Res. 24, 2857–2872 (2010). 34. Kim, H., Boothby, M. J., Ramachandran, S., Lee, D. C. & Ware, H. T. Tough, shape-changing materials: crystallized liquid crystal elastomers. Macromolecules 50, 4267–4275 (2017). 5. Nagya, J. I., Pereda, A. E. & Rashc, J. E. On the occurrence and enigmatic functions of mixed (chemical plus electrical) synapses in the mammalian CNS. Neurosci. Lett. 695, 53–64 (2019). 35. Jiang, H., Li, C. & Huang, X. Actuators based on liquid crystalline elastomer materials. Nanoscale 5, 5225–5240 (2013). 6. Mirvakili, S. M. & Hunter, I. W. Artifcial muscles: mechanisms, applications, and challenges. Adv. Mater. 30, 1704407 (2018). 36. Saed, M. O. et al. Thiol-acrylate main-chain liquid-crystalline elastomers with tunable thermomechanical properties and actuation strain. J. Polym. Sci., Part B 55, 157–168 (2017). g 7. Kanik, M. et al. Strain-programmable ﬁber-based artiﬁcial muscle. Science 365, 145–150 (2019). 8. Lu, H. F., Wang, M., Chen, X. M., Lin, B. P. & Yang, H. Interpenetrating liquid-crystal polyurethane/polyacrylate elastomer with ultrastrong mechanical property. J. Am. Chem. Soc. 141, 14364–14369 (2019). 37. Tu, Z. et al. Biomimetic high performance artiﬁcial muscle built on sacriﬁcial coordination network and mechanical training process. Materialscloud https://doi.org/10.24435/materialscloud:9a-7y (2021). 9. Acome, E. et al. Hydraulically ampliﬁed self-healing electrostatic actuators with muscle-like performance. Science 359, 61–65 (2018). 10. Lin, S., Liu, J., Liu, X. & Zhao, X. Muscle-like fatigue-resistant hydrogels by mechanical training. Proc. Natl Acad. Sci. USA 116, 10244–10249 (2019). ARTICLE ARTICLE p g The authors declare no competing interests. 16. Sun, J. Y. et al. Highly stretchable and tough hydrogels. Nature 489, 133–136 (2012). 17. Yanagisawa, Y., Nan, Y., Okuro, K. & Aida, T. Mechanically robust, readily repairable polymers via tailored noncovalent cross-linking. Science 359, 72–76 (2018). Reprints and permission information is available at http://www.nature.com/reprints g p ( ) 21. Huang, J., Liu, W. & Qiu, X. High performance thermoplastic elastomers with biomass lignin as plastic phase. ACS Sustain. Chem. Eng. 7, 6550–6560 (2019). g p 21. Huang, J., Liu, W. & Qiu, X. High performance thermoplastic elastomers with biomass lignin as plastic phase. ACS Sustain. Chem. Eng. 7, 6550–6560 (2019). 22. Filippidi, E. et al. Toughening elastomers using mussel-inspired iron-catechol complexes. Science 358, 502–505 (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. g p p g 22. Filippidi, E. et al. Toughening elastomers using mussel-inspired iron-catechol complexes. Science 358, 502–505 (2017). 23. Chen, Y., Xu, C., Cao, L., Wang, Y. & Cao, X. PP/EPDM-based dynamically vulcanized thermoplastic oleﬁn with zinc dimethacrylate: preparation, rheology, morphology, crystallization and mechanical properties. Polym. Test. 31, 728–736 (2012). 23. Chen, Y., Xu, C., Cao, L., Wang, Y. & Cao, X. PP/EPDM-based dynamically vulcanized thermoplastic oleﬁn with zinc dimethacrylate: preparation, rheology morphology crystallization and mechanical properties Polym Test 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/. rheology, morphology, crystallization and mechanical properties. Polym. Test. 31, 728–736 (2012). 24. Porchia, M., Pellei, M., Bello, F. D. & Santini, C. Zinc complexes with nitrogen donor ligands as anticancer agents. Molecules 25, 5814 (2020). 25. Boeriu, C. G., Bravo, D., Gosselink, R. J. & Van Dam, J. E. Characterisation of structure-dependent functional properties of lignin with infrared spectroscopy. Ind. Crops Prod. 20, 205–218 (2004). 26. Liu, W. et al. Structure analysis of ethylene/1-octene copolymers synthesized from living coordination polymerization. Eur. Polym. J. Received: 29 October 2020; Accepted: 29 March 2021; Received: 29 October 2020; Accepted: 29 March 2021; 29. Zhai, Q., Qi, H. J. & Xie, T. Recent progress in shape memory polymer: new behavior, enabling materials, and mechanistic understanding. Prog. Polym. Sci. 49, 79–120 (2015). 30. Gao, Y., Liu, W. & Zhu, S. Polyoleﬁn thermoplastics for multiple shape and reversible shape memory. ACS Appl. Mater. Interfaces 9, 4882–4889 (2017). 31. Gao, Y., Liu, W. & Zhu, S. Reversible shape memory polymer from i t lli l ( th l i l t t ) ith d i l 31. Gao, Y., Liu, W. & Zhu, S. Reversible shape memory polymer from semicrystalline poly (ethylene-co-vinyl acetate) with dynamic covalent polymer networks. Macromolecules 51, 8956–8963 (2018). Competing interests p g The authors declare no competing interests. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-021-23204-x Received: 29 October 2020; Accepted: 29 March 2021; Code availability All the code that support the ﬁndings of this study are available from the corresponding author on reasonable request. 10 NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications Additional information Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41467-021-23204-x. 18. Wu, S., Qiu, M., Tang, Z., Liu, J. & Guo, B. Carbon nanodots as high- functionality cross-linkers for bioinspired engineering of multiple sacriﬁcial units toward strong yet tough elastomers. Macromolecules 50, 3244–3253 (2017). Correspondence and requests for materials should be addressed to W.L. or X.Q. Correspondence and requests for materials should be addressed to W.L. or X.Q. Peer review information Nature Communications thanks Mitchell Anthamatten and Ameya Narkar for their contribution to the peer review of this work. Peer reviewer reports are available. Peer review information Nature Communications thanks Mitchell Anthamatten and Ameya Narkar for their contribution to the peer review of this work. Peer reviewer reports are available. 19. Song, P. & Wang, H. High-performance polymeric materials through hydrogen-bond cross-linking. Adv. Mater. 32, 1901244 (2019). y g g , ( ) 20. Zhang, X., Liu, W., Yang, D. & Qiu, X. Biomimetic supertough and strong biodegradable polymeric materials with improved thermal properties and excellent UV-blocking performance. Adv. Funct. Mater. 29, 1806912 (2019). 21. Huang, J., Liu, W. & Qiu, X. High performance thermoplastic elastomers with biomass lignin as plastic phase. ACS Sustain. Chem. Eng. 7, 6550–6560 (2019). 22. Filippidi, E. et al. Toughening elastomers using mussel-inspired iron-catechol complexes. Science 358, 502–505 (2017). 20. Zhang, X., Liu, W., Yang, D. & Qiu, X. Biomimetic supertough and strong biodegradable polymeric materials with improved thermal properties and excellent UV-blocking performance. Adv. Funct. Mater. 29, 1806912 (2019). Reprints and permission information is available at http://www.nature.com/reprints Reprints and permission information is available at http://www.nature.com/reprints Reprints and permission information is available at http://www.nature.com/reprints 54, 160–171 (2014). 27. Tong, Z., Zhou, B., Huang, J., Xu, J. T. & Fan, Z. Q. Regulation of crystallization kinetics, morphology, and mechanical properties of oleﬁnic blocky copolymers. Macromolecules 47, 333–346 (2014). y p y 28. Meng, Y., Jiang, J. & Anthamatten, M. Shape actuation via internal stress-induced crystallization of dual-cure networks. ACS Macro Lett. 4, 115–118 (2015). © The Author(s) 2021 11 NATURE COMMUNICATIONS \| (2021) 12:2916 \| https://doi.org/10.1038/s41467-021-23204-x \| www.nature.com/naturecommunications
https://openalex.org/W4285798853	https://www.researchsquare.com/article/rs-1808975/latest.pdf	English	null	Aetiology of Hospitalized Fever and Risk of Death at Two Rural Tertiary Care Hospitals in Uganda from August 2019 to August 2020	Research Square (Research Square)	2,022	cc-by	6,255	Aetiology of Hospitalized Fever and Risk of Death at Two Rural Tertiary Care Hospitals in Uganda from August 2019 to August 2020 Kenneth Kobba Infectious Diseases Institute Francis Kakooza Infectious Diseases Institute Matthew L. Robinson John Hopkins University School of Medicine Emmanuel Candia Infectious Diseases Institute Jonathan Mayito Infectious Diseases Institute Edgar C. Ndawula Infectious Diseases Institute Abraham Kandathil John Hopkins University School of Medicine Abraham Kandathil John Hopkins University School of Medicine Research Article d under a Creative Commons Attribution 4.0 International License. Read Full License License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/13 Abstract Background: Epidemiology of febrile illness in Uganda is shifting due to increased HIV treatment access, emerging viruses, and increased surveillance. We investigated the aetiology and outcomes of acute febrile illness in adults presenting to hospital using a standardized testing algorithm of available assays in two tertiary care hospitals in Uganda. Methods: We recruited adults with a ≥ 38.0°C temperature or history of fever within 48 hours of presentation. Medical history, demographics, and vital signs were recorded. Testing performed included a complete blood count, renal and liver function, malaria smears, blood culture, and human immunodeficiency virus (HIV). When HIV positive; cryptococcal antigen, CD4 count, and urine lateral flow lipoarabinomannan assay for tuberculosis. Participants were followed during hospitalization and at a 1-month visit. A Cox proportional hazard regression was performed to evaluate for baseline clinical features and risk of death. Results: Of 132 participants, the median age was 33.5 years (IQR 24 to 46) and 58.3% (n=77) were female. Overall, 73 (55.3%) of 132 had a positive microbiologic result. Among those living with HIV, 31 (68.9%) of 45 had at least one positive assay; 16 (35.6%) had malaria, 14 (31.1%) tuberculosis, and 4 (8.9%) cryptococcal antigenemia. The majority (65.9%) were HIV-negative; 42 (48.3%) of 87 had at least one diagnostic assay positive; 24 (27.6%) had positive malaria smears and 1 was Xpert MTB/RIF Ultra positive. Overall, 16 (12.1%) of 132 died; 9 (56.3%) of 16 were HIV-negative, 6 died after discharge. High respiratory rate (≥22 breaths per minute) (hazard ratio [HR] 8.05; 95% CI: 1.81 to 35.69) and low (i.e., <92%) oxygen saturation (HR 4.33; 95% CI: 1.38 to 13.61) were identified to be associated with increased risk of death. Conclusion: In those with hospitalized fever, malaria and tuberculosis were common causes of febrile illness, but most deaths were non-malarial, and most HIV-negative participants did not have a positive diagnostic result. Those with respiratory failure had a high risk of death. Methods From August 2019 to August 2020, we recruited adults 18 years of age or older presenting to Mubende and Arua Regional Referral Hospitals (RRHs) emergency or outpatient departments designated to be hospitalized with a measured or reported temperature ≥ 38.0°C occurring within the past 48 hours or a clinical history consistent with fever within 48 hours of presentation. There were 255 participants consecutively screened with 97 enrolled at Mubende RRH and 35 enrolled at Arua RRH. Participants with COVID-19 diagnoses were in separate designated areas, required scarce personal protective equipment, and were not enrolled. Exclusion criteria included being hospitalized for ≥ 72 hours, receipt of antibiotics, previous participation in the study, being a prisoner, and being a psychiatric patient. The two tertiary care hospitals were chosen as study sites because both sites are located in regions with different ecology and serve different populations. Mubende RRH (bed capacity 173) is located in central Uganda and serves a district population of over 688,000 people.[15] Mubende is situated on a major road between Democratic Republic of the Congo and the capital of Uganda, Kampala. Arua RRH (bed capacity 323) is located in the more arid northwest Uganda (proximal to borders with Democratic Republic of the Congo and with South Sudan), serves over 785,000 people in the districts, and receives referrals from neighbouring countries. [16] After informed consent, participants provided information about sociodemographic and comorbid medical conditions at enrolment. History suggestive of exposure to zoonotic infections was obtained including area of residence, occupation, environmental exposures, and healthcare exposure. [17] Also, recent antibiotic use was documented, relying on patient self- report or any clinical documents they possessed. Vital signs were recorded including temperature, heart rate, respiratory rate, blood pressure, and pulse oximetry; a Glasgow Coma Score (GCS) was calculated. Participants were prospectively followed until hospital discharge. Inpatient antibiotic usage, vital signs monitoring, intensive care unit admission, length of stay, timing of receipt of infectious disease test results, discharge diagnosis, and vital status at discharge were recorded. Upon discharge, participants returned after 3–8 weeks and were interviewed on duration of illness and antimicrobial use and serum, whole blood, and plasma stored for future assessment of seroconversion. If participants were unable to follow-up in- person, they were contacted via telephone to determine vital status. Introduction Systemic causes of fever are notoriously challenging to differentiate clinically in sub-Saharan Africa (SSA), ranging from malaria to Ebola virus disease. Clinicians often rely on local epidemiology to guide empiric treatment decisions. However, during the past decade, the epidemiology of hospitalized, acute febrile illness (AFI) may have may have shifted in Uganda due to improved access to HIV treatment and increased detection of zoonotic infections. Additionally, disease recognition is largely limited to what well performing diagnostics are available. A few decades ago during the early PEPFAR (U.S. President's Emergency Plan for AIDS Relief) era in SSA, HIV coinfections were responsible for the majority of hospitalizations with severe febrile illness and associated with high mortality.[1–3] The introduction of antiretroviral therapy (ART) and the World Health Organization (WHO) recommendation to ‘treat all’ regardless of CD4 T-cell count has reduced mortality.[4–6] However, late presentation remains a tenacious problem with significant opportunistic infection related early mortality after the initiation of ART.[7] Currently, national guidelines recommend universal HIV testing and ART ideally before the development of AIDS. Treating pre-emptively with ART[8], screening for cryptococcal antigen positivity, and urine lipoarabinomannan (LAM) testing in all hospitalized adults with HIV have been shown to decrease early mortality.[9] Since this shift to improved testing and treatment, the effect on the epidemiology of acute febrile illness among hospitalized adults in Uganda has yet to be fully characterized. Due to high biodiversity and population density, many emerging and re-emerging zoonotic infections have been identified in Uganda particularly near the Rift Valley, further expanding the differential diagnosis of undifferentiated fever. With increasing surveillance, more cases of viral hemorrhagic fevers (VHF), including autochthonous and imported Ebola virus disease outbreaks, Crimean-Congo hemorrhagic fever (CCHF) and Rift Valley fever (RVF), have been detected in Uganda [10–14]. Therefore, our acute febrile illness cohort was developed to characterize most common causes of acute febrile illness in Uganda and to determine the frequency of febrile illness due to unknown etiologies. We describe herein the Page 2/13 Page 2/13 Page 2/13 characteristics of hospitalized patients with AFI at two tertiary care centers in Uganda and the diagnostic yield of available testing. characteristics of hospitalized patients with AFI at two tertiary care centers in Uganda and the diagnostic yield of available testing. Methods Laboratory tests that were conducted included complete blood count, serum creatinine, serum alanine transferase (ALT), serum aspartate transferase (AST), blood cultures (one aerobic bottle, Bactec 9050, BD, NJ, United States), hepatitis A IgM (Vaxpert, Inc., Deira Dubai, UAE), hepatitis B sAg (SD Bioline, Abbott diagnostics, Gyeonggi-do, Republic of Korea), and HIV testing with consent (Determine HIV1/2: Abbott, OK, USA; Stat-Pak, Chembio Diagnostics Systems, Inc, NY, USA; SD Bioline: Gyeonggi-do, Republic of Korea), malaria rapid test (CareStart malaria Pf HRP2 antigen RDT), and thick malaria smears. A malaria smear result was stratified by any parasitaemia or ≥ 3000 parasites/microliter [18, 19]. If there was clinical suspicion, testing for tuberculosis was performed using PCR sputum testing (Xpert MTB/RIF Ultra, Cepheid, Sunnyvale, CA, USA). If HIV testing was positive, serum cryptococcal antigen, urine lipoarabinomannan (LAM; Alere Waltham, MA, United States), and plasma CD4 count (BD Facs Calibur, BD, NJ, United States) were measured. HIV testing and counselling were performed per Uganda’s national guidelines. We estimated that a target sample size of 365 patients would estimate the frequency of uncommon endemic viral infections with a 5% prevalence using a precision of 0.025 at 95% confidence. Here we report initial findings from an accrued sample size of 132 participants enrolled to date. Summary statistics were performed for baseline demographics and prevalence of difference diagnoses. Baseline demographics, symptom prevalence, physiologic parameters, clinical laboratory results, quick sequential organ failure assessment (qSOFA; range, 0 [best] to 3 [worst] points) scores [20], and subsequent diagnoses made by the research physicians on-site were summarized and stratified by HIV diagnosis and the presence of a laboratory- confirmed acute infectious disease diagnosis. We stratified these descriptive statistics between those living with HIV or were Page 3/13 newly diagnosed with HIV and those that were HIV-negative. Kaplan-Meier curve of survival stratified by HIV status. The logrank test was also performed to evaluate for a survival different with HIV. Individuals lost to follow-up after discharge were censored at hospital discharge. After checking the proportional hazards assumption, univariate Cox regression was performed to evaluate for a survival difference associated with baseline demographics, HIV status (including both new and known), physiologic parameters, and laboratory parameters. The qSOFA (quick Sepsis Related Organ Failure Assessment) score dichotomized at a score of 2 and components of the qSOFA score were also independently fit. Methods Dichotomized parameters were used at score cut offs (i.e., respiratory rate > 21 breaths per minute, Glasgow coma scale < 15, and systolic blood pressure 100 mmHg). P-values were not corrected for multiple comparisons due to a limited sample size. Participant characteristics Among the participants (35 at Arua RRH and 97 at Mubende RRH), the median age was 33.5 years (interquartile range [IQR]: 24.0 to 46.0); 77 (58.3%) of 132 were female (Table 1). Participants presented from 10 districts in Uganda. Participants were enrolled at a median 4 days (IQR: 3 to 7 days) after onset of symptoms which did not differ significantly by HIV status (Table 2). The majority of those lost to follow-up (3 from Arua RRH and 11 from Mubende RRH), were male (71.4%; 11 of 14 participants) and the median age was 31.5 years (IQR: 27 to 50 years). 27 (20.5%) of 132 participants had a qSOFA severity score of 2 or greater. The most common symptoms included headache (76.0%), and anorexia (58.9%), and nausea or vomiting (58.9%) (Supplementary Table S1). The majority (68.2%) of patients had no known prior medical history. The median white blood count was 4.8 x 103 cells /uL with a median 129.0 x 103 platelets/uL (Supplementary Table S2). Page 4/13 Table 1 h l Table 1 Baseline demographic and physiologic vital sign characteristics Total (N = 132) Arua RRH (N = 35) Mubende RRH (N = 97) HIV-negative (N = 87) HIV-positive (N = 45) Female sex — no. (%) 77 (58.3) 18 (51.4) 59 (60.8) 50 (57.5) 27 (60.0) Age— years, median (IQR) 33.5 (24.0, 46.0) 31.0 (24.0, 47.5) 35.0 (24.0, 45.0) 30.0 (22.0, 44.5) 38.0 (30.0, 48.0) COPD or asthma — no. (%) 1 (0.8) 0 (0.0) 1 (1.0) 1 (1.1) 0 (0.0) Diabetes mellitus — no. (%) 5 (3.8) 2 (5.7) 3 (3.1) 3 (3.4) 2 (4.4) Sickle cell disease/thalassemia— no. Participant characteristics (%) 3 (2.3) 2 (5.7) 1 (1.0) 3 (3.4) 0 (0.0) Duration of symptoms — days, median (IQR) 4.0 (3.0, 7.0) 4.0 (3.0, 7.0) 4.0 (3.0, 7.0) 4.0 (3.0, 7.0) 5.0 (3.0, 7.0) Heart rate — beats per minute, median (IQR) 109.0 (92.8, 119.3) 103.0 (90.5, 113.5) 110.0 (94.0, 122.0) 107.0 (93.0, 119.5) 111.0 (93.0, 118.0) Temperature — degrees Centigrade, median (IQR) 38.3 (38.0, 38.8) 38.300 (38.0, 38.8) 38.2 (38.0, 38.8) 38.3 (38.0, 38.8) 38.1 (38.0, 38.7) Systolic blood pressure —mmHg, median (IQR) 110.5 (98.0, 122.0) 111.0 (95.0, 119.0) 111.0 (100.0, 122.0) 110.0 (100.0, 119.0) 114.0 (98.0, 127.0) Diastolic blood pressure —mmHg, median (IQR) 68.0 (60.0, 78.0) 68.0 (60.5, 74.8) 68.0 (60.0, 78.0) 68.0 (59.0, 75.0) 69.0 (63.0, 80.0) Respiratory rate — breaths per minute, median (IQR) 20.0 (18.0, 24.0) 20.0 (18.0, 26.0) 20.0 (18.0, 24.0) 22.0 (18.0, 25.5) 20.0 (18.0, 24.0) Oxygen saturation — %, median (IQR) 98.0 (96.0, 99.0) 98.0 (96.0, 98.5) 98.0 (96.0, 99.0) 98.0 (96.0, 98.0) 98.0 (96.0, 99.0) Glasgow coma scale — median (IQR) 15.0 (15.0, 15.0) 15.0 (15.0, 15.0) 15.0 (15.0, 15.0) 15.0 (15.0, 15.0) 15.0 (15.0, 15.0) qSOFA score — median (IQR) 1.0 (0.0, 1.0) 1.0 (0.0, 1.0) 1.0 (0.0, 1.0) 1.0 (0.0, 1.0) 1.0 (0.0, 1.0) qSOFA score ≥ 2 — no. (%) 27 (20.5) 8 (22.9) 19 (19.6) 19 (21.8) 8 (17.8) RRH: Regional Referral Hospital; IQR: interquartile range; qSOFA: quick sequential organ failure assessment. Baseline demographic and physiologic vital sign characteristics Overall, 73 (55.3%) of 132 had a positive microbiologic result. The most common positive test results were a malaria thick smear (40/132; 30.3% participants > 0 parasite/high powered field) or malaria rapid diagnostic test (39/132; 29.6% participants). Thirty-five of 40 (87.5%) positive malaria smears were also positive on the malaria RDT (Fig. 1). The second most common diagnosis was tuberculosis from a positive urine LAM (14/45; 31.1% among those with HIV, 10.6% overall cohort) or sputum GeneXpert (2/14; 14.3% among tested, 1.5% overall cohort). Bacteraemia, excluding common contaminants, was identified in 9 (6.8%) of participants. Characteristics of HIV-infected participants Characteristics of HIV-infected participants Overall, 45 (34.1%) were HIV-infected; median CD4 count was 198.0 (IQR: 72.0, 347.0) in the setting of acute illness. There were 8 (17.8%) of 45 participants with a CD4 < 50. Sixteen (35.6%) of 45 were not known to have HIV; 10 were started on ART during hospitalization or within a month after hospitalization. The remaining 7 included 3 who died, 2 who had an unknown treatment history, 1 who was resistant to starting, and 1 who eloped from the hospital and was lost to follow-up. Seventeen (60.0%) of 29 known to have HIV were on ART prior to hospital presentation. Individuals with HIV were older (median 38.0 Page 5/13 Page 5/13 years; IQR: 30.0, 48.0) compared to those without HIV (median 30.0; IQR: 22, 44.5) p = 0.005, otherwise, they had similar baseline demographics (Table 1, Supplementary Table S3). years; IQR: 30.0, 48.0) compared to those without HIV (median 30.0; IQR: 22, 44.5) p = 0.005, otherwise, they had similar baseline demographics (Table 1, Supplementary Table S3). Microbiologic results among patients with HIV Among those living with HIV, 31 (68.9%) of 45 had at least one positive microbiologic assay. Microbiologic and rapid diagnostic test results among those with newly diagnosed or known HIV included positive results for malaria (smear: n = 16, 35.6%; RDT: n = 15, 33.3%). There were 5 (11.1%) with cryptococcal antigenemia, 8 (9.2%) with evidence of hepatitis A and 2 (4.4%) with evidence of hepatitis B (Table 2). Bacteraemia occurred in 5 (11.1%) of 45 patients due to Streptococcus pneumoniae (N = 2), Staphylococcus aureus, Salmonella species, or Proteus vulgaris (Table 2). Fifteen (33.3%) of 45 participants with HIV also had TB; 1 out of 6 participants with sputum samples that tested positive for tuberculosis on Xpert and 14 (31.1%) had positive urine LAM results. Table 2 Diagnostic test results stratified by HIV status and outcome HIV status Outcome Diagnostic test Total (N=132) HIV-negative (N=87) HIV-positive (N=45) p - value Survived (N=102) Died (N=16) p- value Malaria antigen RDT — no. (%) 39 (29.5) 24 (27.6) 15 (33.3) 0.493 33 (32.4) 1 (6.7) 0.037 Malaria smear qualitative result — no. (%) > 0 parasites/HPF 40 (30.3) 24 (27.6) 16 (35.6) 0.43 10 (9.8) 0 (0.0) 0.35 ≥3,000 parasites/HPF 12 (9.1) 10 (11.5) 2 (4.4) 0.22 6 1 Missing values 10 6 4 Serum cryptococcal antigen — no. (%) — — 5 (11.1) — 4/31 (12.9) 1/7 (14.3) 0.922 Urine lipoarabinomannan — no. (%) — — 14 (31.1) — 9/31 (29.0) 4/9 (44.4) 0.157 Sputum GeneXpert — no./total samples (% of cohort) 2/14 (14.3) 1/8 (12.5) 1/6 (16.7) 0.825 2/9 (22.2) 0/3 (0.0) 0.371 Serum hepatitis A IgM — no. (%) 9 (6.9) 1 (3.3) 8 (9.2) 0.139 9 (8.8) 0 (0.0) 0.216 Missing values 1 0 1 Serum hepatitis B sAg — no. (%) 7 (5.3) 5 (5.7) 2 (4.4) 0.752 5 (4.9) 1 (6.2) 0.819 Blood culture positivity 9 (6.8) 4 (4.6) 5 (11.1) 0.273 5 (4.9)*** 2 (12.5) 0.241 Table 2 Diagnostic test results stratified by HIV status and outcome RDT: rapid diagnostic test; HPF: high powered fever; sAg: surface antigen. Fatal cases At one-month follow-up, 16 (13.6%) of 118 died; 10 in hospital and 6 after discharge or hospital transfer. Death occurred a median of 3.0 days (IQR: 1 to 7 days) from enrolment. The median age among those that died was 38.0 years (IQR: 29 to 42.5, range: 18 to 80), and 7 (37.5%) had HIV (Supplementary Table S4). Among those who died, the most common presenting symptom was cough (n = 9, 56.2%). At initial presentation, those that died were more tachypnoeic (median 24.5 breaths per minute, IQR: 22 to 31 breaths per minute) compared to survivors (median 20 breaths per minute; IQR: 18 to 24 breaths per minute). Fatal cases also had lower initial oxygen saturation (median 95.5%; IQR: 92.3 to 98% compared to median 98.0%; IQR: 96.0 to 99.0%) (Supplementary Table S4). The initial qSOFA score was higher among fatal cases (median 1.5; IQR: 1.0 to 2.0) compared to non-fatal cases (median 1.0; IQR: 0.0 to 1.0). Half of fatal cases had an initial qSOFA score ≥ 2 compared to 16.7% of non-fatal cases. Among fatal cases, the initial mean white blood count was 4.4 x103 per microliter (IQR: 3.0 to 7.7 x103 per microliter) and platelet count was 107.0 x103 per microliter (IQR: 71.8 to 155.0 x103 per microliter). The median ALT and AST were 26.0 U/L (IQR: 22.25 to 48.50) and 80.0 U/L (IQR: 34.75 to 215.75), respectively, which were higher (ALT p = 0.031; AST p = 0.001) among fatal cases than non-fatal cases (Supplementary Table S5). The majority (N = 14; 87.5%) of fatal cases had blood cultures without growth. All fatal cases had a negative malaria RDT (and one positive smear) compared to 32.4% of non- fatal cases had a positive malaria smear and 34.3% of non-fatal cases that had a positive malaria RDT (Table 2). There were 4 fatal cases (3 post-discharge) that had HIV and a positive urine LAM, highly suggestive of a diagnosis of tuberculosis. All LAM-positive cases received anti-tuberculosis treatment and one was treated for concomitant cryptococcal meningitis. Using Cox proportional hazards regression, risk of death was not found to be increased by age, sex, or HIV diagnosis (Fig. 2). HIV was not associated with risk of death (HR: 1.70; 95% CI: 0.62 to 4.69; logrank p = 0.32). Fatal cases High respiratory rate (≥ 22 breaths per minute) (hazard ratio [HR] 8.05; 95% CI: 1.81 to 35.69) and low (i.e., < 92%) oxygen saturation (HR 4.33; 95% CI: 1.38 to 13.61) were identified to be associated with increased risk of death, but not heart rate (Fig. 3). Low (< 15) Glasgow Coma Score was associated with an increased risk of death (HR 3.05) but lacked statistical significance (95% CI: 0.86 to 10.80). A high qSOFA (≥ 2 points) was associated with the largest increased risk of parameters evaluated (HR: 5.23; 95%CI: 1.89 to 14.45). Clinical laboratory parameters were not associated with increased risk of death except for AST (HR: 1.03 per 100 U/L; 95% CI: 1.00 to 1.06) and a trend towards significance was observed with ALT (HR: 1.15 per 100 U/L; 95% CI: 0.98 to 1.35). Microbiologic results among patients without HIV Among those without HIV, 42 (48.3%) of 87 had a positive assay. Malaria was common 24 (27.6%) of 87 were smear positive. Rapid diagnostic test results included malaria (RDT: 27.6%), hepatitis A (3.3%), and hepatitis B (5.7%) (Table 2). Clinically relevant blood culture growth occurred in 4 (4.6%) of patients with Streptococcus pneumoniae being most common isolate (N = 2), and an additional isolate each (N = 1) of Citrobacter freundii, and Streptococcus spp. One culture Page 6/13 grew common contaminant coagulase negative staphylococcus, and another grew Micrococcus luteus. There were 1 out of 8 sputum samples that tested positive for tuberculosis on PCR testing. Discussion In this prospective multi-site AFI cohort in rural Uganda, death occurred mostly among young adults with a low prevalence of baseline comorbid illness. Malaria and tuberculosis were common causes of febrile illness, but malaria was not a common cause of death. Among participants that died, clinical and laboratory features suggestive of systemic infections were observed including respiratory distress, elevated liver transaminases; a microbiologic diagnosis was not attained for most cases. Among deaths, over 40% occurred after hospital discharge and tuberculosis was the most common laboratory diagnosis. Our findings emphasize the need for more in-depth assessment of causes and prevention of death due to non- malarial causes of fever among hospitalized adults in Uganda. While HIV infection remains common, the majority of febrile hospitalized adult participants did not have HIV. Our findings are consistent with recent studies suggesting that a lower proportion of febrile patients is due to HIV during the universal antiretroviral therapy era. The prevalence of HIV in this cohort was similar to that of hospitalized patients with febrile illness in the East African region in Tanzania ranging from 32% [21] to 39% in 2007 to 2008 [22]. However, HIV prevalence was 85% Page 7/13 Page 7/13 Page 7/13 among septic patients in Uganda in a 2008 to 2009 cohort [23]. A more recent sepsis cohort from 2017 to 2019 had less but still a majority (55%) of participants had HIV[24]. Similarly, the majority (67%) of participants had HIV in a sepsis cohort in Malawi from 2018 to 2019 [25]. HIV prevalence in our cohort may be lower than previous cohorts in Uganda due to different level of severity in rural settings but could also represent decreased trends after the initiation of a `treat all’ approach. While larger studies and weighting with population-level demographics would be needed for incidence conclusions, this suggests that opportunistic infections may no longer represent the most common causes of febrile illness requiring hospitalization in Uganda. Importantly, however, tuberculosis was identified as a common cause of illness in this cohort and remains a major cause of severe infectious illness regardless of HIV status. Overall, 16 new diagnoses of HIV were identified in this prospective cohort with universal HIV testing, emphasizing the importance of continuing to test all hospitalized febrile patients for HIV. These findings emphasize that HIV testing is high yield for hospitalized adult patients with febrile illness and unknown HIV serostatus in Uganda. Conclusions The majority of deaths occurred among those less than 50 years of age and due to non-malarial febrile illness. Rickettsial and viral aetiologies of severe febrile illness may represent an undetected and pervasive burden of disease[27, 28]. Urgent surveillance efforts and field-ready clinical diagnostics are needed to identify treatable causes of mortality. Discussion Additionally, this provides an opportunity to leverage existing resources for linkage to HIV care per national standards. The deaths that occurred among people with HIV after discharge additionally highlights the need for continuity of care resources particularly among severely ill patients with HIV. Respiratory dysfunction noted by respiratory rate and oxygen saturation was identified as a major risk factor for death in this cohort. While respiratory rate is a known prognosticators as part of the qSOFA score [26], lack of access to respiratory support at these clinical sites may have been a factor leading to fatal outcomes. Similar to other referral hospitals in SSA, access to invasive and non-invasive respiratory support, including highly skilled nursing, high flow oxygen, and ventilatory equipment was not readily available at these hospitals. Improved access to respiratory support may decrease hospitalized deaths due to infection. Our study has notable limitations. First, participants were lost to follow-up and causes of death after discharge were unknown. Participant attrition was in part due to hesitation to return to medical centres during the COVID-19 pandemic. Causes of death were not ascertainable with telephone follow-up. Second, the sample size is limited and associations with mortality that exist may not have been observed in our data. For example, there was no survival difference observed with HIV, but we present here our interim analysis observations and smaller effect sizes may be observed in a larger cohort. Third, diagnostic test results reported here did not have extensive diagnostic testing with serology or multiplex PCR. However, testing was protocolized and in line with recommendations for rapid diagnostic testing and diagnostic assays currently available in Uganda. More expansive diagnostic testing is ongoing to understand the aetiologies of the unknown causes of febrile illness and death including emerging or neglected causes of illness including viral and rickettsial infections. Taken together, these findings highlight the need to strengthen clinical and microbiology laboratory testing capacity to identify the aetiologies of acute febrile illness, particularly in settings where the epidemiology of febrile illness is shifting. These results are mostly from the pre-COVID era and aetiologies of fever may have directly or indirectly changed. Evaluation of optimization of follow-up, treatments, or diagnoses are urgent areas of study to improve outcomes among hospitalized adults in SSA. Authors' information (optional) Not applicable. Declarations Ethics approval and consent to participate This study was conducted in compliance with the Declaration of Helsinki and Good Clinical Practice Guidelines. The study and informed consent process were reviewed and approved by the Joint Clinical Research Centre (JCRC) Research Ethics Committee (JC1518) and the Uganda National Council for Science and Technology (UNCST), HS 371ES, and Johns Hopkins University School of Medicine Internal Review Board (IRB00176961). All participants signed written informed consents prior to study procedures. This study was conducted in compliance with the Declaration of Helsinki and Good Clinical Practice Guidelines. The study and informed consent process were reviewed and approved by the Joint Clinical Research Centre (JCRC) Research Ethics Committee (JC1518) and the Uganda National Council for Science and Technology (UNCST), HS 371ES, and Johns Hopkins University School of Medicine Internal Review Board (IRB00176961). All participants signed written informed consents prior to study procedures. Authors' contributions JM participated in study implementation and drafting the first draft of the manuscript, KK & CE were involved in study implementation and review of manuscript. FK, PWB, RM, KA, MA, AG, MCY and LM participated in study design, sourcing for funding, study implementation and review of the manuscript. Acknowledgements Not applicable. Availability of data and materials The data that support the study findings may be available upon reasonable request to the corresponding author. Consent for publication Not applicable. Funding The study was funded by the CAPA-CT II project which is part of the EDCTP2 programme supported by the European Union. Protocol development was supported by the National Institutes of Health T32 AI007291-27 to PWB. Competing interests All authors declare no conflict of interest 7. Tenforde MW, Walker AS, Gibb DM, Manabe YC: Rapid antiretroviral therapy initiation in low- and middle-income countries: A resource-based approach. PLoS medicine 2019, 16(1):e1002723. 7. Tenforde MW, Walker AS, Gibb DM, Manabe YC: Rapid antiretroviral therapy initiation in low- and middle-income countries: A resource-based approach. PLoS medicine 2019, 16(1):e1002723. 8. Mfinanga S, Chanda D, Kivuyo SL, Guinness L, Bottomley C, Simms V, Chijoka C, Masasi A, Kimaro G, Ngowi B et al: Cryptococcal meningitis screening and community-based early adherence support in people with advanced HIV infection starting antiretroviral therapy in Tanzania and Zambia: an open-label, randomised controlled trial. Lancet 2015. 8. Mfinanga S, Chanda D, Kivuyo SL, Guinness L, Bottomley C, Simms V, Chijoka C, Masasi A, Kimaro G, Ngowi B et al: Cryptococcal meningitis screening and community-based early adherence support in people with advanced HIV infection starting antiretroviral therapy in Tanzania and Zambia: an open-label, randomised controlled trial. Lancet 2015. 9. Gupta-Wright A, Corbett EL, van Oosterhout JJ, Wilson D, Grint D, Alufandika-Moyo M, Peters JA, Chiume L, Flach C, Lawn SD et al: Rapid urine-based screening for tuberculosis in HIV-positive patients admitted to hospital in Africa (STAMP): a pragmatic, multicentre, parallel-group, double-blind, randomised controlled trial. Lancet 2018, 392(10144):292-301. 10. CRIMEAN-CONGO HEMORRHAGIC FEVER - EAST AFRICA: UGANDA (LUWERO) [http://www.promedmail.org/post/5541551] 11. Outbreaks Chronology: Ebola Virus Disease [https://www.cdc.gov/vhf/ebola/outbreaks/history/chronology.html#modalIdString_outbreaks] 11. Outbreaks Chronology: Ebola Virus Disease [https://www.cdc.gov/vhf/ebola/outbreaks/history/chronology.html#modalIdString_outbreaks] 12. Chronology of Marburg Hemorrhagic Fever Outbreaks [https://www.cdc.gov/vhf/marburg/resources/outbreak- table.html] 12. Chronology of Marburg Hemorrhagic Fever Outbreaks [https://www.cdc.gov/vhf/marburg/resources/outbreak- table.html] 13. MacNeil A, Farnon EC, Wamala J: Proportion of Deaths and Clinical Features in Bundibugyo Ebola Virus Infection, Uganda. Emerg Infect Dis 2010 Dec, 16(12):1969-1972. 13. MacNeil A, Farnon EC, Wamala J: Proportion of Deaths and Clinical Features in Bundibugyo Ebola Virus Infection, Uganda. Emerg Infect Dis 2010 Dec, 16(12):1969-1972. 14. Shoemaker TR, Balinandi S, Tumusiime A, al e: Impact of enhanced viral haemorrhagic fever surveillance on outbreak detection and response in Uganda. Lancet Infect Dis 2018, 18(4):373-375. 14. Shoemaker TR, Balinandi S, Tumusiime A, al e: Impact of enhanced viral haemorrhagic fever surveillance on outbreak detection and response in Uganda. Lancet Infect Dis 2018, 18(4):373-375. 15. Mubende Regional Referral Hospital Profile [http://sustainuganda.org/sites/sustainuganda.org/files/Mubende%20Regional%20Referral%20Hospital%20Profile.pdf] 15. Mubende Regional Referral Hospital Profile [http://sustainuganda.org/sites/sustainuganda.org/files/Mubende%20Regional%20Referral%20Hospital%20Profile.pdf] 16. Arua Regional Referral Hospital Profile [http://sustainuganda.org/sites/sustainuganda.org/files/Arua%20Regional%20Referral%20Hospital%20Profile.pdf] 16. Arua Regional Referral Hospital Profile [http://sustainuganda.org/sites/sustainuganda.org/files/Arua%20Regional%20Referral%20Hospital%20Profile.pdf] 16. Arua Regional Referral Hospital Profile [http://sustainuganda.org/sites/sustainuganda.org/files/Arua%20Regional%20Referral%20Hospital%20Profile.pdf] g p [http://sustainuganda.org/sites/sustainuganda.org/files/Arua%20Regional%20Referral%20Hospital%20Profile.pdf] 17. de St Maurice A, Nyakarahuka L, Purpura L, Ervin E, Tumusiime A, Balinandi S, Kyondo J, Mulei S, Tusiime P, Manning C 17. References 1. Gupta A, Nadkarni G, Yang WT, Chandrasekhar A, Gupte N, Bisson GP, Hosseinipour M, Gummadi N: Early mortality in adults initiating antiretroviral therapy (ART) in low- and middle-income countries (LMIC): a systematic review and meta- analysis. PloS one 2011, 6(12):e28691. 1. Gupta A, Nadkarni G, Yang WT, Chandrasekhar A, Gupte N, Bisson GP, Hosseinipour M, Gummadi N: Early mortality in adults initiating antiretroviral therapy (ART) in low- and middle-income countries (LMIC): a systematic review and meta- analysis. PloS one 2011, 6(12):e28691. 2. Gupta RK, Lucas SB, Fielding KL, Lawn SD: Prevalence of tuberculosis in post-mortem studies of HIV-infected adults and children in resource-limited settings: a systematic review and meta-analysis. AIDS (London, England) 2015, 29(15):1987-2002. 2. Gupta RK, Lucas SB, Fielding KL, Lawn SD: Prevalence of tuberculosis in post-mortem studies of HIV-infected adults and children in resource-limited settings: a systematic review and meta-analysis. AIDS (London, England) 2015, 29(15):1987-2002. 3. Cox JA, Lukande RL, Nelson AM, Mayanja-Kizza H, Colebunders R, Van Marck E, Manabe YC: An autopsy study describing causes of death and comparing clinico-pathological findings among hospitalized patients in Kampala, Uganda. PloS one 2012, 7(3):e33685. 3. Cox JA, Lukande RL, Nelson AM, Mayanja-Kizza H, Colebunders R, Van Marck E, Manabe YC: An autopsy study describing causes of death and comparing clinico-pathological findings among hospitalized patients in Kampala, Uganda. PloS one 2012, 7(3):e33685. 3. Cox JA, Lukande RL, Nelson AM, Mayanja-Kizza H, Colebunders R, Van Marck E, Manabe YC: An autopsy study describing causes of death and comparing clinico-pathological findings among hospitalized patients in Kampala, Uganda. PloS one 2012, 7(3):e33685. 4. Group TAS, Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, Ouassa T, Ouattara E, Anzian A, Ntakpe JB et al: A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. The New England journal of medicine 2015, 373(9):808- 822. 4. Group TAS, Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, Ouassa T, Ouattara E, Anzian A, Ntakpe JB et al: A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. The New England journal of medicine 2015, 373(9):808- 822. 4. Group TAS, Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, Ouassa T, Ouattara E, Anzian A, Ntakpe JB et al: A Trial o Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. The New England journal of medicine 2015, 373(9):808 822. Page 9/13 5. References Otieno G, Whiteside YO, Achia T, Kwaro D, Zielinski-Gutierrez E, Ojoo S, Sewe M, Musingila P, Akelo V, Obor D et al: Decreased HIV-associated mortality rates during scale-up of antiretroviral therapy, 2011-2016. Aids 2019, 33(15):2423- 2430. 6. Price AJ, Glynn J, Chihana M, Kayuni N, Floyd S, Slaymaker E, Reniers G, Zaba B, McLean E, Kalobekamo F et al: Sustained 10-year gain in adult life expectancy following antiretroviral therapy roll-out in rural Malawi: July 2005 to June 2014. Int J Epidemiol 2017, 46(2):479-491. 7. Tenforde MW, Walker AS, Gibb DM, Manabe YC: Rapid antiretroviral therapy initiation in low- and middle-income countries: A resource-based approach. PLoS medicine 2019, 16(1):e1002723. de St Maurice A, Nyakarahuka L, Purpura L, Ervin E, Tumusiime A, Balinandi S, Kyondo J, Mulei S, Tusiime P, Manning C et al: Rift Valley Fever: A survey of knowledge, attitudes, and practice of slaughterhouse workers and community members in Kabale District, Uganda. PLoS Negl Trop Dis 2018, 12(3):e0006175. 17. de St Maurice A, Nyakarahuka L, Purpura L, Ervin E, Tumusiime A, Balinandi S, Kyondo J, Mulei S, Tusiime P, Manning C et al: Rift Valley Fever: A survey of knowledge, attitudes, and practice of slaughterhouse workers and community members in Kabale District, Uganda. PLoS Negl Trop Dis 2018, 12(3):e0006175. 18. Mwangi TW, Ross A, Snow RW, Marsh K: Case definitions of clinical malaria under different transmission conditions in Kilifi District, Kenya. J Infect Dis 2005, 191(11):1932-1939. 18. Mwangi TW, Ross A, Snow RW, Marsh K: Case definitions of clinical malaria under different transmission conditions in Kilifi District, Kenya. J Infect Dis 2005, 191(11):1932-1939. 19. Afrane YA, Zhou G, Githeko AK, Yan G: Clinical malaria case definition and malaria attributable fraction in the highlands of western Kenya. Malar J 2014, 13:405. 19. Afrane YA, Zhou G, Githeko AK, Yan G: Clinical malaria case definition and malaria attributable fraction in the highlands of western Kenya. Malar J 2014, 13:405. 20. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche J-D, Coopersmith CM et al: The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016, 315(8):801-810. 20. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche J-D, Coopersmith CM et al: The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016, 315(8):801-810. 21. Nadjm B, Mtove G, Amos B, Walker NF, Diefendal H, Reyburn H, Whitty CJ: Severe febrile illness in adult hospital admissions in Tanzania: a prospective study in an area of high malaria transmission. Trans R Soc Trop Med Hyg 2012, 106(11):688-695. 21. Nadjm B, Mtove G, Amos B, Walker NF, Diefendal H, Reyburn H, Whitty CJ: Severe febrile illness in adult hospital admissions in Tanzania: a prospective study in an area of high malaria transmission. Trans R Soc Trop Med Hyg 2012, 106(11):688-695. Page 10/13 22. 7. Tenforde MW, Walker AS, Gibb DM, Manabe YC: Rapid antiretroviral therapy initiation in low- and middle-income countries: A resource-based approach. PLoS medicine 2019, 16(1):e1002723. Crump JA, Morrissey AB, Nicholson WL, Massung RF, Stoddard RA, Galloway RL, Ooi EE, Maro VP, Saganda W, Kinabo GD et al: Etiology of severe non-malaria febrile illness in Northern Tanzania: a prospective cohort study. PLoS neglected tropical diseases 2013, 7(7):e2324. 23. Jacob ST, Moore CC, Banura P, Pinkerton R, Meya D, Opendi P, Reynolds SJ, Kenya-Mugisha N, Mayanja-Kizza H, Scheld WM: Severe sepsis in two Ugandan hospitals: a prospective observational study of management and outcomes in a predominantly HIV-1 infected population. PloS one 2009, 4(11):e7782. 24. Cummings MJ, Bakamutumaho B, Owor N, Kayiwa J, Namulondo J, Byaruhanga T, Muwanga M, Nsereko C, Baldwin MR, Lutwama JJ et al: Stratifying Sepsis in Uganda Using Rapid Pathogen Diagnostics and Clinical Data: A Prospective Cohort Study. Am J Trop Med Hyg 2021, 105(2):517-524. 25. Lewis JM, Mphasa M, Keyala L, Banda R, Smith EL, Duggan J, Brooks T, Catton M, Mallewa J, Katha G et al: A longitudinal observational study of aetiology and long-term outcomes of sepsis in Malawi revealing the key role of disseminated tuberculosis. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2021. 26. Sabir L, Ramlakhan S, Goodacre S: Comparison of qSOFA and Hospital Early Warning Scores for prognosis in suspected sepsis in emergency department patients: a systematic review. Emerg Med J 2021. 27. Blair PW, Lamorde M, Dumler JS: Rickettsioses and Q Fever in Tanzania: Estimating the Burden of Pervasive and Neglected Causes of Severe Febrile Illness in Sub-Saharan Africa. Am J Trop Med Hyg 2021. 28. Pisharody SR, Matthew; Carugati, Manuela; Nicholson, William; Perniciaro, Jamie; Biggs, Holly; Maze, Michael; Hertz, Julian; Allan, Kathryn; Halliday, Jo; Mmbaga, Blandina; Saganda, Wilbrod; Lwezaula, Bingileki; Kazwala, Rudovick; Cleaveland, Sarah; Maro, Venance; Crump, John: Incidence of acute Q fever and spotted fever group rickettsioses, Kilimanjaro, Tanzania, 2007-2008 and 2012-2014. American Journal of Tropical Medicine & Hygiene 2021. Figures Page 11/13 Page 11/13 Figure 3 Forest plot of univariate unadjusted hazard ratio estimates of demographic characteristics, physiologic parameters, and clinical laboratory results for risk of death. Forest plot of univariate unadjusted hazard ratio estimates of demographic characteristics, physiologic parameters, and clinical laboratory results for risk of death. Figure 2 Kaplan-Meier plot of survival over time stratified by HIV status. Kaplan-Meier plot of survival over time stratified by HIV status. Kaplan-Meier plot of survival over time stratified by HIV status. SupplementaryTables.docx Figure 1 Diagnostic UpSet plot demonstrating the frequency of participants with one or more overlapping positive microbiologic results within the cohort. HRP2: Plasmodium falciparum antigen histidine rich protein 2; sAg: surface antigen; LAM: lipoarabinomannan. Page 12/13 Page 12/13 Page 12/13 Figure 2 Kaplan-Meier plot of survival over time stratified by HIV status. Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. SupplementaryTables.docx Page 13/13 Page 13/13
https://openalex.org/W4388140966	https://www.nature.com/articles/s41541-023-00770-3.pdf	English	null	Efficacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine	npj vaccines	2,023	cc-by	5,852	BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine Hirotaka Fukami 1,2,3, Jun Morinaga 1,2,3✉, Hironori Nakagami 4, Hiroki Hayashi4, Yusuke Okadome1, Eiji Matsunaga1,2,3, Tsuyoshi Kadomatsu 1,3, Haruki Horiguchi1,3,5, Michio Sato1,3, Taichi Sugizaki1,3, Keishi Miyata1,3, Daisuke Torigoe1,6, Masashi Mukoyama2, Ryuichi Morishita7 and Yuichi Oike 1,3,5,6✉ Previously, we reported that an ANGPTL3 vaccine is a hopeful therapeutic option against dyslipidemia. In our current study, we assess durability and booster effects of that vaccine over a period representing a mouse’s lifespan. The vaccine remained effective for over one year, and booster vaccination maintained suppression of circulating triglyceride levels thereafter without major adverse effects on lungs, kidneys, or liver, suggesting vaccine efﬁcacy and safety. npj Vaccines (2023) 8:168 ; https://doi.org/10.1038/s41541-023-00770-3 Here, we assessed durability of the ANGPTL3 vaccine at one year after the ﬁrst vaccination and evaluated effects of booster immunization on anti-E3 antibody titers and lipid proﬁles at 60, 90, and 105 weeks after the ﬁrst vaccination in C57BL/6 J wild-type mice. We also analyzed safety of the E3 vaccine. Dyslipidemia is a critical risk factor for atherosclerosis, which increases risk of CVD development and its progression1,2. A proportion of patients with dyslipidemia, such as those with familial hypercholesterolemia, are not responsive to drugs commonly used to treat dyslipidemia, such as high-dose statins, ezetimibe, or PCSK9 inhibitors. For these patients, there is an urgent need to develop new medications that counfteract dyslipidemia via different mechanisms3. y y In our previous report, we intradermally vaccinated C57BL/6 J wild-type mice with E3 vaccine 3 times (once every two weeks), starting when mice were 8-weeks-old (Fig. 1b). Mice showed signiﬁcantly increased anti-E3 antibody titers in circulation and 16.6% lower circulating TG levels by 30 weeks after the ﬁrst vaccination5. In the current study, to assess extended durability of the E3 vaccine, we continued to observe vaccinated mice for a period exceeding 30 weeks. By 60 weeks after the ﬁrst vaccination, increasing of E3 antibody titers had still been sustained, while circulating TG levels in non-fasting conditions signiﬁcantly decreased in vaccinated compared to control (KLH) groups (Fig. 1c, d), suggesting that the E3 effect would persist for over a year in mice vaccinated at a young age. However, we noted that the therapeutic efﬁcacy in terms of lowering serum TG levels in non-fasting conditions appeared to weaken between 30 weeks (TG levels decreased 27 mg/dl relative to controls) to 60 weeks (TG levels decreased 15 mg/dl relative to controls) (Fig. 1Department of Molecular Genetics, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Chuo-ku, Kumamoto-shi 860-8556, Japan. 2Department of Nephrology, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Chuo-ku, Kumamoto-shi 860-8556, Japan. 3Center for Metabolic Regulation of Healthy Aging (CMHA), Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Chuo-ku, Kumamoto-shi 860-8556, Japan. 4Department of Health Development and Medicine, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita-shi 565-0871, Japan. 5Department of Aging and Geriatric Medicine, Graduate School of Medical Sciences, Kumamoto University, 1-1-1 Honjo, Chuo-ku, Kumamoto-shi 860-8556, Japan. 6Institute of Resource Development and Analysis (IRDA), Kumamoto University, 2-2-1 Honjo, Chuo-ku, Kumamoto-shi 860-0811, Japan. 7Department of Clinical Gene Therapy, Osaka University Graduate School of Medicine, 2-2, Yamadaoka, Suita-shi 565-0871, Japan. ✉email: morijun@kumamoto-u.ac.jp; oike@gpo.kumamoto-u.ac.jp www.nature.com/npjvaccines BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine 1d), as circulating TG levels decreased in parallel with aging of mice. Accordingly, we performed booster E3 vaccinations at 60 weeks after the ﬁrst vaccination (Fig. 1b) and found that booster vaccination at this time point markedly increased anti-E3 antibody titers and lowered circulating TG levels in non-fasting conditions relative to controls (Fig. 1c, d), suggesting that the booster was effective. Next, at 90 weeks after the ﬁrst vaccination, although antibody titers against E3 remained signiﬁcantly increased in vaccinated versus control groups (Fig. 1c), serum TG levels were comparable in E3 and control groups in non-fasting conditions (Fig. 1d). Moreover, booster E3 vaccination at 90 weeks sig- niﬁcantly increased antibody titers and decreased circulating TG levels (Fig. 1c, d). By 105 weeks after the ﬁrst vaccination, when y p The secreted protein ANGPTL3 (angiopoietin-like protein 3) blocks lipoprotein lipase activity, thereby worsening lipid proﬁles including those of circulating low-density lipoprotein cholesterol (LDL-C) and triglyceride (TG). As a result, ANGPTL3 has attracted much attention globally as a new therapeutic target for dyslipidemia, and in 2021, the U.S. Food and Drug Administration approved a monoclonal inhibitory antibody against ANGPTL3 (Evinacumab) as a drug for add-on treatment of adult and pediatric patients ages 12 and above with homozygous familial hypercholesterolemia (HoFH)4. Soon after, we established an anti- dyslipidemia peptide vaccine therapy targeting ANGPTL3 [Epitope 3, E3: EPKSRFAMLD (Fig. 1a)] using mouse models of obese dyslipidemia and familial hypercholesterolemia, as a potentially cost-effective therapeutic strategy against dyslipidemia and associated diseases such as fatty liver and atherosclerosis5. Our study revealed that increased anti-E3 antibody titers and decreased circulating TG levels in mice in non-fasting conditions were maintained until 30 weeks after the ﬁrst vaccination in C57BL/6 J wild-type mice, suggesting that vaccine durability was at least half a year4. However, some patients with dyslipidemia require long-term treatment with lipid-lowering drugs, often into old age. Thus, there is an urgent need to determine whether immunological memory conferred by our E3 vaccine could be prolonged beyond 6 months and continued until old age and whether booster administration could elevate antibody titers in older animals. In addition, evaluation of vaccine safety was also needed in models of aging mice. Published in partnership with the Sealy Institute for Vaccine Sciences Published in partnership with the Sealy Institute for Vaccine Sciences H. Fukami et al. BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine 2 0 6 20 30 60 62 90 92 105 0 50 100 150 200 250 weeks TG (mg/dL) _ _ _ 6 20 30 60 62 90 92 105 0 50000 100000 150000 weeks Antibody titer (OD 50%) KLH E3 EPKSRFAMLD ANGPTL3 vaccine KLH (T-cell epitope) (a) intradermal injection weeks 2 4 0 6 _ : immunization with ANGPTL3 vaccine KLH group (Control) ANGPTL3 (E3) group (b) booster shot booster shot : blood collection 98 70 age in weeks (c) (d) 8 * † † † † † * * * † † * n.s. n.s. : sacrifice † n.s. KLH E3 113 100 : immunization with control vaccine EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD (B-cell epitope) KLH E3 KLH E3 60 62 90 92 105 * Non-fasting TG (mg/dL) Fig. 1 Study protocol and efﬁcacy of the ANGPTL3 vaccine against circulating triglyceride levels in C57BL/6 J mice fed a normal diet. a Schematic of vaccine design and vaccination methods. b Schematic showing vaccine injection protocol. c Circulating antibody titers in mice at 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization. Values are reported as the serum dilution giving half-maximal binding (optical density: OD50%). p < 0.01 for interaction between week and E3 treatment (n = 11 per group). Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. d Non-fasting TG levels circulating in mice at 0, 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization. p < 0.01 for interaction between week and E3 treatment (n = 11 per group). Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. c, d Results are presented as mean ± SEM. n.s. not signiﬁcant. ∗p < 0.05 versus KLH (control) group. †p < 0.01 versus KLH (control) group. Published in partnership with the Sealy Institute for Vaccine Sciences BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine EPKSRFAMLD ANGPTL3 vaccine KLH (T-cell epitope) (a) intradermal injection EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD EPKSRFAMLD (B-cell epitope) _ _ _ weeks 2 4 0 6 _ : immunization with ANGPTL3 vaccine KLH group (Control) ANGPTL3 (E3) group (b) booster shot booster shot : blood collection 98 70 age in weeks 8 : sacrifice 113 100 : immunization with control vaccine 60 62 90 92 105 (b) (a) 6 20 30 60 62 90 92 105 0 50000 100000 150000 weeks Antibody titer (OD 50%) KLH E3 (c) * † † † † † * † KLH E3 0 6 20 30 60 62 90 92 105 0 50 100 150 200 250 weeks TG (mg/dL) (d) * * † † * n.s. n.s. n.s. KLH E3 KLH E3 * Non-fasting TG (mg/dL) (d) (c) weeks weeks Fig. 1 Study protocol and efﬁcacy of the ANGPTL3 vaccine against circulating triglyceride levels in C57BL/6 J mice fed a normal diet. a Schematic of vaccine design and vaccination methods. b Schematic showing vaccine injection protocol. c Circulating antibody titers in mice at 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization. Values are reported as the serum dilution giving half-maximal binding (optical density: OD50%). p < 0.01 for interaction between week and E3 treatment (n = 11 per group). Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. d Non-fasting TG levels circulating in mice at 0, 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization. p < 0.01 for interaction between week and E3 treatment (n = 11 per group). Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. c, d Results are presented as mean ± SEM. n.s. not signiﬁcant. ∗p < 0.05 versus KLH (control) group. †p < 0.01 versus KLH (control) group. vaccinated mice at time of sacriﬁce, based on inspection of those organs. mice would be near the end of life in a natural setting, the increase in anti-E3 antibody titers remained signiﬁcant, and circulating TG levels in non-fasting conditions were relatively lower in E3 compared to control groups, although that decrease was not statistically signiﬁcant (TG levels decreased 11 mg/dl relative to controls, p = 0.13) (Fig. 1c). BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine Renal function was also comparable between these groups (Supplementary Fig. 2g, h). Moreover, we did not observe any overt abnormalities in these organs in Here, we report that intradermal injection of an ANGPTL3 peptide vaccine (E3: EPKSRFAMLD) in C57BL/6 J wild-type mice aged 8, 10, and 12 weeks led to sustained elevation of antibody titers and to a concomitant decrease in triglyceride (TG) levels for up to one year (60 weeks after the initial inoculation). E3 booster administration at weeks 60 and 90 after the primary vaccination markedly increased E3 antibody titers and signiﬁcantly decreased TG levels in these mice. Notably, treatment with the ANGPTL3 vaccine did not promote adverse suppression of circulating lipids in aging mice or result in major organ damage, suggesting vaccine safety. In vaccine therapy for inherited chronic diseases such as familial dyslipidemia, it is essential that therapeutic antibodies continue to be produced throughout life. There is concern that the effective- ness of the vaccine may be reduced in older people, as it is thought that the function of memory B cells declines with age, reducing their ability to produce antibodies6,7. Hence, doubts have always been raised as to whether vaccine therapy for chronic diseases is realistically feasible. Because to date, few studies in humans and mice have investigated how antibody production after vaccination changes in the long term, into old age, or whether vaccine efﬁcacy can be demonstrated when vaccines are given repeatedly over a lifetime. This study has shown that, at least in mice, repeated doses of our E3 vaccine increase the antibody titers corresponding to the administered antigen over the lifetime of the mice. The results show that vaccine-based treatment strategies are feasible for diseases such as familial dyslipidemia, which require lifelong treatment. Next, to assess safety of repeated E3 vaccinations on major organs, we examined mouse liver, which is a major source of ANGPTL3 in vivo, as well as kidneys and lungs. Body weight and the liver weight/body weight ratio of the E3 group were comparable to controls throughout the entire protocol, suggest- ing no signiﬁcant vaccine toxicity (Fig. 2a, b). Also, circulating levels of alanine aminotransferase (ALT) and aspartate amino- transferase (AST), markers of liver damage, were not signiﬁcantly elevated in vaccinated versus control mice at 105 weeks after the ﬁrst vaccination (Fig. 2c, d). npj Vaccines (2023) 168 BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine To further analyze lipid proﬁles, we sacriﬁced these mice at 105 weeks after 15 h of fasting and found that fasting TG, chylomicron(CM), VLDL-C, LDL-C, sd- LDL-C, HDL-C, and total cholesterol levels in circulation of control mice had decreased with age (to 55.4 mg/dL, 0.43 mg/dl, 7.2 mg/ dl, 8.4 mg/dl, 3.9 mg/dl, 44.4 mg/dl, and 61.5 mg/dl, respectively), and that those levels did not differ signiﬁcantly in vaccinated mice (Supplementary Fig. 1a–g). These data overall suggest that the vaccine is safe without excessive suppression of circulating lipids. mice would be near the end of life in a natural setting, the increase in anti-E3 antibody titers remained signiﬁcant, and circulating TG levels in non-fasting conditions were relatively lower in E3 compared to control groups, although that decrease was not statistically signiﬁcant (TG levels decreased 11 mg/dl relative to controls, p = 0.13) (Fig. 1c). To further analyze lipid proﬁles, we sacriﬁced these mice at 105 weeks after 15 h of fasting and found that fasting TG, chylomicron(CM), VLDL-C, LDL-C, sd- LDL-C, HDL-C, and total cholesterol levels in circulation of control mice had decreased with age (to 55.4 mg/dL, 0.43 mg/dl, 7.2 mg/ dl, 8.4 mg/dl, 3.9 mg/dl, 44.4 mg/dl, and 61.5 mg/dl, respectively), and that those levels did not differ signiﬁcantly in vaccinated mice (Supplementary Fig. 1a–g). These data overall suggest that the vaccine is safe without excessive suppression of circulating lipids. Next, to assess safety of repeated E3 vaccinations on major organs, we examined mouse liver, which is a major source of ANGPTL3 in vivo, as well as kidneys and lungs. Body weight and the liver weight/body weight ratio of the E3 group were comparable to controls throughout the entire protocol, suggest- ing no signiﬁcant vaccine toxicity (Fig. 2a, b). Also, circulating levels of alanine aminotransferase (ALT) and aspartate amino- transferase (AST), markers of liver damage, were not signiﬁcantly elevated in vaccinated versus control mice at 105 weeks after the ﬁrst vaccination (Fig. 2c, d). Moreover, histological analysis of liver tissue sections revealed no signiﬁcant signs of inﬂammation or ﬁbrosis in E3-vaccinated relative to control mice (Fig. 2e–g). Accordingly, transcript levels of markers related to inﬂammation or ﬁbrosis were comparable in liver tissues from E3 and control groups (Fig. 2h–k). We also observed no major inﬂammation or ﬁbrosis in sections of kidney and lung tissues in E3-vaccinated compared to KLH control groups, based on histological staining (Supplementary Fig. 2a–f). Published in partnership with the Sealy Institute for Vaccine Sciences BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine g The degree of liver ﬁbrosis is reﬂected by percentages of aniline blue-positive interstitial areas. (H-K) Transcript levels of (h) Il-6, (i) Tnf-α, (j) Col1a1 and (k) Col3a1 in liver tissues (KLH: n = 5, E3: n = 6). KLH group values were set to 1. a, d, g–i, k) Results are expressed as mean ± SEM. b, c, j Results are expressed as median ± IQR. n.s. not signiﬁcant; p < 0.01 versus KLH (control) group. hepatokine synthesized in liver, signiﬁcant elevation of antibody titers against ANGPTL3 required additional E3 injections, suggest- ing that autoimmune hepatitis potentially induced by immunor- eaction against endogenous ANGPTL3 is of minimal concern. To this end, the B-cell epitope (E3: EPKSRFAMLD) of the ANGPTL3 vaccine was designed to include only 10 amino acids (aa) (corresponding to aa 32–41 of human ANGPTL3 protein)5. Based on results reported here, we found that repeated administration of the E3 vaccine throughout life has resulted in safety for major organs. preclinical studies using genetic dyslipidemia models and obesity- related dyslipidemia models other than C57BL/6 J mice fed a normal diet are needed to test optimal vaccine efﬁciency. Furthermore, future clinical studies of human subjects treated with the vaccine are also required. In summary, our anti-dyslipidemic vaccine therapy targeting ANGPTL3 has sustained efﬁcacy for over a year, and additional vaccination restored therapeutic efﬁcacy. Our current study provides insight into the utility and safety of long-term vaccine treatment against lifestyle-related diseases such as dyslipidemia. g Mice used here were wild-type male C57BL/6 J fed a normal diet, and the period of analysis extended from young to old age. Both this study and data from the Jackson Laboratory show that serum TG levels in C57BL/6 J mice increase with growth but spontaneously decrease with aging after 26–52 weeks (The Jackson Laboratory’s database)9, making it difﬁcult to detect lowering of circulating TG levels and other lipid proﬁles by the E3 vaccine in older mice that do not exhibit dyslipidemia. We show that E3 vaccination at 8, 10, 12 weeks of age suppressed non- fasting TG levels in circulation for over a year. Furthermore, additional boosting with the vaccine signiﬁcantly suppressed circulating TG levels in these mice until 100 weeks of age, which is near the end of a mouse’s lifespan. Antibody titers against E3 were high throughout the observation period, indicating that the ANGPTL3 vaccine may induce long-lasting immunological mem- ory. BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine Moreover, histological analysis of liver tissue sections revealed no signiﬁcant signs of inﬂammation or ﬁbrosis in E3-vaccinated relative to control mice (Fig. 2e–g). Accordingly, transcript levels of markers related to inﬂammation or ﬁbrosis were comparable in liver tissues from E3 and control groups (Fig. 2h–k). We also observed no major inﬂammation or ﬁbrosis in sections of kidney and lung tissues in E3-vaccinated compared to KLH control groups, based on histological staining (Supplementary Fig. 2a–f). Renal function was also comparable between these groups (Supplementary Fig. 2g, h). Moreover, we did not observe any overt abnormalities in these organs in y g In general, a therapeutic vaccine against lifestyle-related diseases must induce immunity without activating cytotoxic immune responses or promoting adverse autoimmune effects8. Notably, although ANGPTL3 is an endogenous, secreted npj Vaccines (2023) 168 H. Fukami et al. KLH E3 0 2 4 6 8 Il-6/Rps18 KLH E3 0 10 20 30 40 ALT (IU/L) 0 6 20 30 60 62 90 92 105 0 10 20 30 40 Body Weight (g) KLH E3 KLH E3 0.0 0.5 1.0 1.5 2.0 2.5 Tnf/Rps18 KLH E3 0.0 0.5 1.0 1.5 2.0 Col3a1/Rps18 KLH E3 0 1 2 3 4 Col1a1/Rps18 KLH E3 0 20 40 60 80 100 AST (IU/L) (a) (e) (c) (d) (f) KLH group E3 group n.s. KLH group E3 group n.s. (h) n.s. (i) (j) (k) n.s . s . n . s . n weeks Liver (Hematoxylin-Eosin staining) Liver (Azan-Mallory stain) KLH E3 0 5 10 15 Liver / Body weight (%) KLH E3 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. Liver weight per body weight (%) (b) (g) KLH E3 0 1 2 3 4 5 Percentage of Liver fibrosis (%) n.s. n.s. Fig. 2 Analysis of parameters relevant to safety after repeated administration of the ANGPTL3 vaccine over the lifespan of mice. a Body weight of C57BL/6 J mice before immunization and at 0, 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization (n = 11 per group). All mice were fed a normal diet. Results are presented as mean ± SEM. n.s. not signiﬁcant. ∗p < 0.05 versus KLH (control) group. †p < 0.01 versus KLH (control) group. Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. Vaccine design and synthesis We designed a candidate peptide, E3, which corresponds to a region of the LPL inhibitory domain (32–41 aa)5. After synthesis, the N-terminal of each peptide was conjugated to KLH, and synthetic peptides were puriﬁed by reverse-phase high-perfor- mance liquid chromatography (HPLC) (>98% purity) (Peptide Institute, Osaka, Japan). Before immunization, peptide solutions were mixed with an equal volume of complete/incomplete Freund’s adjuvant (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan). BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine a Body weight of C57BL/6 J mice before immunization and at 0, 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization (n = 11 per group). All mice were fed a normal diet. Results are presented as mean ± SEM. n.s. not signiﬁcant. ∗p < 0.05 versus KLH (control) group. †p < 0.01 versus KLH (control) group. Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. b Liver weight/body weight (%) (KLH: n = 6, E3: n = 5). c Serum alanine aminotransferase (ALT) levels (KLH: n = 6, E3: n = 5). d Serum aspartate aminotransferase (AST) levels (KLH: n = 6, E3: n = 5). e Representative images of Hematoxylin-Eosin-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. f Representative images of Azan-Mallory-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. g The degree of liver ﬁbrosis is reﬂected by percentages of aniline blue-positive interstitial areas. (H-K) Transcript levels of (h) Il-6, (i) Tnf-α, (j) Col1a1 and (k) Col3a1 in liver tissues (KLH: n = 5, E3: n = 6). KLH group values were set to 1. a, d, g–i, k) Fig. 2 Analysis of parameters relevant to safety after repeated administration of the ANGPTL3 vaccine over the lifespan of mice. a Body weight of C57BL/6 J mice before immunization and at 0, 6, 20, 60, 62, 90, 92, and 105 weeks after the ﬁrst E3 immunization (n = 11 per group). All mice were fed a normal diet. Results are presented as mean ± SEM. n.s. not signiﬁcant. ∗p < 0.05 versus KLH (control) group. †p < 0.01 versus KLH (control) group. Portions of plots within dotted squares represent data referenced from our previous report related to the E3 vaccine4. b Liver weight/body weight (%) (KLH: n = 6, E3: n = 5). c Serum alanine aminotransferase (ALT) levels (KLH: n = 6, E3: n = 5). d Serum aspartate aminotransferase (AST) levels (KLH: n = 6, E3: n = 5). e Representative images of Hematoxylin-Eosin-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. f Representative images of Azan-Mallory-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine Vaccine administration may be extended and lower vaccine doses may be effective in controlling lipids until old age, possibilities that should be tested in future analyses. Further Published in partnership with the Sealy Institute for Vaccine Sciences BRIEF COMMUNICATION OPEN Efﬁcacy and safety in mice of repeated, lifelong administration of an ANGPTL3 vaccine b Liver weight/body weight (%) (KLH: n = 6, E3: n = 5). c Serum alanine aminotransferase (ALT) levels (KLH: n = 6, E3: n = 5). d Serum aspartate aminotransferase (AST) levels (KLH: n = 6, E3: n = 5). e Representative images of Hematoxylin-Eosin-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. f Representative images of Azan-Mallory-stained liver sections from KLH (left panel) and E3 (right panel) groups. Scale bars, 500 μm. g The degree of liver ﬁbrosis is reﬂected by percentages of aniline blue-positive interstitial areas. (H-K) Transcript levels of (h) Il-6, (i) Tnf-α, (j) Col1a1 and (k) Col3a1 in liver tissues (KLH: n = 5, E3: n = 6). KLH group values were set to 1. a, d, g–i, k) Results are expressed as mean ± SEM. b, c, j Results are expressed as median ± IQR. n.s. not signiﬁcant; p < 0.01 versus KLH (control) group. 3 KLH E3 0 10 20 30 40 ALT (IU/L) LH 3 KLH E3 0 20 40 60 80 100 AST (IU/L) (c) (d) n.s. n.s. KLH E3 0 5 10 15 Liver / Body weight (%) LH n.s. Liver weight per body weight (%) (b) 0 6 20 30 60 62 90 92 105 0 10 20 30 40 Body Weight (g) (a) weeks n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. (b) (d) (c) (a) (e) (f KLH group E3 group weeks Liver (Hematoxylin-Eosin staining) (e) (f) (e) Liver (Azan-Mallory stain) KLH E3 0.0 0.5 1.0 1.5 2.0 2.5 Tnf/Rps18 group (i) n.s KLH E3 0 1 2 3 4 Col1a1/Rps18 g p (j) . s . n KLH E3 0 2 4 6 8 Il-6/Rps18 g p (h) n.s. (g) KLH E3 0 1 2 3 4 5 Percentage of Liver fibrosis (%) n.s. (i) KLH E3 KLH E3 Fig. 2 Analysis of parameters relevant to safety after repeated administration of the ANGPTL3 vaccine over the lifespan of mice. a Body i h f C BL/ J i b f i i i d d k f h ﬁ E i i i ( ) Fig. 2 Analysis of parameters relevant to safety after repeated administration of the ANGPTL3 vaccine over the lifespan of mice. Published in partnership with the Sealy Institute for Vaccine Sciences Evaluation of ANGPTL3 vaccine durability Eight-week-old male C57BL/6 J wild-type mice (n = 11) were intradermally injected with 100 μg KLH-conjugated E3 peptide, followed by identical immunizations 2 and 4 weeks later. Controls were injected with KLH in Freund’s adjuvant following the same protocol. To evaluate vaccine durability, blood samples from mice in non-fasting conditions were collected from the tail vein at 6, 20, 30, 60, 62, 90, and 92 weeks after the ﬁrst immunization, and antibody titers and non-fasting TG levels were assessed as described below. Real-time polymerase chain reaction analysis 8. Nakamaru, R., Nakagami, H., Rakugi, H. & Morishita, R. Future directions of ther- apeutic vaccines for chronic diseases. Circ J. 84, 1895–1902 (2020). 8. Nakamaru, R., Nakagami, H., Rakugi, H. & Morishita, R. Future directions of ther- apeutic vaccines for chronic diseases. Circ J. 84, 1895–1902 (2020). Total RNA was extracted using TRIzol reagent (Thermo Fisher Scientiﬁc) based on the manufacturer’s protocol. Brieﬂy, DNase- treated RNA was reverse-transcribed using a Prime Script RT reagent Kit (Takara Bio Inc, Shiga, Japan). Quantitative real-time PCR was performed using SYBR Premix Ex TaqII (Takara Bio Inc.). Relative transcript abundance was normalized to that of 18S rRNA levels in mice. Primer sequences are shown in Supplementary Table 1. 9. The Jackson Laboratory. Physiological Data Summary—Aged C57BL/6 (000664). https://www.jax.org/-/media/jaxweb/ﬁles/jax-mice-and-services/phenotypic-data/ aged-b6-physiological-data-summary.pdf. 9. The Jackson Laboratory. Physiological Data Summary—Aged C57BL/6 (000664). https://www.jax.org/-/media/jaxweb/ﬁles/jax-mice-and-services/phenotypic-data/ aged-b6-physiological-data-summary.pdf. Animal studies and immunization The Ethical Committee approved animal studies based on the Animal Experiments of the Kumamoto University Graduate School Published in partnership with the Sealy Institute for Vaccine Sciences npj Vaccines (2023) 168 H. Fukami et al. 4 blue positivity in a deﬁned area in Azan-Mallory-stained sections. All imaging analysis was performed using Photoshop 2022 (version 23.5.5) (Adobe Inc., San Jose, California) of Medical Sciences protocol. Eight-week-old male C57BL/6 J wild- type mice were purchased from Charles River Laboratories (Yokohama, Japan) and bred at 24 °C, in 12-h light-dark cycles, and with free access to normal chow and water. Ten minutes prior to being sacriﬁced, male mice were anesthetized using a combination of 0.3 mg/kg medetomidine, 4.0 mg/kg midazolam, and 5.0 mg/kg butorphanol, delivered via intraperitoneal injection. Blood was then collected from the tail. Following this, the mice were humanely euthanized through cervical dislocation, and their organs were harvested. Statistical analysis Normality of distribution in continuous variables was evaluated by the Kolmogorov–Smirnov test. If the distribution was normal, data was expressed as means ± SEM. Comparisons between two groups were made using Student’s t test. If the distribution was not normal, data was expressed as the median ± interquartile range (IQR), and comparisons between two groups were made using the Mann–Whitney U test. A mixed effect model was applied to repeated-measures data using week effects, vaccine effects and interaction of both as explanatory variables. In this model, ID of individual mice was set as a random intercept. A likelihood ratio test was performed to evaluate statistical signiﬁcance of the interaction. p < 0.05 was considered statistically signiﬁcant. Analyses were performed using GraphPad PRISM version 9.5.1 (GraphPad Software, Inc., La Jolla, California), and STATA MP 17.0 software (StataCorp., College Station, Texas). REFERENCES 1. Grundy, S. M. et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 139, e1082–e1143 (2019). 1. Grundy, S. M. et al. 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ ASPC/NLA/PCNA guideline on the management of blood cholesterol: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 139, e1082–e1143 (2019). 2. Mach, F. et al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modiﬁcation to reduce cardiovascular risk. Eur. Heart J. 41, 111–188 (2020). 2. Mach, F. et al. 2019 ESC/EAS guidelines for the management of dyslipidaemias: lipid modiﬁcation to reduce cardiovascular risk. Eur. Heart J. 41, 111–188 (2020). DATA AVAILABILITY The datasets used and/or analysed during the current study available from the corresponding author on reasonable request. Received: 23 May 2023; Accepted: 13 October 2023; Reporting summary f Further information on research design is available in the Nature Research Reporting Summary linked to this article. Antibody titers were evaluated as described4. In brief, ELISA plates (MaxiSorp Nunc; Thermo Fisher Scientiﬁc, Waltham, Massachu- setts) were coated with 5 mg/mL candidate ANGPTL3 peptides in carbonate buffer overnight at 4 °C. Peptides were conjugated to BSA carrier protein (Peptide Institute). After blocking with PBS containing 5% skim milk, sera were diluted 100- to 312,500-fold in blocking buffer. After overnight incubation at 4 °C and subsequent washing, plates were incubated with horseradish peroxidase (HRP)-conjugated antibodies speciﬁc for mouse IgG (GE Health- care, Chicago, Illinois) for 3 h at room temperature. After washing with PBS, the color was developed using the peroxidase chromogenic substrate 3,3′–5,5′-tetramethyl benzidine (Sigma- Aldrich, St. Louis, Missouri), and the reaction was halted with 0.5 N sulfuric acid. Absorbance at 450 nm was monitored with a microplate reader (iMark, Bio-Rad, Hercules, California). The half- maximal antibody titer was determined according to each sample’s highest value in the dilution range. ACKNOWLEDGEMENTS We thank our colleagues for valuable suggestions and discussion. We also thank K. Tabu, K. Saito, S. Iwaki, Y. Itoh, N. Shirai, and M. Kamada for technical assistance. This work was supported by JSPS KAKENHI (Grant Number 21K20935 and 22K16429) to H.F., by AMED (Grant Number JP23ym0126809) to Y. Oike, by Ono Medical Research Foundation to H.F., by KM Biologics Co., Ltd. to Y. Oike, and by FunPep Co., Ltd. to Y. Oike. Blood chemistry 3. Mercep, I., Strikic, D., Sliskovic, A. M. & Reiner, Z. New Therapeutic Approaches in Treatment of Dyslipidaemia-A Narrative Review. Pharmaceuticals (Basel). 15, https://doi.org/10.3390/ph15070839 (2022). 3. Mercep, I., Strikic, D., Sliskovic, A. M. & Reiner, Z. New Therapeutic Approaches in Treatment of Dyslipidaemia-A Narrative Review. Pharmaceuticals (Basel). 15, https://doi.org/10.3390/ph15070839 (2022). Blood TG levels in non-fasting and fasting conditions were measured using a LabAssay kit (FUJIFILM Wako Pure Chemical Corporation) according to the manufacturer’s protocol. Blood ALT, AST, urea nitrogen (UN) and creatinine levels at sacriﬁce were measured by SRL (Tokyo, Japan), and fasting lipid proﬁle measurements were performed by LipoSEARCH (Immuno-Biologi- cal Laboratories) (Gumma, Japan). 4. Mullard, A. FDA approves ﬁrst anti-ANGPTL3 antibody, for rare cardiovascular indication. Nat. Rev. Drug Discov. 20, 251 (2021). 4. Mullard, A. FDA approves ﬁrst anti-ANGPTL3 antibody, for rare cardiovascular indication. Nat. Rev. Drug Discov. 20, 251 (2021). 5. Fukami, H. et al. Vaccine targeting ANGPTL3 ameliorates dyslipidemia and asso- ciated diseases in mouse models of obese dyslipidemia and familial hypercho- lesterolemia. Cell Rep. Med. 2, 100446 (2021). 6. Cancro, M. P. Age-associated B cells. Annu. Rev. Immunol. 38, 315–340 (2020). 7. Frasca, D., Diaz, A., Romero, M., Garcia, D. & Blomberg, B. B. B cell immunose- nescence. Annu. Rev. Cell Dev. Biol. 36, 551–574 (2020). 7. Frasca, D., Diaz, A., Romero, M., Garcia, D. & Blomberg, B. B. B cell immunose- nescence. Annu. Rev. Cell Dev. Biol. 36, 551–574 (2020). Published in partnership with the Sealy Institute for Vaccine Sciences AUTHOR CONTRIBUTIONS Correspondence and requests for materials should be addressed to Jun Morinaga or Yuichi Oike. Conceptualization: H.F., J.M., H.N., and Y.Oike; Methodology: H.F., J.M., H.N., H.Hayashi, and Y.Oike; Investigation and data collection: H.F., J.M., Y.Okadome, E.M., T.Kadomatsu, H.Horiguchi, and M.S.; Statistical analysis: H.F., J.M., T.S., K.S., and D.T.; Supervision: M.M., R.M., H.N., and Y.Oike; writing—original draft: H.F., J.M., and Y.Oike; writing—review & editing-: H.F., J.M., H.N., and Y.Oike. Reprints and permission information is available at http://www.nature.com/ reprints Histological analysis Parafﬁn-embedded liver tissue was sliced into 4 μm sections and subjected to Hematoxylin-Eosin or Azan-Mallory staining. The degree of ﬁbrosis was calculated based on the proportion (%) of Published in partnership with the Sealy Institute for Vaccine Sciences npj Vaccines (2023) 168 H. Fukami et al. 5 Reprints and permission information is available at http://www.nature.com/ reprints Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. Published in partnership with the Sealy Institute for Vaccine Sciences COMPETING INTERESTS A patent on the vaccine targeting ANGPTL3 as treatment for dyslipidemia and associated disease has been ﬁled (PCT/JP2022/020850; Y.Oike, J.M., H.F.). Y.Oike is ﬁnancially supported by KM Biologics and Funpep. R.M. and H.N. are stockholders of FunPep. H.N. is a scientiﬁc advisor for FunPep. The Department of Health Development and Medicine is an endowed department supported by AnGes, Daicel, and FunPep, and collaborates with these companies. The Department of Clinical Gene Therapy is ﬁnancially supported by Novartis, AnGes, Shionogi, Boehringer, Fancl, Rohto, and FunPep. The remaining authors declare no competing interests. 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/. Published in partnership with the Sealy Institute for Vaccine Sciences ADDITIONAL INFORMATION Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41541-023-00770-3. © The Author(s) 2023 npj Vaccines (2023) 168
https://openalex.org/W4391138824	http://www.clausiuspress.com/assets/default/article/2024/01/23/article_1706007412.pdf	English	null	Research on the Integration of System and Emotion Management in Enterprise Management in the Era of Artificial Intelligence	Accounting and corporate management	2,024	cc-by	4,490	Keywords: Artificial Intelligence, Enterprise Management, Emotion Management, Enterprise System Keywords: Artificial Intelligence, Enterprise Management, Emotion Management, Enterprise System Abstract: With the rapid growth of AI technology, in the field of enterprise management, especially in the era of information technology, the efficiency of human beings in information transmission and processing continues to improve. However, due to the defects of traditional thinking mode such as self-isolation to a certain extent, and the influence of emotional factors on human nature cognition, this article needs to use intelligent, automated and information means to improve the manual operation mode, and apply Artificial Intelligence(AI) technology to enterprise management, integrate system and emotion, and design management models. After that, the running effect of the model is tested. The test results show that the quality control defect rate can be controlled within 2%; satisfaction with humanistic care is up to 94%, and satisfaction with emotional value is up to 96%. This shows that the service quality combining system and emotion has been correspondingly improved. DOI: 10.23977/acccm.2024.060103 ISSN 2523-5788 Vol. 6 Num. 1 DOI: 10.23977/acccm.2024.060103 ISSN 2523-5788 Vol. 6 Num. 1 Accounting and Corporate Management (2024) Clausius Scientific Press, Canada DOI: 10.23977/acccm.2024.060103 ISSN 2523-5788 Vol. 6 Num. 1 Research on the Integration of System and Emotion Management in Enterprise Management in the Era of Artificial Intelligence Ziyu Cai1,a, Xianfa Yang1,b, Changjiang Yu1,c,* 1Law and Business School, Shangrao Normal University, Shangrao, Jiangxi, China a2532571427@qq.com, b2050922751@qq.com, c207593@qq.com Corresponding author Ziyu Cai1,a, Xianfa Yang1,b, Changjiang Yu1,c, 1Law and Business School, Shangrao Normal University, Shangrao, Jiangxi, China a2532571427@qq.com, b2050922751@qq.com, c207593@qq.com Corresponding author Ziyu Cai1,a, Xianfa Yang1,b, Changjiang Yu1,c, 1Law and Business School, Shangrao Normal University, Shangrao, Jiangxi, China a2532571427@qq.com, b2050922751@qq.com, c207593@qq.com Corresponding author 1. Introduction With the rapid growth of AI, enterprise management is facing many new problems, including high employee turnover rate and high staff mobility. In order to solve these problems, it is necessary to conduct in-depth research on the integration of enterprise system and emotion management, so as to improve the work efficiency and loyalty of employees, and promote them to better play their role in the organization. This not only provides a good environment and basic guarantee for China's scientific and technological innovation, but also lays a theoretical foundation for the improvement of enterprise competitiveness, and has important practical value. With the growth of AI technology, there are many new types of problems, such as information security and data leakage, which have a serious impact on human society. In order to deal with these problems, this paper needs to establish a new organizational structure, which can deal with the complex and ever-changing business management system and emotional culture system. In this process, it is also necessary to fully understand the relationship between people and machines, and in-depth understanding of humanized design methods to better manage employees, improve 17 efficiency and respond to various emergencies, so as to promote the innovation of enterprise management system in the era of intelligence. The innovation of this paper is mainly reflected in two aspects. First of all, the research object is targeted, that is, in the era of AI, the integration of enterprise management system and emotion management has been optimized and integrated, which has improved the original corporate culture and organizational structure. Through the application of scientific and technological growth achievements in the new era to practical work to solve practical problems, while providing reference significance and practical experience for other fields of relevant theories and suggestions and guidance, which has important practical value for promoting the construction of China's science and technology innovation country. Secondly, this paper integrates information technology into the organization management system. At present, many large companies in China have established information system platforms based on Internet technology to improve their system operation efficiency and optimize business processes. However, in the era of AI, the organizational structure of enterprises is also changing and improving. 1. Introduction Therefore, from the perspectives of management system, emotional culture and incentive mechanism, this paper proposes to combine information technology with organizational management system to further improve employees' work efficiency and loyalty, and promote the innovation of enterprise management system. 2. Related Work 3.1 AI Technology 3.1 AI Technology AI is a new science and technology based on computer technology, which realizes the ability to recognize and judge things by processing a large number of complex and abstract information. To a certain extent, AI can simulate the way human intelligence thinks and automatically analyze data results. In addition, AI also uses machine language to express and explain the problems and reasoning logical relations involved in the cognitive process of things. The growth of AI comes from the research of human brain technology, which is a new type of science and technology. AI technology refers to the simulation and perception of the human brain through computers, information processing and related intelligent tools, and makes corresponding responses according to its own brain cognitive ability. This technology has a high degree of intelligence and versatility. In the era of AI, enterprise management needs to combine different types of emotional needs to achieve human care and guidance, and also need to pay attention to effective analysis and research on employees' emotional state and behavior, so as to help enterprises develop suitable and targeted incentive mechanisms to promote employees' work enthusiasm. Based on computer, AI technology has made certain progress in information processing, automatic control and intelligent machine learning [11-12]. In the decision-making process, a large number of complex and highly accurate data is essential. In radar signal recognition system, a large amount of data needs to be collected, and sensor technology needs to process and predict this complex, large amount, variety and unstructured information. Therefore, AI can be used to process and predict these data. In the enterprise management of AI, the integration of system and emotion management is of great significance. This fusion combines institutional norms with emotional factors to improve organizational effectiveness and employee satisfaction. The relationship between system implementation efficiency (X) and emotional factor (Y) can be expressed by the following formula: Z b + Y * a = X (1) (1) Z * b + Y * a = X In the formula, a and b are weight parameters, and Z represents other factors that may affect the efficiency of system implementation. This formula can be applied to a series of data samples to calculate the specific value of the weight parameter. By adjusting the parameters, we can find the best balance point between institution and emotion management. 2. Related Work With the rapid growth of AI technology, its understanding and application are getting deeper and deeper. In the context of the rapid growth of information technology, intelligent tools have been produced in large numbers, and the emotional and psychological needs of business management have become crucial. However, most domestic companies have not yet established a perfect and sound human care system and employee incentive mechanism, and have not carried out relevant research work. In contrast, some developed countries pay attention to the use of data mining technology, machine learning theory and AI and other means to achieve the effective combination of customer personalized service. Vladimir Tsyganov's research explored the relationship between social and political stability, voters' emotional expectations and information management [1]. The study of Jorge Fernandez Herrero et al. introduced the steps of applying AI emotional expression recognition software to emotion management in an educational context [2]. Jose Arias-Perez et al. 's research focuses on the organizational emotional capacity and the absorptive capacity orientation of competitors to stimulate the open innovation process [3]. Zahra Sarhadi et al. assessed the impact of knowledge management and emotional intelligence on the work efficiency of Iranian librarians [4]. Satyanarayana Parayitam et al. explored the relationship between role conflict and organizational performance in India, and studied how knowledge management and emotional exhaustion play a moderating role [5]. Yong Liao and Rui Kong analyzed the complexity of iot RFID in the management of fast fashion enterprises [6]. Faizan Ahmed Khan et al. introduced the process discovery and improvement of enterprise management system [7]. Jannis Beese et al studied the impact of enterprise architecture management on the complexity of information system architecture [8]. Raphael David Schilling et al. explored the strategic alignment of enterprise architecture management and how the combination of control mechanisms tracked the ten-year corporate transformation of Deutsche Commerzbank [9]. Islam O. Sulumov et al. studied the problems of enterprise human resource management during the growth of new forms of employment [10]. To sum up, in enterprise management, emotion management mainly exerts influences in the following three aspects: First, it makes integrated analysis and prediction of information resources such as employees, customers and suppliers; the second is to establish a performance management system based on knowledge sharing platform through data mining technology; the third is to realize the 18 incentive and constraint through the performance feedback mechanism. 3. Method 3.1 AI Technology The following formula can be used to accurately measure the comprehensive index of the degree of institutional norms and the influence of emotion management: C * e + B * d + A * c = M (2) (2) In the formula, A represents institutional normality, B represents emotion management factors, and C represents other possible relevant factors. c, d, and e are the weight parameters. Through statistical analysis of the data, the best weight parameters can be determined, and the comprehensive index M can be calculated to evaluate the integration effect of institutional and emotional management. In order to assess the degree of innovation in AI applications, the following formula can be used: F * h + E * g + D * f = I (3) (3) F * h + E * g + D * f = I In the formula, D represents the data analysis effect of AI, E represents the emotional intelligence factor, and F represents other innovation factors. f, g, and h are weight parameters. Through quantification and analysis of relevant data, innovation index I can be calculated to 19 measure the innovation effect of AI on institutional and emotional management in enterprise management. Artificial neurons generate a kind of "memory" by constantly learning new knowledge, so as to realize the interconnection between the human brain and the brain, and automatically regulate the information processing capacity. This kind of neural network is composed of nonlinear functions composed of a large number of repetitive data, which can maintain normal operation, control complex behaviors, realize self-cognition and other functions under uncertain environment [13-14]. In the era of AI, business management is human-centric, treating people as the core elements of machines and information processing systems. In the process of science and technology growth, humanization has become an important direction. Intelligent robot technology enables users to enter a personalized interactive environment, self-learning, self-realization, and contact, communication and interaction with the outside world. In this paper, intelligent robots can perform decision analysis and behavior control by sensing other people's instructions. Intelligent robots can also integrate personal emotions into online platforms, achieving the goal of emotional regulation and expanding interpersonal relationships and social circles. 3.1 AI Technology 2 Integration Model of Institutional and Emotional Management in Enterprise Ma 3.2 Integration Model of Institutional and Emotional Management in Enterprise Management In the enterprise management, the system is an important part of the organization, and the emotion management is the auxiliary service content. In order to realize the integration of the two, it is necessary to establish a perfect and effective communication mechanism and exchange platform to coordinate and link up work. In addition, it is also necessary to improve the level of professional ethics and psychological quality through employee training and other means, and use incentive policies to promote emotional expression and strengthen emotional investment to meet the spiritual needs of employees, so as to achieve the goal of corporate culture construction [15-16]. From another point of view, in order to realize the integration of system and emotion, it is necessary to establish a humanized management system. In the process of enterprise growth, the continuous training of employees, so that they can adapt to the position, and create value for the company, which is determined by the needs of humanity. In order to achieve the coordination between organizational goals and individual behaviors, it is necessary to establish a good communication channel and perfect incentive mechanism to improve the enthusiasm of organization members. Secondly, it is necessary to strengthen the implementation of team building and management system, so that managers can give full play to their own abilities and influence. In enterprise management, system and emotion refer to the behavior and attitude generated by employees in the process of work, as well as the coordination of the relationship between things and others. Figure 1 shows the integration model of institutional and emotional management. Business data The integration of institutional and emotional data Emotional data platform Figure 1: The fusion model of system and emotion management Business data Emotional data platform The integration of institutional and emotional data Figure 1: The fusion model of system and emotion management Figure 1: The fusion model of system and emotion management 20 For a company, it is necessary to establish a system that is people-oriented, meets the needs of humanity and has certain special characteristics. For a company, it is necessary to establish a system that is people-oriented, meets the needs of humanity and has certain special characteristics. 4. Results and Discussion 3.1 AI Technology In addition, starting from the mutual growth of emotions and employees, a complete institutional system can be built to ensure that emotions can play an effective role and effect [17-18]. In business management, the concept of emotion management refers to the two-way interaction between employees and organizations. When the two sides communicate, it can produce a good interaction effect. In this case, it can increase the trust of both managers and the managed, improve work efficiency, achieve common growth goals, improve the overall performance of the company, and promote social harmony and stability. In enterprise management, because different employees have different emotional needs, it is necessary to build an effective incentive model based on the combination of AI technology and management concepts to meet the needs of employees. In the work of system construction and emotion management, taking people as the core is the key. Humanization, scientificity and innovation are the basic conditions for system construction and emotion management. Therefore, this paper can incorporate corporate culture as a humanistic care approach to meet the inner needs of employees and stimulate their enthusiasm for work, so as to create maximum benefits and value for enterprises. 4. Results and Discussion 4.1 Testing the Application Effect of AI in Enterprise Management 21 Table 1: Test specification Testing equipment Equipment specifications Server CPU: Intel Xeon Gold 6248R 2.9GHz, RAM: 128GB, Storage: SSD 1TB Customer service terminal equipment CPU: Intel Core i7-10700K 3.8GHz, RAM: 16GB, Storage: SSD 512GB Industrial control console CPU: Intel Core i5-9600K 3.7GHz, RAM: 8GB, Storage: HDD 1TB Testing instruments Resolution: 1920 * 1080, supports AI algorithm processing/analysis Recruitment platform CPU: Intel Pentium G5400 3.7GHz, RAM: 4GB, storage: HDD 500GB Monitoring system Camera: 1080p Full HD, storage: NAS 4TB Data server CPU: Intel Xeon E5-2690 v4 2.6GHz, RAM: 256GB, Storage: SSD 4TB Logistics management terminal equipment CPU: Intel Celeron J4105 1.5GHz, RAM: 8GB, Storage: SSD 256GB 4.2 Analysis of Test Data Figure 2: Defect rate of quality control The defect rate test of quality control aims to find the root cause by analyzing the error in the production process of the product, and put forward the corresponding solution to improve the management level and service efficiency of the enterprise. In quality control technology, process 1 2 3 4 5 Defect rate 2% 1.50% 1% 2% 0.50% 0% 1% 1% 2% 2% 3% Rate Test times Table 1: Test specification Table 1: Test specification Testing equipment Equipment specifications Server CPU: Intel Xeon Gold 6248R 2.9GHz, RAM: 128GB, Storage: SSD 1TB Customer service terminal equipment CPU: Intel Core i7-10700K 3.8GHz, RAM: 16GB, Storage: SSD 512GB Industrial control console CPU: Intel Core i5-9600K 3.7GHz, RAM: 8GB, Storage: HDD 1TB Testing instruments Resolution: 1920 * 1080, supports AI algorithm processing/analysis Recruitment platform CPU: Intel Pentium G5400 3.7GHz, RAM: 4GB, storage: HDD 500GB Monitoring system Camera: 1080p Full HD, storage: NAS 4TB Data server CPU: Intel Xeon E5-2690 v4 2.6GHz, RAM: 256GB, Storage: SSD 4TB Logistics management terminal equipment CPU: Intel Celeron J4105 1.5GHz, RAM: 8GB, Storage: SSD 256GB 4.2 Analysis of Test Data Figure 2: Defect rate of quality control The defect rate test of quality control aims to find the root cause by analyzing the error in the production process of the product, and put forward the corresponding solution to improve the management level and service efficiency of the enterprise. 4. Results and Discussion 4.1 Testing the Application Effect of AI in Enterprise Management 4. Results and Discussion 4.1 Testing the Application Effect of AI in Enterprise Management 4.1 Testing the Application Effect of AI in Enterprise Management The testing process of the application effect of AI technology in enterprise management includes the following main steps. First, the data needs to be preprocessed, through the analysis of the data, to determine the need for classified statistics, and convert the information into a format suitable for decision makers. Secondly, the model, parameter base and database are established, and the sample features collected by the massive complex system are extracted by the computer, and the relevant rule base is generated to realize the modeling and predictive analysis process. The application of intelligent technology in enterprise management activities can help managers quickly understand the working status and needs of employees, and find problems and deal with them in time by collecting, sorting and analyzing data [19-20]. Among them, artificial neural network technology is an intelligent information model developed based on the cognitive system of human brain. It can automatically obtain various complex behavior patterns and related knowledge system structure images generated in the thinking process of human brain, and use the model to conduct the correlation between decision assistance function and emotion management, so as to provide help or solutions to the emotional and mental state problems that may occur in the work of enterprise employees. The original database is processed and screened effectively by means of intelligent technology and network communication. This paper will conduct 5 rounds of tests on this model. Table 1 shows the test hardware environment. By sorting and extracting a large amount of information, useful data can be obtained for decision-makers' reference, and some non-quantitative or semi-qualitative attribute information can be converted into quantitative or digital forms to assist decision-making and help enterprise managers choose corresponding management strategies to cope with risk problems brought by emergencies of different degrees. 4. Results and Discussion 4.1 Testing the Application Effect of AI in Enterprise Management In quality control technology, process 1 2 3 4 5 Defect rate 2% 1.50% 1% 2% 0.50% 0% 1% 1% 2% 2% 3% Rate Test times Equipment specifications 2 Analysis of Test Data Figure 2: Defect rate of quality control 1 2 3 4 5 Defect rate 2% 1.50% 1% 2% 0.50% 0% 1% 1% 2% 2% 3% Rate Test times Figure 2: Defect rate of quality control The defect rate test of quality control aims to find the root cause by analyzing the error in the production process of the product, and put forward the corresponding solution to improve the management level and service efficiency of the enterprise. In quality control technology, process 22 capability index is very important. In the traditional cost management mode, enterprises mainly rely on post-inspection and regular sampling inspection to carry out product inspection and production management activities. However, the application of intelligent technology can realize the functions of pre-control and in-process supervision. After the intelligent optimization of the management model test, from the data in Figure 2, the quality control defect rate of this paper can be controlled within 2%. Figure 3: Satisfaction with customer service 85% 89% 91% 90% 94% 96% 93% 89% 83% 88% 75% 80% 85% 90% 95% 100% 1 2 3 4 5 Rate Test times Emotional value Humanistic concern Figure 3: Satisfaction with customer service Customer service satisfaction is the result of providing psychological, material and emotional needs to consumers after evaluating product quality and performance. It is designed to meet the needs of consumers and ultimately form a comprehensive evaluation and recognition. In the era of AI, the requirements for employee service level are higher. Employees not only need to have rich work experience, but also need to have certain communication skills. Therefore, enterprises should establish a sound, reasonable and effective customer service system and management system to enhance customer satisfaction. According to the results of the customer service satisfaction survey in Figure 3, the highest degree of satisfaction of humanistic care is 94%, and the highest degree of satisfaction of emotional value is 96%. This shows that the service quality combining system and emotion has been correspondingly improved. 5. Conclusion With the progress of AI, emotion management is constantly introduced into the field of business management. Therefore, how to establish a sound emotion management system has become one of the focuses of current research. This paper aims to analyze the problems and causes of employee emotion management in the current growth stage of Chinese enterprises, and put forward targeted suggestions and countermeasures, so as to provide reference and reference value for improving employee work efficiency, promoting the realization of the company's strategic goals and improving customer satisfaction. This study also provides theoretical support for relevant scholars to promote the improvement of enterprise management system in the era of AI. However, this study has the following shortcomings: when analyzing the fusion relationship between enterprise management system and emotion management in the era of AI, it only considers the influence of AI technology 23 on the organizational structure and function of traditional companies, but ignores the analysis of its internal mechanism. This paper does not study the different needs, motivation characteristics and individual characteristics of employees in this stage from the perspective of employees, and takes corresponding measures according to specific situations. Therefore, there may be certain limitations in practice. In view of the above shortcomings, future research can further strengthen the methods and techniques of data acquisition to improve the credibility of empirical analysis. In addition, it should deeply study the integration mechanism of enterprise management system and emotion management in the era of AI, so as to better transform theory into practice. It can also pay attention to the needs and personalized characteristics of employees, and strengthen communication and interaction with employees to achieve more effective emotional management. Acknowledgement This work was supported by Soft Science Project of Shangrao Science and Technology Bureau “Research on Optimizing Regional Innovation Mechanism of Enterprise General Technology Platform” (2020L015) This work was supported by Soft Science Project of Shangrao Science and Technology Bureau “Research on Optimizing Regional Innovation Mechanism of Enterprise General Technology Platform” (2020L015) References [1] Vladimir Tsyganov. Socio-political stability, voter's emotional expectations, and information management. AI Soc. 38(1): 269-281 (2023). [1] Vladimir Tsyganov. Socio-political stability, voter's emotional expectations, and information management. AI Soc. 38(1): 269-281 (2023). ( ) ( ) [2] Jorge Fernández Herrero, Francisco Gomez-Donoso, Rosabel Roig-Vila. The first steps for adapting an artificial intelligence emotion expression recognition software for emotional management in the educational context. Br. J. Educ. Technol. 54(6): 1939-1963 (2023). [3] José Arias-Pérez, Joaquín Alegre, Cristina Villar. Triggering Open Innovation Processes Through Organizational Emotional Capability and Rival's Absorptive Capacity Orientation. IEEE Trans. Engineering Management 69(2): 388-398 (2022). [4] Zahra Sarhadi, Mojtaba Kaffashan Kakhki, Hassan Behzadi. Productivity story of Iranian librarians: assessing the impact of knowledge management and emotional intelligence. Electron. Libr. 39(6): 846-864 (2021). [5] Satyanarayana Parayitam, Syed Aktharsha Usman, Rajeshwaran Raja Namasivaayam, Mohamed Shaik Naina. Knowledge management and emotional exhaustion as moderators in the relationship between role conflict and organizational performance: evidence from India. J. Knowl. Manag. 25(6): 1456-1485 (2021). [6] Yong Liao, Rui Kong. Complexity Analysis of Internet of Things RFID in the Management of Fast-Fashion Apparel Enterprises. IEEE Consumer Electron. Mag. 12(2): 35-40 (2023). [7] Faizan Ahmed Khan, Farooq Ahmad, Arfat Ahmad Khan, Chitapong Wechtaisong. Process Discovery and Refinement of an Enterprise Management System. Comput. Syst. Sci. Eng. 44(3): 2019-2032 (2023). ] Jannis Beese, Stephan Aier, Kazem Haki, Robert Winter. The impact of enterprise architecture man formation systems architecture complexity. Eur. J. Inf. Syst. 32(6): 1070-1090 (2023). [9] Raphael David Schilling, Kazem Haki, Jannis Beese, Martin Kraus, Stephan Aier, Robert Winter. Strategic alignment of enterprise architecture management - how portfolios of control mechanisms track a decade of enterprise transformation at Commerzbank. Eur. J. Inf. Syst. 32(1): 92-105 (2023). f f y [10] Islam O. Sulumov, Zulay K. Tavbulatova, Gulnaz F. Galieva, Gayane A. Kochyan, Anzaur A. Bzhasso. Human resource management at the enterprise during the growth of new forms of employment. Int. J. Appl. Decis. Sci. 16(3): 300-312 (2023). [11] Guomeng Zhao. An enterprise logistics cost management method under the trend of Industry 4.0. Int. J. Manuf. Technol. Manag. 37(2): 199-210 (2023). u. Research on enterprise financial crisis early warning management based on PLS-BP. Int. J. Wirel. Mo 2): 195-202 (2023). Lihong Zhang. Incentive mechanism system of the management of IC design enterprises. J. Ambient Comput. 14(3): 1311-1322 (2023). z p ( ) ( ) [14] Yu Fu, Xiaoxi Sun, Weiwei Wang. [16] Marc Cowling, Weixi Liu. Access to Finance for Cleantech Innovation and Investment: Evidence from U.K. Small- and Medium-Sized Enterprises. IEEE Trans. Engineering Management 70(3): 963-978 (2023). [17] Usman Ahmed, Jerry Chun-Wei Lin, Gautam Srivastava, Unil Yun. Enhancing Stock Portfolios for Enterprise Management and Investment in Energy Industry. IEEE Trans. Ind. Informatics 19(6): 7667-7675 (2023). [18] Sushil Kumar Singh, Jong Hyuk Park. TaLWaR: Blockchain-Based Trust Management Scheme for Smart Enterprises With Augmented Intelligence. IEEE Trans. Ind. Informatics 19(1): 626-634 (2023). [19] Kaiyang Sun. Analysis of production and organisational management efficiency of Chinese family intelligent manufacturing enterprises based on IoT and machine learning technology. Enterp. Inf. Syst. 16(2): 208-222 (2022). [20] Emmanuel Ramaru, Lalit Garg, Chinmay Chakraborty. A Hybrid Cloud Enterprise Strategic Management System. Int. J. Cloud Appl. Comput. 12(1): 1-18 (2022). References The Optimization of Global Organizational Communication for Enterprise Supply Organization Management by Using Big Data Text Mining. J. Glob. Inf. Manag. 31(3): 1-17 (2023). [15] Ozlem Bak Sarah Shaw Claudia Colicchia Vikas Kumar A Systematic Literature Review of Supply Chain [14] Yu Fu, Xiaoxi Sun, Weiwei Wang. The Optimization of Global Organizational Communication for Enterprise Supply Organization Management by Using Big Data Text Mining. J. Glob. Inf. Manag. 31(3): 1-17 (2023). pp y g g y g g g f g ( ) ( ) [15] Ozlem Bak, Sarah Shaw, Claudia Colicchia, Vikas Kumar. A Systematic Literature Review of Supply Chain Resilience in Small-Medium Enterprises (SMEs): A Call for Further Research. IEEE Trans. Engineering Management 70(1): 328-341 (2023). 24 25
https://openalex.org/W4214592167	https://www.biorxiv.org/content/biorxiv/early/2022/05/14/2022.02.24.481763.full.pdf	English	null	Cryo-electron tomography of Birbeck granules reveals the molecular mechanism of langerin lattice formation	bioRxiv (Cold Spring Harbor Laboratory)	2,022	cc-by	18,590	. CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Cryo-electron tomography of Birbeck granules reveals the 1 molecular mechanism of langerin lattice formation 2 3 Authors: Toshiyuki Oda1, Haruaki Yanagisawa2, Hideyuki Shinmori 3, Youichi Ogawa4, 4 and Tatsuyoshi Kawamura4 5 6 Affiliations: 1 Department of Anatomy and Structural Biology, Graduate School of 7 Medicine, University of Yamanashi, 1110 Shimokato, Chuo, Yamanashi, 409-3898, Japan. 2 8 Department of Cell Biology and Anatomy, Graduate School of Medicine, the University of 9 Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-0033, Japan. 10 3 Faculty of Life and Environmental Science, University of Yamanashi, 4-4-37 Takeda, 11 Kofu, Yamanashi, 400-8510, Japan 12 4 Department of Dermatology, Faculty of Medicine, University of Yamanashi, 1110 13 Shimokato, Chuo, Yamanashi, 409-3898, Japan. 14 15 Correspondence to: toda@yamanashi.ac.jp 16 17 Cryo-electron tomography of Birbeck granules reveals the 1 molecular mechanism of langerin lattice formation 2 1 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Abstract 18 Abstract 18 19 Langerhans cells are specialized antigen-presenting cells localized within the epidermis and 20 mucosal epithelium. Upon contact with Langerhans cells, pathogens are captured by the C- 21 type lectin langerin and internalized into a structurally unique vesicle known as a Birbeck 22 granule. Although the immunological role of Langerhans cells and Birbeck granules have 23 been extensively studied, the mechanism by which the characteristic zippered membrane 24 structure of Birbeck granules is formed remains elusive. In this study, we observed isolated 25 Birbeck granules using cryo-electron tomography and reconstructed the 3D structure of the 26 repeating unit of the honeycomb lattice of langerin at 6.4 Å resolution. We found that the 27 interaction between the two langerin trimers was mediated by docking the flexible loop at 28 residues 258-263 into the secondary carbohydrate-binding cleft. Mutations within the loop 29 inhibited Birbeck granule formation and the internalization of HIV pseudovirus. These 30 findings suggest a molecular mechanism for membrane zippering during Birbeck granule 31 biogenesis and provide insight into the role of langerin in the defense against viral infection. 32 34 2 2 2 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Introduction 35 The epidermis and mucosa provide the first line of defense against pathogens. In these tissues, 36 specialized antigen-presenting cells called Langerhans cells capture pathogens by 37 recognizing their surface mannose-containing oligosaccharides via the C-type lectin langerin 38 (Valladeau et al., 2003; Doebel et al., 2017). Pathogens internalized via langerin are enclosed 39 in a specialized organelle called a Birbeck granule, which has a central striated lamella 40 between closely apposed membranes (Birbeck et al., 1961; Valladeau et al., 2000; 41 McDermott et al., 2004; Valladeau et al., 1999). Once pathogens are captured, Langerhans 42 cells undergo maturation and migrate to peripheral lymph nodes, where they present 43 processed antigens to T cells (Merad et al., 2008; de_Jong MA, 2010; Rhodes et al., 2021). 44 Langerin is a type II transmembrane protein that is composed of a cytoplasmic tail, coiled- 45 coil neck region, and carbohydrate recognition domain (CRD) (Stambach and Taylor, 2003; 46 Valladeau et al., 1999; Valladeau et al., 2000) (Figure 1A). Through the coiled-coil neck 47 region, langerin forms a trimer, which is functionally important for oligosaccharide binding 48 (Stambach and Taylor, 2003; Weis and Drickamer, 1996; Weis and Drickamer, 1994). The 49 calcium-dependent carbohydrate-binding capacity of langerin CRD is essential for the 50 formation of Birbeck granules because the removal of calcium ions causes them to unzip and 51 3 3 disintegrate (Stambach and Taylor, 2003; Bartosik, 1992). It has been proposed that the “face- 52 to-face” interaction between the langerin trimers via CRDs confers the characteristic zippered 53 membranes with a central striation of Birbeck granules (Thepaut et al., 2009) (Figure 1B and 54 1C). However, the precise molecular mechanism of Birbeck granule formation is unknown 55 because of the lack of a high-resolution structure. 56 Langerin’s high affinity for mannose and oligo-mannose allows Langerhans cells to capture 57 a wide range of pathogens, including Saccharomyces cerevisiae, Candida albicans, 58 Mycobacterium leprae, herpes simplex virus, and human immunodeficiency virus-1 (HIV-1) 59 (Tateno et al., 2010; Hunger et al., 2004; de Jong et al., 2010; de Witte et al., 2007; Turville et al., 60 2002). Although Langerhans cells in the mucosa are the initial target of HIV-1 in sexually 61 transmitted infections, the efficiency of Langerhans cell infection is low, requiring high viral 62 loads of HIV-1 to infect cells (Kawamura et al., 2001; Kawamura et al., 2000; Ogawa et al., 2013). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Introduction 35 63 The resistance of Langerhans cells to HIV-1 infection is mediated by the efficient internalization 64 of HIV-1 into Birbeck granules and subsequent autophagic degradation (de Witte et al., 2007; 65 Ribeiro et al., 2016). Therefore, elucidating the mechanism by which HIV-1 is internalized into 66 Birbeck granules is essential to understanding the role of Langerhans cells as a barrier to HIV-1 67 infection. 68 4 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint In the present study, we reconstructed the 3D structure of isolated Birbeck granules using cryo- 69 electron tomography and identified a flexible loop region that is crucial for Birbeck granule 70 formation and virus internalization. 71 72 Results 73 Cryo-electron microscopy of isolated Birbeck granules 74 Exogenous expression of langerin in mammalian cell lines readily reproduces the formation 75 of Birbeck granules (McDermott et al., 2004). We found that the addition of yeast mannan to 76 langerin-expressing cells enhances the formation of large Birbeck granules (Figure 1C). To 77 purify these Birbeck granules, we expressed SNAP-tagged human langerin in 293T cells, 78 biotinylated the tag by incubation with benzylguanine-biotin, and immunoprecipitated the 79 Birbeck granules using streptavidin agarose (Figure 1D and S1A). Under cryo-electron 80 microscopy, the isolated Birbeck granules were observed to be thin lamellar vesicles of 81 assorted sizes, similar to those reported previously (Figure 1E, 1F, and S1B) (Sagebiel and 82 Reed, 1968; Lenormand et al., 2013). Unlike regular 2D crystals of proteins, Birbeck 83 granules appear wavy, suggesting structural heterogeneity. The 2D classification of the 84 segmented images extracted from the micrographs showed nearly three-fold symmetrical 85 In the present study, we reconstructed the 3D structure of isolated Birbeck granules using cryo- 69 electron tomography and identified a flexible loop region that is crucial for Birbeck granule 70 1 Exogenous expression of langerin in mammalian cell lines readily reproduces the formation 75 of Birbeck granules (McDermott et al., 2004). We found that the addition of yeast mannan to 76 langerin-expressing cells enhances the formation of large Birbeck granules (Figure 1C). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Introduction 35 To 77 purify these Birbeck granules, we expressed SNAP-tagged human langerin in 293T cells, 78 biotinylated the tag by incubation with benzylguanine-biotin, and immunoprecipitated the 79 Birbeck granules using streptavidin agarose (Figure 1D and S1A). Under cryo-electron 80 microscopy, the isolated Birbeck granules were observed to be thin lamellar vesicles of 81 assorted sizes, similar to those reported previously (Figure 1E, 1F, and S1B) (Sagebiel and 82 Reed, 1968; Lenormand et al., 2013). Unlike regular 2D crystals of proteins, Birbeck 83 granules appear wavy, suggesting structural heterogeneity. The 2D classification of the 84 segmented images extracted from the micrographs showed nearly three-fold symmetrical 85 5 5 arrangements of pores with a diameter of approximately 10 nm; however, no clear crystalline 86 lattice was observed (Figure 1G). Although most Birbeck granules were oriented in 87 approximately the same direction with respect to the beam axis, some granules were twisted, 88 providing different views that were sufficient for 3D reconstruction (Figure S1C). 89 90 Subtomogram averaging of the repeating unit of Birbeck granules 91 We conducted cryo-electron tomography of the isolated Birbeck granules and reconstructed 92 the repeating units by averaging the subtomograms (Figure 2 and S2A). The initial density 93 map (Figure 2A) shows a langerin trimer (blue) with three inverted trimers (red) attached to 94 the CRDs. However, because of the flexibility between each langerin trimer, the resolution 95 was limited to ~24 Å. To improve the resolution, we generated a mask enclosing the central 96 langerin trimer and three attached CRDs and performed focused refinement and 3D 97 classification (Figure 2B). The resolution was improved to 6.4 Å (Figure S2B), which clearly 98 visualized the alpha helices and enabled us to generate a model based on the previously 99 reported crystal structure of the langerin trimer (Feinberg et al., 2010). We then performed a 100 two-body refinement to visualize the spatial relationship between the two opposing langerin 101 trimers (Nakane et al., 2018). Based on this refinement (Figure 2C), we built a honeycomb 102 arrangements of pores with a diameter of approximately 10 nm; however, no clear crystalline 86 lattice was observed (Figure 1G). Although most Birbeck granules were oriented in 87 approximately the same direction with respect to the beam axis, some granules were twisted, 88 providing different views that were sufficient for 3D reconstruction (Figure S1C). 89 Subtomogram averaging of the repeating unit of Birbeck granules 91 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Subtomogram averaging of the repeating unit of Birbeck granules 91 A previous study found that the secondary calcium-independent 118 binding site has a lower affinity for carbohydrate ligands than the primary calcium-dependent 119 lattice model (Figure 2D and Movie S1), which agrees with the striation of the Birbeck 103 granules observed by ultra-thin section electron microscopy and the ~10 nm pores visualized 104 in the 2D classification images (Figure 1C and 1G) (Birbeck et al., 1961). However, in reality, 105 the honeycomb lattice is non-uniform because of the structural heterogeneity. 106 107 Interaction between the two langerin trimers via flexible loops 108 A i d th l i f th i t ti i b t th t CRD (F 109 Subtomogram averaging of the repeating unit of Birbeck granules 91 We conducted cryo-electron tomography of the isolated Birbeck granules and reconstructed 92 the repeating units by averaging the subtomograms (Figure 2 and S2A). The initial density 93 map (Figure 2A) shows a langerin trimer (blue) with three inverted trimers (red) attached to 94 the CRDs. However, because of the flexibility between each langerin trimer, the resolution 95 was limited to ~24 Å. To improve the resolution, we generated a mask enclosing the central 96 langerin trimer and three attached CRDs and performed focused refinement and 3D 97 classification (Figure 2B). The resolution was improved to 6.4 Å (Figure S2B), which clearly 98 visualized the alpha helices and enabled us to generate a model based on the previously 99 reported crystal structure of the langerin trimer (Feinberg et al., 2010). We then performed a 100 two-body refinement to visualize the spatial relationship between the two opposing langerin 101 trimers (Nakane et al., 2018). Based on this refinement (Figure 2C), we built a honeycomb 102 6 lattice model (Figure 2D and Movie S1), which agrees with the striation of the Birbeck 103 granules observed by ultra-thin section electron microscopy and the ~10 nm pores visualized 104 in the 2D classification images (Figure 1C and 1G) (Birbeck et al., 1961). However, in reality, 105 the honeycomb lattice is non-uniform because of the structural heterogeneity. 106 107 Interaction between the two langerin trimers via flexible loops 108 An in-depth analysis of the interaction region between the two CRDs (Figure 3) revealed that 109 the interaction is mediated by a loop at residues 258-263, which is highly flexible in the 110 crystal structure (Feinberg et al., 2010). The interaction between the two CRDs appears to be 111 maintained by docking the loops of two langerin molecules into each other’s lateral cleft, 112 which constitutes the secondary carbohydrate-binding site (Figure 3A, pink circles) 113 (Chatwell et al., 2008). The interaction between the two CRDs is flexible because the first 114 eigenvector of the multi-body analysis represents an approximately 30° tilting motion with 115 the loop as the fulcrum (Figure 3B, and Movie S2). Interestingly, loop 258-263 of the inverted 116 molecule appears to participate in the secondary carbohydrate-binding site of the upright 117 molecule (Figure 3A). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Interaction between the two langerin trimers via flexible loops 108 CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint binding site (Chatwell et al., 2008). The docked loop 258-263 may provide additional 120 interactions with the hydroxyl group of mannose via Glu261 and Asp263, thus increasing the 121 affinity of the secondary binding site when two langerin trimers form a complex. Furthermore, 122 the Trp264Arg mutation is associated with a lack of Birbeck granules (Verdijk et al., 2005; 123 Figdor et al., 2002). Trp264 may also be involved in docking the loop into the cleft; however, 124 it is more likely that disruption of the hydrophobic core of the CRD is responsible for this 125 defect in Birbeck granule formation (Chatwell et al., 2008). 126 binding site (Chatwell et al., 2008). The docked loop 258-263 may provide additional 120 interactions with the hydroxyl group of mannose via Glu261 and Asp263, thus increasing the 121 affinity of the secondary binding site when two langerin trimers form a complex. Furthermore, 122 the Trp264Arg mutation is associated with a lack of Birbeck granules (Verdijk et al., 2005; 123 Figdor et al., 2002). Trp264 may also be involved in docking the loop into the cleft; however, 124 it is more likely that disruption of the hydrophobic core of the CRD is responsible for this 125 defect in Birbeck granule formation (Chatwell et al., 2008). 126 127 Loop 258-263 is essential for Birbeck granule formation 128 To investigate the importance of loop 258-263 in the biogenesis of Birbeck granules, we 129 mutated three residues in the loop: Met260Ala, Glu261Arg, and Asp263Lys, and generated 130 three mutated langerin constructs, each carrying one (MRGD), two (MRGK), or three 131 mutations (ARGK). We expressed these three langerin mutants in 293T cells and induced 132 Birbeck granule formation by adding yeast mannan (Figure 4A). Under ultra-thin section 133 microscopy, wild-type langerin produced Birbeck granules a few micrometers in length in 134 the presence of mannan (Figure 4A, MEGD WT). As the number of mutations increased, the 135 length of Birbeck granules decreased. With two point mutations (MRGK), some cross- 136 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Interaction between the two langerin trimers via flexible loops 108 An in-depth analysis of the interaction region between the two CRDs (Figure 3) revealed that 109 the interaction is mediated by a loop at residues 258-263, which is highly flexible in the 110 crystal structure (Feinberg et al., 2010). The interaction between the two CRDs appears to be 111 maintained by docking the loops of two langerin molecules into each other’s lateral cleft, 112 which constitutes the secondary carbohydrate-binding site (Figure 3A, pink circles) 113 (Chatwell et al., 2008). The interaction between the two CRDs is flexible because the first 114 eigenvector of the multi-body analysis represents an approximately 30° tilting motion with 115 the loop as the fulcrum (Figure 3B, and Movie S2). Interestingly, loop 258-263 of the inverted 116 molecule appears to participate in the secondary carbohydrate-binding site of the upright 117 molecule (Figure 3A). A previous study found that the secondary calcium-independent 118 binding site has a lower affinity for carbohydrate ligands than the primary calcium-dependent 119 An in-depth analysis of the interaction region between the two CRDs (Figure 3) revealed that 109 the interaction is mediated by a loop at residues 258-263, which is highly flexible in the 110 crystal structure (Feinberg et al., 2010). The interaction between the two CRDs appears to be 111 maintained by docking the loops of two langerin molecules into each other’s lateral cleft, 112 which constitutes the secondary carbohydrate-binding site (Figure 3A, pink circles) 113 (Chatwell et al., 2008). The interaction between the two CRDs is flexible because the first 114 eigenvector of the multi-body analysis represents an approximately 30° tilting motion with 115 the loop as the fulcrum (Figure 3B, and Movie S2). Interestingly, loop 258-263 of the inverted 116 molecule appears to participate in the secondary carbohydrate-binding site of the upright 117 molecule (Figure 3A). A previous study found that the secondary calcium-independent 118 binding site has a lower affinity for carbohydrate ligands than the primary calcium-dependent 119 7 7 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . Loop 258-263 is essential for Birbeck granule formation 128 144 145 Loop 258-263 is crucial for the internalization of HIV pseudovirus 146 To investigate the impact of the mutations in loop 258-263 on virus internalization by Birbeck 147 granules, we added lentiviruses pseudo-typed with the HIV-1 env protein to langerin- 148 expressing 293T cells. Ultra-thin section electron microscopy confirmed internalization of 149 viruses into Birbeck granules (Figure S3A). After 30 min of incubation, we retrieved viruses 150 attached to the surface langerin by washing with EDTA buffer, removed non-specifically 151 attached viruses by trypsinization, and isolated the Birbeck granules using streptavidin- 152 agarose (Figure 5). A langerin mutant that lost its calcium-binding capacity (Stambach and 153 sections became spherical rather than rod-shaped, suggesting that the Birbeck granules failed 137 to extend. With three mutations (ARGK), clusters of Birbeck granules were not observed, 138 with isolated individual granules observed instead. The length of individual Birbeck granules 139 and the total length of granules per area were measured to confirm the above observations 140 (Figure 4B and 4C). In contrast, the number of the Birbeck granules showed less significant 141 differences because numerous short fragments were present in the mutant specimen (Figure 142 4D). These results demonstrate the importance of loop 258-263 in the biogenesis of Birbeck 143 granules. 144 Loop 258-263 is essential for Birbeck granule formation 128 To investigate the importance of loop 258-263 in the biogenesis of Birbeck granules, we 129 mutated three residues in the loop: Met260Ala, Glu261Arg, and Asp263Lys, and generated 130 three mutated langerin constructs, each carrying one (MRGD), two (MRGK), or three 131 mutations (ARGK). We expressed these three langerin mutants in 293T cells and induced 132 Birbeck granule formation by adding yeast mannan (Figure 4A). Under ultra-thin section 133 microscopy, wild-type langerin produced Birbeck granules a few micrometers in length in 134 the presence of mannan (Figure 4A, MEGD WT). As the number of mutations increased, the 135 length of Birbeck granules decreased. With two point mutations (MRGK), some cross- 136 To investigate the importance of loop 258-263 in the biogenesis of Birbeck granules, we 129 mutated three residues in the loop: Met260Ala, Glu261Arg, and Asp263Lys, and generated 130 three mutated langerin constructs, each carrying one (MRGD), two (MRGK), or three 131 mutations (ARGK). We expressed these three langerin mutants in 293T cells and induced 132 Birbeck granule formation by adding yeast mannan (Figure 4A). Under ultra-thin section 133 microscopy, wild-type langerin produced Birbeck granules a few micrometers in length in 134 the presence of mannan (Figure 4A, MEGD WT). As the number of mutations increased, the 135 length of Birbeck granules decreased. With two point mutations (MRGK), some cross- 136 8 8 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint sections became spherical rather than rod-shaped, suggesting that the Birbeck granules failed 137 to extend. With three mutations (ARGK), clusters of Birbeck granules were not observed, 138 with isolated individual granules observed instead. The length of individual Birbeck granules 139 and the total length of granules per area were measured to confirm the above observations 140 (Figure 4B and 4C). In contrast, the number of the Birbeck granules showed less significant 141 differences because numerous short fragments were present in the mutant specimen (Figure 142 4D). These results demonstrate the importance of loop 258-263 in the biogenesis of Birbeck 143 granules. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Loop 258-263 is crucial for the internalization of HIV pseudovirus 146 To investigate the impact of the mutations in loop 258-263 on virus internalization by Birbeck 147 granules, we added lentiviruses pseudo-typed with the HIV-1 env protein to langerin- 148 expressing 293T cells. Ultra-thin section electron microscopy confirmed internalization of 149 viruses into Birbeck granules (Figure S3A). After 30 min of incubation, we retrieved viruses 150 attached to the surface langerin by washing with EDTA buffer, removed non-specifically 151 attached viruses by trypsinization, and isolated the Birbeck granules using streptavidin- 152 agarose (Figure 5). A langerin mutant that lost its calcium-binding capacity (Stambach and 153 9 9 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Taylor, 2003) was used as the negative control (Figure 5, lectin (-)). The HIV pseudoviruses 154 attached to both wild-type and ARGK mutant langerin and the attachment was abolished by 155 the addition of mannan (Figure 5A). This indicated that the mutation of the loop did not 156 disrupt the binding of the glycosylated envelope proteins to the primary carbohydrate- 157 binding sites. In contrast, the internalization of the pseudovirus was significantly inhibited 158 by the ARGK mutation (Figure 5B and 5C). These results suggest that disruption of Birbeck 159 granule formation by the mutation of loop 258-263 inhibited the internalization of HIV 160 pseudovirus. 161 162 Discussion 163 Virus internalization mechanism of Birbeck granules 164 In the present study, we reconstructed the 3D structure of isolated Birbeck granules and found 165 that docking of the flexible loop 258-263 into the secondary carbohydrate-binding cleft is 166 essential for assembly of the langerin lattice. In agreement with this observation, we showed 167 that mutations introduced into the loop inhibited Birbeck granule formation and 168 internalization of HIV pseudovirus. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Loop 258-263 is crucial for the internalization of HIV pseudovirus 146 Based on the structure of the glycosylated HIV env 169 protein (Stewart-Jones et al., 2016; Pan et al., 2020), we propose a model of HIV 170 Taylor, 2003) was used as the negative control (Figure 5, lectin (-)). The HIV pseudoviruses 154 attached to both wild-type and ARGK mutant langerin and the attachment was abolished by 155 the addition of mannan (Figure 5A). This indicated that the mutation of the loop did not 156 disrupt the binding of the glycosylated envelope proteins to the primary carbohydrate- 157 binding sites. In contrast, the internalization of the pseudovirus was significantly inhibited 158 by the ARGK mutation (Figure 5B and 5C). These results suggest that disruption of Birbeck 159 granule formation by the mutation of loop 258-263 inhibited the internalization of HIV 160 pseudovirus. 161 162 Discussion 163 Virus internalization mechanism of Birbeck granules 164 In the present study, we reconstructed the 3D structure of isolated Birbeck granules and found 165 that docking of the flexible loop 258-263 into the secondary carbohydrate-binding cleft is 166 essential for assembly of the langerin lattice. In agreement with this observation, we showed 167 that mutations introduced into the loop inhibited Birbeck granule formation and 168 internalization of HIV pseudovirus. Based on the structure of the glycosylated HIV env 169 protein (Stewart-Jones et al., 2016; Pan et al., 2020), we propose a model of HIV 170 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Virus internalization mechanism of Birbeck granules 164 10 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint internalization by langerin (Figure 6A and 6B). A previous study proposed that N-linked 171 high-mannose oligosaccharides bind to langerin at both primary and secondary carbohydrate- 172 binding sites (Chatwell et al., 2008). Our model requires bending of the coiled-coil necks of 173 the two langerin trimers to form a complex with HIV-1 env via the high-mannose 174 oligosaccharide (Figure 6B). Although the triple coiled-coil structure of the neck domain 175 does not seem to bend strongly, the coiled-coil prediction profile suggests that the N-terminal 176 part of the neck does not necessarily form a rigid triple coiled coil (Lupas et al., 1991) (Figure 177 S1D). Thus, we speculate that the neck region of the langerin trimer is flexible and swings 178 on the cell surface, allowing loop 258-263 of one trimer to dock into the cleft of an adjacent 179 trimer. Once the triad structure, consisting of two langerin trimers and one env protein, is 180 formed, the elasticity of the neck region can deform the cell membrane and initiate 181 invagination. The formation of clathrin pits may also be involved in membrane invagination, 182 because the cytoplasmic ends of Birbeck granules are coated with clathrin (Mc et al., 2002). 183 Additionally, it is possible that other intracellular mechanisms such as the actin cytoskeleton, 184 are involved in this membrane deformation process because annexin A2 is required for the 185 proper formation of Birbeck granules (Thornton et al., 2020). 186 187 187 11 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Significance of langerin-specific secondary carbohydrate-binding cleft 188 It is noteworthy that the docking of loop 258-263 into the secondary calcium-independent 189 carbohydrate-binding cleft, which is not found in other C-type lectins (Chatwell et al., 2008), 190 mediates the assembly of the langerin lattice. In DC-SIGN, which is another C-type lectin 191 expressed in dendritic cells, and is closely related to langerin, the region corresponding to the 192 secondary carbohydrate-binding site of langerin forms a shallow pocket, that is occupied by 193 two calcium ions (Figure 7A) (Guo et al., 2004; Geijtenbeek et al., 2000; Geijtenbeek et al., 194 2000). This structural difference between langerin and DC-SIGN may account for the 195 inability of DC-SIGN to form Birbeck granules despite its high sequence similarity to 196 langerin (Cambi et al., 2009). Another characteristic of the secondary carbohydrate-binding 197 site of langerin is its weak affinity for mannose and maltose (Chatwell et al., 2008). This may 198 reflect that the sugar-binding surface of langerin CRD is less negatively charged than that of 199 DC-SIGN CRD (Figure 7A) (Chatwell et al., 2008). Moreover, the interaction between the 200 secondary binding site and mannose/maltose is mainly mediated by backbone amide and 201 carbonyl groups (Chatwell et al., 2008). It is possible that the side chains of Glu261 and 202 Asp263 of loop 258-263 docked into the cleft of an adjacent langerin CRD provide the 203 negative charge required for stabilizing oligo-mannose bound to the secondary binding site 204 12 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint (Figure 7A and B). In agreement with this hypothesis, in the present study the Asp263Lys 205 mutation caused a significant defect in Birbeck granule formation. The secondary 206 carbohydrate-binding site may only be fully functional when two langerin trimers are bound 207 to each other (Figure 6 and 7B). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Significance of langerin-specific secondary carbohydrate-binding cleft 188 The effect of the Met260Ala mutation on Birbeck granule 208 formation was unexpected, because the mutation was supposed to be silent in terms of side 209 chain charge. The observed structural variation of the langerin trimer complex suggests that 210 the side chain of Met260 acts as a hook to stabilize the interface between the two intersecting 211 loops. 212 Our model of Birbeck granule formation suggests that oligosaccharide ligands are captured 213 by the lateral surface of langerin (Figure 6). In contrast, DC-SIGN binds to dengue virus on 214 its upper surface (Pokidysheva et al., 2006). The difference in the ligand-binding manner 215 between langerin and DC-SIGN may contribute to their different carbohydrate recognition 216 preferences (Valverde et al., 2020; Takahara et al., 2004). Further studies are needed to 217 investigate the physiological significance of the unique carbohydrate-binding mechanism of 218 langerin and the honeycomb lattice structure of Birbeck granules. 219 220 (Figure 7A and B). In agreement with this hypothesis, in the present study the Asp263Lys 205 mutation caused a significant defect in Birbeck granule formation. The secondary 206 carbohydrate-binding site may only be fully functional when two langerin trimers are bound 207 to each other (Figure 6 and 7B). The effect of the Met260Ala mutation on Birbeck granule 208 formation was unexpected, because the mutation was supposed to be silent in terms of side 209 chain charge. The observed structural variation of the langerin trimer complex suggests that 210 the side chain of Met260 acts as a hook to stabilize the interface between the two intersecting 211 loops. 212 Our model of Birbeck granule formation suggests that oligosaccharide ligands are captured 213 by the lateral surface of langerin (Figure 6). In contrast, DC-SIGN binds to dengue virus on 214 its upper surface (Pokidysheva et al., 2006). The difference in the ligand-binding manner 215 between langerin and DC-SIGN may contribute to their different carbohydrate recognition 216 preferences (Valverde et al., 2020; Takahara et al., 2004). Further studies are needed to 217 investigate the physiological significance of the unique carbohydrate-binding mechanism of 218 langerin and the honeycomb lattice structure of Birbeck granules. 219 220 langerin and the honeycomb lattice structure of Birbeck granules. 219 13 13 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Isolation of Birbeck granules 223 The cDNA of human langerin was obtained from the Mammalian Gene Collection library 224 (Dnaform Inc., Yokohama, Japan) and was subcloned into pcDNA3.1-IRES-GFP (Schaefer 225 et al., 2008). A codon-optimized sequence coding SNAP-3×HA-HRV3C cleavage site 226 (obtained from ThermoFisher Scientific, Waltham, MA) was inserted immediately before the 227 start codon of the langerin sequence (Cole, 2013). O6-benzylguanine-biotin was obtained by 228 reacting 16 mM O6-benzylguanine-amine (Accela ChemBio, Shanghai, China) with 50 mM 229 biotin-AC5-Osu (Dojindo, Kumamoto, Japan) and 30 mM triethylamine in 230 dimethylformamide for 24 hours at 30℃. Lenti-X 293T cells (TakaraBio, Tokyo, Japan) were 231 transfected with the langerin-expression plasmid by electroporation using ECM630 232 electroporator (BTX, Holliston, MA) and seeded onto the polyethylenimine-coated dish (PEI, 233 obtained from Fujifilm Wako Chemicals, Osaka, Japan). After 48-hour culturing in 234 DMEM+10% fetal bovine serum (FBS), the culture medium was replaced with serum-free 235 DMEM containing 10 µM O6-benzylguanine-biotin and 10 µg/ml yeast mannan (Nacalai 236 tesque, Kyoto, Japan) and the cells were incubated for 30 min at 37 ℃. The cells were then 237 collected by scraping and washed five times with isotonic HMC buffer (Hepes 30 mM, pH 238 The cDNA of human langerin was obtained from the Mammalian Gene Collection library 224 (Dnaform Inc., Yokohama, Japan) and was subcloned into pcDNA3.1-IRES-GFP (Schaefer 225 et al., 2008). A codon-optimized sequence coding SNAP-3×HA-HRV3C cleavage site 226 (obtained from ThermoFisher Scientific, Waltham, MA) was inserted immediately before the 227 start codon of the langerin sequence (Cole, 2013). O6-benzylguanine-biotin was obtained by 228 reacting 16 mM O6-benzylguanine-amine (Accela ChemBio, Shanghai, China) with 50 mM 229 biotin-AC5-Osu (Dojindo, Kumamoto, Japan) and 30 mM triethylamine in 230 dimethylformamide for 24 hours at 30℃. Lenti-X 293T cells (TakaraBio, Tokyo, Japan) were 231 transfected with the langerin-expression plasmid by electroporation using ECM630 232 electroporator (BTX, Holliston, MA) and seeded onto the polyethylenimine-coated dish (PEI, 233 obtained from Fujifilm Wako Chemicals, Osaka, Japan). After 48-hour culturing in 234 DMEM+10% fetal bovine serum (FBS), the culture medium was replaced with serum-free 235 DMEM containing 10 µM O6-benzylguanine-biotin and 10 µg/ml yeast mannan (Nacalai 236 tesque, Kyoto, Japan) and the cells were incubated for 30 min at 37 ℃. The cells were then 237 collected by scraping and washed five times with isotonic HMC buffer (Hepes 30 mM, pH 238 14 . . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Isolation of Birbeck granules 223 CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 7.2, 150 mM NaCl, 2 mM MgCl2, 2 mM CaCl2, protease inhibitor cocktail (Nacalai tesque)). 239 The washed cells were homogenized with a Dounce glass homogenizer and were further 240 disrupted by sonication for 8 seconds using a Q125 sonicator (Qsonica, Newtown, CT). The 241 cell homogenate was loaded on top of a discontinuous Opti-prep density gradient (10-20- 242 40% layers, ThermoFisher Scientific) and ultracentrifuged for 1 hour at 100,000×g. The 243 fraction at the interface of 20/40% layers was collected and diluted 10-fold with HMC buffer 244 and was again disrupted by sonication for 2 min (120 cycles of 1-sec pulse and 3-sec wait) 245 at 4℃. Large debris was removed by filtration through a 20 µm-pore polyethylene mesh (GL 246 Science, Tokyo, Japan) and the cell lysate was incubated with a 30 µl slurry of streptavidin 247 agarose (Solulink, San Diego, CA) overnight at 4℃. The streptavidin agarose beads were 248 washed five times with HMC buffer and the bound Birbeck granules were eluted by overnight 249 digestion using home-made E. coli overexpressed His×6-HRV3C protease in HMC buffer at 250 4℃. After digestion and elution, HRV3C protease was removed by incubating the elute with 251 Ni-NTA resin (Qiagen, Düsseldorf, Germany). About 50 µl of elute containing released 252 Birbeck granules (approximately 0.05 mg/ml) were obtained from twelve 100-mm dishes of 253 cell culture. 254 255 15 15 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . Isolation of Birbeck granules 223 CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Cryo-electron microscopy of isolated Birbeck granules 256 5 µl of the purified Birbeck granules was applied onto freshly glow-discharged holey carbon 257 grids, Quantifoil R1.2/1.3 Cu/Rh 200 mesh (Quantifoil Micro Tools GmbH, Großlöbichau, 258 Germany), blotted from the backside to concentrate the granules by filtration through the 259 holes, and 4 µl of HMC buffer containing cytochrome c-stabilized 15-nm colloidal gold (BBI 260 Solutions) was applied for attachment of fiducial markers (Oda et al., 2015). Grids were 261 blotted from the backside for 10 seconds at 4℃ under 99% humidity and plunge frozen in 262 liquid ethane using Vitrobot Mark IV (Thermo Fisher Scientific). We removed one of the 263 blotting arms for one-side blotting because Birbeck granules would be absorbed onto the 264 filter paper. Images were recorded using a Titan Krios G3i microscope at the University of 265 Tokyo (Thermo Fisher Scientific) at 300 keV, a Gatan Quantum-LS Energy Filter (Gatan, 266 Pleasanton, CA) with a slit width of 20 eV, and a Gatan K3 Summit direct electron detector 267 in the electron counting mode. The nominal magnification was set to 33,000× with a physical 268 pixel size of 2.67 Å/pixel. Movies were acquired using the SerialEM software (Mastronarde, 269 2005), and the target defocus was set to 3-6 µm. The angular range of the tilt series was from 270 -60° to 60° with 3.0° increments. Each movie was recorded for 0.74 sec with a total dose of 271 1.24 electrons/Å2 and subdivided into 20 frames. The total dose for one tilt series acquisition 272 16 is thus 50 electrons/Å2. 273 274 Data processing 275 Movies were subjected to beam-induced motion correction using MotionCor2 (Zheng et al., 276 2017), and tilt series images were aligned, CTF corrected, and back-projected to reconstruct 277 3D tomograms using the IMOD software package (Kremer et al., 1996). For the subsequent 278 analysis using Relion-4.0 (Kimanius et al., 2021), the tomogram positioning step was skipped 279 to keep the angle offset and the Z-shift zero. Tomograms were 4×binned to reduce loads of 280 the calculation. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Isolation of Birbeck granules 223 The areas of the Birbeck granules were manually selected using the 3dmod 281 model tool (Kremer et al., 1996), and the center coordinates for the subtomograms were 282 determined by placing points at 8-pixel (~8.5 nm) intervals within the selected area using a 283 custom-made Ruby script (Metlagel et al., 2007). For twisting Birbeck granules (Figure S1C), 284 center points were picked manually by selecting the mid-portion of the lamella. Volumes with 285 32×32×32 pixel-dimensions were extracted from 8×binned tomograms and were averaged 286 using the PEET software suite (Nicastro et al., 2006). A randomly selected subtomogram was 287 used for the initial reference. Alignments were repeated five times for 8×binned and twice 288 for 4×binned tomograms with 64×64×64 pixel-dimensions. The coordinates and the rotation 289 17 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint angles of the aligned subtomograms were exported to Relion-4.0. Initial 290 pseudosubtomograms were generated with the box size of 128 pixels, the cropped box size 291 of 64 pixels, and the binning factor of 4. The subtomograms were 3D refined using a spherical 292 mask of 320-Å diameter. After 3D classification, 2×binned pseudosubtomograms were 293 generated with the box size of 256 pixels and the cropped box size of 128 pixels based on the 294 refined coordinates and angles and were 3D refined using the same spherical mask. Finally, 295 un-binned pseudosubtomograms were generated with a box size of 256 pixels and a cropped 296 box size of 128 pixels. 3D refinement was done using a tighter mask enclosing the central 297 upright langerin molecule and three bound CRDs, and low-resolution subtomograms were 298 removed by 3D classification. The alignment was further refined using CTF refinement and 299 frame alignment. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Isolation of Birbeck granules 223 The Coulombic surfaces rendering was conducted using UCSF 308 Chimera and other model presentations, surface rendering, and morph movie generation were 309 conducted using ChimeraX (Goddard et al., 2018). The data processing and the statistics of 310 the model validation are summarized in Figure S2A and Table S1. 311 (Adams et al., 2010). The Coulombic surfaces rendering was conducted using UCSF 308 Chimera and other model presentations, surface rendering, and morph movie generation were 309 conducted using ChimeraX (Goddard et al., 2018). The data processing and the statistics of 310 the model validation are summarized in Figure S2A and Table S1. 311 Isolation of Birbeck granules 223 We initially refined the surrounding inverted langerin molecule by 3D 300 refinement masking out the central upright molecule, but the refinement was biased toward 301 the center. To improve the map quality of the inverted molecule, we applied 3D multi-body 302 refinement (Nakane et al., 2018), which successfully compensated for the flexibility between 303 the upright and the inverted molecules. For model building, the crystal structure of langerin 304 trimer (Feinberg et al., 2010) (PDB ID: 3KQG) was fitted into the map using UCSF Chimera 305 (Pettersen et al., 2004). Then, the fitted model was further refined using PHENIX real-space 306 18 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint refinement tool, and the refined model was validated using the comprehensive validation tool 307 (Adams et al., 2010). The Coulombic surfaces rendering was conducted using UCSF 308 Chimera and other model presentations, surface rendering, and morph movie generation were 309 conducted using ChimeraX (Goddard et al., 2018). The data processing and the statistics of 310 the model validation are summarized in Figure S2A and Table S1. 311 312 Ultrathin section electron microscopy of Birbeck granules 313 Pellets of 293T cells with mannan-induced Birbeck granules were fixed with 4% 314 paraformaldehyde for 1 hour at 4℃ and stained with 1% osmium tetroxide and subsequently 315 with 1% uranium acetate. After dehydration in ethanol and acetone, the samples were 316 embedded in Quetol 812 resin (Nissin EM, Tokyo, Japan). 60-nm ultrathin sections were cut 317 using a ULTRACUT microtome (Reichert Leica) and mounted onto Formvar-coated copper 318 grid. Images were recorded using a JEM-2100F microscope (JEOL, Tokyo, Japan) at 319 University of Yamanashi operated at 200 keV equipped with an F216 CMOS camera (TVIPS 320 GmbH, Gauting, Germany). Length of Birbeck granules were measured using Fiji software 321 (Schindelin et al., 2012), and plots were generated using PlotsOfData web tool (Postma and 322 Goedhart, 2019). For total length of Birbeck granules per 100 µm2, cross sections of cells 323 (Adams et al., 2010). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Ultrathin section electron microscopy of Birbeck granules 313 Pellets of 293T cells with mannan-induced Birbeck granules were fixed with 4% 314 paraformaldehyde for 1 hour at 4℃ and stained with 1% osmium tetroxide and subsequently 315 with 1% uranium acetate. After dehydration in ethanol and acetone, the samples were 316 embedded in Quetol 812 resin (Nissin EM, Tokyo, Japan). 60-nm ultrathin sections were cut 317 using a ULTRACUT microtome (Reichert Leica) and mounted onto Formvar-coated copper 318 grid. Images were recorded using a JEM-2100F microscope (JEOL, Tokyo, Japan) at 319 University of Yamanashi operated at 200 keV equipped with an F216 CMOS camera (TVIPS 320 GmbH, Gauting, Germany). Length of Birbeck granules were measured using Fiji software 321 (Schindelin et al., 2012), and plots were generated using PlotsOfData web tool (Postma and 322 Goedhart, 2019). For total length of Birbeck granules per 100 µm2, cross sections of cells 323 Pellets of 293T cells with mannan-induced Birbeck granules were fixed with 4% 314 paraformaldehyde for 1 hour at 4℃ and stained with 1% osmium tetroxide and subsequently 315 with 1% uranium acetate. After dehydration in ethanol and acetone, the samples were 316 embedded in Quetol 812 resin (Nissin EM, Tokyo, Japan). 60-nm ultrathin sections were cut 317 using a ULTRACUT microtome (Reichert Leica) and mounted onto Formvar-coated copper 318 grid. Images were recorded using a JEM-2100F microscope (JEOL, Tokyo, Japan) at 319 University of Yamanashi operated at 200 keV equipped with an F216 CMOS camera (TVIPS 320 GmbH, Gauting, Germany). Length of Birbeck granules were measured using Fiji software 321 (Schindelin et al., 2012), and plots were generated using PlotsOfData web tool (Postma and 322 Goedhart, 2019). For total length of Birbeck granules per 100 µm2, cross sections of cells 323 19 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint with nuclei of > 5 µm in diameter were selected, and the sum of the lengths of Birbeck 324 granules within the cells were measured. The areas of the cross sections were also measured 325 using Fiji software. 100 µm2 corresponds to the approximate mean of the cross-sectional area 326 of one cell. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint The obtained supernatants were immediately used for subsequent assays. 340 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Ultrathin section electron microscopy of Birbeck granules 313 327 328 Lentivirus preparation 329 Lentiviruses pseudotyped with HIV-1 env were prepared by transfecting the Lenti-X 293T 330 cells with pCMV-dR8.3 Dvpr (Addgene plasmid #8455), pLOX-CW-tdTomato, and pCEP4- 331 AD8gp160 (Adgene plasmid #123260) (Stewart et al., 2003; Heredia et al., 2019). The vector 332 pLOX-CW-tdTomato was generated by replacing the GFP sequence of pLOX-CWgfp 333 (Addgene plasmid #12241) (Salmon et al., 2000) with the tdTomato sequence derived from 334 pCDH-EF1-Luc2-P2A-tdTomato, a gift from Kazuhiro Oka (Addgene plasmid # 72486). The 335 transfections were carried out using the PEI-method with the ratio at PEI: dR8.3: tdTomato: 336 gp160 = 4:1:1:1. 16 hours post-transfection, the cells were washed three times with PBS and 337 incubated in a fresh medium for another 32 hours. The virus-containing medium was 338 collected and removed cell debris by a 1,000×g centrifuge and a 0.45 µm filtration (Millipore). 339 The obtained supernatants were immediately used for subsequent assays. 340 using Fiji software. 100 µm2 corresponds to the approximate mean of the cross-sectional area 326 of one cell. 327 Lentiviruses pseudotyped with HIV-1 env were prepared by transfecting the Lenti-X 293T 330 cells with pCMV-dR8.3 Dvpr (Addgene plasmid #8455), pLOX-CW-tdTomato, and pCEP4- 331 AD8gp160 (Adgene plasmid #123260) (Stewart et al., 2003; Heredia et al., 2019). The vector 332 pLOX-CW-tdTomato was generated by replacing the GFP sequence of pLOX-CWgfp 333 (Addgene plasmid #12241) (Salmon et al., 2000) with the tdTomato sequence derived from 334 pCDH-EF1-Luc2-P2A-tdTomato, a gift from Kazuhiro Oka (Addgene plasmid # 72486). The 335 transfections were carried out using the PEI-method with the ratio at PEI: dR8.3: tdTomato: 336 gp160 = 4:1:1:1. 16 hours post-transfection, the cells were washed three times with PBS and 337 incubated in a fresh medium for another 32 hours. The virus-containing medium was 338 collected and removed cell debris by a 1,000×g centrifuge and a 0.45 µm filtration (Millipore). 339 The obtained supernatants were immediately used for subsequent assays. 340 20 341 Virus attachment and internalization assay 342 Virus attachment and internalization assay 342 293T cells were transfected with wild type langerin, the ARGK mutant, and, as a negative 343 control, a lectin (-) mutant lacking calcium-binding ability. The lectin (-) langerin mutant was 344 generated by introducing three point-mutations at the calcium-binding site: Glu285Lys; 345 Asn287Lys; and Glu293Lys (Chatwell et al., 2008). We conducted surface labeling of 346 langerin-expressing cells and confirmed that these mutations did not alter the surface 347 expression of langerin (Figure S3B). 48 hours post-transfection, the cells were pre-treated 348 with or without 10 µg/ml yeast mannan for 5 min at 37℃. The cells were then incubated in 349 the pseudovirus-containing supernatants plus 10 µM O6-benzylguanine-biotin and 10 µg/ml 350 DEAE-dextran (Sigma Aldrich, St. Louis, MO) with or without 10 µg/ml yeast mannan for 351 30 min at 37℃. After the incubation, supernatant containing unbound viruses were collected. 352 The cells were then washed with TBS-Ca buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 5 353 mM CaCl2) three times, and the viruses attached to langerin at the cell surface was retrieved 354 by washing with TBS-EDTA buffer (20 mM Tris-HCl pH 7.4, 150 mM NaCl, 10 mM EDTA). 355 The collected supernatant of the medium and the TBS-EDTA buffer were centrifuged at 356 1,000×g for 5 min to remove cell debris, and the viruses were sedimented by ultra- 357 21 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint centrifugation at 100,000×g for 3 hours. After the TBS-EDTA washing, non-specifically 358 bound viruses were removed by trypsinization using 0.25 % (w/v) trypsin and 1mM EDTA 359 for 5 min at 37℃ as previously described (Burkard et al., 2014). Detached cells were washed 360 twice with DMEM+10 % FBS, and five times with TBS-Ca. Internalized viruses within 361 Birbeck granules were isolated from the washed cells by streptavidin-agarose precipitation 362 as described above. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Virus attachment and internalization assay 342 Viruses, langerin, and tubulin were detected by immunoblotting using 363 HIV-1 gag p24 antibody (1:1000 dilution, MAB7360, mouse monoclonal, R&D systems, 364 Minneapolis, MN), human CD207 antibody (1:1000 dilution, 11841-1-AP, rabbit polyclonal, 365 Proteintech Group Inc., Rosemont, IL), and tubulin antibody (1:3000 dilution, B-512, mouse 366 monoclonal, Sigma-Aldrich), respectively. ELSA was carried out using HIV Type 1 p24 367 Antigen ELISA kit (ZeptoMatrix, Buffalo, NY). The p24 antibody detected unprocessed 368 (Pr55gag) and fully processed (p24) capsid proteins (Gonelli et al., 2019). Band intensities 369 were measured using ImageJ (Schneider et al., 2012). Surface labeling was carried out by 370 treating the cells with 0.5mM biotin-sulfo-osu (Dojindo, Osaka, Japan) for 30 min at 37℃. 371 Cells were washed three times with TBS and collected by scraping. Surface-expressed 372 proteins were extracted by incubating the cells in TBS plus 1% Triton X-100 for 1 hour at 373 4℃, and immunoprecipitated using streptavidin agarose. ELISA using p24 antibody showed 374 22 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint that >99% of the viruses remained in the supernatant (Figure S3C). 375 376 Isolation of langerin-expressing stable cell lines 377 We isolated a cell line stably expressing langerin by selecting HEK293 cells using G418 378 (Sigma Aldrich). However, the expression levels of the isolated cell line was 10~20% of that 379 of the transiently expressing cells (Figure S3D). These cells can form Birbeck granules in the 380 presence of yeast mannan, but with lower efficiency (Figure S3E and F). Therefore, we chose 381 to use the transient expression system in the main experiments. 382 383 384 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Isolation of langerin-expressing stable cell lines 377 We isolated a cell line stably expressing langerin by selecting HEK293 cells using G418 378 (Sigma Aldrich). However, the expression levels of the isolated cell line was 10~20% of that 379 of the transiently expressing cells (Figure S3D). These cells can form Birbeck granules in the 380 presence of yeast mannan, but with lower efficiency (Figure S3E and F). Therefore, we chose 381 We isolated a cell line stably expressing langerin by selecting HEK293 cells using G418 378 (Sigma Aldrich). However, the expression levels of the isolated cell line was 10~20% of that 379 of the transiently expressing cells (Figure S3D). These cells can form Birbeck granules in the 380 presence of yeast mannan, but with lower efficiency (Figure S3E and F). Therefore, we chose 381 to use the transient expression system in the main experiments. 382 23 23 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Author contributions 385 T.O. conceived the project and designed the experiments. T.O., H.Y., and H.S. performed the 386 experiments. T.O., and H.Y. analyzed the data. T.O., H.Y., Y.O., and T.K. wrote the 387 manuscript. 388 389 Data availability 390 The map and the model will be available on the EMDB upon publication under the following 391 accession numbers: EMD-32906, and PDB 7WZ8. 392 393 Acknowledgements 394 We thank Mrs. Natsuko Maruyama (University of Yamanashi) for technical assistance. This 395 research is partially supported by Platform Project for Supporting Drug Discovery and Life 396 Science Research (Basis for Supporting Innovative Drug Discovery and Life Science 397 Research (BINDS)) from Japan Agency for Medical Research and Development (AMED) 398 under Grant Number JP19am0101115. This research was conducted using the Fujitsu 399 PRIMERGY CX400M1/CX2550M5 (Oakbridge-CX) and the SGI Rackable C2112- 400 4GP3/C1102-GP8 (Reedbush-H/L) in the Information Technology Center, The University of 401 Author contributions 385 T.O. conceived the project and designed the experiments. T.O., H.Y., and H.S. performed the 386 experiments. T.O., and H.Y. analyzed the data. T.O., H.Y., Y.O., and T.K. wrote the 387 manuscript. 388 g We thank Mrs. Natsuko Maruyama (University of Yamanashi) for technical assistance. This 395 research is partially supported by Platform Project for Supporting Drug Discovery and Life 396 Science Research (Basis for Supporting Innovative Drug Discovery and Life Science 397 Research (BINDS)) from Japan Agency for Medical Research and Development (AMED) 398 under Grant Number JP19am0101115. This research was conducted using the Fujitsu 399 PRIMERGY CX400M1/CX2550M5 (Oakbridge-CX) and the SGI Rackable C2112- 400 4GP3/C1102-GP8 (Reedbush-H/L) in the Information Technology Center, The University of 401 24 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Tokyo. This work was supported by the Takeda Science Foundation (to T.O.), the Daiichi 402 Sankyo Foundation of Life Science (to T.O.), the Japan Society for the Promotion of Science 403 (KAKENHI Grant number JP21H02654 to T.O.), and the Naito Foundation (to T.O.). 404 Molecular graphics and analyses performed with UCSF ChimeraX, developed by the 405 Resource for Biocomputing, Visualization, and Informatics at the University of California, 406 San Francisco, with support from National Institutes of Health R01-GM129325 and the 407 Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and 408 Infectious Diseases. 409 Tokyo. This work was supported by the Takeda Science Foundation (to T.O.), the Daiichi 402 Sankyo Foundation of Life Science (to T.O.), the Japan Society for the Promotion of Science 403 (KAKENHI Grant number JP21H02654 to T.O.), and the Naito Foundation (to T.O.). . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 References 412 References 413 Adams, P.D., Afonine, P.V., Bunkoczi, G., Chen, V.B., Davis, I.W., Echol 414 Adams, P.D., Afonine, P.V., Bunkoczi, G., Chen, V.B., Davis, I.W., Echols, N., Headd, 414 J.J., Hung, L.W., Kapral, G.J., Grosse-Kunstleve, R.W. et al. 2010. PHENIX: a 415 comprehensive Python-based system for macromolecular structure solution. Acta. 416 Crystallogr. D. Biol. Crystallogr. 66, 213-221. 417 J.J., Hung, L.W., Kapral, G.J., Grosse-Kunstleve, R.W. et al. 2010. PHENIX: a 415 comprehensive Python based system for macromolecular structure solution Acta 416 J.J., Hung, L.W., Kapral, G.J., Grosse-Kunstleve, R.W. et al. 2010. PHENIX: a 415 comprehensive Python-based system for macromolecular structure solution. Acta. 416 Crystallogr D Biol Crystallogr 66 213 221 417 Crystallogr. D. Biol. Crystallogr. 66, 213-221. Bartosik, J. 1992. Cytomembrane-derived Birbeck granules transport horseradish 418 peroxidase to the endosomal compartment in the human Langerhans cells. J. Invest. 419 Dermatol. 99, 53-58. 420 Birbeck, M.S., Breathnach, A.S., and Everall, J.D. 1961. An electron microscope study of 421 basal melanocytes and high-level clear cells (Langerhans cells) in vitiligo. J. Invest. 422 Dermatol. 37, 51-63. 423 Burkard, C., Bloyet, L.M., Wicht, O., van, Kuppeveld F.J., Rottier, P.J., de, Haan C.A., and 424 Bosch, B.J. 2014. Dissecting virus entry: replication-independent analysis of virus binding, 425 internalization, and penetration using minimal complementation of beta-galactosidase. 426 PLoS. One. 9, e101762. 427 g y p p y g internalization, and penetration using minimal complementation of beta-galactosidase. 426 PLoS. One. 9, e101762. 427 Cambi, A., Beeren, I., Joosten, B., Fransen, J.A., and Figdor, C.G. 2009. The C-type lectin 428 DC-SIGN internalizes soluble antigens and HIV-1 virions via a clathrin-dependent 429 mechanism. Eur. J. Immunol. 39, 1923-1928. 430 Chatwell, L., Holla, A., Kaufer, B.B., and Skerra, A. 2008. The carbohydrate recognition 431 domain of Langerin reveals high structural similarity with the one of DC-SIGN but an 432 additional, calcium-independent sugar-binding site. Mol. Immunol. 45, 1981-1994. 433 Chatwell, L., Holla, A., Kaufer, B.B., and Skerra, A. 2008. The carbohydra 431 Chatwell, L., Holla, A., Kaufer, B.B., and Skerra, A. 2008. The carbohydrate recognition 431 domain of Langerin reveals high structural similarity with the one of DC-SIGN but an 432 domain of Langerin reveals high structural similarity with the one of DC-SIGN but an 432 additional, calcium-independent sugar-binding site. Mol. Immunol. 45, 1981-1994. 433 Cole, N.B. 2013. Site-specific protein labeling with SNAP-tags. Curr. Protoc. Protein. Sci. 434 3 30 1 1 30 1 16 435 Cole, N.B. 2013. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Author contributions 385 404 Molecular graphics and analyses performed with UCSF ChimeraX, developed by the 405 Resource for Biocomputing, Visualization, and Informatics at the University of California, 406 San Francisco, with support from National Institutes of Health R01-GM129325 and the 407 Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and 408 Infectious Diseases. 409 411 25 25 References 412 Site-specific protein labeling with SNAP-tags. Curr. Protoc. Protein. Sci. 434 73, 30.1.1-30.1.16. 435 de Jong, M.A., de, Witte L., Taylor, M.E., and Geijtenbeek, T.B. 2010. Herpes simplex 436 virus type 2 enhances HIV-1 susceptibility by affecting Langerhans cell function. J. 437 Immunol. 185, 1633-1641. 438 de_Jong MA, Geijtenbeek T.B. 2010. Langerhans cells in innate defense against pathogens. 439 Trends. Immunol. 31, 452-459. 440 de Witte, L., Nabatov, A., Pion, M., Fluitsma, D., de Jong M.A., de, Gruijl T., Piguet, V., 441 de Witte, L., Nabatov, A., Pion, M., Fluitsma, D., de Jong M.A., de, Gruijl T., Piguet, V., 441 van, Kooyk Y., and Geijtenbeek, T.B. 2007. Langerin is a natural barrier to HIV-1 442 van, Kooyk Y., and Geijtenbeek, T.B. 2007. Langerin is a natural barrier to 442 26 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint transmission by Langerhans cells. Nat. Med. 13, 367-371. 443 Doebel, T., Voisin, B., and Nagao, K. 2017. Langerhans Cells - The Macrophage in 444 Dendritic Cell Clothing. Trends. Immunol. 38, 817-828. 445 Doebel, T., Voisin, B., and Nagao, K. 2017. Langerhans Cells - The Macro 444 Dendritic Cell Clothing. Trends. Immunol. 38, 817-828. 445 Feinberg, H., Powlesland, A.S., Taylor, M.E., and Weis, W.I. 2010. Trimeric structure of 446 langerin. J. Biol. Chem. 285, 13285-13293. 447 Figdor, C.G., van, Kooyk Y., and Adema, G.J. 2002. C-type lectin receptors on dendritic 448 cells and Langerhans cells. Nat. Rev. Immunol. 2, 77-84. 449 Geijtenbeek, T.B., Kwon, D.S., Torensma, R., van, Vliet S.J., van, Duijnho 450 Middel, J., Cornelissen, I.L., Nottet, H.S., KewalRamani, V.N., Littman, D 451 Middel, J., Cornelissen, I.L., Nottet, H.S., KewalRamani, V.N., Littman, D.R. et al. 2000. 451 DC-SIGN, a dendritic cell-specific HIV-1-binding protein that enhances trans-infection of 452 T cells. Cell. 100, 587-597. 453 Geijtenbeek, T.B., Krooshoop, D.J., Bleijs, D.A., van, Vliet S.J., van, Duijnhoven G.C., 454 Grabovsky, V., Alon, R., Figdor, C.G., and van, Kooyk Y. 2000. DC-SIGN-ICAM-2 455 interaction mediates dendritic cell trafficking. Nat. Immunol. 1, 353-357. 456 interaction mediates dendritic cell trafficking. Nat. Immunol. 1, 353-357. 456 Goddard, T.D., Huang, C.C., Meng, E.C., Pettersen, E.F., Couch, G.S., Morris, J.H., and 457 Ferrin, T.E. 2018. References 412 UCSF ChimeraX: Meeting modern challenges in visualization and 458 analysis. Protein. Sci. 27, 14-25. 459 Goddard, T.D., Huang, C.C., Meng, E.C., Pettersen, E.F., Couch, G.S., Morris, J.H., and 457 Ferrin, T.E. 2018. UCSF ChimeraX: Meeting modern challenges in visualization and 458 Ferrin, T.E. 2018. UCSF ChimeraX: Meeting modern challenges in visualization and 458 analysis. Protein. Sci. 27, 14-25. 459 Gonelli, C.A., Khoury, G., Center, R.J., and Purcell, D.F.J. 2019. HIV-1-based Virus-like 460 Particles that Morphologically Resemble Mature Infectious HIV 1 Virions Viruses 11 461 Gonelli, C.A., Khoury, G., Center, R.J., and Purcell, D.F.J. 2019. HIV-1-based Virus-like 460 Particles that Morphologically Resemble Mature, Infectious HIV-1 Virions. Viruses. 11, 461 507. 462 Particles that Morphologically Resemble Mature, Infectious HIV-1 Virions. Viruses. 11, 461 507. 462 Guo, Y., Feinberg, H., Conroy, E., Mitchell, D.A., Alvarez, R., Blixt, O., Taylor, M.E., 463 Guo, Y., Feinberg, H., Conroy, E., Mitchell, D.A., Alvarez, R., Blixt, O., Taylor, M.E., 463 Weis, W.I., and Drickamer, K. 2004. Structural basis for distinct ligand-binding and 464 targeting properties of the receptors DC-SIGN and DC-SIGNR. Nat. Struct. Mol. Biol. 11, 465 591-598. 466 Guo, Y., Feinberg, H., Conroy, E., Mitchell, D.A., Alvarez, R., Blixt, O., Taylor, M.E., 463 Weis, W.I., and Drickamer, K. 2004. Structural basis for distinct ligand-binding and 464 , , g, , y, , , , , , , , y , , Weis, W.I., and Drickamer, K. 2004. Structural basis for distinct ligand-binding and 464 targeting properties of the receptors DC-SIGN and DC-SIGNR. Nat. Struct. Mol. Biol. 11, 465 591-598. 466 Heredia, J.D., Park, J., Choi, H., Gill, K.S., and Procko, E. 2019. Conformational 467 Engineering of HIV-1 Env Based on Mutational Tolerance in the CD4 and PG16 Bound 468 Heredia, J.D., Park, J., Choi, H., Gill, K.S., and Procko, E. 2019. Conformational 467 Heredia, J.D., Park, J., Choi, H., Gill, K.S., and Procko, E. 2019. Conformational 467 Engineering of HIV-1 Env Based on Mutational Tolerance in the CD4 and PG16 Bound 468 States. J. Virol. 93, e00219-00219. 469 Engineering of HIV-1 Env Based on Mutational Tolerance in the CD4 and PG16 Bound 468 States. J. Virol. 93, e00219-00219. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 et al. 2013. Birbeck granule-like "organized 486 smooth endoplasmic reticulum" resulting from the expression of a cytoplasmic YFP-tagged 487 langerin. PLoS. One. 8, e60813. 488 Kimanius, D., Dong, L., Sharov, G., Nakane, T., and Scheres, S.H.W. 2021. New tools for 481 automated cryo-EM single-particle analysis in RELION-4.0. Biochem. J. 478, 4169-4185. 482 Kremer, J.R., Mastronarde, D.N., and McIntosh, J.R. 1996. Computer visualization of 483 three-dimensional image data using IMOD. J. Struct. Biol. 116, 71-76. 484 L d C S i lh l C Ci i B B di S B i H A i C 485 Kremer, J.R., Mastronarde, D.N., and McIntosh, J.R. 1996. Computer visualization of 483 three-dimensional image data using IMOD. J. Struct. Biol. 116, 71-76. 484 Lenormand, C., Spiegelhalter, C., Cinquin, B., Bardin, S., Bausinger, H., Angenieux, C., 485 Eckly, A., Proamer, F., Wall, D., Lich, B. et al. 2013. Birbeck granule-like "organized 486 smooth endoplasmic reticulum" resulting from the expression of a cytoplasmic YFP-tagged 487 Eckly, A., Proamer, F., Wall, D., Lich, B. et al. 2013. Birbeck granule-like "organized 486 smooth endoplasmic reticulum" resulting from the expression of a cytoplasmic YFP-tagged 487 langerin. PLoS. One. 8, e60813. 488 Eckly, A., Proamer, F., Wall, D., Lich, B. et al. 2013. Birbeck granule-like "organized 486 smooth endoplasmic reticulum" resulting from the expression of a cytoplasmic YFP-tagged 487 langerin. PLoS. One. 8, e60813. 488 Lupas, A., Van, Dyke M., and Stock, J. 1991. Predicting coiled coils from protein 489 sequences. Science. 252, 1162-1164. 490 Lupas, A., Van, Dyke M., and Stock, J. 1991. Predicting coiled coils from protein 489 sequences. Science. 252, 1162-1164. 490 Mastronarde, D.N. 2005. Automated electron microscope tomography using robust 491 prediction of specimen movements. J. Struct. Biol. 152, 36-51. 492 Mastronarde, D.N. 2005. Automated electron microscope tomography using robust 491 prediction of specimen movements. J. Struct. Biol. 152, 36-51. 492 Mc, Dermott R., Ziylan, U., Spehner, D., Bausinger, H., Lipsker, D., Mommaas, M., 493 Mc, Dermott R., Ziylan, U., Spehner, D., Bausinger, H., Lipsker, D., Mommaas, M., 493 Cazenave, J.P., Raposo, G., Goud, B., de, la Salle H. et al. 2002. Birbeck g 494 Cazenave, J.P., Raposo, G., Goud, B., de, la Salle H. et al. 2002. References 412 469 Hunger, R.E., Sieling, P.A., Ochoa, M.T., Sugaya, M., Burdick, A.E., Rea, T.H., Brennan, 470 Hunger, R.E., Sieling, P.A., Ochoa, M.T., Sugaya, M., Burdick, A.E., Rea, T.H., Brennan, 470 P J Belisle J T Blauvelt A Porcelli S A et al 2004 Langerhans cells utilize CD1a and 471 Hunger, R.E., Sieling, P.A., Ochoa, M.T., Sugaya, M., Burdick, A.E., Rea, T.H., Brennan, 470 P.J., Belisle, J.T., Blauvelt, A., Porcelli, S.A. et al. 2004. Langerhans cells utilize CD1a and 471 Hunger, R.E., Sieling, P.A., Ochoa, M.T., Sugaya, M., Burdick, A.E., Rea, T.H., Brennan, 470 P.J., Belisle, J.T., Blauvelt, A., Porcelli, S.A. et al. 2004. Langerhans cells utilize CD1a and 471 l i t ffi i tl t tid ti t T ll J Cli I t 113 701 708 472 P.J., Belisle, J.T., Blauvelt, A., Porcelli, S.A. et al. 2004. Langerhans cells utilize CD1a and 471 langerin to efficiently present nonpeptide antigens to T cells. J. Clin. Invest 472 Kawamura, T., Cohen, S.S., Borris, D.L., Aquilino, E.A., Glushakova, S., Margolis, L.B., 473 Orenstein, J.M., Offord, R.E., Neurath, A.R., and Blauvelt, A. 2000. Candidate 474 microbicides block HIV-1 infection of human immature Langerhans cells w 475 microbicides block HIV-1 infection of human immature Langerhans cells within epithelial 475 27 tissue explants. J. Exp. Med. 192, 1491-1500. 476 Kawamura, T., Qualbani, M., Thomas, E.K., Orenstein, J.M., and Blauvelt, A. 2001. Low 477 levels of productive HIV infection in Langerhans cell-like dendritic cells differentiated in 478 the presence of TGF-beta1 and increased viral replication with CD40 ligand-induced 479 maturation. Eur. J. Immunol. 31, 360-368. 480 Kawamura, T., Qualbani, M., Thomas, E.K., Orenstein, J.M., and Blauvelt, 477 Kawamura, T., Qualbani, M., Thomas, E.K., Orenstein, J.M., and Blauvelt, A. 2001. Low levels of productive HIV infection in Langerhans cell-like dendritic cells differentiated in 478 the presence of TGF-beta1 and increased viral replication with CD40 ligand-induced 479 maturation. Eur. J. Immunol. 31, 360-368. 480 levels of productive HIV infection in Langerhans cell-like dendritic cells d 478 Kimanius, D., Dong, L., Sharov, G., Nakane, T., and Scheres, S.H.W. 2021. New tools for 481 automated cryo-EM single-particle analysis in RELION-4.0. Biochem. J. 478, 4169-4185. 482 Kremer, J.R., Mastronarde, D.N., and McIntosh, J.R. 1996. Computer visualization of 483 three-dimensional image data using IMOD. J. Struct. Biol. 116, 71-76. 484 Lenormand, C., Spiegelhalter, C., Cinquin, B., Bardin, S., Bausinger, H., Angenieux, C., 485 Eckly, A., Proamer, F., Wall, D., Lich, B. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint specific genes. Mol. Ther. 2, 404-414. 541 References 412 Birbeck granules are 494 subdomains of endosomal recycling compartment in human epidermal Langerhans cells, 495 hich form here Langerin acc m lates M l Bi l C ll 13 317 335 496 subdomains of endosomal recycling compartment in human epidermal Langerhans cells, 495 which form where Langerin accumulates. Mol. Biol. Cell. 13, 317-335. 496 y g p p g , which form where Langerin accumulates. Mol. Biol. Cell. 13, 317-335. 496 McDermott, R., Bausinger, H., Fricker, D., Spehner, D., Proamer, F., Lipsker, D., 497 Cazenave, J.P., Goud, B., De, La Salle H., Salamero, J. et al. 2004. Reprod 498 Langerin/CD207 traffic and Birbeck granule formation in a human cell line model. J. 499 Invest. Dermatol. 123, 72-77. 500 Merad, M., Ginhoux, F., and Collin, M. 2008. Origin, homeostasis and function of 501 Langerhans cells and other langerin-expressing dendritic cells Nat Rev Immunol 8 935- 502 Merad, M., Ginhoux, F., and Collin, M. 2008. Origin, homeostasis and function of 501 Merad, M., Ginhoux, F., and Collin, M. 2008. Origin, homeostasis and function of 501 Langerhans cells and other langerin-expressing dendritic cells. Nat. Rev. Immunol. 8, 935- 502 947. 503 Merad, M., Ginhoux, F., and Collin, M. 2008. Origin, homeostasis and function of 501 Langerhans cells and other langerin-expressing dendritic cells. Nat. Rev. Immunol. 8, 935- 502 947. 503 Langerhans cells and other langerin-expressing dendritic cells. Nat. Rev. Immunol. 8, 935- 502 947. 503 Metlagel, Z., Kikkawa, Y.S., and Kikkawa, M. 2007. Ruby-Helix: an implementation of 504 helical image processing based on object-oriented scripting language. J. Struct. Biol. 157, 505 95-105. 506 Metlagel, Z., Kikkawa, Y.S., and Kikkawa, M. 2007. Ruby-Helix: an implementation of 504 helical image processing based on object-oriented scripting language. J. Struct. Biol. 157, 505 95-105. 506 Nakane, T., Kimanius, D., Lindahl, E., and Scheres, S.H. 2018. Characterization of 507 Nakane, T., Kimanius, D., Lindahl, E., and Scheres, S.H. 2018. Characterization of 507 molecular motions in cryo-EM single-particle data by multi-body refinement in RELION. 508 molecular motions in cryo-EM single-particle data by multi-body refinement in RELION. 508 28 Elife. 7, e36861. 509 Nicastro, D., Schwartz, C., Pierson, J., Gaudette, R., Porter, M.E., and McI 510 Nicastro, D., Schwartz, C., Pierson, J., Gaudette, R., Porter, M.E., and McIntosh, J.R. 2006. 510 The molecular architecture of axonemes revealed by cryoelectron tomography. Science. 511 313, 944-948. 512 Nicastro, D., Schwartz, C., Pierson, J., Gaudette, R., Porter, M.E., and McIntosh, J.R. 2006. References 412 510 The molecular architecture of axonemes revealed by cryoelectron tomography. Science. 511 313, 944-948. 512 The molecular architecture of axonemes revealed by cryoelectron tomography. Science. 511 313 944 948 512 The molecular architecture of axonemes revealed by cryoelectron tomography. Science. 511 313, 944-948. 512 Oda, T., Yanagisawa, H., and Kikkawa, M. 2015. Detailed structural and biochemical 513 characterization of the nexin-dynein regulatory complex. Mol. Biol. Cell. 26, 294-304. 514 characterization of the nexin-dynein regulatory complex. Mol. Biol. Cell. 26, 294-304. 514 Ogawa, Y., Kawamura, T., Matsuzawa, T., Aoki, R., Gee, P., Yamashita, A., Moriishi, K., 515 Yamasaki, K., Koyanagi, Y., Blauvelt, A. et al. 2013. Antimicrobial peptide LL-37 516 produced by HSV-2-infected keratinocytes enhances HIV infection of Langerhans cells. 517 Cell. Host. Microbe. 13, 77-86. 518 Ogawa, Y., Kawamura, T., Matsuzawa, T., Aoki, R., Gee, P., Yamashita, A., Moriishi, K., 515 Yamasaki, K., Koyanagi, Y., Blauvelt, A. et al. 2013. Antimicrobial peptide LL-37 516 Ogawa, Y., Kawamura, T., Matsuzawa, T., Aoki, R., Gee, P., Yamashita, A., Moriishi, K., 515 Yamasaki, K., Koyanagi, Y., Blauvelt, A. et al. 2013. Antimicrobial peptid 516 Yamasaki, K., Koyanagi, Y., Blauvelt, A. et al. 2013. Antimicrobial peptide LL-37 516 produced by HSV-2-infected keratinocytes enhances HIV infection of Langerhans cells. 517 Cell. Host. Microbe. 13, 77-86. 518 produced by HSV-2-infected keratinocytes enhances HIV infection of Langerhans cells. 517 Cell. Host. Microbe. 13, 77-86. 518 Pan, J., Peng, H., Chen, B., and Harrison, S.C. 2020. Cryo-EM Structure of 519 Pan, J., Peng, H., Chen, B., and Harrison, S.C. 2020. Cryo-EM Structure of Full-length 519 HIV-1 Env Bound With the Fab of Antibody PG16. J. Mol. Biol. 432, 1158-1168. 520 Pan, J., Peng, H., Chen, B., and Harrison, S.C. 2020. Cryo EM Structure of Full length 519 HIV-1 Env Bound With the Fab of Antibody PG16. J. Mol. Biol. 432, 1158-1168. 520 HIV-1 Env Bound With the Fab of Antibody PG16. J. Mol. Biol. 432, 1158-1168. 520 Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., 521 and Ferrin, T.E. 2004. UCSF Chimera--a visualization system for exploratory research and 522 analysis. J. Comput. Chem. 25, 1605-1612. 523 Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M 521 Pokidysheva, E., Zhang, Y., Battisti, A.J., Bator-Kelly, C.M., Chipman, P.R., Xiao, C., 524 Gregorio, G.G., Hendrickson, W.A., Kuhn, R.J., and Rossmann, M.G. 2006. References 412 544 Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., 545 Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B. et al. 2012. Fiji: an open-source platform 546 for biological-image analysis. Nat. Methods. 9, 676-682. 547 Schneider, C.A., Rasband, W.S., and Eliceiri, K.W. 2012. NIH Image to ImageJ: 25 years 548 of image analysis. Nat. Methods. 9, 671-675. 549 Stambach, N.S., and Taylor, M.E. 2003. Characterization of carbohydrate recognition by 550 551 Stambach, N.S., and Taylor, M.E. 2003. Characterization of carbohydrate recognition by 550 langerin, a C-type lectin of Langerhans cells. Glycobiology. 13, 401-410. 551 Stambach, N.S., and Taylor, M.E. 2003. Characterization of carbohydrate r 550 langerin, a C-type lectin of Langerhans cells. Glycobiology. 13, 401-410. 551 Stewart, S.A., Dykxhoorn, D.M., Palliser, D., Mizuno, H., Yu, E.Y., An, D.S., Sabatini, 552 D.M., Chen, I.S., Hahn, W.C., Sharp, P.A. et al. 2003. Lentivirus-delivered stable gene 553 silencing by RNAi in primary cells. RNA. 9, 493-501. 554 wart-Jones, G.B., Soto, C., Lemmin, T., Chuang, G.Y., Druz, A., Kong, R., Thoma P.V., Wagh, K., Zhou, T., Behrens, A.J. et al. 2016. Trimeric HIV-1-Env Structures Define 556 Glycan Shields from Clades A, B, and G. Cell. 165, 813-826. 557 Glycan Shields from Clades A, B, and G. Cell. 165, 813-826. 557 Takahara, K., Yashima, Y., Omatsu, Y., Yoshida, H., Kimura, Y., Kang, Y.S., Steinman, 558 R.M., Park, C.G., and Inaba, K. 2004. Functional comparison of the mouse DC-SIGN, 559 Takahara, K., Yashima, Y., Omatsu, Y., Yoshida, H., Kimura, Y., Kang, Y 558 R.M., Park, C.G., and Inaba, K. 2004. Functional comparison of the mouse DC-SIGN, 559 R.M., Park, C.G., and Inaba, K. 2004. Functional comparison of the mouse 559 SIGNR1, SIGNR3 and Langerin, C-type lectins. Int. Immunol. 16, 819-829. 560 Tateno, H., Ohnishi, K., Yabe, R., Hayatsu, N., Sato, T., Takeya, M., Narimatsu, H., and 561 Hirabayashi, J. 2010. Dual specificity of Langerin to sulfated and mannosylated glycans via 562 a single C-type carbohydrate recognition domain. J. Biol. Chem. 285, 6390-6400. 563 y y Hirabayashi, J. 2010. Dual specificity of Langerin to sulfated and mannosylated glycans via 562 Hirabayashi, J. 2010. Dual specificity of Langerin to sulfated and mannosylated glycans via 562 a single C-type carbohydrate recognition domain. J. Biol. Chem. 285, 6390-6400. 563 Thepaut, M., Valladeau, J., Nurisso, A., Kahn, R., Arnou, B., Vives, C., Saeland, S., Ebel, 564 C., Monnier, C., Dezutter-Dambuyant, C. et al. 2009. References 412 Cryo-EM 525 t ti f d i i l ith th b h d t iti d i f 526 Pokidysheva, E., Zhang, Y., Battisti, A.J., Bator-Kelly, C.M., Chipman, P.R., Xiao, C., 524 Gregorio, G.G., Hendrickson, W.A., Kuhn, R.J., and Rossmann, M.G. 2006. Cryo-EM 525 reconstruction of dengue virus in complex with the carbohydrate recognition domain of 526 DC-SIGN. Cell. 124, 485-493. 527 Postma, M., and Goedhart, J. 2019. PlotsOfData-A web app for visualizing data together 528 with their summaries. PLoS. Biol. 17, e3000202. 529 Rhodes, J.W., Botting, R.A., Bertram, K.M., Vine, E.E., Rana, H., Baharlo 530 Rhodes, J.W., Botting, R.A., Bertram, K.M., Vine, E.E., Rana, H., Baharlou, H., Vegh, P., 530 O'Neil, T.R., Ashhurst, A.S., Fletcher, J. et al. 2021. Human anogenital monocyte-derived 531 dendritic cells and langerin+cDC2 are major HIV target cells. Nat. Commun. 12, 2147. 532 O'Neil, T.R., Ashhurst, A.S., Fletcher, J. et al. 2021. Human anogenital mo 531 O'Neil, T.R., Ashhurst, A.S., Fletcher, J. et al. 2021. Human anogenital monocyte-derived 531 dendritic cells and langerin+cDC2 are major HIV target cells. Nat. Commun. 12, 2147. 532 dendritic cells and langerin+cDC2 are major HIV target cells. Nat. Commu 532 Ribeiro, C.M., Sarrami-Forooshani, R., Setiawan, L.C., Zijlstra-Willems, E.M., van, 533 Hamme J.L., Tigchelaar, W., van, der Wel N.N., Kootstra, N.A., Gringhuis, S.I., and 534 Geijtenbeek, T.B. 2016. Receptor usage dictates HIV-1 restriction by human TRIM5alpha 535 in dendritic cell subsets. Nature. 540, 448-452. 536 Sagebiel, R.W., and Reed, T.H. 1968. Serial reconstruction of the characteristic granule of 537 the Langerhans cell. J. Cell. Biol. 36, 595-602. 538 Salmon, P., Oberholzer, J., Occhiodoro, T., Morel, P., Lou, J., and Trono, D. 2000. 539 Reversible immortalization of human primary cells by lentivector-mediated transfer of 540 specific genes. Mol. Ther. 2, 404-414. 541 Reversible immortalization of human primary cells by lentivector-mediated transfer of 540 29 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint Schaefer, M.R., Williams, M., Kulpa, D.A., Blakely, P.K., Yaffee, A.Q., and Collins, K.L. 542 2008. A novel trafficking signal within the HLA-C cytoplasmic tail allows regulated 543 expression upon differentiation of macrophages. J. Immunol. 180, 7804-7817. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Valladeau, J., Duvert-Frances, V., Pin, J.J., Dezutter-Dambuyant, C., Vincent, C., 573 Massacrier, C., Vincent, J., Yoneda, K., Banchereau, J., Caux, C. et al. 1999. The 574 References 412 Structural studies of langerin and 565 Birbeck granule: a macromolecular organization model. Biochemistry. 48, 2684-2698. 566 Thornton, S.M., Samararatne, V.D., Skeate, J.G., Buser, C., Luhen, K.P., Taylor, J.R., Da, 567 Silva D M and Kast W M 2020 The Essential Role of anxA2 in Langerhans Cell 568 Thepaut, M., Valladeau, J., Nurisso, A., Kahn, R., Arnou, B., Vives, C., Saeland, S., Ebel, 564 C., Monnier, C., Dezutter-Dambuyant, C. et al. 2009. Structural studies of langerin and 565 Birbeck granule: a macromolecular organization model. Biochemistry. 48, 2684-2698. 566 C., Monnier, C., Dezutter-Dambuyant, C. et al. 2009. Structural studies of langerin and 565 Birbeck granule: a macromolecular organization model. Biochemistry. 48, 2684-2698. 566 Birbeck granule: a macromolecular organization model. Biochemistry. 48, 2684-2698. 566 Thornton, S.M., Samararatne, V.D., Skeate, J.G., Buser, C., Luhen, K.P., Taylor, J.R., Da, 567 Silva D.M., and Kast, W.M. 2020. The Essential Role of anxA2 in Langerhans Cell 568 Birbeck Granules Formation. Cells. 9, 974. 569 Turville, S.G., Cameron, P.U., Handley, A., Lin, G., Pohlmann, S., Doms, R.W., and 570 Cunningham, A.L. 2002. Diversity of receptors binding HIV on dendritic cell subsets. Nat. 571 Immunol. 3, 975-983. 572 Valladeau, J., Duvert-Frances, V., Pin, J.J., Dezutter-Dambuyant, C., Vincent, C., 573 Valladeau, J., Duvert-Frances, V., Pin, J.J., Dezutter-Dambuyant, C., Vincent, C., 573 Massacrier, C., Vincent, J., Yoneda, K., Banchereau, J., Caux, C. et al. 1999. The 574 Massacrier, C., Vincent, J., Yoneda, K., Banchereau, J., Caux, C. et al. 199 574 30 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint monoclonal antibody DCGM4 recognizes Langerin, a protein specific of Langerhans cells, 575 and is rapidly internalized from the cell surface. Eur. J. Immunol. 29, 2695-2704. 576 Valladeau, J., Ravel, O., Dezutter-Dambuyant, C., Moore, K., Kleijmeer, M., Liu, Y., 577 Duvert-Frances, V., Vincent, C., Schmitt, D., Davoust, J. et al. 2000. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 Langerin, a novel C- 578 type lectin specific to Langerhans cells, is an endocytic receptor that induces the formation 579 of Birbeck granules. Immunity. 12, 71-81. 580 Valladeau, J., Dezutter-Dambuyant, C., and Saeland, S. 2003. Langerin/CD207 sheds light 581 on formation of birbeck granules and their possible function in Langerhans cells. Immunol. 582 Res. 28, 93-107. 583 Valverde, P., Martinez, J.D., Canada, F.J., Arda, A., and Jimenez-Barbero, J. 2020. 584 Molecular Recognition in C-Type Lectins: The Cases of DC-SIGN, Langerin, MGL, and L- 585 Sectin. Chembiochem. 21, 2999-3025. 586 Verdijk, P., Dijkman, R., Plasmeijer, E.I., Mulder, A.A., Zoutman, W.H., Mieke, 587 Mommaas A., and Tensen, C.P. 2005. A lack of Birbeck granules in Lange 588 Mommaas A., and Tensen, C.P. 2005. A lack of Birbeck granules in Langerhans cells is 588 associated with a naturally occurring point mutation in the human Langerin gene. J. Invest. 589 Dermatol. 124, 714-717. 590 Weis, W.I., and Drickamer, K. 1994. Trimeric structure of a C-type mannose-binding 591 protein. Structure. 2, 1227-1240. 592 Weis, W.I., and Drickamer, K. 1996. Structural basis of lectin-carbohydrate recognition. 593 Annu. Rev. Biochem. 65, 441-473. 594 Zheng, S.Q., Palovcak, E., Armache, J.P., Verba, K.A., Cheng, Y., and Agard, D.A. 2017. 595 Zheng, S.Q., Palovcak, E., Armache, J.P., Verba, K.A., Cheng, Y., and Agard, D.A. 2017. 595 MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron 596 microscopy. Nat. Methods. 14, 331-332. 597 Zheng, S.Q., Palovcak, E., Armache, J.P., Verba, K.A., Cheng, Y., and Agard, D.A. 2017. 595 MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron 596 MotionCor2: anisotropic correction of beam-induced motion for improved cryo-electron 596 microscopy. Nat. Methods. 14, 331-332. 597 31 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 32 599 32 32 . of the structure, some classes showed a porous structure with a honeycomb-like lattice. 617 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint Figure 1: Isolation of Birbeck granules. (A) Domain organization of langerin. Human 600 langerin is a 328-amino acids protein composed of an N-terminal cytoplasmic tail, a 601 transmembrane domain (TM), a coiled-coil neck, and a C-terminal carbohydrate-recognition 602 domain (CRD). We fused a SNAP-tag at the N-terminus of langerin and inserted an HRV3C 603 protease cleavage site between the tag and the langerin sequences. For streptavidin-mediated 604 precipitation of Birbeck granules, the SNAP-tag was biotinylated (star). (B) Model of 605 langerin oligomerization within Birbeck granules. Langerin trimers bind to each other face- 606 to-face via the CRDs, bringing the two layers of the plasma membrane closer together. (C) 607 Birbeck granules formed in 293T cells overexpressing langerin. Addition of yeast mannan 608 induced the formation of Birbeck granules a few micrometers long. Inset shows a magnified 609 view of Birbeck granules. (D) SDS-PAGE of isolated Birbeck granules. Purified langerin 610 (arrowhead) was released from streptavidin-agarose by HRV3C digestion. M: molecular 611 weight marker; and BG: isolated Birbeck granules. (E) Cryo-electron microscopy of isolated 612 Birbeck granules. Black square indicates the position of the close-up view shown in F. (F) 613 Wavy lamellar structure of the Birbeck granule. Black dots were gold nanoparticles used as 614 fiducial markers. (G) Class averages of the projection images of Birbeck granules. The image 615 dimension is 34 nm2. Although 2D classification did not converge well due to the continuity 616 33 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint the structure, some classes showed a porous structure with a honeycomb-like lattice 617 618 34 34 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 619 35 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 2: Averaged subtomograms of Birbeck granules. (A) Result of the initial 3D 620 refinement. The central langerin trimer (blue) binds to the three inverted langerin trimers 621 (red). (B) Result of the focused refinement. Improved resolution (6.4 Å) allowed secondary 622 structure modeling. The gray mesh and the cyan surface indicate low and high threshold 623 isosurfaces, respectively. (C) Result of the two-body refinement. The second body (orange) 624 was refined with respect to the first body (cyan). (D) Honeycomb model of the langerin lattice 625 within Birbeck granules. This honeycomb lattice is an ideal model, with the assumption of 626 Figure 2: Averaged subtomograms of Birbeck granules. (A) Result of the initial 3D 620 refinement. The central langerin trimer (blue) binds to the three inverted langerin trimers 621 (red). (B) Result of the focused refinement. Improved resolution (6.4 Å) allowed secondary 622 structure modeling. The gray mesh and the cyan surface indicate low and high threshold 623 isosurfaces, respectively. (C) Result of the two-body refinement. The second body (orange) 624 was refined with respect to the first body (cyan). (D) Honeycomb model of the langerin lattice 625 within Birbeck granules. This honeycomb lattice is an ideal model, with the assumption of 626 structural uniformity. (A-D) left, top views; right, side views. 627 629 36 36 . . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint References 412 ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 3 3 References 412 CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 37 0 630 37 37 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 3: Binding interface between two langerin trimers. (A) Two langerin trimers 631 interact via the loops at residues 258-263. Amino acid codes and residue numbers are 632 indicated. Yellow and pink circles indicate the primary and secondary carbohydrate-binding 633 sites. The orientations of the side chains are speculative because the resolution of these maps 634 (6.4 Å) was insufficient for precise modeling. Calcium ions (green) were placed based on the 635 crystal structure (PDB ID: 3KQG). Top and side views are displayed in the top and bottom 636 rows, respectively. Broken squares indicate the positions of the magnified views on the right. 637 (B) Flexibility between the two trimers. The first eigenvector of the structural variation of 638 the inverted trimer (red) relative to the upright trimer (blue) shows an approximately 30° 639 rotation about the loop 258-263 (see Movie S2). 640 642 38 38 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. 643 39 39 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 4: Mutations to the loop 258-263 disrupted Birbeck granule formation. (A) Ultra- 644 ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 4: Mutations to the loop 258-263 disrupted Birbeck granule formation. (A) Ultra- 644 thin electron microscopy of Birbeck granules with and without yeast mannan in langerin- 645 overexpressing 293T cells. Residues at 260-263 (Met-Glu-Gly-Asp, MEGD) were mutated 646 to Met-Arg-Gly-Asp (MRGD), Met-Arg-Gly-Lys (MRGK), or Ala-Arg-Gly-Lys (ARGK). 647 The Mannan (+) WT image has a lower magnification than the others in order to show the 648 entire Birbeck granules. (B-D) Quantification of Birbeck granule formation. Horizontal lines 649 indicate the median values. Single asterisks and the double asterisk indicate statistically 650 significant differences with p < 0.01 and p = 0.0101, respectively. NS indicates no statistically 651 significant differences. p values were calculated using Bonferroni-corrected Student’s t-tests. 652 (B) Lengths of individual Birbeck granules were measured. N = 337 (WT mannan-), 386 653 (WT mannan+), 348 (MRGD), 233 (MRGK), and 70 (ARGK). (C) Sum of the length of 654 Birbeck granules per 100 µm2 cell area. Cross-sections of cells with nuclei of > 5 µm in 655 diameter were selected and the total lengths of the Birbeck granules were measured within 656 the cells. N = 20 for all the samples. (D) Number of the Birbeck granules per 100 µm2 cell 657 area. N = 20 for all the samples. 658 660 40 40 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 61 61 661 41 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 5: Mutation to the loop 258-263 inhibited the internalization of pseudoviruses. 662 g p p HIV-1 pseudoviruses were added to langerin-expressing 293T cells. Yeast mannan (10 663 µg/ml) was added to block the lectin-dependent binding of pseudoviruses. A langerin mutant 664 lacking calcium binding ability (lectin (-)) was used as the negative control. (A) Immunoblots 665 of pseudoviruses attached to the cell surface. Unbound and attached viruses were collected 666 from the supernatant of the culture medium and TBS-EDTA buffer, respectively. Samples of 667 unbound viruses were diluted 50-fold to adjust the band intensities. The expression levels of 668 SNAP-tagged langerin show that the numbers of transfected cells were approximately the 669 same in each experiment. Pr55gag and p24 indicate unprocessed and fully-processed capsid 670 proteins, respectively. (B) Immunoblots of internalized pseudoviruses. Birbeck granules were 671 isolated by precipitation using streptavidin-agarose, and intracellular viruses and langerin 672 were detected by their respective antibodies. Tubulins in the whole-cell lysates were detected 673 for loading controls. (C) Quantification of internalized viruses using p24 ELISA. Horizontal 674 lines indicate the mean. NS and Asterisk indicate no significant difference and statistically 675 significant differences (p < 0.001) calculated using Bonferroni-corrected Student’s t-tests 676 (N=4), respectively. 677 678 42 42 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 9 Figure 6: Model of HIV internalization by langerin. (A) Glycosylated structure of HIV-1 0 9 679 Figure 6: Model of HIV internalization by langerin. (A) Glycosylated structure of HIV-1 680 43 43 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint envelope (env, gray) (Constructed based on PDB ID: 5FYJ and 6ULC) bound to the langerin 681 trimer complex. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 5: Mutation to the loop 258-263 inhibited the internalization of pseudoviruses. 662 Branches of the high-mannose oligosaccharide interacting with the primary 682 (yellow) and secondary (pink) carbohydrate-binding sites. The broken square indicates the 683 position of the close-up view on the right. (B) Schematics of HIV internalization by langerin. 684 The inset on the top shows an overview of antigen processing within a Langerhans cell. 685 Viruses are captured by langerin, internalized into a Birbeck granule, and degraded by the 686 autophagic system. The red square indicates the viral entry process described in the lower 687 panels. The high-mannose oligosaccharide on HIV-1 env first binds to the primary 688 carbohydrate-binding site, and one of the branches of the oligo-mannose weakly interacts 689 with the secondary binding site. The flexible neck region allows docking of the loop 258-263 690 of an adjacent trimer into the secondary carbohydrate-binding cleft, reinforcing the binding 691 of the oligo-mannose. As the elastic neck regions straighten, the plasma membrane is 692 deformed into a lamellar invagination, which internalizes the virus. 693 695 44 44 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: oRxiv preprint . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint 45 96 696 45 . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 7: Coulombic surfaces of the carbohydrate-binding sites. (A) Comparison of the 697 surface charge between the CRDs of langerin and DC-SIGN (PDB ID: 1SL4). Blue and red 698 coloring indicates positive and negative charges, respectively. The primary and secondary 699 carbohydrate-binding sites are indicated by yellow and pink arrowheads, respectively. The 700 region of DC-SIGN corresponding to the secondary carbohydrate-binding sites of langerin is 701 occupied by calcium ions (green) and does not form a cleft. . CC-BY 4.0 International license available under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (which this version posted May 14, 2022. ; https://doi.org/10.1101/2022.02.24.481763 doi: bioRxiv preprint Figure 5: Mutation to the loop 258-263 inhibited the internalization of pseudoviruses. 662 (B) Surface model of the langerin 702 trimer complex. Mannose ligands are placed based on a previously proposed model (Chatwell 703 et al., 2008). (A-B) Broken squares indicate the position and the orientation of the side (left) 704 and the top (right) views. Oligosaccharide ligands are shown in gray. 705 707 46
https://openalex.org/W2055011650	https://vuir.vu.edu.au/40267/1/142-648-1-PB.pdf	English	null	Understanding Well-Being in Multi-Levels: A review	Health, culture and society	2,013	cc-by	9,954	Understanding Well-Being in Multi-Levels: Understanding Well-Being in Multi-Levels: A review This is the Published version of the following publication The publisher’s official version can be found at http://hcs.pitt.edu/ojs/index.php/hcs/article/view/142 Note that access to this version may require subscription. Downloaded from VU Research Repository https://vuir.vu.edu.au/40267/ A review Instead, th acknowledges the multi-levels of well-being are closely tied and should be taken into accoun well-being is concerned. We will provide a brief review of the two major approaches – su well-being and quality of life – of well-being before the multi-level approach is introduced. The and challenges of the multilevel approach will be discussed. Keywords: well-being; subjective well-being (SWB); positive psychology; quality of life (QOL); multi-level approach Health, Culture and Society ds: well-being; subjective well-being (SWB); positive psychology; quality of life (QOL); multi-level approach A review E. C. W Ng & A. T. Fisher Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu New articles in this journal are licensed under a Creative Commons Attribution 3.0 United States License. This journal is published by the University Library System of the University of Pittsburgh as part of its D-Scribe Digital Publishing Program, and is cosponsored by the University of Pittsburgh Press. New articles in this journal are licensed under a Creative Commons Attribution 3.0 United States License. This journal is published by the University Library System of the University of Pittsburgh as part of its D-Scribe Digital Publishing Program, and is cosponsored by the University of Pittsburgh Press. les in this journal are licensed under a Creative Commons Attribution 3.0 United States License. Abstract Well-being is not only an emerging research agenda, but also a critical issue concerning the individual a well as the societal development, because how the issue is viewed has a huge theoretical as well a practical, even policy, implication. In academic, while some argue that well-being is in the subjectiv perception of one’s life or psychological functioning, others argue that well-being is in the objectiv conditions and the broader environment. This paper, drawing on psychology tradition, tries to go beyon the dichotomy of well-being as either an individual attribute or external conditions. Instead, this artic acknowledges the multi-levels of well-being are closely tied and should be taken into accounts whe well-being is concerned. We will provide a brief review of the two major approaches – subjectiv well-being and quality of life – of well-being before the multi-level approach is introduced. The strengt and challenges of the multilevel approach will be discussed. Keywords: well-being; subjective well-being (SWB); positive psychology; quality of life (QOL); multi-level approach Abstract Well-being is not only an emerging research agenda, but also a critical issue concerning the indiv well as the societal development, because how the issue is viewed has a huge theoretical as practical, even policy, implication. In academic, while some argue that well-being is in the su perception of one’s life or psychological functioning, others argue that well-being is in the o conditions and the broader environment. This paper, drawing on psychology tradition, tries to go the dichotomy of well-being as either an individual attribute or external conditions. I. Introduction What is a good life or well-being, what factors influence it, whether it can be promoted and how to attain it are all contemporary questions appealing to wide range of profesisonals. Philosophers and academics may debate whether well-being is concerned with people’s happiness or is a human virtue, subjective feeling or an objective condition, out of intellectual curiosity or in the interests of gaining insights so as to garner and develop appropriate intervention. Policy makers may be more concerned about the operations of the concept of well-being and how it can be addressed and integrated into public policy design. The general public, however, may not be as concerned with how the term is defined or whether policy has explicitly addressed the issue: what the public are interested in is how and what they can do to enhance their quality of life or to feel content, publicly and privately. Mental health and well-being is everybody’s business (VicHealth, 2009). Although such diverse viewpoints are but an aspect of public discourse about the issue, they are sufficient to provide us with an idea that well-being is a multifaceted phenomenon. But is well-being hedonic or eudemonic in nature (Keyes, 2006; Ryan & Deci, 2001)? Is it attributed to individual perception or an objective condition (Cummins, 1997; Diener, Sapyta, & Suh, 1998)? Who can judge and decide precisely what well-being is? Can it be changed? And if it can be changed, is intervention by way of the individual or society (Lyubomirsky, Sheldon, & Schkade, 2005; Sirgy et al., 2006)? Well-being is therefore not only an emerging research agenda, but also a critical issue concerning individual as well as societal development. This said, how the issue is viewed has important implications for theory, practice and policy. And in a culture of surplus (Diener, 2009; Eckersley, 2004), the question of well-being has emerged to resonate with a modern need to live a life of quality, existentially so, and not merely to exist biologically. Despite health and well-being attracting government attention owing to policy implications, health has long been perceived as a state free of disease (Frank, 2002). In much the same way, mental health has often been equated with the absence of mental illness – while the positive notions of mental health (e.g., well-being) have remained elusive with little empirical investigation (Keyes & Waterman, 2003). A review E. C. W Ng & A. T. Fisher E. C. W Ng & A. T. Fisher Abstract Well-being is not only an emerging research agenda, but also a critical issue concerning the individual as well as the societal development, because how the issue is viewed has a huge theoretical as well as practical, even policy, implication. In academic, while some argue that well-being is in the subjective perception of one’s life or psychological functioning, others argue that well-being is in the objective conditions and the broader environment. This paper, drawing on psychology tradition, tries to go beyond the dichotomy of well-being as either an individual attribute or external conditions. Instead, this article acknowledges the multi-levels of well-being are closely tied and should be taken into accounts when well-being is concerned. We will provide a brief review of the two major approaches – subjective well-being and quality of life – of well-being before the multi-level approach is introduced. The strength and challenges of the multilevel approach will be discussed. Keywords: well-being; subjective well-being (SWB); positive psychology; quality of life (QOL); multi-level approach Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu Understanding Well-Being in Multi-Levels: II. Subjective Well-Being There are two main streams of psychological research in understanding subjective well-being — hedonic and eudaimonic (Keyes, Shmotkin, & Ryff, 2002; Ryan & Deci, 2001). The hedonic stream of subjective well-being is rooted in the Greek philosopher Aristippus, who considered that life’s ideal is to maximize the amount of pleasure and happiness. The affiliated psychological construct, normally termed SWB – subjective well-being (Diener, 1984) – concerns feelings towards life and the evaluation of affective states, including overall happiness and life satisfaction as well as positive and negative affective experiences. Life satisfaction, the cognitive perspective of SWB, is the person’s evaluation of his or her own life wherein happiness, as balancing between positive and negative affect, reflects a person’s immediate experience of pleasant or unpleasant moments. Thus, SWB takes into consideration the affective as well as the cognitive components of people’s emotional experience towards life. In contrast, eudaimonic subjective well-being proposes that true happiness is not just following human felt desire, but sooner “found in the expression of virtues – that is, in doing what is worth doing” (Ryan & Deci, 2001, p. 145). Related to this, is the term PWB (psychological well-being) which assesses a person’s psychological and social functioning as the criteria of well-being. Compared to SWB, which specifically focuses on subjectivity, PWB is a person’s potential to realize a meaningful life and to meet real life challenges. The major proponent of PWB is Carol Ryff (Ryff, 1989; Ryff & Keyes, 1995), who argued that SWB and PWB are related but distinct conceptions of well-being. For Ryff, PWB signals that a person functions positively in six aspects of life: looking at oneself in an appropriate way (self-acceptance); having good relations with others (positive relations with others); having a sense of growth (personal growth); realizing one’s life is meaningful (purpose in life); being capable of managing the surrounding world (environmental mastery), and possessing a sense of self-determination (autonomy). Other theorists define fully functioning people as those who can actualize themselves, in the area of autonomy, competence and relatedness (Ryan & Deci, 2001), who live according to the true self (Waterman, 1993), or who strive for a good balance between happiness and meaningfulness (McGregor & Little, 1998). Keyes (1998), in following on from Ryff, argued that to attain well-being people have to function effectively in regard to social challenges and tasks. I. Introduction Since the first publication on positive mental health (Jahoda, 1958), conceptualizations and measures of well-being have progressed substantially over the past 50 years (Diener & Suh, 2000b; Keyes & Waterman, 2003). In this study, we seek to draw upon contemporary psychological literature and provide further thought on the conceptualization of well-being, in the hope of encouraging further discussion of such a socially critical issue. The two major approaches – subjective well-being and quality of life – will be first reviewed before proposing the approach and concept of multi-level well-being. Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 308 308 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 308 II. Subjective Well-Being Keyes suggested five dimensions of social well-being – social integration; social contribution; social coherence; social actualization and social acceptance. Social integration is whether one is capable of being a part of a society; social contribution is the degree to which a person can offer something valuable to the world; social coherence is whether a person finds the events and eventualities around them understandable and coherent; social actualization refers to whether a person can envision the future and potential of a society, and social acceptance is whether a person can trust and accept both the good and bad of human nature. The argument is this: attainment of well-being goes beyond the private sphere in order to resonate effectively and coherently in the public domain. Keyes further suggested that people who attain a high level in all the above aspects (SWB, PWB and social well-being) can be described as having complete mental health, and as flourishing, in contrast to languishing (Keyes & Lopez, 2002). III. The Positive Psychology Movement In line with the growing study of subjective well-being, positive psychology is another significant presence within the field of psychology that seeks to bring to light the brighter side of human nature, scientifically and phenomenologically. Seligman and Csikszentmihalyi (2000) contended that psychology has to shift its focus from merely correcting the weaknesses of an individual to exploring human strengths and virtues. The underlying goal not only aims to prevent further serious mental problems, but also to build and develop flourishing and fully functioning people. These goals have been advanced over the last two decades by the endeavours in research of positive emotion Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 309 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 309 (Fredrickson, 2000, 2001; Isen, 1987); optimism (Scheier & Carver, 1985); learned optimism (Seligman, 1991); hope (Snyder, 1994, 2000); courage (Rate, Clarke, Lindsay, & Sternberg, 2007; O' Byrne et al. 2000 cited in Snyder & Lopez, 2007); wisdom (Sternberg, 1990, 1998); gratitude (Emmons & McCullough, 2004); forgiveness (McCullough, 2000; McCullough, Pargament, & Thoresen, 2000); mindfulness (Langer, 2002); flow (Csikszentmihalyi, 1990), and spirituality (Larson, 1996; Pargament, 1997, 2002). (Fredrickson, 2000, 2001; Isen, 1987); optimism (Scheier & Carver, 1985); learned optimism (Seligman, 1991); hope (Snyder, 1994, 2000); courage (Rate, Clarke, Lindsay, & Sternberg, 2007; O' Byrne et al. 2000 cited in Snyder & Lopez, 2007); wisdom (Sternberg, 1990, 1998); gratitude (Emmons & McCullough, 2004); forgiveness (McCullough, 2000; McCullough, Pargament, & Thoresen, 2000); mindfulness (Langer, 2002); flow (Csikszentmihalyi, 1990), and spirituality (Larson, 1996; Pargament, 1997, 2002). Peterson and Seligman developed the VIA Classification of Strengths (2004) in contrast to the DSM categories1 (Snyder & Lopez, 2007) which are so prevalent in the conceptualisation and practice of much psychology and psychiatry. Although the universality of such diagnostic nomenclature is still under debate, the fruits in this area of research are abundant and the advancement in empirical research of human strengths is impressive. Nevertheless, it should be noted that as with PWB, studies on different forms of human strengths is still in the developmental stage and that most research on well-being continues to focus on SWB of the hedonic tradition. As consequence, eudaimonic approaches only began to receive critical attention after the 1980s (Keyes et al., 2002). 1 The Diagnostic and Statistical Manual of Mental Disorders (DSM) is a set of standard criteria used widely in mental health for the classification of mental disorders. The first version of DSM was published in 1952 and the current version is the DSM-5 (fifth edition). In order to provide researchers and practitioners with the same language set through which they could talk about the positive, Peterson and Seligman developed the VIA system as a counterpart to DSM. Despite enjoying a longer history and recognition, the DSM system shares with VIA, criticism as to the validity and reliability of the diagnostic categories proposed and employed. III. The Positive Psychology Movement The differentiation of SWB from PWB or other human strengths, not only shows the basic differences of hedonic and eudaimonic approaches to well-being and their underlying philosophies and ethos (Ryan & Deci, 2001), but such differences also lead to more in-depth discussions about whether well-being should be defined by experiencing individuals or observers (by experts or a normative value). What’s more there is the question of as to whether well-being can be universally or culturally applied, that is, whether there is a cultural specifity to well-being in as much as there possibly exists a cultural and anthropological variance to psychiatric categories. Although SWB can be criticised as self-indulgent, Diener and his colleagues have south to reiterate that SWB pays more respect to people’s own values and judgments, avoiding the hegemony of external judgment (Diener et al., 1998; Diener & Suh, 1997). Thus, Diener et al. (1998) remark that PWB, presumably universal, may indeed vary across cultures, and that while PWB may contribute to the understanding of well-being, SWB is still essential to the debate and reality of wellness. In view of the widespread use of SWB and the emerging development of PWB as indicators in well-being research, more understanding and conceptualization on the subjective contexts of well-being will be advanced. It also seems that by complementing SWB with PWB, a better picture of the complex construct of human well-being can be built (Keyes et al., 2002; Ryan & Deci, 2001). IV. Quality of life Another approach towards understanding what a good life or well-being is has been conducted by researchers in the area of quality of life (QOL) studies. QOL is a complex and still debatable construct that has been employed by different disciplines to delineate what a “good life” is, individually and collectively (Carr, 2004; Rapley, 2003; Sirgy et al., 2006). QOL research is debatable precisely because there is, as of yet, little or no consensus on whether quality of life should be objectively or subjectively measured, whether it should be defined by researchers or participants, or whether the concept of QOL can be applied and defined universally or culturally. Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 310 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 31 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 3 What’s more, QOL has been associated with the social indicator movement (Bauer, 1966),2 popular in the 1960s, which sought to provide a broader measure of the quality of life of a nation in addition to the economic and monetary indicators employed (such as GDP or income levels). While QOL takes the external condition more seriously, recent QOL research has emphasized individual and subjective perspectives of people’s lives (Rapley, 2003). Some even suggest that QOL, before all else, is primarily subjective (Haas, 1999). Nevertheless, whether in terms of the individual or the collective, it is broadly agreed that QOL is a multidimensional construct, embracing both objective and subjective domains, despite there being no common consensus regarding the number of dimensions to the definition of QOL (Costanza et al., 2007; Cummins, 1997; Diener & Suh, 1997; Rapley, 2003; Schalock & Felce, 2004). With a goal to informing intervention and social policy, certain national well-being indicators focus on a comprehensive range of factors that may affect the quality of life of an individual and of society as a whole (Rapley, 2003). These include indicators related to housing; crime and safety; work; education; health, as well as family and community. Drawing on 11 basic human needs such as security; affection; leisure; reproduction and participation, Costanza et al. (2007) constructed a QOL framework in relation to various kinds of capital which can be advanced by policy implementation. 2 Raymond A. Bauer was considered one of the key proponents of social indicators during the initial stage of development. His keynote work remains Social Indicators (M.I.T Press, 1967), which documented the effects of the space exploration program upon American society. The interest in social trend analysis was reawakened in the Kenndy years and later developed as a public concern regarding domestic social problems such as poverty, race, health, unemployment and housing (Rossi & Gilmartin, 1980, p.2) IV. Quality of life Thus the content of QOL, in contrast to SWB, is more compatible with policy design, as it deals directly with objective socio-economic environments. From the field of psychology, Cummins (1997) has proposed the following definition: Quality of life is both objective and subjective, each axis being the aggregate of seven domains: material well-being, health, productivity, intimacy, safety, community, and emotional well-being. Objective domains comprise culturally-relevant measures of objective well-being. Subjective domains comprise domain satisfaction weighted by their importance to the individual. (Cummins, 1997, p. 6) Similar to Cummins, Schalock (2000) proposes a similar QOL conceptualization that is specifically relevant, but not exclusive, to people with intellectual disabilities. Drawing on a wide range of literature, Schalock proposed eight dimensions (emotional well-being; interpersonal relations; material well-being; personal development; physical well-being; self-determination; social inclusion and rights) as the core domain of quality of life. In each dimension, subjective personal appraisal and objective functional assessment were measured independently. While Cummins (1997) covered more culturally relevant domains (e.g., safety and community), Schalock (2000) emphasized autonomy and individual rights, especially related to his target concerns about people with intellectual disabilities. Both authors however take into consideration the importance of objective and subjective measurement, at the individual level. Schalock and Felce (2004) even suggested that objective indicators of life experiences should be adopted instead of SWB, that is, if the environmental impact on people is to be evaluated in a more sensitive, socially conducive way. Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 311 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 311 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.e Yet, compared to the social indicator tradition as initiated by Bauer in the 1960s which emphasized more on the objective external circumstances, Cummins and Schalock both pay more detailed attention to people’s inner perceptions and experience of satisfaction. In contrast to the scope of SWB or PWB which remain limited to feeling and functioning, the quality of life definitions proposed by Cummins and Schalock concern both the subjective and objective evaluation of ones’ situation, and in multiple domains that are culturally important and normatively relevant, this, bridging the traditionally objective measures of social situations at the population level with the subjective evaluation of individuals. IV. Quality of life Such conceptualizations imply that objective population indicators affect objective individual domains, which in turn affect the SWB of people. As Michalos (in Sirgy et al., 2006) has put forward, the construction of QOL models is a political and philosophical exercise rooted in specific kinds of values. Theoretically, there are rich accounts of the approaches in quality of life with different underlying philosophies: utility; needs; prudential values and capabilities, as summarized by Phillips (2006). According to Phillips, those embracing utilitarianism envision and strive for societies which enable their citizens to achieve the greatest happiness. In contrast, those who adopt prudential values emphasize whether a life is good based not through people’s desires, but through values which render humanity better. Such values include those of the utilitarian approach – enjoyment and the avoidance of pain – but also those of accomplishment; basic capabilities; understanding; liberty; self respect; and deep personal relations, all of which are believed to enhance life in a general, intelligible way. While this concept is simple, little idea is provided as to how quality of life can be defined. The focus on capabilities signals a hybrid of the previous three approaches and regards quality of life as being about a person’s “ability to do valuable acts or to reach valuable state of well-being” (Phillips, 2006, p. 91). This understanding of QOL takes both the external environment, human desires and needs into consideration. Nevertheless, what constitutes the elementary capability set varies according to different scholars.3 3 For example, Sen's approach gives more weight to the freedom to choose among various alternatives, while Nussbaum's approach emphasizes moral imperatives sooner than freedoms and choice (Phillip, 2006). V. Health Related Quality of Life QOL can be seen to become a more complicated and complex matter when extended to the domain of Health related Quality of Life (HRQOL). There have always been strong associations between health and a good life, whether through the optic of philosophy, psychology or medicine, and the WHO has long defined health in terms of a more holistic perspective, namely a, “state of complete physical, mental and social well-being” (Larson, 1996, p. 141). As a result, health status, well-being, QOL and health related QOL are “often used interchangeably in the health field” (Sirgy et al., 2006, p. 401). Such is Bowling’s (1997) and Fitzpatrick’s (1996) definition of health related quality of life, which incorporates a wide range of components into human functioning, including physical health; emotional status, or social roles. What’s more such generic measures of health and QOL are generally used in national health surveys to monitor a nation’s health status (Rapley, 2003; Sirgy et al., 2006), and despite being narrower than the concept of QOL, HRQOL conceptualizations and measurements, especially the wide-ranging and multidimensional ones, can be understood as quality of life in general (Michalos, 2004; Phillips, 2006). As a logical consequence to such research, Michalos suggested that we abandon the term “health related quality of life” for “clarity”. The wide range and amorphous nature of HRQOL has little use for health professionals, especially if they are seeking evidence to inform their decisions about life and death or curing illness, or evaluating different medical Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 312 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu procedures and policies. A more impersonal, scientific and objective approach to HRQOL has therefore been proposed in the field. Some of the examples are Quality Adjusted Life Years (QALY), Disability Adjusted Life Years (DALY), and the time-trade off method (Phillips, 2006; Rapley, 2003). These approaches are useful because they provide a straightforward statistical method that can inform health practitioners in making medical decisions after comparing the relative cost and effect of different procedures as well as the risk involved. Such approaches are, however, debatable because the patients’ point of view is not regarded, this flagging the ethical issue involved around the relationship between value of life and quality of life (Phillips, 2006; Rapley, 2003). VI. Community and Ecological Levels of QOL Despite social indicator research having brought social circumstances into scrutiny, beyond monetary concerns, little attention had been paid to community contexts in which QOL is embedded. Raphael et al. (2001) have recognized the complex role of community structures in individual well-being warning that neo-liberalism has both threatened and deteriorated community infrastructures and as a result the building up of strength and support between and within community membership has waned. The emerging concern of community indicators also highlights the importance of satisfaction in the community level to the people’s overall well-being (Dluhy & Swartz, 2006; Wiseman, 2009; Wong, 2002). Rapley (2003) argued for example, that large scale indices (e.g., GDP; a fair and crime free society; vigorous cultural life) are too far removed from the concerns of local communities and their members. Thus, work on community indicators can bridge the gap and provide valuable information to the policy designer at the local level, with the aim to link community development to the locally meaningful indices of QOL. Whilst some proposed community QOL to be community members’ satisfaction with government services, business services and non-profit services (Sirgy & Rahtz, 2006), there are numerous other indicators adopted by different communities addressing various needs across various regions (Dluhy & Swartz, 2006; Sawicki, 2002; Wiseman, 2009). Nowadays, scholars suggest that community indicators be developed through community consultation, with intellectual input from academia (Dluhy & Swartz, 2006; Rapley, 2003; Sawicki, 2002; Wong, 2002). In addition to the individual and national level, a community indicator approach suggests that there is an intermediate layer to which we have to pay attention for the community can play a significant role in affecting our well-being. In summary, research has presented to us both ends of conceptualizing QOL as a state-of-states versus a state-of-persons. It also exhibits the endeavour and necessity of an interdisciplinary approach to investigating such a complex issue given that a wealth of factors and variables define the normative and given of modern daily life. Nevertheless, as Rapley (2003) argues the large scale indices (at the national level) may mask the real differences between population subgroups and they are far from the concerns of individuals or local communities. Given this, social community indicators are proving the more favourable option. V. Health Related Quality of Life The dilemmas of using QALY brings out the critical issue that being physically healthy does not equate to having health related quality of life, let alone overall quality of life (Michalos, 2004). This said, there are greater contributory factors to defining the how and what of quality of life. VII. Quality of Life and Subjective Well-Being In the discussion of QOL comparisons have been made between the concepts of Quality of Life and Subjective Well-being, particularly with reference to Cummins and Schalock. Whereas subjective well-being (SWB) emphasizes the personal level, including life satisfaction, happiness and human functioning; quality of life (QOL) is a Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 313 Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 313 broader concept, including life in multiple dimensions, embracing both subjective and objective domains, in terms of the individual reality and that of the population. This said, subjective well-being can be seen as a subset of quality of life, not accounting for external circumstances. Because of the widespread use of SWB indicators in subjective well-being research, subjective well-being is dominated by the narrow focus on the hedonic approach of well-being, ignoring the eudemonic approach of well-being at the personal level. In spite of this, Phillips (2006) still argues that QOL should start with subjective well-being, because it takes an individual’s judgment seriously and as a constitutive factor. Quality of life, however, does not stop there for many studies have shown that subjective and objective well-being is not necessarily correlated (e.g., Phillips, 2006; Schalock & Felce, 2004): by merely focusing on the subjective dimension without attending to the social dimensions indeed presents limits. Descriptions such as “happy slave” or a “fool’s paradise” are employed to bring out the contradiction of someone who has adapted well with a high level of SWB, yet who suffers high levels of deprivation at the same time (Phillips, 2006; Sirgy et al., 2006). As a result, researchers have progressively proposed that the most assured way of measuring QOL is to combine both objective and subjective indicators because they each provide valuable information that can enhance quality of life and inform policy intervention (Costanza et al., 2007; Diener & Suh, 1997; Schalock & Felce, 2004). In relation to Phillip’s (2006) four main approaches of QOL, SWB is closely aligned to the utilitarian and individualistic perspective, in that it accords with the maximization of pleasure and the avoidance of pain (Diener & Suh, 1997; Phillips, 2006). PWB, or other human strengths, are similar to what is advocated by the prudential values and capabilities approach in which realizing one’s potential and striving to make human life better is the goal. Nevertheless, the prudential values and capabilities approach takes the objective environment as a cogent factor and the vision towards a collective goal is explicit. Understanding Well-Being in Multi-Levels Although the basic needs approach is one way to consider QOL, its minimum standards towards human existence furnish little evidence with regards to human nature, its evolution and possible development (Phillips, 2006). VIII. A Multi-Level Approach to Well-Being The above discussion suggests that well-being is not only multi-dimensional, but also spans across multiple levels. Focusing on the individual or external condition is not only limited, but such an approach would deprive us of capturing the richness and dynamic nature of such an important concept. Rather than targeting well-being in a specific level, community psychologists recognized that individuals cannot be understood in isolation from their broader environment. Likewise, well-being should also be understood in multiple levels simultaneously. Cowen’s (1991; 2000) concept of “wellness” and Prilleltensky et al. (Nelson & Prilleltensky, 2005; Prilleltensky & Prilleltensky, 2006) proposed that well-being should go beyond the personal and intra-psychic, and require divergent solutions across different levels. Cowen (1991; 2000) explicated that wellness exists not as an either-or-not, but along a continuum and can be nurtured or inhibited by external circumstance. In addition, wellness depends on many aspects of life so that it cannot be solved by any simplistic solution, but only by comprehensive strategies targeting various levels. Cowen specifically highlighted several sources of influence which had a tremendous impact on peoples’ wellness: (a) the family context in which a child can naturally develop through their infancy; (b) educational experience; (c) significant social setting or mediating structure in which people interact with one another; and, (d) larger society. Cowen (1991; 2000), in the promotion of wellness, proposed several concepts (competence, resilience, social setting modification and empowerment) to be explored in more depth, so that more understanding and subsequent intervention targeting different key sources of influence can be obtained. Prilleltensky et al. further argued that well-being is deeply embedded in the social environment including Prilleltensky et al. further argued that well-being is deeply embedded in the social environment Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 314 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 31 family, community and society (Nelson & Prilleltensky, 2005; Prilleltensky & Prilleltensky, 2006). In their words, “well-being is not either personal, organizational, or collective, but the integration of them all. For any one of these spheres – personal, organizational, or collective – to experience well-being, the other two need to be in equally good shape” (Prilleltensky & Prilleltensky, 2006, p. 11). VIII. A Multi-Level Approach to Well-Being Thus, to understand well-being through a holistic perspective, Nelson and Prilleltensky (2005) suggested the personal (such as self-esteem, independence), interpersonal (having a supportive relationship and freely engaging in society) and collective levels of well-being (able to acquire the community resources). Prilleltensky and Prilleltensky (2006) explained that without the contribution of persons who have optimism, self-efficacy and an equal sense of purpose, social groups cannot function properly. Similarly, people cannot flourish without an immediate environment that is stimulating, respecting, affirmative and reflective, as well as a broader society that is resourceful, fair and free of corruption. In other words, to enable people to be optimistic, having self-efficacy, determination and personal growth, at the personal level, an enabling organizational structure and a community with justice is necessary, if not a must. Such a proposition is aligned with the most recent research on child development. For example, Shonkoff, and Deborah (2000) assert that in the facilitation of healthy child development, there is the importance of knowledge and intervention targeting various levels from the neuron to the neighbourhood, because both the inner biological operation and environmental influence matters. Leadbeater Schellenbach, Maton, and Dodgen (2004) likewise highlight that the building of the competence of individual children cannot be isolated from the family, community and policy development. IX. Comparisons between Different Conceptualizations of Well-Being A multi-level understanding of well-being can be compared with subjective well-being and QOL in several aspects. Firstly, Prilleltensky’s (2006) multi-level model of well-being is explicitly collective in nature, where individual enjoyment is not the primary research concern. Although an individual’s self-determination, freedom and personal growth are highlighted, it should be proportionate in value to other integrated levels, such as collaboration, respect for diversity or social justice. Thus, Prilleltensky’s model is far removed from the SWB or utilitarian approach to QOL, and more in line with the eudaimonic aspect of subjective well-being (PWB), and with the prudential value approach to quality of life. Secondly, a multi-level understanding of well-being complements traditional disciplines (such as psychology, which merely focuses on understanding the personal, and sociology which focuses on the macro level of analysis): a multi-level perspective assesses the critical role of intermediate structures, which can play an enhancing or inhibiting role in bridge building between the individual and society. Thirdly, and most importantly, a multi-level approach to well-being acknowledges the importance of the interrelationship of different levels as a complexity, which is not estranged from social epidemiology’s claim that social relations matter in health. Lastly, the multi-level understanding of well-being provides a comprehensive, inter-disciplinary framework that we can build on and collaborate with diverse disciplines in the pursuit of well-being and in the address of modern life. X. The Strength of Multi-Level Understandings of Well-Being A multi-level understanding of well-being avoids psychologization of the issue. In psychology, well-being has long been studied as a common given at the individual level. It is has been demonstrated, however, that individual well-being (e.g. having goals, coping strategies or a higher level of SWB) does not arise in a vacuum, but is sooner Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 315 Understanding Well-Being in Multi-Levels Understanding Well-Being in Multi-Levels 315 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu closely related to the socio-cultural environment. Despite the fact that empirical studies exist correlating SWB with environmental variables (such as wealth and culture, (Diener & Suh, 1999, 2000a), psychological research maintains a focus on the association between SWB and other psychological attributes (e.g. personality). In the study of health and well-being, most research has been conducted at the individual level while social determinants (Marmot & Wilkinson, 1999) or the social origins (Eckersley, 2001) of health and well-being has been late in receiving warranted scientific attention. The scholarly tendencies appear to suggest that well-being is largely individual, with less value attached to the validity of the social environment for such research. A multi-level research paradigm, inspired by Urie Bronfenbrenner’s (1979) ecological model of human development, can thus provide a timely, although not definitive, starting point to approaching the phenomenon and complex of well-being and to understanding the complexity of field relationships, factors and influences multi-level analysis can engage with (figure 1): Figure 1: Bronfenbrenner’s Ecology of Human Development (1979) Figure 1: Bronfenbrenner’s Ecology of Human Development (1979) A multi-level perspective is a contemporary imperative rather than an option. O’Neil (2005) cogently argued that the way we look at problems or an issue affects how we study them and try to solve them, with ethical Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 316 implications. If well-being is to be conceptualized just within subjective or macro limits, the results will be strategies merely targeting specific spheres of human experience, without attending to the intertwining relationship between people and external circumstances. Likewise, if well-being is to be conceptualized just within the subjective mind of a person, with little or no regard to external circumstances, the result may prove nothing more than a “happy slave” or a “fool’s paradise”: if well-being is positioned beyond the individual’s perception and based on objective and external characteristics (such as crime-rates, unemployment, housing etc), as is proposed by the social indicator approach, the consequence is merely political advocacy. Understanding Well-Being in Multi-Levels People, moreover, will also be falsely perceived as passive recipients of environmental influence, as reflected in the neo-Pavlovianism of the 1930s (headed by Karl Lashley, John Watson and Donald Hebb) which diminished the role of human agency and positioned the individual as the passive receiver of stimuli. In contrast, a multi-level understanding of well-being requires a systemic and integrational perspective where a transactional and reciprocal influence between the multiple levels should be explored in order to attain a clear and consistent understanding of well-being and its complexity. In addition, subsequent intervention would also require multiple solutions, rather than single solutions, to fully address well-being as a phenomenon of multiple, integrated levels. Thus, understanding well-being in the multi-level perspective prepares us for a more well-grounded knowledge, the challenges of our modernity as well as intervention tools (such as including both behavioural change and policy change strategies as different components in a community-wide intervention) to promote people's well-being. A multi-level understanding of well-being highlights the mediating role of immediate settings, experiences, organizations or communities. This matches with QOL research focusing on organization (Sirgy et al., 2006) and community social indicators (Dluhy & Swartz, 2006; Sawicki, 2002; Wiseman, 2009). In view of the fact that more citizens have increasingly less trust towards government and bureaucracy, Berger and Neuhaus (1977) suggested that a mediating structure is needed to stand between individuals and larger institutions, in order for social policy to be more socially conducive and less detached from the realities of individual and community life. In other words, immediate settings can help in mediating between micro and macro aspects of life where the functionability of organizations determines the extent of impact of a megastructure upon the individual. Understanding how mediating structures work in relation to well-being, therefore not only sharpens our understanding towards well-being, but also provides us with ideas for the brokerage of new interventions (such as changing the organizational culture). Lastly, a multi-level understanding of well-being signals the importance of the inter-relationship between person and environment. This is in line with recent advancements in the field of psychology and public health where a recent Institute of Medicine report (IOM) report highlighted the interrelationship between mental, emotional and behavioural disorders as normally stemming from a set of common conditions (National Research Council & Institute of Medicine, 2009). XI. The Challenges of a Multi-Level Understanding of Well-Being Given the strength of the multi-level approach, there are also challenges in understanding well-being in multi-level and taking the ecological context seriously. The challenge is not only conceptual, but also methodological. Conceptually, ‘context’ is a slippery concept as there is a diversity of contexts – immediate settings, institutions or organizations, communities, societies or cultural environments (Trickett, 1996). Furthermore, person and context are mutually intertwined (Cole, 1996; Rogoff, 2003) so much as that the environment and the individual cannot be easily separated. Thus, if well-being is understood in terms of a multi-level, to what precisely does well-being refer? The ambiguous meaning of the concept also deprives us appropriate intervention. One solution is to find a proper unit of analysis (Van der Veer, 2001) for each specific study or action, so that the phenomenon can be understood clearly and coherently. Methodologically, as noted by Bronfenbrenner (1995), psychologists are still unfamiliar with the method that enables the study of people contextually and in terms of multi-level (especially the interplay between the person and the environment and the complex psychological processes therein involved). Some social scientists tend to regard the environment as independent and as a factor or variable external to the individual; however, to understand people in an ecological context we cannot overlook the intertwined relationship between people and the environment. Furthermore, when we examine the ecological context (as with QOL research), it is unclear as to whether the objective versus subjective or structural versus cultural dimensions should be the focus. Hays (1994) has already posited the difficulty in separating culture and social structures. As to untangling the two, if they are inherently intertwined, how to empirically study them are the real challenge facing researchers. Structure and culture are elusive concepts, and can be unfamiliar to psychologists. As psychology is an empirical science, conventional scientific approaches (such as surveys or statistical methods) are often used to study and quantify the causal relationship of social phenomenon. Some methods (such as questionnaires), however, are notoriously ill-equipped for capturing the nuances and detail in subjective experiences, or in the collective meaning of culture. Thus, to be able to study the interaction and reciprocal nature of multi-level well-being, researchers have to equip themselves with multiple tools of methodologies so as to avoid the disciplinary boundary, narrowness or even distortions in research. Understanding Well-Being in Multi-Levels As Biglan, Flay, Embry, & Sandler (2012) advocate, we have to support and foster a nurturing environment to promote human well-being. Such environments minimize biologically and psychologically toxic events, sooner promoting and reinforcing prosocial behaviours, limiting potential occurrences of problem behaviors, and fostering psychological flexibility. In epidemiological studies, the famous Whitehall studies of British civil servants (Marmot, Shipley, & Rose, 1984; Marmot & Wilkinson, 1999) likewise demonstrated the impact of social hierarchy on ill health. Berkman and Glass (2000) illustrated how social networks have a direct impact on population health by providing social support; mutual influence; engagement opportunities; close personal contact and resources, and material goods. Diez-Roux called for bringing back “context” to epidemiology (1998), and suggest using multilevel analysis to uncover the complexity and reciprocal relationships between factors operating at different levels (2000). A wide range of studies also show that social relations matter in health and well-being. Helliwell and his colleagues (2002; Helliwell & Putnam, 2004) also substantiate that community well-being (manifested in higher social capital) is strongly linked to SWB. Baumeister and Leary (1995) and Myers (1999) ascertained that humans have a fundamental need to belong and that good interpersonal relationships are conducive to health and subjective well-being. Thus, all these findings have shown that different facets of well-being are not separate, but closely related. Understanding Well-Being in Multi-Levels 317 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu The author/s of the present article also believe/s that the well-being of a person is not a standalone entity declaring health status and a personal psychological state. Rather, well-being is manifested across various levels and closely associated with the person’s relation to oneself, others and society. XI. The Challenges of a Multi-Level Understanding of Well-Being In academic disciplines, although there are some (psychologists) who argue that well-being is in the subjectively perceived or a gestalt of psychological functioning, others argue that well-being is closely related to, if not in, the objective conditions and broader environment of a person. While subjective well-being or other human characteristics of strength and virtue (Diener, 1984; Diener, Suh, Lucas, & Smith, 1999; Peterson & Seligman, 2004; Seligman & Csikszentmihalyi, 2000; Snyder & Lopez, 2002) have been studied widely in recent decades, the social indicator movement or the recent notion of quality of life has proved itself to be more aligned with, and used at, the policy level (Rapley, 2003; Sirgy et al., 2006). Our study suggests that we indeed have to go beyond the dichotomy of well-being as either a construct of individual attributes or external conditions. Rather, research needs to acknowledge, if not embrace, the dynamic interaction and reciprocal nature of multi-level well-being. This understanding is not only an ethical imperative, but also a sublimate of empirical evidence which posits personal well-being and the social environment. Despite the conceptual and methodological challenge facing researchers adopting the multi-level approach, understanding well-being in such a way can critically provide a more valid picture and understanding of human being’s emotional and social situation. Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 318 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 3 References Bauer, R. A. (1966). Social indicators. Cambridge, Mass.: M.I.T. Press. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachment as a fundamental human motivation. . Psychological Bulletin, 117, 497-529. Berger, P. L., & Neuhaus, R. J. (1977). To empower people. Washington, D. C.: American Enterprise Institute. Berkman, L. F., & Glass, T. (2000). Social integration, social network, social support, and health. In I. Kawachi & L. F. Berkman (Eds.), Social epidemiology (pp. 137-173). New York ; Oxford: Oxford University Press. Biglan, A., Flay, B. R., Embry, D. D., & Sandler, I. N. (2012). The critical role of nurturing environments for promoting human well-being. American Psychologist, 67, 257-271. Bowling, A. (1997). Measuring health : A review of quality of life measurement scales (2nd ed.). Buckingham ; Philadelphia: Open University Press. Bronfenbrenner, U. (1979). The Ecology of Human Development. Harvard University Press. Bronfenbrenner, U. (1995). Developmental ecology through space and time: A future perspective. In P. Moen, G. H. Elder, K. Lüscher & U. Bronfenbrenner (Eds.), Examining lives in context : Perspectives on the ecology of human development (1st ed. ed.). Washington, DC: American Psychological Association. Carr, A. (2004). Positive psychology : the science of happiness and human strengths. London; New York: Brunner-Routledge. Cole, M. (1996). Cultural psychology : a once and future discipline. Cambridge, Mass.: Belknap Press of Harvard University Press. Costanza, R., Fisher, B., Ali, S., Beer, C., Bond, L., Boumans, R., et al. (2007). Quality of life: An approach integrating opportunities, human needs, and subjective well-being. Ecological Economics 61, 267-276. Cowen, E. L. (1991). In pursuit of wellness. American Psychologist, 46(4), 404-408. Csikszentmihalyi, M. (1990). Flow : The psychology of optimal experience (1st ed.). New York: Ha Cummins, R. A. (1997). Comprehensive quality of life scale - Adult. Melbourne: School of Psychology, Deakin University. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95(3), 542-575. Diener, E. (2009). Positive psychology: Past, present, and future. In C. R. Snyder & S. J. Lopez (Eds.), Oxford handbook of positive psychology (2nd ed., pp. 7-12). Oxford ; New York: Oxford University Press. Diener, E., Sapyta, J. J., & Suh, E. (1998). Subjective well-being is essential to well-being. Psychological Inquiry, 9, 33-37. Diener, E., & Suh, E. (1997). Measuring quality of life: Econnomic, social, and subjective indicators. Social Indicators Research, 40, 189-216. Diener, E., & Suh, E. M. (1999). National differences in subjective well-being. In D. Kahneman, E. References Diener & N. Schwarz (Eds.), Well-being : The foundations of hedonic psychology. New York: Russell Sage Foundation. Diener, E., & Suh, E. M. (2000a). Culture and subjective well-being. Cambridge, Mass. ; London: MIT Press. Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 319 Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu Diener, E., & Suh, E. M. (2000b). Measuring subjective well-being to compare the quality of life of cultures. In E. Diener & E. M. Suh (Eds.), Culture and subjective well-being. Cambridge, Mass. ; London: MIT Press. Diener, E., Suh, E. M., Lucas, R. E., & Smith, H. L. (1999). Subjective well-being: Three decades of progress. Psychological Bulletin, 125(2), 276-302. Diez-Roux, A. V. (1998). Bringing context back into epidemiology: Variables and fallacies in multilevel analysis. American Journal of Public Health, 88(2), 216-222. Diez-Roux, A. V. (2000). Multilevel analysis in public health research. Annual Review of Public Hea Dluhy, M., & Swartz, N. (2006). Connecting knowledge and policy: The promise of community indicators in the United States. Social Indicators Research, 79, 1-23. Eckersley, R. (2001). Culture, health and well-being. In R. Eckersley, J. M. Dixon & R. M. Douglas (Eds.), The social origins of health and well-being. Cambridge ; Melbourne: Cambridge University Press. Eckersley, R. (2004). Well and Good. Melbourne: The Text Publishing Company. Eckersley, R. (2004). Well and Good. Melbourne: The Text Publishing Company. Emmons, R. A., & McCullough, M. E. (2004). The psychology of gratitude. Oxford: Oxford Univer & McCullough, M. E. (2004). The psychology of gratitude. Oxford: Oxford University Press. Fitzapatrick, R. (1996). Alternative approaches to the assessment of health-related quality of life. In A. Offer (Ed.), Pursuit of the Quality of Life. Oxford: Oxford University Press. Frank, P. (2002). The preamble of the Constitution of the World Health Organization. Bulletin of the World Health Organization, 80, 982. Fredrickson, B. L. (2000). Cultivating positive emotions to optimize health and well-being. Prevention and Treatment 3. http://journals.apa.org/prevention/volume3/toc-mar07-00.html. Fredrickson, B. L. (2001). The role of positive emotions in positive psycology: The broaden-and-build therogy of positive emotions. American Psychologist, 56(3), 218-226. nough, U. (2000). Vertical and horizontal transcendence. Zygon: Journal of religion and science, 35 Gorsuch, R. L. (2006). Measurement and research design in studying spiritual development. In E. C. Roehlkepartain, P. E. King, L. Wagener & P. L. References Benson (Eds.), The handbook of spiritual development in childhood and adolescence. Thousand Oaks, Calif. ; London ; New Delhi: SAGE. Haas, B. K. (1999). A multidisciplinary concept analysis of quality of life. Western Journal of Nursing Research, 21, 728-742. Hays, S. (1994). Structure and agency and the sticky problem of culture. Sociological Theory, 12(1), 57-72. Helliwell, J. F. (2002). Globalization and well-being. Vancouver: UBC Press. Helliwell, J. F. (2002). Globalization and well-being. Vancouver: UBC Press. Helliwell, J. F., & Putnam, R. D. (2004). The social context of well-being. Philosophical Transactions of the Royal Society B, 359(1449), 1435-1446. Isen, A. M. (1987). Postive affect, cognitive processes, and social behavior. Advances in Experimental Social Psychology, 20, 205-253. Jahoda, M. (1958). Current concepts of positive mental health. New York,: Basic Books. Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 320 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 32 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 3 Jones, K. E. (1998). A study of the difference between Faith Maturity Scale and Multidimensional Self Concept Scale scores for youth participating in two denominational ministry projects. ProQuest Information & Learning, US. Keyes, C. L. M. (2006). Subjective well-being in mental health and human development research worldwide: An introduction. Social Indicators Research, 77, 1-10. Keyes, C. L. M., & Lopez, S. J. (2002). Toward a science of mental health: Positive directions in diagnosis and treatment. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology. Oxford England ; New York: Oxford University Press. Keyes, C. L. M., Shmotkin, D., & Ryff, C. D. (2002). Optimizing well-being: The empirical encounter of two traditions. Journal of Personality and Social Psychology, 82(6), 1007-1023. Keyes, C. L. M., & Waterman, M. B. (2003). Dimensions of well-being and mental health in adulthood. In B. M. H. (Ed.), Well-being : Positive development across the life course Mahwah, N.J.: Lawrence Erlbaum Associates. Langer, E. (2002). Well-being: Mindfulness versus positive evaluation. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology. Oxford: Oxford University Press. Larson, J. S. (1996). The world health organization's definition of health: Social versus spiritual health. Social Indicators Research, 38, 181-192. Leadbeater, B., Schellenbach, C. J., Maton, K., & Dodgen, D. W. (2004). References Research and policy for building strengths: Processes and contexts of individual, family and community development. In K. I. Maton (Ed.), Investing in children, youth, families, and communities: strengths-based research and policy (1st ed.). Washington, DC: American Psychological Association. Lyubomirsky, S., Sheldon, K. M., & Schkade, D. (2005). Pursuing happiness: The architecture of sustainable change. Review of General Psychology, 9(2), 111-131. Marmot, M. G., Shipley, M. J., & Rose, G. (1984). Inequalities in death - specific explanations of a general pattern. Lancei, i, 1003-1006. Marmot, M. G., & Wilkinson, R. G. (Eds.). (1999). Social Determinants of Health. New York: Oxford University Press. McCullough, M. E. (2000). Forgiveness as a human strength: Theory, measurement, and links to well-being. Journal of Social and Clinical Psychology, 19, 43-55. McCullough, M. E., Pargament, K. I., & Thoresen, C. E. (2000). Forgiveness : Theory, research, and practice. New York: Guilford Press. McGregor, I., & Little, B. R. (1998). Personal projects, happiness, and meaning: On doing well and being yourself. Journal of Personality and Social Psychology, 74, 494-512. Michalos, A. C. (2004). Social indicators research and health-related quality of life. Social Indicators Research, 65(1), 27-72. Myers, D. G. (1999). Close relationships and quality of life. In D. Kahneman, E. Diener & N. Schwarz (Eds.), Well-being : The foundations of hedonic psychology. New York: Russell Sage Foundation. National Research Council & Institute of Medicine. (2009). Preventing mental, emotional, and behavioral disorders among young people: Progress and possibilities. Washington, DC: National Academic Press. Understanding Well-Being in Multi-Levels Understanding Well-Being in Multi-Levels Understanding Well-Being in Multi-Levels 321 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 321 Nelson, G. B., & Prilleltensky, I. (2005). Community Psychology: In pursuit of Liberation and Wellbeing. New York: Palgrave Macmillan. O' Neill, P. (2005). The ethics of problem definition. Canadian Psychology, 46(1), 13-20. Pargament, K. I. (1997). The psychology of religion and coping : Theory, research, practice. New York: Guilford Press. Pargament, K. I. (2002). The bitter and sweet: An evaluation of the costs and benefits of religiousness. Psychological Inquiry, 13(3), 168-181. Peterson, C., & Seligman, M. E. P. (2004). Character strengths and virtues : A handbook and classification. New York: Oxford University Press. Phillips, D. (2006). Quality of life : Concept, policy and practice. London ; New York: Routledge. 6). Quality of life : Concept, policy and practice. London ; New York: Routledge. Prilleltensky, I., & Prilleltensky, O. (2006). Promoting well-being : Linking personal, organizational, and community change. Hoboken, N.J.: John Wiley. Raphael, D., Renwick, R., Brown, I., Steinmetz, B., Sehdev, H., & Phillips, S. (2001). Making the links between community structure and individual well-being: Community quality of life in Riverdale, Toronto, Canada. Health and Place, 7(179-196). Rapley, M. (2003). Quality of life research : a critical introduction. London: SAGE. Rate, C. R., Clarke, J. A., Lindsay, D. R., & Sternberg, R. J. (2007). Implicit theories of courage. The Journal of Positive Psychology, 2(2), 80 - 98. Rogoff, B. (2003). The cultural nature of human development. New York: Oxford University Press. Rossi, R. J., & Gilmartin, K. J. (1980). The handbook of social indicators : sources, characteristics, and analysis. New York: Garland STPM Press. Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on hedonic and eudaimonic well-being. Annual Review of Psychology, 52, 141-166. Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 57, 1069-1081. Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being revisited. Journal of Personality and Social Psychology, 69(4), 719-727. Sawicki, D. S. (2002). Improving community indicator systems: Injecting more social science into the folk movement. Planning Theory and Practice, 3(1), 13-32. Schalock, R. L. (2000). Three decades of quality of life. In M. L. Wehmeyer & J. R. Understanding Well-Being in Multi-Levels Patton (Eds.), Mental retardation in the 21st century. Austin, TX: Pro-Ed. Schalock, R. L., & Felce, D. (2004). Quality of life and subjective well-being: Conceptual and measurement issues. In E. Emerson, C. Haton, T. Thompson & T. R. Parmenter (Eds.), The international handbook of applied research in intellectual disabilities. . Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychoogy, 4(219-247). Health, Culture and Society Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 322 Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 322 Seligman, M. E. P. (1991). Learned optimism. New York: Knopf. Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction. American Psychologist, 55(1), 5-14. Shonkoff, J. P., & Phillips, D. (2000). From neurons to neighborhoods : the science of early child development. Washington, D.C.: National Academy Press. Sirgy, M. J., Michalos, A. C., Ferriss, A. L., Easterlin, R. A., Patrick, D., & Pavot, W. (2006). The quality-of-life (QOL) research movement: Past, present, and future. Social Indicators Research, 76, 343-466. Sirgy, M. J., & Rahtz, D. (2006). A measure and method to assess subjective community quality of life. In M. J. Sirgy, D. Rahtz & D. Swain (Eds.), Community Quality of Life Indicators: Best Cases II (pp. 61-74). Netherlands: Springer. Snyder, C. R. (1994). The psychology of hope: You can get there from here. New York: Free Press. Snyder, C. R. (2000). Handbook of hope : Theory, measures, and applications. Burlington: Elsevier. Snyder, C. R., & Lopez, S. J. (2002). Handbook of positive psychology. Oxford England ; New York: Oxford University Press. Snyder, C. R., & Lopez, S. J. (2007). Positive psychology : The scientific and practical explorations of human strengths. Thousand Oaks, Calif.: SAGE Publications. Sternberg, R. J. (1990). Wisdom : Its nature, origins, and development. Cambridge ; New York: Cambridge University Press. Sternberg, R. J. (1998). A balance theory of wisdom. Review of General Psychology, 2, 347-365. Trickett, E. J. (1996). A future for community psychology: The contexts of diversity and the diversity of contexts. American Journal of Community Psychology, 24(2), 209-234. Van der Veer, R. (2001). The idea of units of analysis: Vygotsky's contribution. In S. Chaiklin (Ed.), The theory and practice of cultural-historical psychology. Oxford: Aarhus University Press. VicHealth. (2009). Mental health and wellbeing: Everybody's business. Understanding Well-Being in Multi-Levels VicHealth Letter, 34, 2. Waterman, A. S. (1993). Two conceptions of happiness: Contrasts of personal expressiveness (eudaimonia) and hedonic enjoyment. Journal of Personality and Social Psychology, 64, 678-691. Wiseman, J. (2009). Community wellbeing indicators as tools for citizen engagement and priority setting: Learning from community indicators Victoria. Paper presented at the The Good Life Conference, Griffith University, Brisbane, 12 February, 2009. Wong, C. (2002). The crossroad of community indicators. Planning Theory and Practice, 3(2), 259-260. Understanding Well-Being in Multi-Levels Volume 5, No. 1 (2013) \| ISSN 2161-6590 (online) \| DOI 10.5195/hcs.2013.142 \| http://hcs.pitt.edu 323
W2782865230.txt	https://sjes.springeropen.com/track/pdf/10.1007/BF03399309	en		Discussion of “Measuring the Natural Output Level by DSGE Models: An Empirical Investigation for Switzerland” by Stefan Leist and Klaus Neusser	Zeitschrift für schweizerische Statistik und Volkswirtschaft/Schweizerische Zeitschrift für Volkswirtschaft und Statistik/Swiss journal of economics and statistics	2,010	cc-by	2,051	Discussion of “Measuring the Natural Output Level by DSGE Models: An Empirical Investigation for Switzerland” by Stefan Leist and Klaus Neusser Jean-Marc Natal The natural level of output1 is a concept of paramount importance for monetary policymakers. Besides ensuring zero2 inflation, it is the best allocation3 that stabilization policy can achieve (Goodfriend and King, 2001; Woodford, 2003) and thus a natural benchmark for central banks. Yet, the natural level of output cannot be observed. Identifying and estimating it is a difficult endeavor that requires a general equilibrium model of the economy.4 This paper makes a first serious attempt at conducting such an exercise for Switzerland and should be praised for that. However – and despite the use of state of the art econometrics – I argue that the failure of the model to explain variations in domestic inflation casts some doubts on its ability to really identify and estimate natural output in Switzerland. A Closer Look at the Results The authors’ model-based estimation of the natural level of output results in a measure of the output gap that is at odds with traditional (HP) filtered measures. As shown in Figure 3(a) of the paper, the Swiss natural output gap does not seem to fluctuate at all, meaning that the level of output corresponds, by and large, to the natural level of output over the estimation period. Although this result could be interpreted as a confirmation that monetary policy in Switzerland was conducted optimally during this period5, some caution is warranted. 1 2 3 4 Defined as the level of output that prevails under imperfect competition in the absence of price and wage rigidities. More precisely, it ensures constant inflation at the pre-specified steady-state value, which can be different from zero in certain applications. At least within a New-Keynesian environment, which is the framework chosen by the authors. With the obvious caveat that the estimates will depend strongly on the model specification. © Swiss Society of Economics and Statistics 2010, Vol. 146 (1) 305–311 306 Jean-Marc Natal If the estimated natural output gap (LN) is to be considered a preferable measure of slack in the economy, it must be the case that it improves upon filtered alternatives (Hodrick-Prescott (HP), multivariate HP (MHP), linear detrending (LT), production function (PF)) when it comes to explaining domestic inflation. This must be true, at least, within the model environment. Solving forward6 the NKPC (equation (3.3) in the paper) yields: πH ,t = ∞ κ Et ∑ ( γ f ) j xt + j , τ + λ j =0 (1) which shows that fundamental7 domestic inflation is a function of current and future output gaps. Figure 1 compares fundamental domestic inflation rates – obtained using alternative measures of the output gap – to actual domestic inflation data. The contrast between the rate of inflation implied by the natural output gap and by more traditional filtered methods is striking.8 The bold red line is the historical quarterly domestic inflation rate, the bold green line represents the fundamental inflation rate based on the natural output gap (LN) and the thin lines represent the fundamental inflation rates based on the more traditional filtered methods. Although low during most of the period, domestic inflation still displays very clear cycles and even – on two occasions, in 2000 and 2008 – crosses the 2% (in annualized terms) threshold.9 Surprisingly maybe, and despite the ad hoc nature of their detrending procedure, traditional measures of the output gap give rise to estimates of domestic inflation that are more or less in line with actual inflation, whereas the authors’ computed natural output gap is clearly off the mark, predicting constant inflation of 1% over the whole period. How can we explain such a surprising result? In my view, there are two issues, mainly, related to the model specification and the estimation strategy, that I will address in turns in the next two paragraphs. 5 6 7 8 9 Of course, an alternative interpretation is that the shocks have been particularly small since 1996. By assuming no indexation, like the authors do. The term “fundamental inflation” was coined by J. Galí and relates to the isomorphic formula used to compute the fundamental value of an asset based on its future discounted returns. Note that the fundamental domestic inflation rate is computed by using, in each case, equation (1), the authors’ estimates of κ, τ, λ and γ f and by assuming the same AR(1) process for all output gaps, natural and filtered alike. It is worth mentioning here that the SNB defines price stability in terms of yearly CPI inflation below 2%, and not in terms of quarterly domestic inflation, as depicted here to comply with the model definition of inflation. Discussion of “Measuring the Natural Output Level by DSGE Models” 307 Figure 1: Quarterly Domestic Inflation Rates, as Implied by Natural or Filtered Output Gaps domestic inﬂation (data) fundamental inﬂation (PF) fundamental inﬂation (MHP) fundamental inﬂation natural output gap (LN) fundamental inﬂation (HP) fundamental inﬂation (LT) 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 –0.1 –0.2 II/1997 II/1999 II/2001 II/2003 II/2005 II/2007 II/2009 Issue #1: Model Specification My first concern is that the model may be too simple to be considered a realistic model of an open economy. Especially if it is to be taken to the Swiss data. The authors assume complete markets, perfect risk-sharing, the law of one price between foreign and domestic goods and a unique, worldwide productivity trend, which allows them to derive a very simple and elegant expression for the natural level of output, deemed proportional and negatively correlated to world output (as implied by equation (3.5)).10 10 Previous experiments with DSGE-CH, a dsge model of the Swiss economy developed at the SNB, have shown that i) local currency pricing in export and import markets, ii) real wage rigidity and iii) the incorporation of oil as an input to production and consumption are crucial features of a model able to replicate the dynamics of output and inflation in Switzerland. 308 Jean-Marc Natal But perfect risk sharing also implies that current accounts must be zero at all times, which stands in stark contrast with Switzerland’s experience over the last fifteen years, when current account surpluses have kept increasing to reach exceptional levels by international standards. Moreover, trend productivity is lower in Switzerland than in the rest of the world, a characteristic that is partly reflected in lower real interest rates. Finally, ample empirical evidence as shown that the law of one price does not hold in the short run. In order to get reasonable parameter estimates despite the simplicity of the model, something has to give. The authors chose to treat foreign variables as unobserved states, estimated with a Kalman filter. I argue below that this approach has important drawbacks in the present context as important information to the identification and estimation of natural output is ignored. An obvious alternative is to relax the simplifying assumptions. Clearly, this would complicate the analysis. Natural output would still be a function of foreign output, but also of domestic productivity, the real exchange rate and Switzerland’s net foreign asset position. In addition, a closed form solution for the natural level of output would no longer be available. But this extra complexity would allow to loosen the straightjacket that the present version of the model imposes on the data and to use a larger information set to help identifying and estimating natural output. Issue #2: Estimation Strategy The authors rely on five quarterly time series (CPI inflation, 3M Libor, changes in the terms of trade, Swiss real GDP growth (ΔGDP), change in nominal exchange rate) over the period 1997:2–2009:2 to estimate the model parameters and to filter out Switzerland’s natural output gap, foreign output (y ∗ ) and inflation, as well as the common worldwide technological stochastic trend (z) using the Kalman filter. This is a lot of information to be extracted from only five observable time series! In particular, the state-space representation of the model shows that ΔGDP is related to two AR(1) processes for y ∗ and z, both of which are treated as unobservable. The difficulty is that in the absence of more structural differentiation, the Kalman filter is not able to separately estimate the variance and persistence of these two processes, thus leaving them unidentified.11 This in turn affects the 11 See Hamilton (1994), chapter 13, for a description of the identification problem with the Kalman filter. Discussion of “Measuring the Natural Output Level by DSGE Models” 309 estimation of the natural output gap since the natural level of output is a negative, linear function of y ∗ in the model. Some reverse engineering can illustrate the extent of the problem. If the output gap was closed during the whole period, as estimated by the authors, this would imply that Swiss real GDP was also negatively correlated with foreign output (as imposed by equation (3.5)), a pattern clearly at odds with actual data12, as Figure 2 shows. Figure 2 : Swiss and Foreign GDP Growth 3 2 1 0 –1 –2 –3 Switzerland GDP growth (yoy) Euro area GDP growth (yoy) –4 –5 1985 1990 1995 2000 2005 2010 Artificially increasing the number of unobserved state variables, as is done in this paper, relaxes the (too restrictive) constraints imposed by the model on the data but at the risk of compromising its ability to identify the truly unobservable natural output gap. I would therefore encourage the authors to at least use available information on y ∗ when estimating the model. More generally, they should make 12 Most of Switzerland’s trade is conducted with the Euro area. Using OECD GDP instead would not alter the message. 310 Jean-Marc Natal use of all available information on foreign and domestic variables13, or show that the model – if it is to be trusted as a reliable model of the Swiss economy – is able to capture reasonably well the dynamic behavior of these series. Is Closing the Natural Output Gap Always Optimal ? The natural level of output is an important benchmark for monetary policy. However, as some recent papers have shown, closing the natural output gap by aiming at zero inflation in all periods may not be the optimal stabilization policy if, for instance, the cyclical fluctuations are due to markup or distortionary tax shocks (Woodford, 2003), if oil price shocks interact with real rigidities (Blanchard and Galí, 2007), or if they lead to time-varying cost shares (Natal, 2009). In these cases a monetary policy trade-off arises so that perfectly stabilizing inflation (however defined) may turn out to be quite costly. How costly will depend on the type and the size of the shocks hitting the economy, on the differences between the responses of natural and efficient14 output and on the welfare cost of inflation. Whether or not a closed natural output gap is the best allocation possible in Switzerland is thus an empirical question that can only be answered by estimating a micro-founded model of the Swiss economy. It is a challenging endeavor and this paper is a welcome first step in that direction. Literature Blanchard, Olivier J., and Jordi Galí (2007), “Real Wage Rigidities and the New Keynesian Model”, Journal of Money, Credit and Banking, Blackwell Publishing, 39 (1), pp. 35–65. Goodfriend, Marvin, and Robert G. King (2001), “The Case for Price Stability”, NBER Working Paper No. W8423. 13 The authors mention their reluctance to add measurement errors to the model. I would argue that allowing for different productivity trends and markup shocks in foreign and domestic economy would go a long way towards solving the problem. An alternative would be to rely on limited information estimation methods like GMM or SMM (Ruge-Murcia, 2002). 14 The level of output that would prevail under perfect competition and flexible prices and wages. Discussion of “Measuring the Natural Output Level by DSGE Models” 311 Hamilton, James D. (1994), Time Series Analysis, Princeton University Press. Natal, Jean-Marc (2009), “Monetary Policy Response to Oil Price Shocks”, Federal Reserve Bank of San Francisco Working Paper 2009-16. Ruge-Murcia, F. J. (2002), “Methods to Estimate Dynamic Stochastic General Equilibrium Models”, Economics Working Paper Series, University of California at San Diego. Woodford, Michael (2003), Interest and Prices: Foundations of a Theory of Monetary Policy, Princeton University Press.
https://openalex.org/W4300545140	https://zenodo.org/record/5903039/files/1990%20%20Whale%20Entrapments%20in%20inshore%20fishing%20gear.pdf	English	null	Whale Entrapments in Inshore Fishing Gear and a Summary of the Entrapment Assistance Program in Newfoundland and Labrador During 1990. A Preliminary Report to the Fisheries Development Branch of the Department of Fisheries and Oceans, and the Newfoundland and Labrador Department of Fisheries.	Zenodo (CERN European Organization for Nuclear Research)	1,990	cc-by	6,185	Jon Lien, Julie Huntington, Wayne Ledwell, Tim Huntsman Whale Resea,ch G,oup, 230 Mount 5cio Road Memo,ial Unive,sity of Newfoundland St. John's, Newfoundland Ale 557 Whale Ent,apments in Insho,e Fishing Gea, and a 5umma,y of the Ent,apment Assistance P,og,am in Newfoundland and Lab,ado, Du,ing 1990. Whale Ent,apments in Insho,e Fishing Gea, and a 5umma,y of the Ent,apment Assistance P,og,am in Newfoundland and Lab,ado, Du,ing 1990. A P,elimina,y Repo,t to the Fishe,ies Development B,anch of the Depa,tment of Fishe,ies and Oceans, and the Newfoundland and Lab,ado, Depa,tment of Fishe,ies 31 Decembe, 1990 Executive Summary Assistance was provided to inshore fishermen in Newfoundland and Labrador who incidentally caught large whales and sharks. The animals were removed from the fishing gear quickly, and gear damage and animal mortality minimized. Humpback whales are the most common problem species. This year a total of 75 animals were reported entrapped. Mortality of humpbacks caught in gear was higher than in previous years eN = 10; 13%) because a large number of entrapments occurred during a week when the capelin fishery was at its peak, and also some codtraps were not hauled regularly because of rough weather; 7 of 10 humpbacks died during this week. The numbers of humpbacks caught this year represents a new record for one season; record number of entrapments have now occurred for three years in succession. The numbers of humpbacks reported caught in this period is greater than numbers entrapped during 1979-1980 when humpback abundance inshore was at its peak due to a collapse of capelin stocks. Several research programs are under way to alleviate the number of collisions which occur including a large scale test of acoustical "alarm" devices which assist the whales in locating codtraps. Research has also begun to evaluate reasons for the increase in the humpback problem which has been evident over the past three years. Three other large species of cetaceans were entrapped in fishing gear during 1990, including minke whales eN = 14), finbacks eN = 2) and a single northern bottlenose whale. Entrapments of small cetaceans are also common in inshore fishing gear. Harbour porpoise have been recently classified as "threatened" by COSEWIC in our waters because of the extent of the fishery by-catch and the problem has been given priority status by the International Whaling Commission. other species such as white-beaked dolphins, white-sided dolphins and beluga are also caught regularly. In the past these have been frequently reported to the Entrapment Assistance Program. This year a cooperative project investigated the extent of small cetacean by-catch and a long-term monitoring program was initiated by Memorial University, D.F.O., C.W.S. and the World Wildlife Fund. Leatherback turtles and some sharks were also reported entrapped in fishing gear on occasion but numbers were not high. Incidental catches of seals were also reported to the Entrapment Assistance Program but these data are being systematically monitored by several other programs. This was the eleventh year the Entrapment Assistance Program operated. Executive Summary During that period, mortality of humpbacks has been reduced from about 50% prior to the program, to an average of 9.4% for the past several years; cost of the program per reported entrapment has been about $220. which has been estimated to save fishermen from $250-1,100 per accident; there has not been a single accident or injury which has resulted from dealing with the 887 whale entrapments reported during the period; and most importantly, fishermen continue to value and cooperate with the program. This was the eleventh year the Entrapment Assistance Program operated. During that period, mortality of humpbacks has been reduced from about 50% prior to the program, to an average of 9.4% for the past several years; cost of the program per reported entrapment has been about $220. which has been estimated to save fishermen from $250-1,100 per accident; there has not been a single accident or injury which has resulted from dealing with the 887 whale entrapments reported during the period; and most importantly, fishermen continue to value and cooperate with the program. Table of Contents Executive Summary Page 2 Table of Contents 4 List of Tables 4 List of Figures 5 Acknowledgements 5 Introduction to the Entrapment Assistance Program 7 Methods 7 Results during 1990 8 Discussion 13 References 17 Tables and Figures 24 List of Tables Page Table 1 Humpback whales reported entrapped in inshore 25 fishing gear during 1990. 2 Minke whales reported entrapped in inshore 29 fishing gear during 1990. 3 Other species of cetaceans reported entrapped 30 in inshore fishing gear during 1990. 4 Stranded cetaceans reported during 1990. 31 5 Ice entrapments of cetaceans reported during 1990. 33 6 Sharks reported incidentally caught in inshore fishing gear during 1990. 34 7 Pinnipeds reported incidentally entrapped in 35 inshore fishing gear during 1990. 8 Turtles reported incidentally entrapped in 36 inshore fishing gear during 1990. Table of Contents Executive Summary Page 2 Table of Contents 4 List of Tables 4 List of Figures 5 Acknowledgements 5 Introduction to the Entrapment Assistance Program 7 Methods 7 Results during 1990 8 Discussion 13 References 17 Tables and Figures 24 List of Figu,es List of Figu,es Page Fig. 1 Locations of ,epo,ted humpback in insho,e fishing gea, du,ing whale 1990. ent,apments 37 2 Numbe, of humpback whales ,epo,ted ent,apped in insho,e fishing gea,: 1979-1990. 38 3 Pe,centage mo,tality of humpback whales ent,apped in insho,e fishing gea, (p,e 1979 39 - 1990). List of Tables Page Table 1 Humpback whales reported entrapped in inshore 25 fishing gear during 1990. 2 Minke whales reported entrapped in inshore 29 fishing gear during 1990. 3 Other species of cetaceans reported entrapped 30 in inshore fishing gear during 1990. 4 Stranded cetaceans reported during 1990. 31 5 Ice entrapments of cetaceans reported during 1990. 33 6 Sharks reported incidentally caught in inshore fishing gear during 1990. 34 7 Pinnipeds reported incidentally entrapped in 35 inshore fishing gear during 1990. 8 Turtles reported incidentally entrapped in 36 inshore fishing gear during 1990. Acknowledgements Funds fo, the Ent,apment Assistance P,og,am a,e p,ovided by the Fishe,ies Development B,anch of the Depa,tment of Fishe,ies and Oceans and the Newfoundland and Lab,ado, Depa,tment of Fishe,ies and we g,atefully acknowledge thei, suppo,t, encou,agement and advice. Funds fo, p,ojects which supplement the Ent,apment Assistance P,og,am we,e p,ovided by the Canadian Cent,e for Fishe,ies Innovation, the Wo,ld Wildlife Fund and the Museum of Natu'e. We would like to thank Sue Ca,ve" K,istina Cu"en,Ch,istina Folge" Janice Jones, Ralph Kullencamp, Elling Lien, Dawn Nelson, ShellY Richa,dson, Rosie Seton, Ch,is Spence" Sean Todd, Amy Ve,hulst, and Dena Weisman fo, assistance with field wo,k. Jacques Guigne, John Guzzwell, F,ank Chopin and And,ew Smith aided us 1f1 labo,ato,y wo'k. Ga,y Stenson at Science B,anch, D.F.O. and John Cha,dine of C.W.S. also gave advice and assisted us in many ways. Qu, g,eatest debt, and the most thanks must again go to the insho,e fishe,men of Newfoundland and Lab,ado,. Thei, coope,ation and help, often in a pe,iod of ve,y ha,d wo,k and disappointment, a,e t,uly 'ema'kable. The Ent,apment Assistance P,og,am wo,ks because of thei, effo,ts and we a,e thankful to them fo, thei, coope,ation. .., I Introduction The Entrapment Assistance Program has operated since 1979 to assist the inshore fishermen of Newfoundland and Labrador with problem whales and sharks. Large whales and sharks collide with fishing gear and become entrapped; removing them quickly and safely is difficult for fisherman. The program is designed to minimize the gear and down-time losses which occur when these large animals collide and are entrapped by fishing gear, and to minimize unnecessary mortality to cetaceans due to entrapment. Results from the Entrapment Assistance Program in previous years have been summarized in annual reports. (Lien 1980; Lien and Aldrich 1982; Lien et al. 1982; Lien et al. 1983; Lien et al. 1984; Lien, walter and Harvey-Clark 1985; Lien et al. 1986; Lien, Papineau and Dugan 1987; Lien, Ledwell and Nauen 1988; Lien, Ledwell and Huntington 1989b). Methods During 1990, the Entrapment Assistance Program operated using methods similar to those of previous years (Lien 1988). For the third year in succession, a trained crew of fishermen were hired specifically for dealing with entrapments. Entrapment Assistance was widely advertised by a variety of publications and advertisements, through Fisheries and Oceans field offices and by Newfoundland and Labrador Department of 8 Fisheries field workers. A card advertising the program was mailed to all fishermen in one of the regular D.F.O. Newfoundland Region mailings. Fishermen could call a 24-hour toll-free phone service for advice and assistance. If assistance was requested, a crew was dispatched to help remove the entrapped animals as quickly as possible. During the busiest period of the summer a second crew was ready and responded to calls. Calls were received which reported a variety of problems in addition to the entrapment of large whales and sharks, including marine mammal strandings, ice entrapments of cetaceans and entrapment of marine turtles and seals. Many fishermen also called or stopped by the Whale Research Group at M.U.N. for advice on building "alarms·. As appropriate, fishermen were given information, access to tools or field assistance. Results Humpback Whales: Humpback whales (Megaptera novaengliae) remain the major problem species for inshore fishermen; 54% of entrapment reports involved this species. A listing of the dates, locations, means of entrapment and outcome of reported humpback entrapments is presented in Table 1 . The location of these entrapments is shown in Figure 1. 9 There were a total of 75 humpback entrapments reported during 1990. All entrapments were made by fishermen except two which were reported through Canadian Coast Guard lighthouse keepers. Some entrapments undoubtedly were not reported, but the number of such incidents is low, and under-reporting probably would be about 10%. Mortality of entrapped humpbacks was high this year (13%); 10 animals died as a result of entrapment. Seven of the ten deaths occurred during the week of 23-30 June. This was a period when the capelin fishery was at its peak in several areas; codtraps were in place prior to this time but were not always fished each day. AdditionallY, weather during this week was exceptionally bad and fishermen in several areas were not able to check traps each day. Entrapment assistance was given successfully in 36% of the reported entrapments; animals were released alive and completelY free of gear (29.3%) or released from the gear but had rope or net fragments that remained on the animal after release (6.7%). Self releases occurred in 33% of the reported entrapments, before human intervention or assistance. Humpbacks towed gear off in 17.3% of reported entrapments; the gear in these instances was usually groundfish gill nets. Most entrapments occurred in codtraps (45%), fleets of 10 groundfish gillnets (31%) or salmon gillnets (15%). A few entrapments were reported in capelin traps, crab pots, mackerel traps and experimental gear (9%). The fishing season was generally "late" in 1990 - so were entrapments of humpbacks: 14.6% were reported in May; 29.3% in June; 45.3% in July; 17.3% in August and 6.7% in September. Minkes: A total of 14 minke whales (Balaenoptera acutorostrata) were reported entrapped in fishing gear (Table 2). Two of these reports (14.3%) did not come from fishermen that caught them but were reported stranded or floating, with fishi\lg gear on the body, by others. Thus under-reporting of this species is higher than for humpbacks; under-reporting of smaller cetacean by-catch is typically higher than for larger whales. Results Mortality is also higher; 11 of the 14 animals died as a result of entrapment (78.6%); 3 animals (21.4%) were released with assistance. Codtraps were most frequently the fishing gear involved in minks whale entrapments (50%); groundfish gillnet entrapments were also common (28.6%). Most (85.7%) minke entrapments occurred in July (Table 2). Minkes: A total of 14 minke whales (Balaenoptera acutorostrata) were reported entrapped in fishing gear (Table 2). Two of these reports (14.3%) did not come from fishermen that caught them but were reported stranded or floating, with fishi\lg gear on the body, by others. Thus under-reporting of this species is higher than for humpbacks; under-reporting of smaller cetacean by-catch is typically higher than for larger whales. Mortality is also higher; 11 of the 14 animals died as a result of entrapment (78.6%); 3 animals (21.4%) were released with assistance. Codtraps were most frequently the fishing gear involved in minks whale entrapments (50%); groundfish gillnet entrapments were also common (28.6%). Most (85.7%) minke entrapments occurred in July (Table 2). Other Species of Cetaceans: Three additional species of large cetaceans were caught in inshore fishing gear during 1990 (Table 3). A northern bottlenose (Hyperoodon ampullatus) Other Species of Cetaceans: Three addition large cetaceans were caught in inshore fishing gear during 1990 (Table 3). A northern bottlenose (Hyperoodon ampullatus) was incidentally entrapped and released alive from a squid trap. A fin whale (Balaenoptera physalus) was caught in Harbour Deep, 11 W.8.; this same animal was again caught near Coachman's Cove, W.8. and partially released. The animal had earlier stranded in Trinity, T.B. in September and had been towed off the beach but released towing ropes and floats. A beluga (Delphinapterus leucus) was caught in St. Lunaire in mid-June. Strandings: There were two mass strandings of cetaceans reported during 1990. A group of four sperm whales (Physeter catadon) began stranding in Black Tickle, Labrador. One of the four was beached for a while but re-floated on high tide; all animals were successfully driven from the cove and to sea during this tide. A second stranding of pilot whales (Globicephala melas) occurred sometime during late summer near St. Anthony; these four animals were not discovered until December (Table 4). There were two potential live strandings of northern cetaceans. A group of northern bottlenose were in the Dildo, T.8. area for a number of weeks. During this time several observers believed that the animals were stranding. Similarly a narwhal (Monodon monoceros) observed in the same area was reported as behaving in an unusual manner and near stranding. In Conception Bay, a Sowerby's beaked whale (Mesop!odon bidens) stranded in early June (Table 4). All other strandings were of single animals and most were quite decomposed. A total of six humpbacks were stranded; we 12 cannot be sure if these are any of the same animals that died as a result of entrapment in fishing gear (Table 4). Ice Entrapments: Two ice entrapments of dolphin species were reported (Table 5). In both cases groups of animals were covered by ice; no exact counts of the animals that died were made. Sharks: There were few sharks reported to the Entrapment Assistance Program during 1990 (Table 6). A total of 6 basking sharks (Cetorhinus maximus) and 3 porbeagles (Lamna nasus) were reported entrapped. Without a market, shark by-catch is greatly under-reported. Pinnipeds: Pinnipeds which were reported to the Entrapment Assistance Program as incidentally caught in fishing gear are listed in Table 7. Pinniped by-catch is being monitored by D.F.O. and in a joint M.U.N./D.F.O./C.W.S by-catch program. Therefore few animals were reported to the Entrapment Assistance Program. Other Species of Cetaceans: Three addition Turtles: A program with the World Wildlife Fund encourages the reporting of leatherback turtle (DermochelYs coriacea) sightings and their live release if entrapped in fishing gear. In 1990 there were 6 sightings and 6 gear entrapments reported (Table 8). Two of the animals died as a result of entrapment. 13 Discussion Humpbacks: Entrapments of humpbacks in inshore fishing gear continues at record levels in Newfoundland and Labrador. Trends in the numbers of humpbacks incidentally caught and mortality over the past decade is presented in Figure 2. A peak of entrapments occurred ln 1979-1980 due to a collapse of capelin (Mallotus villosus) stocks (Whitehead and Carscaden 1985); over the past three years there have also been increases in reported entrapments. Mortality of humpbacks resulting from incidental entrapment is a function of the time the animal spends in the gea1-. Prior to the Entrapment Assistance Program mortality was estimated at about 50% (Lien 1980); in years when, due to funding, the program began late mortality was also very high (Lien et al. 1988). Entrapment mortality was high this year during a week in which the capelin fishery and bad weather impeded daily checking of codtraps. Over the past five years entrapment mortality of humpbacks has averaged 9.4%. The Entrapment Assistance Program continues to receive excellent cooperation from fishermen. since 1979 the Whale Research Group has dealt with 887 whale entrapments at an average cost of 5220. Estimated savings for fishermen are difficult to calculate but range from 5250. - 51,100. per entrapment (Lien 14 1980) . There has never been a serious accident in the eleven years of the Entrapment Assistance Program. The numbers of entrapments during the peak of the inshore fishery now requires the use of two entrapment crews to service entrapment calls in a timely manner. The growth of reports appears to be a real increase and not the result of differential reporting tendencies by fishermen. Causes for the increase are not clear but increase in the inshore fishing effort and in the numbers of humpbacks are likely reasons. There is extremely poor documentation of trends in either of these factors, however. During 1990 work continued on the development of acoustical alarms to reduce the number of whale collisions with fishing gear (Guigne, Lien and Guzzwell 1990a; Guigne et al. 1990b; Lien, Guigne and Todd 1990e). There have been many efforts, most inconclusive, to modify fishing gear in order to reduce by-catch of cetaceans (Todd and Nelson 1990). Early "alarm" designs used in the Newfoundland and Labrador inshore fishery have been shown useful in reducing the probability of collisions and the amount of damage which results if a collision does occur (Lien, Todd and Guigne 1990a). Discussion Several studies have shown the perceptual problems that humpbacks have in detecting and locating nets due to the minimal acoustical target that nets provide and the interaction of these sounds with ambient sounds and the presence of bait (Nelson 1990; Lien, Todd and Guigne 1990a; Todd et al. 1990). The 15 new "alarm" devices are much louder and produce a sound which should optimize detection under these constraints (Lien, Guigne and Todd 1990e). Early tests of humpbacks reactions to the sounds showed that the animals were curious about the devices and approached them; further, there was good evidence that the whales could quickly learn to associate the devices with biologically important events or stimuli (Lien et al. 1990c). By next summer we will have produced 300 prototype "alarms" which will be tested on codtraps in areas where major whale damage occurs. Following successful testing, "alarms" will be commercially produced in Newfoundland. Additional tests of the "alarms" will be conducted in South Africa, Australia and the North Pacific. Small Cetacean By-Catch: The incidental captures of small cetaceans has not been adequately studied in Newfoundland and Labrador. Lien (1987; Lien et al. 1989c) found the catches of harbour porpoise (Phocoena phocoena) to be very large. In 1990 COSEWIC listed the harbour porpoise as "threatened" in waters off Eastern Canada based on studies in Gaspe and Quebec (Fontaine et al. 1990), the Bay of Fundy (Read et al. 1990) and Newfoundland and Labrador (Lien et al. 1990b). In the entire NW Atlantic stocks of harbour porpoise are poorly known but it is believed that in most areas, fishery by-catch exceeds the maximum 1% take which might be sustained (I.W.C. 1990). 16 Based on these data, the I.W.C. Workshop on the Mortality of Cetaceans in Passive Fishing Nets and Traps (I.W.C. 1990) held in La Jolla, California in October 1990 concluded that the NW Atlantic stock of harbour porpoise were unable to sustain current levels of removal caused by net and trap fisheries and recommended that immediate steps be taken to reduce mortality of harbour porpoises. Based on these recommendations Memorial University, Fisheries and Oceans and the Canadian Wildlife Service began studies of the by-catch of harbour porpoise as well as seals and seabirds. This work has been supported by the World Wildlife Fund (Canada). Additional studies on the biology of the harbour porpoise (S. Richardson, personal communication) are under way at Memorial University. Discussion In addition, an education program is being planned and implemented. 17 References References Fontaine, Pierre-Michel, C. Barrette, M.O. Hammill and M. Kingsley (1990) Incidental catches of harbour porpoise (Phocoena phocoena) in the Gulf of St. Lawrence and the St. Lawrence River Estuary, Quebec, Canada. Abstracts of the I.W.C. Symposium on Mortality of Cetaceans in Passive Fishing Nets and Traps, La Jolla, California., 20-21 October 1990, p. 10. Fontaine, Pierre-Michel, C. Barrette, M.O. Hammill and M. Kingsley (1990) Incidental catches of harbour porpoise (Phocoena phocoena) in the Gulf of St. Lawrence and the St. Lawrence River Estuary, Quebec, Canada. Abstracts of the I.W.C. Symposium on Mortality of Cetaceans in Passive Fishing Nets and Traps, La Jolla, California., 20-21 October 1990, p. 10. Guigne, J. Y., J. Lien and J. Guzzwell (1990a) Development of acoustic protection for fixed fishing gear to minimize incidental catches of marine mammals. First interim progress report for the Centre for Fisheries Innovation. Centre for Cold Ocean Engineering (C-CORE), St. John's, Newfoundland, June 1990, 14 pp. Guigne, J.Y., J. Lien, J. Guzzwell and A. smith (1990b) Development of acoustic protection for fixed fishing gear to minimize incidental catches of marine mammals: Second progress report for the Centre for Fisheries Innovation, Centre for Cold Ocean Engineering (C-CORE), St. John's, Newfoundland, August 1990, 14 pp. International Whaling Commission (1990) Executive Summary of the Workshop on the Mortality of Cetaceans in Passive Fishing 18 Nets and Traps, Cambridge, U.K., 7 pp. Lien, J. (1980) Whale collisions with fishing Newfoundland. Report to Fisheries and Oceans Newfoundland Region, 31 December, 316 pp. gear in Canada: Lien, J. and D. Aldrich (1982) Damage to inshore fishing gear in Newfoundland by sharks and whales during 1981. CAFSAC WP/82/104, 46 pp. Lien, J., J. Dong, L. Baraff, J. Harvey and K. chu (1982) Whale entrapments in inshore fishing gear during 1982: A preliminary report to Fisheries and Oceans Canada, St. John's, Newfoundland, 20 September, 36 Pp. Lien, J. S. Staniforth, L. Fawcett, R. Vaughan, and J. Dong (1983) Whale and shark entrapments in inshore fishing gear during 1983. Report to Fisheries and Oceans Canada, 26 pp. Lien, J. S. Staniforth, L. Fawcett, R. Vaughan, and J. Dong (1983) Whale and shark entrapments in inshore fishing gear during 1983. Report to Fisheries and Oceans Canada, 26 pp. Lien, J., L. Dix, E. Lee and H. Walter (1984) Whale and shark entrapments in inshore fishing gear during 1984. Report to Fisheries and Oceans Canada, 21 pp. Lien, J., H. References Walter and C. Harvey-Clark (1985) Whale and shark entrapments in inshore fishing gear reported during 1985. Report to Fisheries and Oceans Canada - Newfoundland Region, Lien, J., H. Walter and C. Harvey-Clark (1985) Whale and shark entrapments in inshore fishing gear reported during 1985. Report to Fisheries and Oceans Canada - Newfoundland Region, 19 21 Pp. 21 Pp. Lien, J., K. B,eeck, D. Pinsent and H. Walte, (1986) Whale and sha,k ent,apments in insho,e fishing gea, du,ing 1986: A p,elimina,y ,epo,t to Fishe,ies and Oceans Canada, St. John's, Newfoundland, 33 pp. Lien, J., K. B,eeck, D. Pinsent and H. Walte, (1986) Whale and sha,k ent,apments in insho,e fishing gea, du,ing 1986: A p,elimina,y ,epo,t to Fishe,ies and Oceans Canada, St. John's, Newfoundland, 33 pp. Lien, J. J. Papineau and L. Dugan (1987) Incidental ent,apments of cetaceans, sha,ks and ma,ine tu,tles in insho,e fishing gea, ,epo,ted du,ing 1987 in Newfoundland and Lab,ado,. Repo,t to the Depa,tment of Fishe,ies and Oceans Newfoundland Region and the Newfoundland and Lab,ado, Depa,tment of Fishe,ies, 30 Decembe" 42 pp. Lien, J. (1987) Incidental catches of ha,bou, po,poise (Phocoena phocoena) in Newfoundland. CAFSAC WP/87/168, 6 pp. Lien, J. (1987) Incidental catches of ha,bou, po,poise (Phocoena phocoena) in Newfoundland. CAFSAC WP/87/168, 6 pp. Lien, J. (1988) P,oblems of Newfoundland fishe,men with la,ge whales and sha,ks du,ing 1987 and a ,eview of incidental ent,apment in insho,e fishing gea, du,ing the past decade. The Osp,ey, 19 (1), 30-38; 19 (2) 65-71. Lien, J. (1988) P,oblems of Newfoundland fishe,men with la,ge whales and sha,ks du,ing 1987 and a ,eview of incidental ent,apment in insho,e fishing gea, du,ing the past decade. The Osp,ey, 19 (1), 30-38; 19 (2) 65-71. Lien, J., W. Ledwell, and J. Nauen (1988) Incidental ent,apments in insho,s fishing gsa, du,ing 1988: A p,slimina,y ,epo,t to the Newfoundland and Lab,ado, Depa,tment of Fishe,ies and Fisheries and Oceans Canada Newfoundland Region, 15 Lien, J., W. Ledwell, and J. Nauen (1988) Incidental ent,apments in insho,s fishing gsa, du,ing 1988: A p,slimina,y ,epo,t to the Newfoundland and Lab,ado, Depa,tment of Fishe,ies and Fisheries and Oceans Canada Newfoundland Region, 15 20 December, 30 pp. Lien, J., J. Guigne and F. Chopin (1989a) Development of acoustic protection for fixed fishing gear to minimize incidental catches of marine mammals. Canadian Centre for Fisheries Innovation, St. John's, Newfoundland, 24 pp. Lien, J., W. Ledwell and J. References Huntington (1989b) Whale and shark entrapments in inshore fishing gear in Newfoundland and Labrador. Report to the Newfoundland and Labrador Department of Fisheries and the Department of Fisheries and Oceans: Newfoundland Region, 10 January, 14 pp. Lien, J., G.B. Stenson, and I. Ni (1989c) A review of incidental entrapment of seabirds, seals and whales in inshore fishing gear in Newfoundland and Labrador: A problem for fishermen and fishing gear designers. In Proceedings of the World Symposium on Fishing Gear and Fishing Vessel Design. G Fox and J. Huntington (eds.) Marine Institute, St. John's, Newfoundland, pp. 67-71. Lien, J., S. Todd, and J. Guigne (1990a) Inferences about perception in large cetaceans, especially humpback whales, from incidental catches in fixed fishing gear, enhancement of nets by "alarm" devices and the acoustics of fishing gear. In Sensory Abilities of Cetaceans, J. Thomas and R. 21 Kastelein (eds.), Plenum Press, N.Y., In press. Lien, J., G.B. Stenson, S. Carver and J. Chardine (1990b) How many did you catch? The effect of methodology on by-catch reports obtained from fishermen. International Whaling Commission Workshop on the Incidental Capture of Cetaceans in Passive Fishing Gear, La Jolla, California, Oct. 1990 and submitted to I.W.C. Special Issue on Gillnet By-catch. Lien, J., A. Verhulst, Tim Huntsman, J. Jones and R. Seton (1990c) Reactions of Humpback Whales to Novel Sounds: Curiosity and Conditioning. International Whaling Commission Workshop on the Incidental Capture of Cetaceans in Passive Fishing Gear, La Jolla, California, Oct. 1990, and submitted to I.W.C. Special Issue on Gillnet By-catch. Lien, J., G.B. Stenson, W. Ledwell and J. Huntington (1990d) Incidental catches of marine mammals and damage to inshore fishing gear In Newfoundland and Labrador (1979-1990). Manuscript for World Fisheries Congress, 14-19 April 1991, Athens Greece. Lien, J., G.B. Stenson, W. Ledwell and J. Huntington (1990d) Incidental catches of marine mammals and damage to inshore fishing gear In Newfoundland and Labrador (1979-1990). Manuscript for World Fisheries Congress, 14-19 April 1991, Athens Greece. Lien, J., J. Guigne, and S. Todd (1990e) Acoustical redesign of fixed nets and traps to minimize incidental catches of marine mammals. Manuscript fOT World Fisheries Congress, 14 22 19 Ap,il 1991, Athens G'eece. Nelson, Dawn (1990) A ,eview of gea, and animal cha,acte,istics ,esponsible fo, incidental catches of ma,ine mammals in fishing gea,. Manusc,ipt fo, the Wo,ld Fishe,ies Cong,ess, Athens, G,eece, Ap,il 1991,25 pp. References Nelson, Dawn (1990) A ,eview of gea, and animal cha,acte,istics ,esponsible fo, incidental catches of ma,ine mammals in fishing gea,. Manusc,ipt fo, the Wo,ld Fishe,ies Cong,ess, Athens, G,eece, Ap,il 1991,25 pp. Read, A., L. Mu,ison, P. Be,gg,en, T. Woodley, A. Westgate and D. Gaskin (1990) A tangled web: Ha,bou, po,poises and gillnets in the Bay of Fundy. Abst,acts of the I.W.C. Symposium on Mo,tality of Cetaceans in Passive Fishing Nets and T,aps. La Jolla, Calif., 20-21 Octobe" 1990, p. 40. Todd, S.K., J. Lien, J. Guigne and P. Hunt (1990) Acoustic pe,ception of bait and capelin nets by humpback whales, Megapte,a novaeangliae, with ,efe,ence to fo,aging ability and incidental collisions with fishing gea,. Manusc,ipt submitted to Canadian Jou,nal of Zoology. Todd, S.K., J. Lien, J. Guigne and P. Hunt (1990) Acoustic pe,ception of bait and capelin nets by humpback whales, Megapte,a novaeangliae, with ,efe,ence to fo,aging ability and incidental collisions with fishing gea,. Manusc,ipt submitted to Canadian Jou,nal of Zoology. Todd, S.K. and D. Nelson (1990) A ,eview of modifications to the webbing and setting st,ategies of passive fishing gea, to ,educe incidental by-catch of cetaceans. Submitted to Inte,national Whaling Commission Special Issue on Gillnet by-catch. Whitehead, H. and J.E. Ca,scadden (1985) p,edicting insho,e whale 23 abundance of whales and capelin off the Newfoundland coast. Canadian Journal of Fisheries and Aquatic Science, 42 (5) 976-981. Tabl e I: Humpback whal es reported entrapped in inshore fishing gear during 1990. Location Type of Gear Comments 4 May Southern Hbr, P.B. Codtrap Self release 17 May Hermitage Gillnets Self release 3 June Southern Harbour Codtrap Dead 16 June Bay Bulls Salmon nets Released alive 17 June Old Bonavista Salmon nets Towed gear off 17 June Bay Bulls Salmon nets Towed gear off 18 June Ochre Pit Cove Codtrap Released alive 19 June Lord's Cove Codtrap Self release 19 June Bay Bulls Salmon nets Self release 19 June st. Shotts Groundfish gillnet Released alive mooring 20 June Lower Island Cove Codtrap Released alive 20 June Baine Harbour Codtrap Self release 21 June Frenchman's Cove Codtrap Dead 23 June O'Donnells, 5MB Caplin trap Released alive 23 June Heart's Content Caplin trap Released alive 24 June st. References Bernard's Codtrap Dead 25 June Petit Forte, PB Codtrap Dead 27 June Pouch Cove Codtrap Dead 27 June Bonavista Codtrap Self release Table 1 (continued): 28 June Lower Island Cove Codtrap Dead 28 June Lower Island Cove Codtrap Dead 30 June st. Brendan's Groundfish gillnets Partial release 7 June st. Bride's Groundfish gillnets Released alive 1 July Bay Bulls Codtrap Self release 2 July Petty Harbour Codtrap Released alive 2 July st. Shotts Groundfish gillnets Released alive 2 July Frenchman's Cove Groundfish gillnets Partial release 3 July Heart's Content Codtrap, salmon Released alive and groundfish gillnets 4 July Gaskiers Codtrap Self release 4 July Cape Broyle Salmon nets Towed gear off 5 July Lower Island Cove Codtrap Self release 7 July Bauline South Codtrap moorings, Partial release Salmon and Groundfish gillnets 9 July Mobile Salmon nets (7) Dead 10 July st. Vincent's, 5MB Codtrap Self release 10 July Little Paradise, PB Groundfish gillnets Released alive 10 July st. Mary's Codtrap Released alive 11 July Herring Neck, NDB Groundfish gillnets Released alive 15 July Melrose Harbour Gillnets Self release 16 July Bay Bulls Salmon nets Towed gear off 16 July st. Bride's Gill nets Towed gear off 16 July Lower Island Cove Codtrap Released alive Table 1 (continued): 19 July Little Bay Islands Salmon nets Self release 19 July Whale's Gulch Groundfish gillnets Self release 20 July Mobile Experimental gear Self release 22 July Cotrell's Cove Uncertain Partial release 23 July Duntara Codtrap Self release 23 July st. Shotts Codtrap Released alive 24 July Morton's Harbour Codtrap, flatfish Released alive gillnets 26 July Joe Batt's Arm Groundfish gillnets Released alive 26 July Twillingate Codtrap Self release 28 July st. Anthony Codtrap Dead 29 July Fleur de Lys Codtrap Self release 30 July st. Anthony Salmon nets Released alive 30 July Ladle Cove Codtrap Released alive 30 July Little Bay East, FB Mackerel trap Self release 31 July Aspen Cove Groundfish gillnets Self release 31 July Lord's Cove Codtrap Self release 2 August st. References Lawrence Codtrap Self release 5 August Twillingate Codtrap Self release 7 August Barr'd Islands Codtrap Released alive 9 August Renews Codtrap Self release 9 August Trepassey Groundfish gillnets Towed gear off 9 August Twillingate Unidentified gear Self release in mouth 14 August Mary's Harbour Crab pots Partial release 15 August Deep Bay, Fogo Groundfish gillnets Towed gear off Table 1 (continued): 16 August Seldom, Fogo 16 August Ladle Cove 22 August Elliston 24 August Seal Island Labrador 31 August Cape st. Francis (68 nm off) 9 Sept Hermitage Bay 14 Sept Funk Island 17 Sept st. Bernard's Sept Tadoussac, Que. 28 Sept Miquelon Groundfish gillnets Towed gear off Groundfish gillnets Self release Groundfish gillnets Released alive Codtrap Released alive Crabpots Dead Groundfish gillnets Towed gear off Salmon nets Towed gear off Unidentified Towed gear off Gillnet Unidentified Towed gear off Gillnet Gillnets Towed gear off Table 1 (continued): 16 August Seldom, Fogo 16 August Ladle Cove 22 August Elliston 24 August Seal Island Labrador 31 August Cape st. Francis (68 nm off) 9 Sept Hermitage Bay 14 Sept Funk Island 17 Sept st. Bernard's Sept Tadoussac, Que. 28 Sept Miquelon Groundfish gillnets Towed gear off Groundfish gillnets Self release Groundfish gillnets Released alive Codtrap Released alive Crabpots Dead Groundfish gillnets Towed gear off Salmon nets Towed gear off Unidentified Towed gear off Gillnet Unidentified Towed gear off Gillnet Gillnets Towed gear off Table 2: Minke whales reported entrapped in inshore fishing gear during 1990 Location Type of gear Comments 3 July Bay Roberts Capelin trap Dead 5 July st. Bride's Groundfish gillnets Released alive 10 July Manuels Groundfish gillnets Released alive 10 July Chance Cove, TB Codtrap Dead 11 July Summerville, BB Codtrap Dead 15 July Ferryland ? Dead 17 July Renews Codtrap Dead 19 July Terrenceville ? Dead 22 July Dildo Codtrap Dead 25 July Nipper's Harbour Codtrap Dead 27 July Lord's Cove Codtrap Dead 28 July Twillingate Codtrap Dead 28 Sept Southern Harbour Groundfish gillnets Dead 14 Nov Brigus South Groundfish gillnets Released alive Table 3: Other species of cetaceans reported entrapped in inshore fishing gear during 1990. Date Location Species Comments 13 June St. Lunaire Beluga Dead in lumpnet 5 Sept Dildo, T.B. Northern Released alive Bottlenose from a squid trap. 29 Sept Harbour Deep Fin Towed gillnets off. 8 Oct Baie Ve,te Fin Released alive from variety of gear .. Towed gear off. References Table 4: Stranded cetaceans reported during 1990. Date Location Species Comments 30 April Northwest River, Labrador ? June Whale's Gulch 9 June Seal Cove 12 June Southern Harbour 28 June Garnish 2 July Mobile 3 July Garnish 9 July Baie Verte 22 July Off Bonavista 31 July Twillingate 31 July Aspen Cove 4 August Salt Cove, Burin 1 Sept Dildo 10 Sept Trinity Harbour 28 Sept Dildo 10m baleen whale (Sei ?) Beluga Sowerby's Porpoise Humpbac k (?) Humpback Humpback (?) Humpbac k (?) Humpbac k (?) Humpback(?) Beluga White sided dolphin Northern Bottlenose Finback Narwhale Dead; old 3m est Female; examined Dead; codtrap Dead Floating Dead Reported about to \ strand, but did not Dead Dead Alive; acting strangely Dead; old Acting strangly Towed off; alive but trailing gear Sighting Table 4: Stranded cetaceans reported during 1990. 30 Sept Black Tickle, Lab 4 Sperm Driven out Table 4: continued. 20 Oct Bay Komo Blue Stranding ? Grand Codroy Blue Stranding 14 Dec. St. Anthony 4 Pilot Dead; old. Table 4: continued. 20 Oct Bay Komo Blue Stranding ? Grand Codroy Blue Stranding 14 Dec. St. Anthony 4 Pilot Dead; old. Table 5: Ice entrapments of cetaceans reported during 1990. Location Species Comments 6 Feb. Jamestown Dolphins Died in ice Spp. ? March Rencontre East Dolphins Died in ice SPP. ? Comments Table 6: shar ks reported incidentally caught in inshore fishing gear during 1990. Table 6: shar ks reported incidentally caught in inshore fishing gear during 1990. Table 6: shar ks reported incidentally caught in inshore fishing gear during 1990. Location Species Type of gear 3 July Torbay Basking Codtrap 4 July Portugal Cove South Basking Codtrap 30 July St. Bride's Basking Groundfish gillnets 4 August Spotted Islands, Porbeagle Groundfish gillnets Labrador 4 Sept Whiteaway, TB Basking Squid trap 14 Sept Eastport, BB Basking Mackerel net 28 Sept Snug Harbour, Lab. Basking Unknown 4 Oct Dunville, PB Porbeagle Longlines 8 Oct Random Island Porbeagle Longlines Table 7: Pinnipeds reported incidentally entrapped in inshore fishing gear or stranded in 1990. References * Date Location Species Comments 19 May Change Islands Hood seal Dead 24 May Renews Ring seal Alive at OSC 28 May Terrenceville Walrus Released alive 9 July Holyrood Harp seal Dead 11 Nov Spaniard's Bay Jar seal Alive at OSC 18 Nov Chapel Cove Hood seal Dead 24 Dec Spanard's Bay Harp seal Alive- released * Note that these records are simply the ones which are reported on the Entrapment Assistance Program phone. There are many others and this table does not represent a true picture of pinniped entrapments and strandings. Table 8: Turtles reported incidentally caught in inshore fishing gear during 1990. Location Type of gear Comments 2 Aug La Scie Groundfish gill nets Released alive 10 Aug Lord's Cove Sighting 16 Aug Lamaline Sighting 20-23 Lamaline Sightings Aug 4 Sept Grand La Pierre Mackerel gillnets Dead 12 Sept Burin Groundfish gillnets Released alive 17 Sept Seal Cove, WB Unidentified Released alive 25 Sept Happy Adventure Sighting 27 Sept Tilti ng, Fogo Sighting 1 Oct Leading Tickles Herring nets Released alive 8 Oct Kelly's Island Sighting M a y h a v e rope in mouth 16 Oct Old Perlican Longlines Dead Released alive M a y h a v e rope in mouth Dead
https://openalex.org/W3006571835	https://europepmc.org/articles/pmc6958584?pdf=render	English	null	Why do women deliver where they had not planned to go? A qualitative study from peri-urban Nairobi Kenya	BMC pregnancy and childbirth	2,020	cc-by	8,311	Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 https://doi.org/10.1186/s12884-019-2695-7 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 https://doi.org/10.1186/s12884-019-2695-7 (2020) 20:30 Open Access Abstract Background: In urban Kenya, couples face a wide variety of choices for delivery options; however, many women end up delivering in different facilities from those they had intended while pregnant. One potential consequence of this is delivering in facilities that do not meet minimum quality standards and lack the capacity to provide treatment for obstetric and neonatal complications. Methods: This study investigated why women in peri-urban Nairobi, Kenya deliver in facilities they had not intended to use. We used 60 in-depth audio-recorded interviews in which mothers shared their experiences 2–6 months after delivery. Descriptive statistics were used to summarize socio-demographic characteristics of participants. Qualitative data were analyzed in three steps i) exploration and generation of initial codes; ii) searching for themes by gathering coded data that addressed specific themes; and iii) defining and naming identified themes. Verbatim excerpts from participants were provided to illustrate study findings. The Health Belief Model was used to shed light on individual-level drivers of delivery location choice. Results: Findings show a confluence of factors that predispose mothers to delivering in unintended facilities. At the individual level, precipitate labor, financial limitations, onset of pain, complications, changes in birth plans, undisclosed birth plans, travel during pregnancy, fear of health facility providers, misconception of onset of labor, wrong estimate of delivery date, and onset of labor at night, contributed to delivery at unplanned locations. On the supply side, the sudden referral to other facilities, poor services, wrong projection of delivery date, and long distance to chosen delivery facility, were factors in changes in delivery location. Lack of transport discouraged delivery at a chosen health facility. Social influences included others’ perspectives on delivery location and lack of aides/escorts. Conclusions: Results from this study suggest that manifold factors contribute to the occurrence of women delivering in facilities that they had not intended during pregnancy. Future studies should consider whether these changes in delivery location late in pregnancy contribute to late facility arrival and the use of lower quality facilities. Deliberate counseling during antenatal care regarding birth plans is likely to encourage timely arrival at facilities consistent with women’s preferences. Keywords: Delivery facility choice; Access to care, Maternal and child health, Peri-urban Kenya Why do women deliver where they had not planned to go? A qualitative study from peri-urban Nairobi Kenya V. Naanyu1* , V. Mujumdar2, C. Ahearn3, M. McConnell4 and J. Cohen4 V. Naanyu1* , V. Mujumdar2, C. Ahearn3, M. McConnell4 and J. Cohen4 * Correspondence: vnaanyu@mu.ac.ke; vnaanyu@gmail.com * Correspondence: vnaanyu@mu.ac.ke; vnaanyu@gmail.com 1Department of Health Policy and Management, School of Public Health, College of Health Science, Moi University, Eldoret, Kenya Full list of author information is available at the end of the article * Correspondence: vnaanyu@mu.ac.ke; vnaanyu@gmail.com 1Department of Health Policy and Management, School of Public Health, College of Health Science, Moi University, Eldoret, Kenya Full list of author information is available at the end of the article © The Author(s). 2020 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. Study design and setting Our qualitative project occurred after the pilot random- ized control trial was conducted. In the larger study, 550 pregnant women were recruited in their 5th–7th gesta- tional month and randomized into one of three groups. One group received an unconditional cash transfer given during the 7th month of pregnancy with a label that “this is intended to help you deliver in the facility of your choice.” One group received the same unconditional cash transfer with a label, but was also asked to pre-commit to a delivery facility in their 7th month of pregnancy and was given a conditional cash transfer after delivery if they de- livered in a facility of their choice. A third group served as a control. Women were surveyed twice during pregnancy and then once shortly after delivery. In the informal settlements of Nairobi, Kenya, the ma- ternal and newborn mortality rate is among the highest in the world, despite most women delivering in a facility [15–17]. Hundreds of public and private maternity facil- ities of widely-varying quality and cost operate in Nairobi. There is wide variation in quality across these facilities with some providing very high-quality care while others offering sub-standard care [18]. Many pregnant women in Nairobi have limited and inaccurate information about the quality of delivery facility options available [19]. Lim- ited reliable information about delivery facility options during pregnancy can potentially lead to delivery in facil- ities lacking quality standards and capacity to provide care for emergency obstetric complications [19, 20]. The informal settlements surrounding Nairobi are within 12–15 km of the city center and are primarily made up of low-income residential estates shared with industrial enterprises, especially in locations closer to the city center. These areas are characterized by a large number of public, private, and faith-based health facilities ranging from small pharmacies and outpatient care to large hospitals with maternity wards. These facilities also range widely in cost, size, and services available. Previous research from this trial demonstrated that maternity facilities used by women in the sample varied widely in terms of quality of care, with some facilities well-equipped to handle emergencies and complications for the mother and baby, and others extremely ill-equipped [8]. g y p The complex delivery facility landscape in Nairobi con- tributes to fragmented and often unplanned maternity- care seeking. © The Author(s). 2020 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. Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Page 2 of 9 Page 2 of 9 Background helped women deliver in their desired facility, reduced last minute decisions, and encouraged more timely ar- rival at higher-quality delivery facilities [18]. This study was conducted as a complementary qualitative investiga- tion to this randomized controlled trial and was intended to investigate why women so often deliver in a facility where they had not planned to go. g If women could access a package of effective and affordable interventions for safe childbirth in a timely manner, it is es- timated that 80% of the 350,000 maternal deaths, over 45% of the 1.2 million intrapartum-related stillbirths, and up to 72% of the 3.1 million neonatal deaths occurring annually could be avoided [1–3]. Although the proportion of preg- nant women who deliver in a facility (rather than home or with a traditional birth attendant) has been increasing rap- idly in low-income countries [4], the majority of deliveries occur in very low-quality facilities [5, 6]. While a number of studies have found that pregnant women have strong pref- erences for facilities that provide high quality technical and inter-personal quality care [7–12], other studies have found that facility distance and cost, as well as factors such as fa- milial influences, and fear of discrimination, influence facil- ity decisions [13, 14]. Characteristics of study participants A total of 60 participants successfully engaged in the inter- views. The characteristics of the sample (measured at the time of baseline survey in the larger randomized trial) were relatively evenly distributed between the arms of the study (Table 1). The total sample had a median age of 26 years, 86.7% were married; 60.0% had a high level of liter- acy, 60.0% had some level of secondary schooling, and 33.0% were delivering their first child during this study. Study design and setting Normally, women use multiple providers for prenatal care, deliver with a different provider, and choose a delivery facility very late in pregnancy [7]. Many factors have been shown to influence the choice of delivery loca- tion in the informal settlements of Nairobi, such as educa- tion, wealth, counseling during antenatal care, pregnancy “wantedness,” and parity [21–23]. While several studies have explored the determinants of facility choice in Nairobi and elsewhere [24], there has been very little re- search on how and why pregnant women switch from a chosen delivery location, and why they often change plans late in pregnancy or during labor. Using 60 in-depth interviews, the study sought to an- swer the following question: Why do some women end up delivering where they had not planned to go during preg- nancy? We apply the Health Belief Model (HBM) to iden- tify individual level factors that influenced delivery location [25–27]. The HBM postulates six constructs that predict health behavior where people prevent ill-health; 1) If they consider themselves to be susceptible, 2) If they be- lieve the condition may have serious consequences, 3) If they believe that they can do something to reduce their susceptibility, or the severity of the illness, 4) If there may be desirable outcomes, 5) If they recognize barriers that would affect their health action, and 6) If they believe they can successfully complete the action despite barriers. The sixth construct suggests internal and external cues can in- fluence final uptake of a health action [26, 28]. We conducted a randomized controlled trial in peri- urban Nairobi to explore the impact of cash transfers to pregnant women that incorporated several different in- centives designed to improve planning and help women deliver in their desired facility. That study found that more than half of women deliver in a maternity facility they were not even considering in their 8th month of pregnancy and nearly two thirds deliver in a facility that was not where they wanted to deliver. The cash transfers Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Page 3 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Study participants of described themes. Additionally, verbatim quotes were selected and used to illustrate study findings. The Health Belief Model was then used to explain individual level factors that influenced birth location. This qualitative component of the study enrolled a sub- sample of 60 randomly selected participants from the original study [8], with 20 women enrolled from each of the three treatment arms to participate in in-depth inter- views conducted two to 6 months after their deliveries. We specially targeted women who did not deliver at their intended location and those who were considering multiple delivery facilities. The interviews took place after the original study was completed. Using the list of participants from the main study, women from the three arms were invited by phone and in person to participate. Seventeen women contacted were not engaged in the in- terviews, 10 because of relocation out of the study area, 5 because they were too busy or not available and 2 be- cause of refusal to participate. These women were re- placed with other participants from the same arm. Data collection procedures The lead author, who is proficient in qualitative methodology, trained female research assistants on consent and in-depth interviewing techniques. Research assistants administered one-on-one in-depth interviews lasting approximately one hour at the home of the respondents. An interview guide was used to lay out the open-ended topics in English and Kiswa- hili. The interview guide used in this study has been attached as an Additional file 1. The interviews covered questions about how women decide where to deliver in an environment with many choices, questions about beliefs, values, and cul- tural expectations regarding child birth, questions about birth planning, a detailed description of their labor/birth experi- ence, and reasons why those with a birth plan did not deliver where they had initially planned to go. However, this paper focuses only on data on why women deliver where they had not planned to go. Interviewers were trained to hold a “nat- ural dialogue” with the women, allow free flow of the women’s narratives, and probe for relevant topic areas. Each interview was audio-recorded for later transcription and translation to English. Informed consent was signed prior to all study procedures. Women and their families take time to think about an- ticipated deliveries and some make elaborate birth plans. However, many end up delivering in a facility they were not planning to use or in some cases not even consider- ing. Emergent themes were classified into four themes including: 1) individual level factors, 2) health facility Table 1 Participant Characteristics (N = 60) Variable Age of Respondent, median (IQR) 26 (22–29) Age of Respondent, mean (range) 26.3 (19–42) Marital Status, n (%) Single 5 (8.3) Partner 2 (3.3) Married 52 (86.7) Separated 1 (1.7) Literacy, n (%) High level of literacy 36 (60.0) Moderate/low level of literacy 24 (40.0) Education, n (%) Primary school or less 18 (30.0) Some level of secondary school 36 (60.0) Post-secondary school 6 (10.0) Income, median (IQR) in thousands Monthly earning (Participant) 0 (0–2) Monthly earning (Husband) 12 (7.6–24) Number of facilities considered at baseline Mean (SD) 3.5 (1.3) First Birth, n (%) Yes 18 (30) No 42 (70) Table 1 Participant Characteristics (N = 60) Analysis In this way, some women ended up delivering at home. Women who failed to attend ante-natal care (ANC) shied away from health facility services. It was common for health workers in the maternity unit to require all delivering women to bring their ANC cards in order to facilitate continuity of care from ANC to delivery and onwards to postnatal care. Women who feared health workers due to ANC non-attendance there- fore avoided giving birth at their chosen location. ‘This is due to the early contractions that may make a woman think they’re false alarms only to find out its labor, so at the end of the day they deliver at home.’ (19006) ‘If she had not been attending clinic, she will be afraid to go because she will be asked for her [ANC] book.’ (14071) ‘If she had not been attending clinic, she will be afraid to go because she will be asked for her [ANC] book.’ (14071) Delivery services and associated logistics often require substantial financial resources. For instance, women often must pay transport fares to the health facility, must purchase supplies needed during delivery, and pay for other hospital costs. A woman could choose to deliver at a favorite private health facility only to end up at a pub- lic health facility due to financial constraints: Sudden intense pain, confusion, and unexpected com- plications that occurred during child birth could make a woman change birth location. Those who experienced complications readily decided to change to a new deliv- ery facility where they could get appropriate care. Women described how intense labor pains incapacitated them to a point of confusion about location of birth: ‘Another thing is money; suppose you had planned to go to a private hospital and you don’t have money? Then you’ll just have to go to a public one.’ (25051) ‘Another thing is money; suppose you had planned to go to a private hospital and you don’t have money? Then you’ll just have to go to a public one.’ (25051) ‘You see during labor; one usually gets confused due to the overwhelming pain so you may end up in a different place.’ (04011) There were reports about women who did not disclose their birth plans to their friends and relatives. Once in labor, such women relied heavily on whoever was escort- ing them to a delivery location. Analysis After data collection, all transcribed data were read and categorized into meaningful units that were conse- quently coded using NVivo software. Analysis involved application of both a priori codes (from the question guide) and emergent inductive codes. The study followed Braun and Clarke’s steps of data analysis (2006) which include becoming familiar with data, generating initial codes across the data set and grouping coded data, searching for themes by gathering data that were rele- vant to each theme, reviewing themes, defining and naming themes identified, and producing an analysis re- port and selecting appropriate, vivid quotes in support Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Page 4 of 9 unplanned locations especially if they did not know their EDD: factors, 3) influence of significant others, and 4) lack of transport. unplanned locations especially if they did not know their EDD: The first category includes individual-level factors in- fluencing a woman’s ability to deliver at a chosen loca- tion. Women with birth plans that were not well- considered were reported as more likely to deliver in any health facility within easy reach. This was because a woman without a solid decision on delivery location was ready to deliver at any available health facility: ‘If a woman does not know when she is due, labor comes in suddenly and you will just go to the nearest hospital.’ (27009) ‘One [reason] is because maybe she had travelled and did not know her days were due.’ (10002) ‘Because she had not made a concrete decision, it is like she is gotten unaware hence she ends up going to any place and delivers there.’ (25045) ‘Because she had not made a concrete decision, it is like she is gotten unaware hence she ends up going to any place and delivers there.’ (25045) ‘Sometimes, it could be due to untimely labor which can take place before you reach the intended hospital so you will be forced to go to the convenient hospital.’ (17021) ‘Sometimes, it could be due to untimely labor which can take place before you reach the intended hospital so you will be forced to go to the convenient hospital.’ (17021) Misunderstanding of signs of labor resulted in delivery in unplanned locations. Women were said to ignore contractions that they assumed were false signs of labor. Consequently, some realized too late that they were in active phases of labor. Analysis You find that the hospital you went to does not have what you need.’ (17017) As labor commenced, friends and family were reported as raising new or impulsive ideas on where the mother was to deliver. Spouses were especially influential and could change delivery location on the day of delivery: ‘Maybe the baby has a problem, maybe the positioning of the baby is not proper, and it will force you to go elsewhere.’ (14066) ‘Maybe the baby has a problem, maybe the positioning of the baby is not proper, and it will force you to go elsewhere.’ (14066) ‘You may want to go to this hospital but your husband wants you to go to the other one. You know he has also collected information from outside so there will be an agreement. But mostly you will settle on where he decided because likely he is the one who will pay the bills.’ (04001) On some occasions, health care workers were reported as giving wrong expected dates of delivery. As a result, women went into labor at a time and date that they did not expect. This was disconcerting especially when women made trips to their health facility of choice and were told their EDD was in the near future. As one woman shared, some were sent back home, only to end up delivering near their homes: ‘At times maybe your husband forces you to go to a hospital, you had not planned for.’ (04003) ‘Sometimes, whoever is taking you to give birth decides to take you to a different hospital, not the one you had planned to go.’ (05060) ‘On the day of delivery, I went to Kiambu and I was told that I was not yet ready to deliver and when I came back the pain got worse and I had to deliver at a nearby hospital.’ (09003) Some women lacked a close family aide to escort them to the maternity service of choice. This made them opt for any nearby convenient facilities that offered delivery ser- vices. Some relied on the choices made by neighbors who volunteered to get them to the closest health facility: Distribution of health facilities was also a concern. Some ma- ternity services were located far away from study communities. Analysis They did not fuss about their earlier choices; instead they surrendered decision making on the delivery location to those supporting them: ‘Confusion due to the labor pains … That pain malfunction’s the brain; it kind of affects her psychologically … You feel like you are going to die, it is painful.’ (01008) ‘Maybe she had not told anyone about where she wanted to go and when her time comes, those who are to take her don’t know of her plans so they take her elsewhere.’ (08006) The second category of factors captures health facility influences. Women choose a new delivery location when they observe poor handling of maternity clients by health facility staff. As they seek care and interact with others, they notice disrespectful maternity care at the health facilities. This deters some women who may have planned to deliver at those facilities: For women who were unsure about their expected date of delivery (EDD), the labor was unexpected. As they went about their daily business, labor began and they ended up going to the most convenient location within easy reach. Moreover, those who travelled dur- ing advanced pregnancy were likely to give birth in ‘One may get discouraged at how the doctors handle patients and just leave for another hospital.’ (23006) Page 5 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 your mind and go to another hospital. People can influence you a lot.’ (04001) your mind and go to another hospital. People can influence you a lot.’ (04001) Health facilities did not provide the same level of ma- ternity services. As such, some women attempted to de- liver at their intended facility were referred on to other facilities due to complications that arose in the course of labor. The health care provider wanted to ensure such women got required services accordingly: ‘Or maybe you had already prepared yourself to go to a particular hospital then you realize it’s not what you really had planned for from what you have heard, it will force you to change and go somewhere else even if it is far.’ (14066) ‘They are usually referred during maybe an emergency. Analysis Upon onset of labor, a long distance to chosen the chosen de- livery facility made it difficult for some women to actually de- liver where they had intended. Furthermore, onset of labor at night time meant far off chosen facilities could not be used – it was more convenient to utilize nearby health facilities. ‘Well, in case of early labor and you cannot help yourself, your neighbor may take you to the closest but not of your choice.’ (27001) ‘Well maybe, the person to take you [to the health facility] may not be there and another one is available. So, you will end up going to the nearest hospital.’ (01066) ‘You know it depends, one can go into labor and she will feel that the baby is near and she cannot get to the hospital she had chosen so she will just go to the nearest.’ (21015) Transport was the final factor discussed as influencing access to chosen facilities. Women who lacked readily available means of transport ended up delivering in places they had not chosen. This was a particular barrier for women who had chosen health facilities that were far from their homes. Moreover, some taxi drivers were re- ported as unwilling to transport women who were in labor to health facilities in order to avoid the possibility of delivery happening in their vehicles while in transit. ‘Her husband saw that the only place we can get help is the nearest private because it was at night. It was 2am at night.’ (14081) A third group of factors captures cues and stimuli from others including husbands, neighbors, and any other es- cort to the delivery location. Rumors and reports about maternity services could dissuade a mother from fulfilling her initial birth plan. Two quotes are illustrative: ‘Something else you can plan to go to a hospital which is far away but due to lack of means of transport you go to a closer hospital.’ (05030) ‘Then there is also that influence from people. You may have decided till the last minute then you change Page 6 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Page 6 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 ‘Also means of transport. Discussion In this study, a number of factors seem to influence de- livery at an unplanned facility. At the individual level, several cognitive factors corresponding to the HBM do- mains of perceived risk, perceived benefits, perceived barriers, along with the cues to action and self-efficacy, were reported as influencing final birth location. In addition, emotional factors seem to make mothers deliver where they had not intended to go. For instance, fear of health workers who reprimand women aggres- sively for failing to attend antenatal care makes women avoid health facilities. Past literature reports that disres- pect and abuse during childbirth is prevalent in Kenya and barriers to skilled birth delivery include fear of being mistreated or neglected by health workers [20, 34, 35]. According to the HBM concept of perceived risk, people prevent ill-health if they consider themselves to be suscep- tible. Findings suggest women opt for a new delivery loca- tion when they sense danger as they observe their body changes, experience sudden intense pain and confusion, or go into abrupt labor. In addition, onset of labor at night time and unexpected birth complications are perceived to be risky and contribute to changes in delivery location. These findings have also been seen in other settings where nighttime delivery and unexpected birth complications have been shown to influence delivery location [29]. The socioeconomic background of the women influ- enced their ability to deliver at a chosen facility. Finan- cial constraints were a barrier for women who needed money for delivery-preparation, transportation fares, and costs of delivery services. Other studies have shown that hidden costs associated with free maternity care are an obstacle and they often include transportation expenses, costs of food, and incidentals [36, 37]. y Perceived benefits closely align with perceived risks. When women believe that they can do something to reduce their susceptibility of a risky delivery, they make changes to the birth plan. As the EDD nears, women observe their preg- nancy and reflect on consequences of giving birth in certain locations of their choice. Consequently, they may change birth location in order to ensure safer delivery in a facility that has skilled staff and the necessary maternity infrastruc- ture for mother and newborn. This has been seen across the world, particularly for those who are choosing between a home or facility-based delivery [30]. Social networks were also a factor influencing the final birth location. Analysis Some taxi drivers refuse to take the women [to health facilities] so you have to go to the nearest.’ (02012) Self-efficacy applies to a woman’s belief that they she can successfully deliver at a place of choice despite bar- riers. A lack of firm decision-making and planning dur- ing pregnancy leads to women delivering where they had not intended to go. Research has indicated that women’s decision-making power influences delivery location in other settings as well [32, 33]. Strengthening self-efficacy by bolstering the confidence of women to make deci- sions surrounding their delivery location could reduce the lack of decision-making and help women choose to deliver where they want. ‘Maybe the taxi or means of transport are not easily available or the hospitals are quite far.’ (11022) ‘Maybe the taxi or means of transport are not easily available or the hospitals are quite far.’ (11022) Conclusion Much of the current literature on maternity care-seeking focuses on the determinants of facility choice without con- sidering the factors that may contribute to women deliver- ing in facilities that differ from their intentions. In this study, we explored factors that seem to influence delivery at a location not intended for among women in peri- urban Nairobi, Kenya. Numerous cognitive, health sys- tems, and other factors influence delivery facility choice in peri-urban Nairobi. This study suggests many reasons for change in birth location are beyond the delivering woman. Significant barriers, including lack of transport at odd hours, lack of support at time of labor and delivery, dis- tance to desired facility, swiftness of labor are all challen- ging to balance, especially for low income women in informal settlements. It is possible that if nurses spend more time counseling women during ANC regarding the development and implementation of a birth plan, more women would be able to deliver in facilities that are con- sistent with their preferences and arrive at the facility earl- ier in labor. Further research should appreciate localized barriers to fulfillment of birth plans and consequently use controlled trials to explore how well diverse interventions reduce delivery at unexpected locations. Accordingly, this study on individual patient choice should be accompanied by a strong health systems approach that includes a strong lens on policy implications. Health system factors influencing delivery choice have been reported in past studies including lack of sufficient staff and training, insufficient referral systems, costs associated with de- livery, and distance to health facility [13, 34, 36, 47, 48]. Women are discouraged from delivering where they had planned to go due to poor handling of maternity clients by health facility staff. Research has suggested that poor provider attitude and behavior have been barriers to facility based de- livery which may lead to women delivering were they did not intend to go [49]. Findings suggest that health care workers’ use of wrong expected dates of delivery, and their sudden re- ferral of mothers to other facilities, makes some women de- liver where they had not planned to go. A recent study conducted in Zambia showed that approximately half of the women who attended an antenatal care visit were made aware of their EDD which enabled them to plan for their de- livery. Conclusion Either having the wrong EDD or not having any infor- mation on their EDD can be a major barrier to the planning process for women which may lead to many delivering where they had not intended to go [50]. This study had strengths and limitations. While a number of previous studies have evaluated the reasons why women choose certain delivery facilities, this is one of the first stud- ies to investigate why women may not deliver in places they had hoped and intended to go. We evaluate this question in the context of peri-urban Nairobi, Kenya where the facil- ity used for delivery can have important implications for the quality of care received in delivery and subsequent health outcomes for mother and baby. This research has important implications because of the implications for ma- ternal and child health outcomes when delivery occurs in facilities that cannot meet minimal quality and safety re- quirements. The sample of 60 recently delivered women allowed for varying viewpoints on the subject matter as they all reflected on why they delivered where they had not intended to go. However, the study also had limitations. Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12884-019-2695-7. Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12884-019-2695-7. Supplementary information Supplementary information accompanies this paper at https://doi.org/10. 1186/s12884-019-2695-7. Additional file 1. Study interview guide. Discussion BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Page 7 of 9 Page 7 of 9 environmental barriers to delivery at a chosen location. Distance and travel barriers, including the costs associ- ated with transportation, have been cited frequently as barriers to delivering at health facilities or delivering where women intended. In some settings, the mode of transportation is limited to motorbike or walking, which is not a possibility for some women who are already in labor [13, 29, 45]. Additionally, a recent study found that in an evaluation of 43 Demographic and Health Surveys from different countries in Africa and Asia, transporta- tion plays a major role in where women choose to de- liver, which was reflected in this study [46]. If women go to whatever place is closest when their birth plan fails, they are less likely to go to a facility that is good quality. Women living in informal settlements, such as those in our study, are often closest to low quality facilities and delivering in such circumstances is likely to have import- ant implications for outcomes for the mother and baby. We interviewed a random sub-sample of study participants, stratified by the three arms of the intervention. This study would have benefitted from including a purposive sample of women stratified by socio-demographics (e.g. women who were single versus those who were married or first- time mothers versus those who had multiple children) and purposively reaching those who had changed from the intended facility of delivery to another one, and had experi- enced bad delivery outcomes. More research should also explore how well interventions aimed at individual, health system, and other factors improve women’s ability to de- liver in their desired maternity facilities. Abbreviations AMREF: African Medical and Research Foundation; ANC: Ante Natal Care; EDD: Expected Date of Delivery; HBM: Health Belief Model Discussion Husbands, neighbors, and other escorts to the delivery location were reported as over-riding women’s choices about where to deliver. A study conducted in Uganda found that when a woman consulted with her spouse, friend, or family member, she was more likely to deliver with a skilled birth attendant [38] and other studies have shown the importance of male partners in the decision making process of where to deliver [39, 40]. In some cases, women lacked an aide to escort them so they opted for any nearby convenient facilities that offered delivery services. Previous research has shown that this has been found to be particularly problematic during the night time when there is no escort available to bring women to the health facility for delivery [41]. Lack of perceived severity with respect to progress in labor applies to themes on misunderstanding signs of labor, precipitate labor, as well as travel close to EDD. If a woman does not perceive her progress in labor and/or potential birth complication risks as far along or severe enough, she may not leave enough time to reach the de- livery location of her choice, leading her to be forced to deliver where she had not intended. Similar results have been found in various settings where “quick labor” forced some to change delivery location or to deliver at home instead of a facility [29, 31]. Moreover, limited sharing of birth plan information and knowledge about pregnancy and labor seemed to influence cues to action. Mothers’ lack of communication with birth partners on choice of facility, poor birth plans, indeterminate EDD, and misunderstanding of signs of labor are factors that would influence cues to action. Women face barriers to obtaining care whether they live in rural, urban, or peri-urban settings [42–44]. Usu- ally the issue of distance to maternity facilities is thought to be a problem in rural areas but our study suggests women in urban areas - where distances are not actually far - experience this barrier too. While most women in the informal settlements of Nairobi have a high-quality facility within five kilometers, higher-quality facilities are much more likely to be located outside of the settle- ments in the city center, requiring women to travel fur- ther to reach them [8]. Long distance and lack of means of transport on the day of delivery were prominent Naanyu et al. Acknowledgements We wish to thank the research assistants for delivering well on data collection activities. We also thank reviewers who helped us improve the initial manuscript. Page 8 of 9 Page 8 of 9 Page 8 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Funding ld We would also like to thank the Bill and Melinda Gates Foundation (Grant number 01818000165) for funding through the Grand Challenges Explorations program. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funder. The findings and conclusions in this paper are those of the authors and do not necessarily represent the official position of the funding agencies. 13. Bohren MA, Hunter EC, Munthe-Kaas HM, Souza JP, Vogel JP, Gülmezoglu AM. Facilitators and barriers to facility-based delivery in low- and middle-income countries: a qualitative evidence synthesis. Reprod Health. 2014;11(1):71. 14. Mason L, Dellicour S, Ter Kuile F, Ouma P, Phillips-Howard P, Were F, et al. Barriers and facilitators to antenatal and delivery care in western Kenya: a qualitative study. BMC Pregnancy Childbirth. 2015;15:26. 15. Matthews Z, Channon A, Neal S, Osrin D, Madise N, Stones W. Examining the “Urban Advantage” in Maternal Health Care in Developing Countries. PLoS Med. 2010. https://doi.org/10.1371/journal.pmed.1000327. Availability of data and materials The data used in this study can easily reveal some of the study participants due to the personal narratives used. To avoid breach of confidentiality, data are available from the corresponding author on reasonable request. 16. Ziraba AK, Madise N, Mills S, Kyobutungi C, Ezeh A. Maternal mortality in the informal settlement of Nairobi city: what do we know? Reprod Health. 2009a;6:6. 17. Ziraba AK, Mills S, Madise N, Saliku T, Fotso JC. The state of emergency obstetric care services in Nairobi informational settlement and environs: Results from a maternity health facility survey. BMC Health Serv Res. 2009b;12(9):46. Consent for publication Not Applicable. 20. Abuya T, Warren CE, Miller N, Njuki R, Ndwiga C, Maranga A, et al. Exploring the prevalence of disrespect and abuse during childbirth in Kenya. PLoS One. 2015;10(4):e0123606. https://doi.org/10.1371/journal.pone.0123606. Competing interests The authors declare that they have no competing interests 21. De Jongh T, Gurol-Urganci I, Allen E, Jiayue Zhu N, Atun R. Barriers and enablers to integrating maternal and child health services to antenatal care in low and middle income countries. Bjog. 2016;123(4):549–57. Author details 1Department of Health Policy and Management, School of Public Health, College of Health Science, Moi University, Eldoret, Kenya. 2Department of Obstetrics and Gynecology, Thomas Jefferson University Hospital, Philadelphia, PA, USA. 3Department of HIV, ID and Global Medicine, UCSF, San Francisco, California, USA. 4Department of Global Health and Population, Harvard T.H. Chan School of Public Health, Boston, USA. Ethics approval and consent to participate Written informed consent was obtained from study participants. The study was approved by the Institutional Review Boards of African Medical and Research Foundation, # P144/2014 and the Harvard T. H. Chan School of Public Health, # 3265–01. 18. Cohen J, Rothschild C, Golub G, Omondi GN, Kruk ME, Mcconnell M. Measuring the impact of cash transfers and behavioral ‘nudges’ on maternity care in Nairobi, Kenya. Global Health Policy. 2017b;36:1956–64. 19. Siam ZA, McConnell M, Golub G, et al. Accuracy of patient perceptions of maternity facility quality and the choice of providers in Nairobi, Kenya: a cohort study. BMJ Open. 2019;9:e029486. Authors’ contributions 10. Kruk ME, Paczkowski M, Mbaruku G, Pinho HD, Galea S. Women’s preferences for place of delivery in rural Tanzania: a population-based discrete choice experiment. Am J Public Health. 2009;99(9):1666–72. VN, MM and JC contributed to the design, data collection, and analysis of data. VM, CA and VM drafted the first draft of the manuscript. All authors contributed in interpretation of study findings and agree with the findings presented in this paper. They all contributed to the writing and editing of the final manuscript. All authors read and approved the final manuscript. 11. Larson E, Hermosilla S, Kimweri A, Mbaruku GM, Kruk ME. Determinants of Perceived Quality of Obstetric Care in Rural Tanzania: A Cross-Sectional Study. BMC Health Serv Res. 2014;14:483. 12. Larson E, Vail D, Mbaruku GM, Kimweri A, Freedman LP, et al. Moving Toward Patient-Centered Care in Africa: A Discrete Choice Experiment of Preferences for Delivery Care among 3003 Tanzanian Women. PLoS One. 2015;10(8):e0135621. References 24. Fotso JC, Mukiira C. Perceived quality of and access to care among poor urban women in Kenya and their utilization of delivery care: harnessing the potential of private clinics? Health Policy Plan. 2012;27(6):1. 1. Colbourn T. Achieving progress for mothers, babies, and children in Kenya. Lancet Glob Health. 2017;5(8):e732–3. 1. Colbourn T. Achieving progress for mothers, babies, and children in Kenya. Lancet Glob Health. 2017;5(8):e732–3. 1. Colbourn T. Achieving progress for mothers, babies, and children in Kenya. Lancet Glob Health. 2017;5(8):e732–3. 2. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Look PF. WHO analysis of causes of maternal death: A systematic review. Lancet. 2006;367(9516):8. 2. Khan KS, Wojdyla D, Say L, Gülmezoglu AM, Look PF. WHO analysis of causes of maternal death: A systematic review. Lancet. 2006;367(9516):8. 25. Becker MH. The health belief model and personal health behavior. Health Educ Monogr. 1974;2:149. 3. Sialubanje C, Massar K, Hamer DH, Ruiter RAC. Understanding the psychosocial and environmental factors and barriers affecting utilization of maternal healthcare services in Kalomo, Zambia: a qualitative study. Health Educ Res. 2014;29(3):521–32. 3. Sialubanje C, Massar K, Hamer DH, Ruiter RAC. Understanding the psychosocial and environmental factors and barriers affecting utilization of maternal healthcare services in Kalomo, Zambia: a qualitative study. Health Educ Res. 2014;29(3):521–32. 26. Champion V, Skinner CS. The health belief model. In: Rimer BK, Glanz K, Viswanath K, editors. Health behavior and health education. 4th ed. San Francisco: Jossey-Bass; 2008. p. 45–65. 27. Loke AY, Davies L, Li S. Factors influencing the decision that women make on their mode of delivery: the Health Belief Model. BMC Health Serv Res. 2015;15:274. 4. Victora CG, Barros AJD, Axelson H, Bhutta ZA, Chopra M, Franca GVA, et al. How changes in coverage affect equity in maternal and child health intervention in 35 Countdown to 2015 countries: an analysis of national surveys. Lancet. 2012;380(9848):1149–56. 4. Victora CG, Barros AJD, Axelson H, Bhutta ZA, Chopra M, Franca GVA, et al. How changes in coverage affect equity in maternal and child health intervention in 35 Countdown to 2015 countries: an analysis of national surveys. Lancet. 2012;380(9848):1149–56. 28. Rosenstock IM, Stecher VJ, Becker MJ. Social Learning theory and the Health Belief Model. Health Educ Q. 1988;15(2):8. 29. Amooti-Kaguna B, Nuwaha F. Factors influencing choice of delivery sites in Rakai district of Uganda. Soc Sci Med. 2000;50(2):10. 5. Kruk ME, Leslie HH, Verguet S, Godfrey MM, Adanu RMK, Langer A. Author details 1D f 22. Fotso JC, Ezeh A, Madise N, Ziraba A, Ogollah R. What does Access to Maternal Care Mean Among the Urban Poor? Factors Associated with Use of Appropriate Maternal Health Services in the Slum Settlements of Nairobi, Kenya. Matern Child Health J. 2008;13(1):7. 23. Sialubanje C, Massar K, van der Pijl MSG, Kirch EM, Hamer DH, Ruiter RAC. Improving access to skilled facility-based delivery services: Women’s beliefs on facilitators and barriers to the utilisation of maternity waiting homes in rural Zambia. Reprod Health. 2015;12:61. https://doi.org/10.1186/s12978-015- 0051-6. Received: 4 March 2019 Accepted: 23 December 2019 References Quality of basic maternal care functions in health facilities of five African countries: an analysis of national health system surveys. Lancet Glob Health. 2016;4(11):845–55. basic maternal care functions in health facilities of five African countries: an analysis of national health system surveys. Lancet Glob Health. 2016;4(11):845–55. analysis of national health system surveys. Lancet Glob Health. 2016;4(11):845–55. 30. Belay A, Sendo E. Factors determining choice of delivery place among women of child bearing age in Dega Damot District, north west of Ethiopia: a community based cross- sectional study. BMC Pregnancy Childbirth. 2016; 16(1):229. 6. Sharma J, Leslie HH, Kundu F, Kruk ME. Poor Quality for Poor Women? Inequities in the Quality of Antenatal and Delivery Care in Kenya. PLoS One. 2017;12(10):e0171236. 6. Sharma J, Leslie HH, Kundu F, Kruk ME. Poor Quality for Poor Women? Inequities in the Quality of Antenatal and Delivery Care in Kenya. PLoS One. 2017;12(10):e0171236. 31. Keri L, Kaye D, Sibylle K. Referral practices and perceived barriers to timely obstetric care among Ugandan traditional birth attendants (TBA). Afr Health Sci. 2010;10(1):75–81. 7. Cohen J, Golub G, Kruk ME, McConnell M. Do active patients seek higher quality prenatal care?: A panel data analysis from Nairobi, Kenya. Prev Med. 2016;92:74–81. 7. Cohen J, Golub G, Kruk ME, McConnell M. Do active patients seek higher quality prenatal care?: A panel data analysis from Nairobi, Kenya. Prev Med. 2016;92:74–81. 32. Hailu D, Berhe H. Determinants of institutional childbirth service utilisation among women of childbearing age in urban and rural areas of Tsegedie district, Ethiopia. Midwifery. 2014;30(11):8. 8. Cohen JL, Lofgren K, McConnell M. Precommitment, Cash Transfers, and Timely Arrival for Birth: Evidence from a Randomized Controlled Trial in Nairobi Kenya. Am Econ Rev. 2017a;107(5):501–5. 33. Taghipour A, Sadat Borghei N, Latifnejad Roudsari R, Keramat A, Jabbari Nooghabi H. Psychological Empowerment Model in Iranian Pregnant Women. Int J Community Based Nurs Midwifery. 2016;4(4):339–51. 9. Kruk ME, Hermosilla S, Larson E, Mbaruku GM. Bypassing primary care clinics for childbirth: a cross-sectional study in the Pwani region, United Republic of Tanzania. Bull World Health Organ. 2014;92(4):246-53. Page 9 of 9 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 34. Anastasi E, Borchert M, Campbell OM, Sondorp E, Kaducu F, Hill O, Lange IL. References Lee AC, Lawn JE, Cousens S, Kumar V, Osrin D, Bhutta ZA, Darmstadt GL. Linking families and facilities for care at birth: What works to avert intrapartum-related deaths? Int J Gynecol Obstet. 2009;107(Suppl 1):S65-88. 48. Uzondu CA, Doctor HV, Findley SE, Afenyadu GY, Ager A. Female Health Workers at the Doorstep: A Pilot of Community-Based Maternal, Newborn, and Child Health Service Delivery in Northern Nigeria. Glob Health Sci Pract. 2015;3(1):97–108. 49. Mannava P, Durrant K, Fisher J, Chersich M, Luchters S. Attitudes and behaviours of maternal health care providers in interactions with clients: a systematic review. Global Health. 2015;11(1):36. 49. Mannava P, Durrant K, Fisher J, Chersich M, Luchters S. Attitudes and behaviours of maternal health care providers in interactions with clients: a systematic review. Global Health. 2015;11(1):36. 50. Stekelenburg J, Kyanamina S, Mukelabai M, Wolffers I, van Roosmalen J. Waiting too long: low use of maternal health services in Kalabo, Zambia. Trop Med Int Health. 2004;9(3):390–8. 50. Stekelenburg J, Kyanamina S, Mukelabai M, Wolffers I, van Roosmalen J. Waiting too long: low use of maternal health services in Kalabo, Zambia. Trop Med Int Health. 2004;9(3):390–8. References Losing women along the path to safe motherhood: why is there such a gap between women’s use of antenatal care and skilled birth attendance? A mixed methods study in northern Uganda. BMC Pregnancy Childbirth. 2015;15(1):287. 35. Bohren MA, Vogel JP, Hunter EC, Lutsiv O, Makh SK, Souza JP, et al. The mistreatment of women during childbirth in health facilities globally: a mixed-methods systematic review. PLoS Med. 2015;12(6). mixed-methods systematic review. PLoS Med. 2015;12(6). 36. Nahar S, Costello A. The Hidden Cost of Free Maternity Care in Dhaka, Bangladesh. Health Policy Plan. 1988;13(4):5. 37. Ravit M, Philibert A, Tourigny C, Traoré M, Coulibaly A, Dumont A, et al. The Hidden costs of a free caesarean section policy in West Africa (Kayes Region, Mali). Matern Child Health J. 2015;19(8):1734–43. 38. Kabakyenga JK, Östergren PO, Turyakira E, Pettersson KO. Influence of Birth Preparedness, Decision-Making on Location of Birth and Assistance by Skilled Birth Attendants among Women in South-Western Uganda. PLoS One. 2012;7(4):e35747. 39. Danforth E, Kruk ME, Rockers PC, Mbaruku G, Galea S. Household Decision- making about Delivery in Health Facilities: Evidence from Tanzania. J Health Popul Nutr. 2009;27(5):7. 40. Tancred T. Birth preparedness and place of birth in Tandahimba district, Tanzania: what women prepare for birth, where they go to deliver, and why. BMC Pregnancy Childbirth. 2017;16(1):165. 41. Kumbani L, Bjune G, Chirwa E, Malata A, Odland JØ. Why some women fail to give birth at health facilities: a qualitative study of women’s perceptions of perinatal care from rural Southern Malawi. Reprod Health. 2013;10(9):9-10. 42. Aryeetey R, Aikins M, Dako-Gyeke P, Adongo P. Pathways utilized for antenatal health seeking among women in the Ga East District, Ghana. Ghana Med J. 2015;49(1):44–9. 43. Ebonwu J, Mumbauer A, Uys M, Wainberg ML, Medina-Marino A. Determinants of late antenatal care presentation in rural and peri-urban communities in South Africa: A cross-sectional study. PLoS One. 2018;13(3):e0191903. 44. Munguambe K, Boene H, Vidler M, Bique C, Sawchuck D, Firoz T, et al. Barriers and facilitators to health care seeking behaviours in pregnancy in rural communities of southern Mozambique. Reprod Health. 2016;13(Suppl 1):31. https://doi.org/10.1186/s12978-016-0141-0. 45. Thaddeus S, Maine D. Too far to walk: Maternal mortality in context. Soc Sci Med. 1994;38(8):19. 46. Montagu D, Sudhinaraset M, Diamond-Smith N, Campbell O, Gabrysch S, Freedman L, Kruk M, Donnay F. Where women go to deliver: understanding the changing landscape of childbirth in Africa and Asia. Health Policy Plan. 2017;czx060;1146–52. 47. Naanyu et al. BMC Pregnancy and Childbirth (2020) 20:30 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/W4229448515	https://link.springer.com/content/pdf/10.1007/s12020-022-03065-0.pdf	English	null	Arterial function, biomarkers, carcinoid syndrome and carcinoid heart disease in patients with small intestinal neuroendocrine tumours	Endocrine	2,022	cc-by	7,757	* Niina Matikainen niina.matikainen@hus.ﬁ Abstract Purpose Carcinoid heart disease (CHD) is a life-threatening complication of carcinoid syndrome (CS) characterised by tricuspid regurgitation (TR). However, there is an unmet need for earlier diagnosis of CHD. We cross-sectionally assessed the prevalence and potential predictive or diagnostic markers for CS and CHD in a contemporary cohort of patients with small intestinal neuroendocrine tumours (SI-NETs). Methods Biochemical characteristics, hepatic tumour load, measures of arterial and endothelial function, atherosclerosis, and transthoracic echocardiography were analysed in a prospective cross-sectional setting. g p y y p p g Results Among the 65 patients studied, 29 (45%) had CS (CS+ ), and 3 (5%) CHD. CS+ was characterised by signiﬁcantly higher hepatic tumour load, S-5-HIAA and fP-CgA, higher frequency of diarrhoea and ﬂushing, and more frequent PRRT compared to CS−(for all, P < 0.05). Central systolic, central mean, and central end-systolic blood pressures were sig- niﬁcantly higher in CS+ than in CS−(for all, P < 0.05). Subjects with grades 2–4 TR had higher hepatic tumour burden, fP-CgA, and S-5-HIAA compared to those with grades 0–1 TR, but measures of vascular function did not differ. fP-CgA (P = 0.017) and S-5-HIAA (P = 0.019) but not proBNP increased signiﬁcantly according to the severity of TR. Conclusion Although CS is common, the prevalence of CHD was found to be lower in a contemporary cohort of SI-NET patients than previously anticipated. Measures of arterial or endothelial function or carotid atherosclerosis do not identify subjects with mild TR. Echocardiography remains the most sensitive means to diagnose CHD in CS patients with high tumour burden and elevated CgA and 5-HIAA. Keywords Arterial function ●Carcinoid heart disease ●Carcinoid syndrome ●Small intestinal neuroendocrine tumour ● Transthoracic echocardiography Keywords Arterial function ●Carcinoid heart disease ●Carcinoid syndrome ●Small intestinal neuroendocrine tumour ● Transthoracic echocardiography Arterial function, biomarkers, carcinoid syndrome and carcinoid heart disease in patients with small intestinal neuroendocrine tumours Iiro Kostiainen1 ●Noora Karppinen1 ●Piia Simonen2 ●Milla Rosengård-Bärlund1 ●Riikka Lindén3 ● Maija Tarkkanen4,5 ●Daniel Gordin6,7 ●Janne Rapola2 ●Camilla Schalin-Jäntti 1 ●Niina Matikainen 1 Iiro Kostiainen1 ●Noora Karppinen1 ●Piia Simonen2 ●Milla Rosengård-Bärlund1 ●Riikka Lindén3 ● Maija Tarkkanen4,5 ●Daniel Gordin6,7 ●Janne Rapola2 ●Camilla Schalin-Jäntti 1 ●Niina Matikainen 1 Received: 14 February 2022 / Accepted: 21 April 2022 / Published online: 10 May 2022 © The Author(s) 2022 Endocrine (2022) 77:177–187 https://doi.org/10.1007/s12020-022-03065-0 Endocrine (2022) 77:177–187 https://doi.org/10.1007/s12020-022-03065-0 ORIGINAL ARTICLE 4 Comprehensive Cancer Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 5 Finnish Medicines Agency FIMEA, Helsinki, Finland 6 Department of Nephrology, Abdominal Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 7 Minerva Institute for Medical Research, Helsinki, Finland 4 Comprehensive Cancer Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland These authors contributed equally: Camilla Schalin-Jäntti, Niina Matikainen 1 Endocrinology, Abdominal Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland Introduction The incidence of small intestinal neuroendocrine tumours (SI-NETs) originating from the enterochromafﬁn cells in the ileum (midgut) has risen in recent years [1]. In the 2012 SEER database, the incidence was 1.3/100 000 [2]. These authors contributed equally: Camilla Schalin-Jäntti, Niina Matikainen 4 Comprehensive Cancer Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 1 Endocrinology, Abdominal Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 5 Finnish Medicines Agency FIMEA, Helsinki, Finland 6 Department of Nephrology, Abdominal Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 2 Cardiology, Heart and Lung Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland 7 Minerva Institute for Medical Research, Helsinki, Finland 3 Radiology, HUS Diagnostic Center, Helsinki University Hospital and University of Helsinki, Helsinki, Finland Endocrine (2022) 77:177–187 178 Advanced disease that is usually metastatic often leads to carcinoid syndrome (CS). CS lacks a standard deﬁnition but is characterised by typical symptoms of diarrhoea, ﬂushing, bronchial constriction and elevated concentrations of ser- otonin and serotonin metabolites, the most important of which is 5-hydroxyindoleacetic acid (5-HIAA) [3, 4]. SI- NETs with substantial tumour burden and high levels of circulating serotonin and its metabolites are associated with carcinoid heart disease (CHD), a condition that has been found to cause substantial morbidity and mortality in the form of right-sided heart failure [1]. In contemporary patient series, frequency of CHD among patients with CS varies between 21 and 37% [5, 6]. irreversible ﬁbrosis related right-sided valvular deforma- tion with frequent tricuspid regurgitation and pulmonary regurgitation/stenosis [17–19]. More pronounced valvular dysfunction may lead to increased right ventricular strain and eventual right-sided heart failure. CHD is associated with diminished survival [20], with a recent study demonstrating an approximately two-year reduction in median survival when compared to patients characterized by CS only [5]. There is thus a need for earlier diagnosis via speciﬁc prognostic markers or novel diagnostic tools for CHD, which is currently detected in a late phase mostly after irreversible right-side valvular damage has occurred. The aim of our study was to evaluate the prevalence of CHD and study possible predictive biochemical and cardiovascular markers for CS and CHD in a contemporary cohort of patients with SI-NET at a tertiary centre. Our aim was also to ﬁnd possible subtle alterations in biomarkers, arterial function measurements, or echocardiography that could represent early diagnostic markers for CHD. Introduction The pathophysiology of valvular injury in CHD is yet to be fully understood [1]. Previous studies have identiﬁed high circulating serotonin concentrations as the major culprit behind the development of CHD. This relationship has been revealed in animal studies, in which the long-term administration of high-dose serotonin or the deﬁciency of the 5-HIAA transporter gene has been shown to result in the formation of carcinoid-like plaques on cardiac valves [7, 8]. Furthermore, a stimulatory action of serotonin on subendocardial cell proliferation has been demonstrated in cell culture studies [9]. In humans, the essential role of serotonin in the development of CHD is based on indirect evidence only. Human heart valves express mRNA for serotonin receptors (5-HT1B, 1D, 2 A, and 2B) [10]. In clinical studies, concentrations of urinary 5-HIAA have been found to be higher among patients with CHD than those without cardiac involvement. Furthermore, increased urinary 5-HIAA levels have been associated with the pro- gression of CHD [11]. Endothelial function All laboratory analyses were performed at HUSLAB, the laboratory of Helsinki University Hospital. FP-CgA was measured with immunoradiometric assay, urinary and serum 5-HIAA with liquid chromatography-mass spectro- metry [21], and P-proBNP with immunochemiluminometric assay. S-5-HIAA and fP-CgA were available from all patients. Urinary 5-HIAA was available from 36 patients (55%) and proBNP from 63 patients (97%) because of logistic problems. Primary tumours were graded according to the 2019 WHO classiﬁcation of tumours of the digestive system using pathology report data [22]. The reactive hyperaemia index (RHI), a measure of endo- thelial function, was calculated using measurements from a peripheral arterial tonometry (PAT) device placed on the tip of each index ﬁnger (Endo-PAT2000, Itamar Medical, Caesarea, Israel). The PAT device applies uniform pressure to the surface of the distal ﬁnger, allowing for measurement of pulse volume changes in the ﬁnger [27]. Baseline pulse amplitude was measured from each ﬁngertip for 5 minutes. Arterial ﬂow was interrupted for 5 minutes by a cuff placed on a proximal forearm (Hokanson AG101, D.E. Hokanson Inc., Bellevue, WA, USA) at whichever occlusion pressure was higher between 200 mm Hg and 60 mm Hg plus sys- tolic blood pressure. Pulse amplitude was recorded elec- tronically in both ﬁngers and analysed by a computerised, automated algorithm (Itamar Medical) that provided the average pulse amplitude for each 30-second interval after forearm cuff deﬂation for up to 5 minutes. To evaluate the vascular response in relation to baseline, with adjustment for systemic effects and skewed data, the hyperaemic response was expressed as the natural logarithm of the ratio of post-deﬂation to baseline pulse amplitude in the hyper- aemic ﬁnger divided by the same ratio in the contralateral ﬁnger, which served as a control. Patients and study design The study population included patients with a histologi- cally conﬁrmed diagnosis of SI-NET who were treated at the Helsinki University Hospital in the Departments of Endocrinology and Oncology between May 2016 and November 2017. All participants gave their written informed consent. Subjects with hereditary tumour pre- disposition syndromes were excluded. The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Helsinki Uni- versity Hospital. Aside from the cardiac valves, endothelial cells also express various serotonin receptors [12, 13]. These ser- otonin receptors are important controllers of arterial function and tone and mediate both vasoconstrictive and vasodilatory arterial effects under normal metabolic con- ditions [14–16]. Therefore, a long-term excess of circu- lating serotonin could affect vascular endothelial function. In addition to serotonin, SI-NETs can also produce other vasoactive substances, such as substance P, neurokinin A, neuropeptide K, histamine, prostaglandins, bradykinin, activin A, connective tissue growth factor, and trans- forming growth factor beta [1]. These, or other yet unrecognised humoral agents, may be alternative aetiolo- gical factors or may act synergistically with serotonin to induce the lesions found in the heart and possibly else- where in the circulatory system. However, it remains unknown whether patients with SI-NETs are characterised by impaired endothelial function. This was a prospective, cross-sectional study including arterial function measurements, transthoracic echocardio- graphy, and biochemical measurements of chromogranin A (fP-CgA), urinary and serum 5-HIAA and P-proBNP of the patients. The most recent imaging study performed as part of the clinical follow-up was used to assess hepatic tumour burden. A questionnaire assessing CS symptoms was performed at the study visit. Prevalence of tumour- related symptoms, duration of disease, treatment mod- alities, and pathology reports were retrieved from the electronic patient records. CS was deﬁned as the presence of related symptoms (i.e., diarrhoea, ﬂushing, or CHD) in conjunction with S-5-HIAA concentrations above the upper limit of normal (ULN, reference range <123 nmol/l) or S-5-HIAA con- centrations higher than three times the ULN, regardless of symptoms. Although early diagnosis and timely surgical inter- vention in CHD patients offer a survival beneﬁt, the diagnosis is based on echocardiography ﬁndings of Endocrine (2022) 77:177–187 179 Echocardiography Pulse wave velocity (PWV) pressure waveforms were recorded sequentially at the carotid, femoral, and radial arteries to measure arterial stiffness in the large (aortic) and intermediate (brachial)-sized arteries. With a simultaneous ECG recording of the R wave as a refer- ence frame, the system software calculated the PWV [24]. The differences in the carotid to femoral and car- otid to radial path length were estimated from the dis- tance from the sternal notch to the femoral and carotid palpable pulse. Transthoracic echocardiography (TTE) was performed on the high-end Philips EPIQ 7 cardiac ultrasound system (Philips Ultrasound Inc., Bothell, WA, USA) using the X5-1 Matrix Array Transducer. Recordings were saved in DICOM format for analysis with QLab (Philips Medical Systems). TTE examinations were carried out in a standard manner with the patient in the supine left lateral position. Standard parasternal long-and short-axis views, together with apical 4-, 3-, and 2-chamber views, were recorded. An additional parasternal long axis view for optimal visualisation of the tricuspid valve was recorded. Pulsed wave Doppler was used to measure the tricuspid inﬂow. Continuous wave Doppler was used for measuring the tricuspid regurgitation (TR) gradient as well as the pulmonary valve gradient. Colour Doppler was used for evaluating tricuspid and pul- monary valve regurgitation. Two cardiologists evaluated the echocardiographic imaging results afterwards without knowing the results of each other’s analysis. The augmentation index (AIx) and subendocardial viability ratio (SEVR) were derived from measurements. AIx, adjusted at 75 beats per minute, is the most com- monly used parameter for arterial stiffness in small arteries (resistance vessels). SEVR was used to approximate sub- endocardial perfusion of the heart. Arterial stiffness Arterial stiffness was measured by applanation tonometry from the radial artery with a pen-like micromanometer (SPC-301; Millar Instruments, Texas, USA). A model of the central pressure waveform was synthesised with Sphyg- moCor software (SphygmoCor; ATCOR Medical, Sydney, Australia) using a validated generalised transfer function as previously described [23]. A mean of two measurements was used in the analysis. Laboratory methods Endothelial function Carotid intima-media thickness Intima-media thickness (IMT) of the common carotid arteries (CCA) was measured with a multiarray echo tracking system (ArtLab) based on classical high- resolution echo tracking technology (WallTrack system), which generates high precision and reproducibility [25, 26]. Measurements were performed on both the right and left CCA. A mean of two measurements was used in the analysis. The diagnosis of CHD was based on echocardiographic grading of TR, leaﬂet mobility, and morphological abnormalities of the leaﬂets. From these parameters, we calculated a score that has shown good discrimination between individuals with and without CHD [28]. A score of three was used as a cut-off point for CHD, with previous work showing high sensitivity and speciﬁcity [29]. Carotid intima-media thickness Endocrine (2022) 77:177–187 180 Table 1 Patient characteristics Variable All patients (n = 65) CS− (n = 36) CS+ (n = 29) p value (comparison to CS−) CS+ excluding CHD (n = 26) p value (comparison to CS−) CHD (n = 3) Age (years) 64.2 ± 8.9 64.7 ± 8.8 63.6 ± 9.1 0.649 64 ± 9.4 0.836 59.7 ± 5.5 Sex (male/female) 33/32 (51/49%) 19/17 (53/47%) 14/15 (48/52%) 0.805 12/14 (46/54%) 0.797 2/1 (67/33%) BMI 26.6 ± 4.7 27.2 ± 4 25.9 ± 5.5 0.141 26.4 ± 5.5 0.343 21.4 ± 2.9 Duration of disease (months) 76 ± 58 83 ± 62 67 ± 53 0.300 72 ± 53 0.583 20 ± 14 Primary tumour Ki-67 (%) 3.2 ± 3.1 3.3 ± 2.9 3.1 ± 3.3 0.588 3.1 ± 3.4 0.578 3 ± 2.8 Primary tumour gradea NET G1 41 (67%) 21 (64%) 20 (71%) 19 (73%) 1 (50%) NET G2 20 (33%) 12 (36%) 8 (29%) 7 (27%) 1 (50%) NET G3 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) CgA (nmol/l) 5 (1.6–4100) 2.9 (1.6–16) 19 (4.2–4100) <0.001 16 (4.2–4100) <0.001 250 (96–400) Serum 5-HIAA (nmol/l) 138 (37–7470) 79 (37–582) 442 (132–7470) <0.001 405 (132–7170) <0.001 3220 (1940–7470) Urinary 5-HIAA (µmol/24 h) 32 (11–1621) 25 (11–127) 84 (19–1621) 0.001 76 (19–1621) 0.001 1354 (1354) proBNP (ng/l) 81 (7–5907) 55 (16–1573) 93 (7–5907) 0.146 87 (7–1175) 0.296 1283 (86–5907) Locally advanced or metastatic disease 58 (89%) 29 (81%) 29 (100%) 0.014 26 (100%) 0.035 3 (100%) Any distant metastases 51 (78%) 24 (67%) 27 (93%) 0.014 24 (92%) 0.029 3 (100%) Liver metastases 50 (77%) 24 (67%) 26 (90%) 0.040 23 (89%) 0.122 3 (100%) Hepatic tumour load <0.001 0.001 0% 23 (35.4%) 19 (52.8%) 4 (13.8%) 4 (15.4%) 0 (0%) 1–10% 23 (35.4%) 12 (33.3%) 11 (37.9%) 11 (42.3%) 0 (0%) 11–25% 9 (13.8%) 4 (11.1%) 5 (17.2%) 4 (15.4%) 1 (33.3%) 26–50% 7 (10.8%) 1 (2.8%) 6 (20.7%) 6 (23.1%) 0 (0%) >50% 3 (4.6%) 0 (0%) 3 (10.3%) 1 (3.8%) 2 (66.7%) Flushing 23 (35%) 8 (22%) 15 (52%) 0.019 13 (50%) 0.031 2 (67%) Diarrhoea 39 (60%) 17 (47%) 22 (76%) 0.024 20 (77%) 0.035 2 (67%) Bowel movements/day 3.8 (1.0–7.0) 1.0 (0.8–5.3) 5.5 (3.0–7.0) 5.3 (2.6–7.0) 7.0 (7.0–7.0) Somatostatin analogue treatment 58 (89%) 30 (83%) 28 (97%) 0.120 25 (96%) 0.222 3 (100%) Peptide receptor radionuclide therapy 24 (37%) 8 (22%) 16 (55%) 0.010 15 (58%) 0.007 1 (33%) Interferon therapy 12 (18%) 4 (11%) 8 (28%) 0.114 8 (31%) 0.101 0 (0%) Primary tumour resected 58 (89%) 33 (92%) 25 (86%) 0.691 25 (96%) 0.633 0 (0%) Liver metastases resected 12 (19%) 8 (22%) 4 (14%) 0.524 4 (15%) 0.538 0 (0%) Other metastases resected 14 (22%) 8 (22%) 6 (21%) 1.000 6 (23%) 1.000 0 (0%) Liver thermoablation 3 (5%) 3 (8%) 0 (0%) 0.247 0 (0%) 0.258 0 (0%) Values are presented as means ± SD, median (range), or n (proportion), as appropriate. Carotid intima-media thickness Emphasis in bold denotes statistical signiﬁcance (P < 0.05) aGraded according to the 2019 WHO classiﬁcation of tumours of the digestive system [22]. Primary tumours for grading were available from 61 subjects Values are presented as means ± SD, median (range), or n (proportion), as appropriate. Emphasis in bold denotes statistical signiﬁcance (P < 0.05) aGraded according to the 2019 WHO classiﬁcation of tumours of the digestive system [22]. Primary tumours for grading were available from 61 subjects Statistical analysis An abdominal radiologist (R.L.) reassessed the hepatic tumour burden of the SI-NET patients using available radiologic follow-up imaging, which was tailored to the individual treatment scheme. Assessment was based on CT (n = 47, 72.3%), MRI (n = 13, 20.0%), or 68Ga- DOTANOC-PET-CT (n = 5, 7.7%). Tumour burden was estimated using a visual semi-quantitative approach. This method has previously been applied in other studies on patients with NETs [30, 31]. Four to six scan slices with the most extensive disease burden were selected and scored visually. Hepatic tumour burden was divided into ﬁve categories: 0, <10, 10–25, 25–50, and >50%. Data are presented as means and standard deviation for continuous variables and medians and ranges for non- normally distributed variables. Proportions were calculated for categorical data. Statistical analysis was performed with IBM SPSS Statistics 25. The chi-squared test was used to calculate differences in the categorical variables between groups. Mean ranks between groups were compared using the Mann–Whitney U test or the Kruskal–Wallis test for comparison of more than two groups. Correlations were analysed with Pearson’s correlation coefﬁcient. Stepwise logistic regression was performed with binary logistic regression with a forward conditional method. Statistical analysis Endocrine (2022) 77:177–187 181 Table 2 Arterial function measurements Variable All patients (n = 65) CS−(n = 36) CS+ (n = 29) CS+ excluding CHD (n = 26) CHD (n = 3) Subendocardial viability ratio, SEVR 151.8 ± 25.6 151.9 ± 29 151.6 ± 21.1 149.4 ± 20.7 170.6 ± 16.5 Augmentation index, C-APHG HR75 AIx 23.9 ± 9.2 23.7 ± 8.8 24.1 ± 9.7 25.4 ± 8.9 13.1 ± 11.4 Central systolic pressure (mmHg) 130 ± 17 127 ± 15 133 ± 20 136 ± 18 109 ± 20 Central diastolic pressure (mmHg) 80 ± 10 78 ± 9 81 ± 11 83 ± 10 67 ± 12 Central mean pressure (mmHg) 100 ± 12 99 ± 11 103 ± 13 105 ± 11 83 ± 12 Brachial pulse wave velocity (m/s) 8.5 ± 1.2 8.3 ± 1.1 8.6 ± 1.3 8.6 ± 1.3 8.9 ± 0.9 Aortic pulse wave velocity (m/s) 9.7 ± 2.7 9.8 ± 3 9.7 ± 2.5 9.8 ± 2.6 8.6 ± 1.9 Reactive hyperaemia index, RHI 2.2 ± 0.7 2.3 ± 0.7 2.1 ± 0.7 2.2 ± 0.8 1.7 ± 0.3 Intima-media thickness, left mean (µm) 683 ± 133 693 ± 140 670 ± 124 679 ± 129 601 ± 11 Intima-media thickness, right mean (µm) 633 ± 134 633 ± 149 634 ± 114 649 ± 108 500 ± 73 Mean common carotid artery diameter (mm) 7.7 ± 1 7.7 ± 1.1 7.6 ± 0.9 7.7 ± 0.9 6.8 ± 0.3 Values are presented as mean ± SD. Emphasis in bold denotes statistical signiﬁcance (P < 0.05) when compared to CS−group Reported P values are two-sided, with a P value of <0.05 considered statistically signiﬁcant. with capecitabine). CS+ patients had received PRRT more often than those who were CS−(15/26 (58%) vs. 8/ 36 (22%), P = 0.01). Basic characteristics CS+ patients had a signiﬁcantly higher hepatic tumour bur- den (P < 0.001, distribution is shown in Table 1) when compared to those who were CS−. Prevalence of ﬂushing (50% vs. 8%, P = 0.031) and diarrhoea (77% vs. 47%, P = 0.035) and concentrations of S-5-HIAA (838 ± 1459 nmol/l vs. 101 ± 89 nmol/l, P < 0.001), dU-5-HIAA (226 ± 451 µmol vs. 32.4 ± 24.8 µmol, P = 0.001), and fP-CgA (184 ± 800 nmol/l vs. 3.8 ± 2.7 nmol/l, P < 0.001) were higher in CS+ patients as compared to those who were CS−. The inclusion of the three CHD patients in the analysis did not change the results (Table 1). Of the 65 patients with SI-NET, 55% (46/65) did not have CS (CS−), 45% had CS (CS+ ), and 5% (3/65) had CHD. The mean age, disease duration, BMI, and gender distribution did not differ between the groups. The patient characteristics are described in Table 1. Fifty- eight (89%) of the patients had at least locally spread disease, 51 (78%) had distant metastases, and 50 (77%) had liver metastases. The frequency of both distant metastases in general and liver metastases was sig- niﬁcantly higher in CS+ patients as compared to those who were CS−(Table 1). When all patients were included in the analysis, S-5- HIAA correlated signiﬁcantly with tumour load (r = 0.582, P < 0.001), fP-CgA (r = 0.677, P < 0.001), P-proBNP (r = 0.597, P < 0.001) and AIx (r = −0,264, P = 0.035). When CHD patients were excluded from the analysis, S-5- HIAA correlated signiﬁcantly only with tumour load (r = 0.454, P < 0.001), and fP-CgA (r = 0.883, P < 0.001). Treatments for SI-NET The primary tumour had been resected in 58 (89%) of the patients. Twelve (19%) patients had undergone surgical resection of hepatic metastases. Resection of other metas- tases (including lymph node metastases) was performed in 14 (22%) patients. Three patients (5%) had received liver thermoablation therapy. Measures of arterial function The results of the arterial function measurements are described in Table 2. When CHD patients were excluded, CS+ patients had signiﬁcantly higher central systolic pressure (136 ± 18 mmHg vs. 127 ± 15 mmHg, P = 0.042), central mean pressure (105 ± 11 mmHg vs. 99 ± 11 mmHg, P = 0.029), and central end-systolic pressure (121 ± 15 mmHg vs. Measures of arterial function 113 ± 13 mmHg, P = 0.030) Fifty-eight patients (89%) were on somatostatin ana- logue treatment, 24 (37%) had undergone peptide receptor radionuclide therapy (PRRT; with 177Lu- DOTATATE), with a median of 6.5 cumulative cycles (range 3–10), 12 (18%) received interferon alpha, and three (5%) mTOR inhibitor or chemotherapy (ever- olimus, temozolomide, or temozolomide in combination Endocrine (2022) 77:177–187 182 Variable All patients CS− (n = 36b) CS+ (n = 29b) CS+ excluding CHD (n = 26b) CHD (n = 3) Tricuspid valve regurgitation None 9 (14.8%) 8 (23.5%) 1 (3.7%) 1 (4.2%) 0 (0%) Trace 23 (37.7%) 13 (38.2%) 10 (37%) 10 (41.7%) 0 (0%) Mild 24 (39.3%) 12 (35.3%) 12 (44.4%) 12 (50%) 0 (0%) Moderate 3 (4.9%) 1 (2.9%) 2 (7.4%) 1 (4.2%) 1 (33.3%) Severe 2 (3.3%) 0 (0%) 2 (7.4%) 0 (0%) 2 (66.7%) Pulmonic valve regurgitation None 39 (63.9%) 24 (72.7%) 15 (53.6%) 15 (60%) 0 (0%) Trace 6 (9.8%) 1 (3%) 5 (17.9%) 5 (20%) 0 (0%) Mild 13 (21.3%) 8 (24.2%) 5 (17.9%) 5 (20%) 0 (0%) Moderate 1 (1.6%) 0 (0%) 1 (3.6%) 0 (0%) 1 (33.3%) Severe 2 (3.3%) 0 (0%) 2 (7.1%) 0 (0%) 2 (66.7%) Tricuspid valve leaﬂet mobility Increased 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Normal 58 (93.5%) 35 (100%) 23 (85.2%) 23 (95.8%) 0 (0%) Mildly reduced 2 (3.2%) 0 (0%) 2 (7.4%) 1 (4.2%) 1 (33.3%) Moderately reduced 1 (1.6%) 0 (0%) 1 (3.7%) 0 (0%) 1 (33.3%) Severely reduced 1 (1.6%) 0 (0%) 1 (3.7%) 0 (0%) 1 (33.3%) Pulmonic valve leaﬂet mobility Increased 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Normal 56 (94.9%) 32 (100%) 24 (88.9%) 24 (100%) 0 (0%) Mildly reduced 2 (3.4%) 0 (0%) 2 (7.4%) 0 (0%) 2 (66.7%) Moderately reduced 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) Severely reduced 1 (1.7%) 0 (0%) 1 (3.7%) 0 (0%) 1 (33.3%) Right ventricle area, systolic (cm2)a 12 ± 4.5 11.8 ± 4.7 12.4 ± 4.4 12.5 ± 4.4 11.2 ± 5.3 Right ventricle basal dimension, diastolic (mm)a 35.1 ± 6 35.3 ± 6.4 34.8 ± 5.5 34.1 ± 5 40 ± 7.9 Right ventricle mid-cavity dimension, diastolic (mm)a 31.6 ± 6.4 32.2 ± 6.4 30.9 ± 6.4 30.7 ± 6.2 32 ± 9.8 Right ventricle longitudinal dimension, diastolic (mm)a 62.8 ± 7.1 62.6 ± 7.7 63 ± 6.5 62.9 ± 6 64.3 ± 11.1 Right atrium area, systolic (cm2)a 16.1 ± 4.5 15.7 ± 4.8 16.5 ± 4.1 15.4 ± 2.5 25.7 ± 2.1 Tricuspid annular plane systolic excursion, TAPSE (mm) 22.1 ± 3.6 21.6 ± 3.1 22.7 ± 4.2 23.3 ± 3.1 18.3 ± 9 Values are presented as means ± SD or n (proportion), as appropriate aEvaluated from apical four-chamber view. Predictors of tricuspid regurgitation compared to those who were CS−. Of note, the carotid IMT and endothelial function (RHI) did not differ between the groups. The degree of TR was correlated with hepatic tumour load (r = 0.27, P = 0.040). There was a statistically signiﬁcant difference for S-5-HIAA (P = 0.017) and fP-CgA (P = 0.009) according to the severity of TR (shown in Fig. 1) when all patients were included in the analysis. Both CgA and 5-HIAA increased with more severe TR. How- ever, no such difference was noted for P-proBNP (P = 0.199). Measures of arterial function bEchocardiographic measurements completely or partly available for given n Table 3 Transthoracic echocardiographic measurements related to right side of the heart aEvaluated from apical four-chamber view. bEchocardiographic measurements completely or partly available for given n aEvaluated from apical four-chamber view. bEchocardiographic measurements completely or partly available for given n Endocrine (2022) 77:177–187 183 function did not differ between groups stratiﬁed by TR grade. We used a logistic stepwise regression model to study the association of variables with TR. Regurgitation was categorised into two groups: those with no regurgitation and trace regurgitation (grades 0–1) and those with mild or d t it ti ( d 2 3) CHD ti t load, Ki-67 of primary tumour, ejection duration, SEVR, AIx, central systolic pressure, central diastolic pressure, central pulse pressure, central end-systolic pressure, bra- chial and aortic PWV, RHI, removal of primary tumour, use of somatostatin analogues, PRRT, fP-CgA, and P-proBNP. Fig. 1 Boxplots of S-5-HIAA, P-proBNP, and fP-CgA by degree of tricuspid valve regurgitation with none/trace/ mild/moderate/severe corresponding to 0/1/2/3/4, respectively function did not differ between groups stratiﬁed by TR grade. load, Ki-67 of primary tumour, ejection duration, SEVR, AIx, central systolic pressure, central diastolic pressure, central pulse pressure, central end-systolic pressure, bra- chial and aortic PWV, RHI, removal of primary tumour, use of somatostatin analogues, PRRT, fP-CgA, and P-proBNP. load, Ki-67 of primary tumour, ejection duration, SEVR, AIx, central systolic pressure, central diastolic pressure, central pulse pressure, central end-systolic pressure, bra- chial and aortic PWV, RHI, removal of primary tumour, use of somatostatin analogues, PRRT, fP-CgA, and P-proBNP. We used a logistic stepwise regression model to study the association of variables with TR. Regurgitation was categorised into two groups: those with no regurgitation and trace regurgitation (grades 0–1) and those with mild or moderate regurgitation (grades 2–3). CHD patients were excluded from the analysis, and thus no study subjects displayed severe (grade 4) mitral regurgitation. The step- wise regression model (R2 = 0.255) included sex (P = 0.012) and the presence of CS (P = 0.041). The model excluded age, BMI, duration of disease, hepatic tumour Echocardiography Three patients were diagnosed with CHD. Selected TTE results related to the function and anatomy of the right side of the heart are presented in Table 3. Additional variables evaluated included thick- ening, mobility, and regurgitation of the aortic and mitral valves and stenosis of the aortic, mitral, and pulmonic valves. To further study the predictors of CHD, we divided the CS−and CS+ subjects into those without (grades 0–1) and with TR (grades 2–4). Subjects with TR of grades 2–4 were characterised by signiﬁcantly higher hepatic tumour load, fP-CgA, and S-5-HIAA, and lower BMI when compared to TR of grades 0–1 (P = 0.049, P = 0.033, P = 0.030, and P = 0.028, respectively). However, the measures of arterial When patients with CHD were excluded, none of the echocardiographic measurements demonstrated statistically signiﬁcant differences in the comparison of the CS+ and CS−groups. Endocrine (2022) 77:177–187 Discussion We here report, for the ﬁrst time, simultaneous measure- ments of echocardiography and arterial and endothelial function in patients with SI-NETs. Further, we describe the Endocrine (2022) 77:177–187 184 CHD development is not related to the vascular function parameters measured in the present study. Therefore, endothelial function tests or systemic vascular resistance measures are not able to discriminate the early asympto- matic phase of CHD, at least in a cross-sectional setting. prevalence of CS and CHD in our cohort of patients who had received contemporary treatments at a tertiary centre. Our results indicate that metastatic disease and CS are highly prevalent among such SI-NET patients, affecting almost 50% of the patients. However, CHD was found to be less common than previously reported [5, 6, 11, 28]. In the present study, those with grades 2–4 TR were characterised by high liver tumour burden and high levels of S-5-HIAA and fP-CgA. However, we could not identify speciﬁc pre- dictors for TR among the detailed vascular function tests performed, nor could such predictors be identiﬁed by echocardiography. Neuroendocrine tumour cells secrete CgA and serotonin, the precursor of 5-HIAA, which thus reﬂect the disease burden [4] while proBNP is a marker of cardiac failure, mainly secreted by atrial and ventricular myocytes in response to cardiac wall stress [33]. In many studies in patients with neuroendocrine tumours with liver metastases and/or CS, transthoracic echocardiography ﬁndings have been correlated with 24-h urinary 5-HIAA levels, but serum N-terminal proBNP has shown more mixed results [20]. Our ﬁndings conﬁrm the previous ﬁndings that CS and CHD are characterised by increased hepatic tumour burden and concentrations of fP-CgA. Chronic and excessive exposure to circulating serotonin is considered one of the most critical factors contributing to CHD [6, 34]. Since most of the studies have evaluated 24-h urinary 5-HIAA levels [11, 28, 35–39], it must be noted that only a few have reported plasma 5-HIAA concentrations in CHD, and our results validate these ﬁndings [40]. In the present study, increases of fP-CgA and S-5-HIAA were statistically sig- niﬁcantly associated with the degree of TR, whereas increases of proBNP were not. Our data suggest that follow- up echocardiography is the most sensitive means to identify CHD among patients with SI-NET characterised by sig- niﬁcant hepatic tumour burden and increased fP-CgA and 5-HIAA. Although biomarkers are convenient, our study found no single screening marker for early phase of CHD [41]. Discussion There is an increasing armamentarium of biomarkers with potential prognostic utility in SI-NET, but further studies are needed to establish their role in CHD [42]. CS is characterised by increased concentrations of cir- culating vasoactive substances, most importantly serotonin, that, besides affecting mood, satiety, and gastrointestinal function, is known to be a key regulator of vascular tone, possessing both vasoconstrictive and vasodilatory proper- ties [14]. Increased serotonin concentrations have been described in cases of arterial hypertension, carotid athero- sclerosis, and coronary artery disease [32]. However, its role in vascular pathology in SI-NET—other than right-side valvular disease—is not well known. We utilised robust, validated methodologies to assess arterial (applanation tonometry) and endothelial function (PAT) and the degree of atherosclerosis (CIMT). Surprisingly, we did not observe clear differences in these measures between those with and without CS. The results are interesting and suggest that different pathophysiological mechanisms may be taking place on top of those found in typical cardiovascular dis- ease. However, the ﬁndings need to be replicated, and further studies in other cohorts are needed to gain more insights into the mechanisms involved. Of note, subjects with CS had higher central systolic, mean, and systolic BP compared to those not having CS, indicating that CS is characterised by elevated pressure circumstances in the large arteries. CHD is typically diagnosed 1.5 to 2 years after the diagnosis of SI-NET [11, 19]. Our patient cohort had a rather long disease duration of 76 months. Despite multiple treatment modalities, almost 90% of the patients had at least locally advanced disease and 45% had CS. Regardless of the long disease duration and metastasis, the prevalence of CHD among these SI-NET patients was low at a rate of only 4.6%. A recent systematic review indicated that the inci- dence of CHD varies widely from 3% to 65% between studies, with older studies tending to report higher estimates of prevalence and incidence. This could reﬂect more effective options used in the treatment of advanced SI-NET and CS in the more recent studies [20]. In contrast, the three subjects diagnosed with CHD demonstrated decreased central arterial pressure and an approximately 50% decrease in the augmentation index as compared to those without CHD. Our data suggest that these hemodynamic changes develop after the onset of severe TR in CHD. However, as the number of patients with CHD was small, these ﬁndings need to be conﬁrmed. References 1. C. Palaniswamy, W.H. Frishman, W.S. Aronow, Carcinoid heart disease. Cardiol. Rev. 20, 167–176 (2012). https://doi.org/10. 1097/CRD.0b013e31824c866e 2. M.R. Lee, C. Harris, K.J. Baeg, A. Aronson, J.P. Wisnivesky, M.K. Kim, Incidence trends of gastroenteropancreatic neu- roendocrine tumors in the United States. Clin. Gastroenterol. Hepatol. 17, 2212–2217.e1 (2019). https://doi.org/10.1016/j. cgh.2018.12.017 In summary, we have demonstrated that although CS was highly prevalent among contemporary SI-NET patients treated with current modern therapies at a ter- tiary referral centre, CHD was rarely encountered. TR in CS patients was not reﬂected in the measurements of vascular resistance, PWV, or central arterial pressure. Rather, TR was related to hepatic tumour burden and ele- vated plasma CgA and 5-HIAA concentrations, i.e., to the severity of CS. However, we found no single ideal novel diagnostic marker for CHD, and thus a high degree of clinical awareness still stands as the key to early recogni- tion of CHD. Transthoracic echocardiography remains the ﬁrst-line imaging modality for the assessment of the severity of tricuspid and pulmonary valve diseases to establish the diagnosis of CHD. 3. B. Niederle, U.-F. Pape, F. Costa, D. Gross, F. Kelestimur, U. Knigge, K. Öberg, M. Pavel, A. Perren, C. Toumpanakis, J. O’Connor, D. O’Toole, E. Krenning, N. Reed, R. Kianmanesh, Vienna Consensus Conference participants: ENETS Consensus Guidelines Update for Neuroendocrine Neoplasms of the Jejunum and Ileum. Neuroendocrinology 103, 125–138 (2016). https://doi. org/10.1159/000443170 4. T. Ito, L. Lee, R.T. Jensen, Carcinoid-syndrome: recent advan- ces, current status and controversies. Curr. Opin Endocrinol. Diabetes Obes. 25, 22–35 (2018). https://doi.org/10.1097/MED. 0000000000000376 5. R. Fijalkowski, D. Reher, A. Rinke, T.M. Gress, J. Schrader, R.P. Baum, D. Kaemmerer, D. Hörsch, Clinical features and prognosis of patients with carcinoid syndrome and carcinoid heart disease: a retrospective multicentric study of 276 patients. Neuroendocri- nology. (2021). https://doi.org/10.1159/000518651 6. E. Baron, C. Szymanski, H. Hergault, C. Lepère, O. Dubourg, M. Hauguel-Moreau, N. Mansencal, Progression of Carcinoid Heart Disease in the Modern Management Era. J. Am. Heart Assoc. e020475 (2021). https://doi.org/10.1161/JAHA.120.020475 Acknowledgements We thank Heikki Koistinen, MD, PhD, Leo Niskanen MD, PhD, Tiinamaija Tuomi MD, PhD, and Hanna Mäen- pää, MD, PhD for clinical help in performing the study, and the study nurse Päivi Koskinen for technical assistance. 7. B.I. Gustafsson, K. Tømmerås, I. Nordrum, J.P. Loennechen, A. Brunsvik, E. Solligård, R. Fossmark, I. Bakke, U. Syversen, H. Waldum, Long-term serotonin administration induces heart valve disease in rats. Circulation 111, 1517–1522 (2005). https://doi. Discussion The CS−and CS+ groups did not differ according to cardiovascular risk factors. Assessed factors included smoking status, LDL cholesterol, previous diagnosis of diabetes mellitus, hypertension and coronary artery disease (data not shown). The carotid IMT was comparable between the groups, suggesting that high serotonin levels or CS are not associated with accelerated vascular atherosclerosis. The therapeutic options to treat CHD are limited. They consist of tumour debulking surgery, medication to reduce serotonin concentrations (i.e., somatostatin analogues), PRRT, and treatments for right-sided heart failure, includ- ing valve surgery. The diagnosis of CHD is associated with a diminished prognosis when compared to SI-NET patients In further analysis, we did not detect any differences in vascular function or carotid atherosclerosis between sub- jects with grades 0–1 or 2–4 TR, suggesting that the risk of Endocrine (2022) 77:177–187 185 without CHD [20], and the mortality of CHD patients is also high following valve surgery [43]. Early recognition of SI-NET patients with high risk for CHD is thus important to improve their prognosis. In our study, almost 90% of the patients used somatostatin analogues that have efﬁcacy in preventing CHD [44]. Furthermore, 37% of the study sub- jects had received PRRT, which may be even more effective than somatostatin analogues for the treatment of CS [45]. The use of PRRT was more than double among CS+ , which, together with a large entity of contemporary treat- ments for SI-NET, may be linked to a lower prevalence of CHD than that reported in historical cohorts. 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/. Our study has some limitations, which include the lim- ited number of patients and the cross-sectional design. References org/10.1161/01.CIR.0000159356.42064.48 Funding This work was supported by an unrestricted research grant from Ipsen Nordic and Helsinki University Hospital research grants (grant numbers TYH2019254 to CSJ, TYH2020402 to NM and M1021YLI31 to NM), Finska Läkaresällskapet (to CSJ), the Minerva Foundation Institute for Medical Research (to DG), and the Academy of Finland (grant number UAK1021MRI, to DG). Open Access funding provided by University of Helsinki including Helsinki Uni- versity Central Hospital. 8. A. Mekontso-Dessap, F. Brouri, O. Pascal, P. Lechat, N. Hanoun, L. Lanfumey, I. Seif, N. Benhaiem-Sigaux, M. Kirsch, M. Hamon, S. Adnot, S. Eddahibi, Deﬁciency of the 5-hydroxytryptamine transporter gene leads to cardiac ﬁbrosis and valvulopathy in mice. Circulation 113, 81–89 (2006). https://doi.org/10.1161/ CIRCULATIONAHA.105.554667 9. N.M. Rajamannan, N. Caplice, F. Anthikad, T.J. Sebo, T.A. Orszulak, W.D. Edwards, J. Tajik, R.S. Schwartz, Cell prolifera- tion in carcinoid valve disease: a mechanism for serotonin effects. J. Heart Valve Dis. 10, 827–831 (2001) Discussion Also, the disease duration and treatment modalities may have affected the outcome, as there are reports that SI-NET treatments may decrease 5-HIAA levels, but CHD may still progress [11]. Another problem is that the deﬁnition of CS varies between studies and uniform diagnostic criteria are awaited [46]. Compliance with ethical standards 10. A.J. Kaumann, F.O. Levy, 5-hydroxytryptamine receptors in the human cardiovascular system. Pharmacol Ther. 111, 674–706 (2006). https://doi.org/10.1016/j.pharmthera.2005.12.004 Conﬂict of interest The authors declare no competing interests. Conﬂict of interest The authors declare no competing interests. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. 11. J.E. Møller, H.M. Connolly, J. Rubin, J.B. Seward, K. Modesto, P.A. Pellikka, Factors associated with progression of carcinoid Endocrine (2022) 77:177–187 186 heart disease. N. Engl. J. Med. 348, 1005–1015 (2003). https:// doi.org/10.1056/NEJMoa021451 27. N.M. Hamburg, M.J. Keyes, M.G. Larson, R.S. Vasan, R. Schnabel, M.M. Pryde, G.F. Mitchell, J. Sheffy, J.A. Vita, E.J. Benjamin, Cross-sectional relations of digital vascular function to cardiovascular risk factors in the Framingham Heart Study. Circulation 117, 2467–2474 (2008). https://doi.org/10.1161/ CIRCULATIONAHA.107.748574 g 12. R. Pakala, J.T. Willerson, C.R. Benedict, Mitogenic effect of serotonin on vascular endothelial cells. Circulation 90, 1919–1926 (1994). https://doi.org/10.1161/01.cir.90.4.1919 13. A.G. Ramage, C.M. Villalón, 5-hydroxytryptamine and cardio- vascular regulation. Trends Pharmacol Sci. 29, 472–481 (2008). https://doi.org/10.1016/j.tips.2008.06.009 28. G. Westberg, B. Wängberg, H. Ahlman, C.H. Bergh, M. Beckman- Suurküla, K. Caidahl, Prediction of prognosis by echocardiography in patients with midgut carcinoid syndrome. Br. J. Surg. 88, 865–872 (2001). https://doi.org/10.1046/j.0007-1323.2001.01798.x 14. T. Itoh, J. Kajikuri, Characteristics of the actions by which 5-HT affects electrical and mechanical activities in rabbit jugular vein. Br. J. Pharmacol. 164, 979–991 (2011). https://doi.org/10.1111/j. 1476-5381.2011.01373.x 29. R. Dobson, D.J. Cuthbertson, J. Jones, J.W. Valle, B. Keevil, C. Chadwick, G.P. Poston, M.I. Burgess, Determination of the optimal echocardiographic scoring system to quantify carcinoid heart disease. Neuroendocrinology 99, 85–93 (2014). https://doi. org/10.1159/000360767 15. J.H. Myers, T.E. Mecca, R.C. Webb, Direct and sensitizing effects of serotonin agonists and antagonists on vascular smooth muscle. J. Cardiovasc. Pharmacol. 7(Suppl 7), S44–S48 (1985). https:// doi.org/10.1097/00005344-198500077-00014 30. M.E. Caplin, M. Pavel, J.B. Ćwikła, A.T. Phan, M. Raderer, E. Sedláčková, G. Cadiot, E.M. Wolin, J. Capdevila, L. Wall, G. Rindi, A. Langley, S. Martinez, J. Blumberg, P. Ruszniewski, CLARINET Investigators: Lanreotide in metastatic enter- opancreatic neuroendocrine tumors. N. Engl. J. Med. 371, 224–233 (2014). https://doi.org/10.1056/NEJMoa1316158 16. D.J. Sung, H.J. Noh, J.G. Kim, S.W. Park, B. Kim, H. Cho, Y.M. Bae, Serotonin contracts the rat mesenteric artery by inhibiting 4- aminopyridine-sensitive Kv channels via the 5-HT2A receptor and Src tyrosine kinase. Exp. Mol. Med. 45, e67 (2013). https://doi. org/10.1038/emm.2013.116 31. A. Rinke, H.-H. Müller, C. Schade-Brittinger, K.-J. Klose, P. Barth, M. Wied, C. Mayer, B. Aminossadati, U.-F. Pape, M. Bläker, J. Harder, C. Arnold, T. Gress, R. Conﬂict of interest The authors declare no competing interests. Arnold, PROMID Study Group: Placebo-controlled, double-blind, prospective, randomized study on the effect of octreotide LAR in the control of tumor growth in patients with metastatic neuroendocrine midgut tumors: a report from the PROMID Study Group. J. Clin. Oncol. 27, 4656–4663 (2009). https://doi.org/10.1200/JCO.2009.22.8510 17. A. Zaidi, D. Oxborough, D.X. Augustine, R. Bedair, A. Harkness, B. Rana, S. Robinson, L.P. Badano, Echocardiographic assess- ment of the tricuspid and pulmonary valves: a practical guideline from the British Society of Echocardiography. Echo. Res. Pract. 7, G95–G122 (2020). https://doi.org/10.1530/ERP-20-0033 18. T. Baron, J. Bergsten, A. Albåge, L. Lundin, J. Sörensen, K. Öberg, F.A. Flachskampf, Cardiac imaging in carcinoid heart disease. JACC Cardiovasc. Imaging 14, 2240–2253 (2021). https://doi.org/10.1016/j.jcmg.2020.12.030 32. M. Rieder, N. Gauchel, C. Bode, D. Duerschmied, Serotonin: a platelet hormone modulating cardiovascular disease. J. Thromb. Thrombolysis 52, 42–47 (2021). https://doi.org/10.1007/s11239- 020-02331-0 19. K. Oleinikov, A. Korach, D. Planer, D. Gilon, S. Grozinsky- Glasberg, Update in carcinoid heart disease - the heart of the matter. Rev. Endocr. Metab. Disord. 22, 553–561 (2021). https:// doi.org/10.1007/s11154-020-09624-y 33. C. Hall, Essential biochemistry and physiology of (NT-pro)BNP. Eur. J. Heart Fail 6, 257–260 (2004). https://doi.org/10.1016/j. ejheart.2003.12.015 20. A. Buchanan-Hughes, A. Pashley, M. Feuilly, F. Marteau, D.M. Pritchard, S. Singh, Carcinoid heart disease: prognostic value of 5-hydroxyindoleacetic acid levels and impact on survival: a sys- tematic literature review. Neuroendocrinology 111, 1–15 (2021). https://doi.org/10.1159/000506744 34. E. Goldberg, J.B. Grau, J.H. Fortier, E. Salvati, R.J. Levy, G. Ferrari, Serotonin and catecholamines in the development and progression of heart valve diseases. Cardiovasc Res. 113, 849–857 (2017). https://doi.org/10.1093/cvr/cvx092 21. A. Becker, C. Schalin-Jäntti, O. Itkonen, Comparison of Serum and Urinary 5-Hydroxyindoleacetic Acid as Biomarker for Neu- roendocrine Neoplasms. J. Endocr. Soc. 5, bvab106 (2021). https://doi.org/10.1210/jendso/bvab106 35. W.D. Denney, W.E. Kemp, L.B. Anthony, J.A. Oates, B.F. Byrd, Echocardiographic and biochemical evaluation of the develop- ment and progression of carcinoid heart disease. J. Am. Coll. Cardiol 32, 1017–1022 (1998). https://doi.org/10.1016/s0735- 1097(98)00354-4 22. I.D. Nagtegaal, R.D. Odze, D. Klimstra, V. Paradis, M. Rugge, P. Schirmacher, K.M. Washington, F. Carneiro, I.A. Cree, WHO Classiﬁcation of Tumours Editorial Board: The 2019 WHO classiﬁcation of tumours of the digestive system. Histopathology 76, 182–188 (2020). https://doi.org/10.1111/his.13975 36. T.N. Weingarten, M.D. Abel, H.M. Connolly, D.R. Schroeder, H. V. Schaff, Intraoperative management of patients with carcinoid heart disease having valvular surgery: a review of one hundred consecutive cases. Anesth Analg 105, 1192–1199 (2007). https:// doi.org/10.1213/01.ane.0000284704.57806.0b. table of contents 23. M. Karamanoglu, M.F. O’Rourke, A.P. Conﬂict of interest The authors declare no competing interests. Avolio, R.P. Kelly, An analysis of the relationship between central aortic and peripheral upper limb pressure waves in man. Eur. Heart J. 14, 160–167 (1993). https://doi.org/10.1093/eurheartj/14.2.160 37. N. Mansencal, E. Mitry, J.-B. Bachet, P. Rougier, O. Dubourg, Echocardiographic follow-up of treated patients with carcinoid syndrome. Am. J. Cardiol 105, 1588–1591 (2010). https://doi.org/ 10.1016/j.amjcard.2010.01.017 24. I.B. Wilkinson, S.A. Fuchs, I.M. Jansen, J.C. Spratt, G.D. Murray, J.R. Cockcroft, D.J. Webb, Reproducibility of pulse wave velocity and augmentation index measured by pulse wave analysis. J. Hypertens 16, 2079–2084 (1998). https://doi.org/10.1097/ 00004872-199816121-00033 38. S. Bhattacharyya, C. Toumpanakis, D. Chilkunda, M.E. Caplin, J. Davar, Risk factors for the development and progression of car- cinoid heart disease. Am. J. Cardiol 107, 1221–1226 (2011). https://doi.org/10.1016/j.amjcard.2010.12.025 25. A. Paini, P. Boutouyrie, D. Calvet, M. Zidi, E. Agabiti-Rosei, S. Laurent, Multiaxial mechanical characteristics of carotid plaque: analysis by multiarray echotracking system. Stroke 38, 117–123 (2007). https://doi.org/10.1161/01.STR.0000251796.38954.b2 39. V. Rodríguez Laval, M. Pavel, I.G. Steffen, A.D. Baur, L.M. Dilz, C. Fischer, K. Detjen, V. Prasad, A. Pascher, D. Geisel, T. Denecke, Mesenteric ﬁbrosis in midgut neuroendocrine tumors: functionality and radiological features. Neuroendocrinology 106, 139–147 (2018). https://doi.org/10.1159/000474941 26. J.M. Meinders, P.J. Brands, J.M. Willigers, L. Kornet, A.P. Hoeks, Assessment of the spatial homogeneity of artery dimen- sion parameters with high frame rate 2-D B-mode. Ultrasound Med. Biol. 27, 785–794 (2001). https://doi.org/10.1016/s0301- 5629(01)00351-9 40. R. Dobson, M.I. Burgess, J.W. Valle, D.M. Pritchard, J. Vora, C. Wong, C. Chadwick, B. Keevi, J. Adaway, U. Hofmann, G.J. Poston, D.J. Cuthbertson, Serial surveillance of carcinoid heart Endocrine (2022) 77:177–187 187 disease: factors associated with echocardiographic progression and mortality. Br J. Cancer 111, 1703–1709 (2014). https://doi. org/10.1038/bjc.2014.468 disease. Eur. Heart J. Cardiovasc. Imaging 23, 524–531 (2022). https://doi.org/10.1093/ehjci/jeab049 p g j j 44. A.R. Hayes, J. Davar, M.E. Caplin, Carcinoid heart disease: a review. Endocrinol Metab. Clin. North Am. 47, 671–682 (2018). https://doi.org/10.1016/j.ecl.2018.04.012 41. J. Davar, H.M. Connolly, M.E. Caplin, M. Pavel, J. Zacks, S. Bhattacharyya, D.J. Cuthbertson, R. Dobson, S. Grozinsky-Glas- berg, R.P. Steeds, G. Dreyfus, P.A. Pellikka, C. Toumpanakis, Diagnosing and managing carcinoid heart disease in patients with neuroendocrine tumors: an expert statement. J. Am. Coll Cardiol 69, 1288–1304 (2017). https://doi.org/10.1016/j.jacc.2016.12.030 45. W.T. Zandee, T. Brabander, A. Blažević, N.S. Minczeles, R.A. Feelders, W.W. de Herder, J. Hoﬂand, Peptide receptor radio- nuclide therapy With 177Lu-DOTATATE for symptomatic control of refractory carcinoid syndrome. J. Clin. Endocrinol. Metab. 106, e3665–e3672 (2021). https://doi.org/10.1210/ clinem/dgab289 42. M. Kalligeros, L. Conﬂict of interest The authors declare no competing interests. Diamantopoulos, C. Toumpanakis, Biomarkers in small intestine NETs and carcinoid heart disease: a compre- hensive review. Biology 10, 950 (2021). https://doi.org/10.3390/ biology10100950 46. J. Hoﬂand, A.D. Herrera-Martínez, W.T. Zandee, W.W. de Her- der, Management of carcinoid syndrome: a systematic review and meta-analysis. Endocr. Relat. Cancer 26, R145–R156 (2019). https://doi.org/10.1530/ERC-18-0495 43. J. Bergsten, F.A. Flachskampf, L. Lundin, P. Öhagen, A. Albåge, A 33-year follow-up after valvular surgery for carcinoid heart
https://openalex.org/W2561109103	https://www.meatjournal.ru/jour/article/download/32/39	Latin	null	PRINCIPLES OF DETERMINATION OF VALUE IN USE FOR MEAT AND MEAT PRODUCTS BASED ON QUALITY INDICATORS — THE COEFFICIENTS OF CONSUMER PROPERTIES	Teoriâ i praktika pererabotki mâsa	2,016	cc-by	17,149	Аннотация Price setting is as complex process, which requires taking into con- sideration many factors. In different periods, the systems of price setting existed in the meat sector of the agro-industrial complex, which took into account only the external factors: competition, val- ue of goods and production costs. The system of price formation that was in existence in Russia up to now was based only on the cost- based principle. Transition to formation of the free market prices practically has not led to changes in the methodological approaches in price setting and has not influenced their structure. The current price formation system in the meat sector of the agro-industrial complex does not correspond to the contemporary requirements of the economic science. Thus, it is an obstacle on the way of intro- duction of the objective economic laws in conditions of the market relations. It is possible to achieve production efficiency with such use of the existing resources when the differentiated production costs are proportional to the utility of these resources. The utility of products is determined by a complex of properties that reflect their value in use. The main qualitative parameters are consumer prop- erties of products. The main internal factor influencing the price parameters is the qualitative composition of raw material. In order to create parity in price setting for different groups of products, the unified method of price equivalence with regard to the qualitative parameters of the raw material constituent was developed. Qual- ity characteristics of meat products are composed of the product structure, morphology and chemical composition, and, finally, co- efficients of consumer properties calculated with consideration for all above-mentioned factors. Определение цены — это сложный процесс, требующий учета большого количества факторов. В мясной отрасли АПК в различные периоды существовали системы постро- ения цен, которые учитывали только внешние факторы: конкуренцию; ценностную значимость товара; издержки производства. Действовавшая до последнего времени си- стема ценообразования в России основывалась только на затратном принципе. Переход на формирование свободных рыночных цен практически не привел к изменению методо- логических подходов при определении цен и не повлиял на их структуру. Система ценообразования, сложившаяся в на- стоящее время в мясной отрасли АПК, не отвечает совре- менным требованиям экономической науки. В связи с этим она является тормозом на пути внедрения объективных экономических законов в условиях рыночных отношений. Достижение эффективности производства возможно при таком использовании имеющихся ресурсов, при котором дифференцированные производственные затраты пропор- циональны полезности этих ресурсов. 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА УДК/UDC: 637.5:338.518 DOI 10.21323/2414-438X-2016-1-3-81-95 PRINCIPLES OF DETERMINATION OF VALUE IN USE FOR MEAT AND MEAT PRODUCTS BASED ON QUALITY INDICATORS — THE COEFFICIENTS OF CONSUMER PROPERTIES ПРИНЦИПЫ ОПРЕДЕЛЕНИЯ ПОТРЕБИТЕЛЬНОЙ СТОИМОСТИ МЯСА И МЯСНЫХ ПРОДУКТОВ НА ОСНОВЕ ПОКАЗАТЕЛЕЙ КАЧЕСТВА — КОЭФФИЦИЕНТОВ ПОТРЕБИТЕЛЬСКИХ СВОЙСТВ Ключевые слова: мясо, мясные продукты, потребительная стоимость, показатели качества, полезность, потреби- тельские свойства. Ключевые слова: мясо, мясные продукты, потребительная стоимость, показатели качества, полезность, потреби- тельские свойства. Аннотация Полезность продук- тов определяется комплексом свойств, отражающих их потребительную стоимость. Основными качественными параметрами являются потребительские свойства про- дуктов. Основным внутренним фактором, влияющим на ценовые параметры, является качественный состав сырья. С целью создания паритета при определении цен на различ- ные группы продуктов разработана единая методика цено- вой эквивалентности с учетом качественных параметров сырьевой составляющей. Качественные характеристики мясной продукции складываются из структуры продуктов, их морфологии, химического состава и, в конечном счете, рассчитываемых с учетом всех перечисленных факторов коэффициентов потребительских свойств. ПРИНЦИПЫ ОПРЕДЕЛЕНИЯ ПОТРЕБИТЕЛЬНОЙ СТОИМОСТИ МЯСА И МЯСНЫХ ПРОДУКТОВ НА ОСНОВЕ ПОКАЗАТЕЛЕЙ КАЧЕСТВА — КОЭФФИЦИЕНТОВ ПОТРЕБИТЕЛЬСКИХ СВОЙСТВ Neburchilova N.F., Petrunina I.V. The V.M. Gorbatov All-Russian Meat Research Institute, Moscow, Russia Neburchilova N.F., Petrunina I.V. The V.M. Gorbatov All-Russian Meat Research Institute, Moscow, Russia Keywords: meat, meat products, value in use, quality indicators, utility, consumer properties. Актуальность Эту теорию он изложил в вышедшей в 1854 году книге «Развитие законов общественного обмена и вытека- ющих отсюда правил человеческой деятельности»1. Теория Г. Госсена базируется на том, что главным мотивом, определяющим поведение человека являет- ся стремление к получению максимума полезности. In the middle of the 1850s, the German economist Hermann Heinrich Gossen began to work on his own economic theory. He described this theory in his book The Development of the Laws of Human Intercourse and The Consequent Rules of Human Action1 published in 1854. The Gossen’s theory is based on the assumption that the main stimulus determining human behavior is striving to achieve the maximum utility. In this connection, the main task of the economic science is formation of the rules of maximization (increasing of the total) utility. However, his theory was not recognized by his contemporaries. The Gossen’s book was found by prof. Adamson and republished in 1889 and then in 1927, and since then became widely known. В связи с этим главной задачей экономической науки является формирование правил максимизации (уве- личение общей) полезности. Однако его теория не получила признания у современников. Книга Госсе- на была найдена профессором Адамом и переиздана в 1889 году, а затем в 1927 году и с тех пор получила широкую известность. Л. Вальрас, У. Джевонс и другие экономисты на- чали широко пропагандировать теорию предельной полезности, которую рассматривал Госсен. Исполь- зуя понятия полезности и, в частности предельной полезности, можно определять потребительские предпочтения. Полезность — это степень удовлет- ворения, полученного человеком от потребления ка- кого-то блага, таким образом, оценка потребителем степени полезности различных товаров (например, товар X лучше, чем товар У) и является потребитель- ским предпочтением. L. Walras, W. Jevons and other economists began to widely propagandize the theory of marginal utility, which Gossen had examined. Using the concept of utility, in particular, marginal utility, it is possible to determine consumer preferences. Utility is a degree of satisfaction received by a person from consumption of a specific good or service and, therefore, an evaluation of a degree of utility of different goods (for example, a good X is better than a good Y) by a consumer and a consumer preference. At the end of the 19th century, the representatives of the Austrian school of marginalism (C. Menger, E. Böhm- Bawerk, F. Wieser) suggested that a specific quantitative measure of utility existed for each individual. 1 Книга Госсена 1854 года «Die Entwickelung der Gesetze des menschlichen Verkehrs, und der daraus fließenden Regeln für menschliches Handeln». В 1983 году был опубликован перевод на английский язык. На русский язык название можно перевести как «Разработка законов общественного обмена и вытекающих из них правил человеческой деятельности». Эту работу можно найти в переводе на русский язык в книге «Мировая экономическая мысль. Сквозь призму веков». В 5 т. — М., 2005. — Т. 2. 1 The book «Die Entwickelung der Gesetze des menschlichen Verkehrs, und der daraus fließenden Regeln für menschliches Handeln» by Gossen (1854). In 1983, the translation to English was published. This work can be found in translation to Russian in the book World economic thought. Through the prism of centuries. In 5 volumes — М., 2005. — Vol. 2.l Актуальность Introduction of the market relations in the country was accompanied by the liberalization of prices, which differed from the early performed reforms in the field of price formation. Practically, the methodology of price setting has changed. Free price formation has become one of the most complex problems in the sector’s economics. Введение рыночных отношений в стране сопрово- ждалось либерализацией цен, которая отличалась от ранее проводимых реформ в области ценообразова- ния. Практически изменилась методология определе- ния цен. Свободное ценообразование стало одной из сложных проблем в отраслевой экономике. В насто- 81 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 At present, different types of prices are used in the RF, the main of which are purchasing, wholesale and retail prices. Purchasing prices are set for agricultural products, wholesale or transfer prices are set for products manufactured by industrial enterprises and retail prices are prices of products sold to population. ящее время в РФ используются различные виды цен, основными их которых являются закупочные, опто- вые и розничные. Закупочные цены устанавливаются на сельско- хозяйственную продукцию, оптовые или отпускные цены — на продукцию, которую выпускают промыш- ленные предприятия, и розничные цены по которым товары реализуют населению. The system of price formation, which was established in the meat sector of the agro-industrial complex, does not correspond to the contemporary requirements of the economic science. Thus, it is an obstacle on the way of introduction of the objective economic laws in the conditions of the market relations. В мясной отрасли АПК сложившаяся система це- нообразования не отвечает современным требовани- ям экономической науки. И в связи с этим является тормозом на пути внедрения объективных экономи- ческих законов в условиях рыночных отношений. В середине пятидесятых годов ХIХ века Госсен Герман Генрих — немецкий экономист занялся раз- работкой собственной экономической теории. Эту теорию он изложил в вышедшей в 1854 году книге «Развитие законов общественного обмена и вытека- ющих отсюда правил человеческой деятельности»1. Теория Г. Госсена базируется на том, что главным мотивом, определяющим поведение человека являет- ся стремление к получению максимума полезности. В связи с этим главной задачей экономической науки является формирование правил максимизации (уве- личение общей) полезности. Однако его теория не получила признания у современников. Книга Госсе- на была найдена профессором Адамом и переиздана в 1889 году, а затем в 1927 году и с тех пор получила широкую известность. В середине пятидесятых годов ХIХ века Госсен Герман Генрих — немецкий экономист занялся раз- работкой собственной экономической теории. Актуальность Со- гласно этой теории величина ценности каждого това- ра или услуги определяется величиной их пользы для конкретного потребителя. При этом имеется в виду не величина полезности как таковая, а предельная полез- ность товара или услуги. With an increase of the total amount of goods or servic- es available to a consumer, the marginal utility decreases and the total utility increases. A choice of a consumer is always limited by an income as well as by prices. The equal positive effect for a buyer can be expressed by a ratio: С увеличением общего количества товаров или услуг, которыми располагает потребитель, предельная полезность уменьшается, а общая — увеличивается. Выбор потребителя всегда ограничен доходом, а так- же ценами. Равный полезный эффект для покупателя можно выразить соотношением: MUа = MUв Pа Pв , Па = Пв Ца Цв , where: MUа and MUb — marginal utility of goods А and В, Pа and Pв — prices on goods А and В. где: Па и Пв — предельная полезность товаров А и В, Ца и Цв — цены товаров А и В. где: Па и Пв — предельная полезность товаров А и В, Ца и Цв — цены товаров А и В. Покупатель стремится получить максимальную полезность, покупая такой товар или услугу, когда от- ношение предельной полезности к цене максимально близки. A consumer strives to achieve the maximum utility buying a product or services when the ratio of marginal utility to a price is maximally close. Полезность продуктов — это степень удовлетво- рения потребностей человека в конкретном товаре. Полезность является субъективным понятием. В Сло- варе экономических терминов определено понятие полезности. Пищевая ценность продуктов — это ком- плекс полезных свойств, определяющих их биологиче- скую и энергетическую ценность и обеспечивающих физиологические потребности человека в энергии и в основных питательных веществах. Product utility is a degree of satisfaction of a human need for a specific product. Utility is a subjective concept. The dictionary of the economic terms defines the concept of utility. Food value of products is a complex of useful properties that determine their biological and energy value and satisfy the physiological requirements of an individual in energy and the main nutrients. Therefore, it is necessary to distinguish what determines a product price — supply (cost) or demand (utility). Актуальность The Austrian scientists introduced in their analysis the term marginalism (from French marginal — ultimate), which is a direction of the economic theory widely applied in the analysis of the regularities of the economic processes based on the use of the ultimate values. Accordingly, all В конце XIX в. представители австрийской школы маржинализма (К. Менгер, Ф. Визер, Е. Бём-Баверк), полагали, что для каждого человека существует опре- деленный количественный измеритель полезности. Австрийские ученые ввели в свой анализ термин маржинализм» (от французского marginal — предель- ный) — направление экономической теории, кото- рое широко применяется в анализе закономерности экономических процессов на основе использования предельных величин. Соответственно все основные 82 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА main categories in the marginalistic theory are based on the use of the quantitative analysis, a leading role in which is given to the use of the concept of borders or margins. It is such categories as marginal productivity, marginal costs, marginal utility and so on. категории в маржиналистской теории основаны на применении количественного анализа, при котором ведущая роль отводится использованию понятия пре- делов. Это такие категории, как предельная произво- дительность, предельные издержки, предельная по- лезность и пр. В экономической теории различают две формы полезности: общую и предельную. Общая полезность (TU — total utility) — это сумма предельных полезно- стей (или полезность всех имеющихся в наличии това- ров и услуг). Предельная полезность товара или услу- ги (MU — marginal utility) — это полезность единицы (наименьшая польза) из имеющегося запаса данного вида товара или услуги [1]. The economic theory distinguishes two forms of utility: total and marginal. Total utility (TU) is a sum of marginal utilities (or utility of all available goods or services). Marginal utility (MU) of a good or service is the utility of a unit (the least utility) from an available supply of this type of goods or services [1]. The theoretical development of the issue of utility was carried out by the economic scientists W. Jevons, C. Menger, F. von Wieser, E. von Bawerk, L. Walras. According to this theory, the value of each good or service is determined by the degree of their utility to an individual consumer. With that, it is the marginal utility of goods or services that is meant and not a degree of utility per se. Теоретическую разработку проблемы полезности осуществили ученые-экономисты У. Джевонс, К. Мен- гер, Ф. фон Визер, Е. фон Бём-Баверк, Л. Вальрас. Методы Методы Methods The methods of price substantiation, which are based on the main price forming factors, present a foundation for the methodology of price formation (Fig. 1). Основу методологии ценообразования составляют методы обоснования цен, которые базируются на ос- новных ценообразующих факторах (рис. 1). Figure 1. Methods of price formation \| Рис. 1. Методы ценообразования Methods for price formation \| Методы ценообразования With orientation on production costs \| С ориентацией на издержки производства With orientation on quality and consumer properties of a product (parametric models) \| С ориентацией на качество и потребительские свойства продукции (параметрические методы) With orientation on demand, level of competition \| С ориентациоей на спрос, уровень конкуренции Cost-based \| Затратный Aggregate \| Агрегатный Structural analogy \| Структурной аналогии Following the leader \| Следования за лидером Perceived value of a good \| Ощущаемой ценности товара Specifc indicators \| Удельных показателей Scoring \| Балловый Correlation and regression \| Корреляционно-регрессионный Methods for price formation \| Методы ценообразования With orientation on production costs \| С ориентацией на издержки производства With orientation on demand, level of competition \| С ориентациоей на спрос, уровень конкуренции With orientation on quality and consumer properties of a product (parametric models) \| С ориентацией на качество и потребительские свойства продукции (параметрические методы) Specifc indicators \| Удельных показателей Perceived value of a good \| Ощущаемой ценности товара Figure 1. Methods of price formation \| Рис. 1. Методы ценообразования Figure 1. Methods of price formation \| Рис. 1. Методы ценообразования Одним из наиболее распространенных методов ценообразования, которые ориентированы на издер- жки производства, в отечественной производствен- ной практике, является затратный метод. Суть его за- ключается в том, что к рассчитанной себестоимости единицы продукции добавляется заданный размер прибыли, а также косвенные налоги и неналоговые платежи, непосредственно увеличивающие цену: In the national production practice, one of the most common methods of price formation that are oriented on production costs is the cost-based method. The essence of the method is that the specified size of profit, as well as indirect taxes and non-tax payments, are added to the calculated unit cost of a product directly increasing the price: Ц = С + П + Н, P = C + P + N, где: С — себестоимость единицы товара; П — прибыль в расчете на единицу товара; Н — косвенные налоги и неналоговые платежи в цене товара. Актуальность With that, it is necessary to understand what is primary: utility of a product as a «function» of cost or, on the contrary, cost is a «function» of utility of a product. Up to now, the pos- sibility to combine two approaches to price formation by uniting in a price an «objectivity» (cost) and «subjectivity» (utility) has not been examined in the framework of the modern economic theory. Таким образом, необходимо различать, что опре- деляет цену товара — предложение (стоимость) или спрос (полезность). При этом важно понимать, что яв- ляется первичным полезность товара как «функция» стоимости или, напротив, стоимость есть «функция» полезности товара. До настоящего времени в рамках современной экономической теории возможность объединить оба подхода к ценообразованию, совме- стив в цене «объективность» (стоимость) и «субъек- тивность» (полезность) товара не рассматривалась. 83 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА At present, meat and meat products are realized at free- of-control prices, which include a product cost, profit and value-added tax; with that, the real size of profit depends on the level of the agreed price with consideration for market conditions, i.e., the demand of a specific good. Мясо и мясные продукты в настоящее время ре- ализуются по свободным ценам, которые включают в себя себестоимость продукции, прибыль, налог на добавленную стоимость, при этом фактический раз- мер прибыли зависит от уровня согласованной цены с учетом конъюнктуры рынка, то есть спроса на кон- кретный товар. i A foundation in the practice of price setting is a product cost as a lower limit of a price and demand for a product as an upper limit of a price. В практике построения цен базой служит себестои- мость продукции — как нижний предел цены и спрос на продукцию — как верхний предел цены. In this connection, the prices now are set by the cost- based method without consideration for any characteris- tics of a product per se. В связи с этим цены в настоящее время строятся по затратному методу, без учета каких-либо характе- ристик самого продукта. The quality indicators by the categories of meat fatness and grades of meat products do not reflect the true belong- ing of a product to one or another quality group.i Показатели качества по категориям упитанности мяса, сортов мясных продуктов не отражают истин- ной принадлежности продукта к той или иной качест- венной группе. Nowadays, classification of meat by kinds, categories and grades is based on quality determined with regard to the energy value, which is characterized by the quantity of energy released by a human organism from food products for provision of its physiological functions. In this connec- tion, when developing the normative-technical documen- tation on meat of all kinds of slaughter animals and meat products, the main criteria was the indicator of the energy value, which was detected mainly by the presence of fat in a product. Классификация мяса по видам, категориям, сортам в настоящее время устанавливается на основе качества, определяемого с учетом энергетической ценности, ко- торая характеризуется количеством энергии высвобо- ждаемой в организме человека из пищевых продуктов для обеспечения его физиологических функций. В свя- зи с этим при разработке нормативно-технической документации на мясо всех видов убойных животных и мясные продукты основным критерием качества яв- лялся показатель энергетической ценности, которая определялась в основном наличием жира в продукте. Methods where: С — unit cost of a product; P — profit per unit; N — in- direct taxes and non-tax payments in the price of a product. The cost-based method is popular not only in the national but also in the foreign practice of price formation. h The cost-based method is popular not only in the national but also in the foreign practice of price formation. This is associated with several reasons. First, manufacturers are better informed about their costs than about consumer demand. Thus, the cost-based method is considered quite simple. In addition, according to the expert opinion, this method is the fairest both with respect to a seller and a buyer. Затратный метод является популярным не только в отечественной, но и в зарубежной практике ценоо- бразования. Это связано с рядом причин. Во-первых, производители всегда лучше осведомлены о своих затратах, чем о потребительском спросе. Поэтому за- тратный метод считается достаточно простым. А так- же, по мнению специалистов, этот метод является на- иболее справедливым как по отношению к продавцу, так и к покупателю. Однако в настоящее время, в условиях рынка на- иболее важную роль при формировании цены играет полезность товара. Дело в том, что покупателя инте- ресует не товар как таковой, а то, в какой степени он будет удовлетворять его потребности, то есть полез- ность товара [2]. Таким образом, в настоящее время необходимо учитывать все методы построения цен. Nowadays, however, in the market conditions, the most important role in price formation plays utility of a good. This is because consumers are interested not in a good per se, but in a degree of satisfaction of their needs, that is in utility of a good [2]. Today, therefore, it is necessary to consider all methods for price formation. 84 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 Up to date, no country in the world has taken into ac- count the consumer properties of a product in assessment of its possible cost. However, it is necessary to note that at the beginning of 2000s, the Australian scientists in their scientific research to some extent touched upon the neces- sity to consider the qualitative characteristics of a product in price setting and pointed at the necessity to determine a price level with respect to the consumer properties of meat products. До настоящего времени ни в одной стране мира при оценке возможной стоимости продукта не учи- тывались его потребительские свойства. Однако сле- дует отметить, что в начале двухтысячных годов ав- стралийские ученые в своих научных исследованиях в определенной степени затрагивают необходимость учета при построении цен качественных характери- стик продуктов и указывают на необходимость опре- деления уровня цен с учетом потребительских свойств мясных продуктов. N.J. Simmons, C.C. Daly, C.R. Mudford, I. Richards et al. in their work published in Meat Science [3] pointed to the necessity to take into account qualitative characteristics of products (i.e., their consumer properties) when setting prices on meat. The authors suggested that the component of food raw material being a constituent of the true value in use has to be derived by the precise assessment of an in- dividual muscle and these data can be converged to detect the value of a slaughter animal carcass in total. Как пишут в своих исследованиях Н. Дж. Симмонс, С.С. Дейли, С.Р. Мудфорд, И. Ричардс и другие, резуль- таты которых опубликованы в статье журнала «Меаt Science», в которой указывают на необходимость при построении цен на мясо и мясные продукты принимать во внимание качественные характеристики продуктов, то есть их потребительские свойства [3]. Авторы ут- верждают, что «компонент пищевого сырья, входящий в истинную потребительную стоимость, должен быть получен путем точной оценки отдельной мышцы, и эти данные можно свести воедино, чтобы определить стои- мость туши убойного животного в целом». However, their research is extended only to include meat from different species of slaughter animals. The Aus- tralian scientists assessed meat quality of individual cuts, which were obtained in cutting, and natural meat prod- ucts, i.e., semi-prepared products not subjected to thermal treatment. Однако их исследования распространяются только на мясо различных видов убойных животных. Оценку качества мяса австралийские ученые провели по от- дельным отрубам, получаемым при разделке и на на- туральные мясные продукты, то есть полуфабрикаты, не подвергнутые термической обработке. THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 According to the results, the Meat Standards Australia (MSA), which is an eating quality grading system was de- veloped [4]. In 2014–2015, the Australian Beef Eating Qual- ity Audit was carried out for the first time. Assessment of beef was performed according to MSA. Over 3.2 million cattle were assessed on compliance to MSA (meat color, pH, carcass fat). Similar audits are planned to be per- formed at least up to 2020 [5].h По результатам были разработаны австралийские стандарты на мясо (Meat Standards Australia — MSA) — система оценки потребительских характеристик [4]. В 2014–2015 гг. впервые в истории страны проводился Аудит Австралийской говядины по потребительским характеристикам (Australian Beef Eating Quality Audit). Оценка говядины осуществлялась по MSA. Было оце- нено 3.2 миллиона голов крупного рогатого скота на соответствие MSA (цвет мяса, рН, жир на туше). По- добные аудиты планируется проводить как минимум до 2020 года [5]. The research studies that have been carried out by the specialists of V.M. Gorbatov VNIIMP for many years have allowed establishing the qualitative characteristics of meat and meat products on the basis of their structure, tissue and chemical composition, which eventually gave an op- portunity to detect a complex of consumer characteristics of the specific meat products. Исследования, которые проводились на про- тяжении многих лет специалистами ВНИИМП им. В.М. Горбатова, позволили установить, качествен- ные характеристики мяса и мясной продукции на ос- нове их структуры, тканевого и химического состава, что, в конечном счете, дало возможность определить комплекс потребительских свойств конкретных мяс- ных изделий. Accounting of consumer properties, products in the form of the coefficients of consumer properties makes it possible to determine promptly a basis for a price on a meat constituent for all types of products manufactured in the meat sector enterprises. Therefore, a discrepancy be- tween the prices on livestock raw material and products of its processing can be eliminated according to the principle of price equivalence with regard to the quality parameters. Учет потребительских свойств, продукции в форме коэффициентов потребительских свойств позволяет оперативно определять основу цены мясной состав- ляющей всех видов продукции выпускаемой на пред- приятиях мясной промышленности. Таким образом, может быть ликвидировано несоответствие в соот- ношении цен на животноводческое сырье и продукты его переработки по принципу ценовой эквивалентно- сти с учетом качественных параметров. 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА To detect the consumer value of products, it is necessary to pay attention to several indicators, which characterize the biological value and optimal physiological utility of a product, its correspondence to the normal requirements of a human body with consideration for the physico-chemical indicators. Food value of meat and its organoleptic indica- tors are closely linked with the properties and quantitative ratio of the tissues in meat and their chemical composi- tion, i.e., content of proteins, fats and carbohydrates. у Для определения потребительной ценности про- дуктов необходимо принимать во внимание целый ряд показателей, которые характеризую биологи- ческую ценность и оптимальную физиологическую полезность продукта, его соответствие нормальным потребностям организма человека с учетом физи- ко-химических показателей. Пищевая ценность мяса и его органолептические показатели тесно связаны со свойствами и количественным соотношением тканей в мясе и их химическим составом, то есть содержани- ем белков, жиров и углеводов. The chemical methods for detection of meat and meat product quality, which have been widely used since the middle of the 19th century, have laid a foundation for re- vealing the content and quantitative measurement of their elementary constituents. However, characterization of quality only by the chemical composition, i.e., taking into account only the content of fat, total protein, moisture and ash, is not sufficient. Therefore, the indicators such as the content of total protein by composition (complete (muscle) and incomplete (connective tissue)) and fat by the types of fatty acids (saturated and unsaturated) have been studied in more details. Химические методы определения качества мясно- го сырья и мясных продуктов, широко используемые с середины девятнадцатого века, заложили основу для познания состава и количественного измере- ния их элементарных компонентов. Однако характе- ристика качества лишь по химическому составу, то есть с учетом только содержания — жира, общего белка, влаги и золы уже не достаточна. Поэтому бо- лее детально исследуются такие показатели, как со- держание общего белка по составу — полноценный (мышечный) и неполноценный (соединительноткан- ный), а жира — по видам жирных кислот: насыщен- ных и ненасыщенных. The main principle of the optimum for a set of consum- er properties has to be equivalence of quality assessment of raw material and finished products and evaluation of their consumer characteristics. Основным принципом оптимальности набора по- требительских свойств должна стать эквивалентность оценок качества сырья и готовой продукции и их по- требительских характеристик. 85 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА Due to the fact that the coefficients of the consumer properties according to product types allow significant simplification of raw material calculation, operativity of managerial decision making will be ensured by acceler- ated calculation of variable material costs, the conditions for express calculations of a marginal profit value will be created in the framework of introduction of a managerial accounting system in enterprises of the meat sector of the agro-industrial complex. В условиях постоянно меняющихся цен на рын- ке стабильный показатель качества сырья и готовой продукции обеспечит возможность оперативно ре- шать проблемы по разработке ассортиментной зада- чи, плана продаж, производственной программы с це- лью получения высоких экономических показателей, обеспечивающих безубыточность производства и ста- бильность финансового положения предприятия. В связи с тем, что коэффициенты потребительских свойств по видам продукции позволяют значительно упростить сырьевые расчеты, это обеспечит оператив- ность принятия управленческих решений, путем уско- ренных расчетов переменных материальных затрат, создадут условия для экспресс расчетов величины маржинального дохода, в рамках внедрения системы управленческого учета на предприятиях мясной отра- сли АПК. In terms of making effective managerial decisions upon introduction of accounting by the centers of responsibility, the coefficients will provide an opportunity for solving as- sortment tasks with consideration for the state of the raw material potential, marketing preferences and, eventually, will create conditions for strengthening competiveness in manufacturing specific types of production. Для принятия эффективных управленческих реше- ний при внедрении учета по центрам ответственности коэффициенты обеспечат возможность решения ас- сортиментных задач с учетом состояния сырьевого по- тенциала, маркетинговых предпочтений и в конечном счете создадут условия повышения конкурентоспособ- ности производства конкретных видов продукции. Recently, the specialists of the Institute have developed the methodological approach to determination of the cri- teria for assessing the consumer properties of meat raw material and raw material constituent of meat products for detection of finished product costs. В настоящее время специалистами института раз- работан методологический подход к определению критериев оценки потребительских свойств мясного сырья и сырьевой составляющей мясной продукции для определения стоимости готовых продуктов. The basis of the methodological approach is the prin- ciple of price equivalence of meat and meat raw material with consideration for qualitative parameters, where the main constituent should be accounting of consumer prop- erties when forming the product cost [6]. В основу методологического подхода положен принцип ценовой эквивалентности мяса и мясного сырья с учетом качественных параметров, где основ- ной составляющей должен стать учет потребительских свойств, при формировании стоимости продуктов [6]. THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 In the conditions of the constantly changing prices in the market, a stable indicator of raw material and fin- ished product quality will enable a prompt solution to the problems regarding development of assortment tasks, sale plans, production programs with the aim of obtaining high economic indicators that ensure breakeven of production and stability of financial situation of an enterprise. 86 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА In the meat industry, the main parameters determin- ing product quality, which present consumer properties of products, are the indicators of morphology and chemical composition of raw material and raw material constituent. В мясной промышленности основными параметра- ми определяющими качество продукции, которые со- ставляют потребительские свойства продукции, явля- ются показатели морфологии и химического состава сырья и сырьевой составляющей. Morphology is the content of muscle, connective, fatty and bone tissues. Meat morphology is detected in the pro- cess of cutting, boning and trimming of bone-in meat, which is obtained as a result of slaughter and processing of slaughter animals. The chemical composition is determined by the content of proteins and fats. The main proteins being constituents of meat products are muscle and connective tis- sue proteins. The analysis of fat-containing products is car- ried out by the histological method with account for content of fatty acids (poly- and monounsaturated). Морфология — это содержание мышечной, соеди- нительной, жировой и костной тканей. Морфология мяса определяется в процессе разделки, обвалки и жи- ловки мяса на костях, которое получают в результате убоя и переработки убойных животных. Химический состав определяется содержанием белков и жиров. К основным белкам, входящим в состав мясных про- дуктов относятся: мышечные и соединительноткан- ные белки. Анализ жиросодержащих продуктов проводится гистологическим методом, с учетом содер- жания жирных кислот (поли- и мононенасышенных). The main species of slaughter animals in the meat in- dustry are cattle, sheep and goats, and pigs. р В мясной промышленности основными видами убойного скота являются крупный и мелкий рогатый скот, свиньи. The criteria for evaluation of all constituent parts are established to determine the coefficients of the consumer properties of bone-in meat. Bone-in meat (meat in car- casses) presents a complex of four types of tissues: muscle, connective tissue, fatty and bone tissues. Для определения коэффициентов потребительских свойств мяса на костях установлены критерии оценки всех составных частей. Мясо на костях (мясо в тушах) 87 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 A h fi представляет собой совокупность четырех видов тка- ней: мышечной, соединительной, жировой и костной. skeleton of a of each typ лете на основе At the first stage, the calculations for determination of the structure of the bony skeleton of all species of slaughter animals were carried out and the specific weight of each type of bones was established [7, 8]. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch Cb = Pb ÷ Pcarcass (2) 𝐶 – the coefficient of the consumer properties of bone tis t in a bone, g/100 g; Pcarcass – the protein content in a car owing stage the connective tissue of slaughter animals w At the following stage, the connective tissue of slaugh- ter animals was examined. To detect the mass fraction of the connective tissue in a carcass, the connective tissue ex- tracted in trimming simultaneously with the muscle tis- sue and the connective tissue inside the muscle tissue (i.e., intramuscular connective tissue) were taken into account.h фициент потребительских свойств соединительной ткани льной ткани, г на 100 г; Бт – содержание белка в туше, г н лен коэффициент на жировую ткань мяса КРС енности полиненасыщенных жирных кислот (в перес нности общей белковой составляющей говядины (в пе с учетом показателя усвояемости говяжьего жира (пут мпературы плавления для полного усвоения жиров его жира): 𝑘𝑘𝑘𝑘ж гов = ПНЖКАСВ гов × 9,0 БАСВ гов × 4,0 × 370С 𝑡𝑡пл гов На следующем этапе рассмотрена соединительная ткань мяса убойных животных. Для определения мас- совой доли соединительной ткани в туше учитывает- ся соединительная ткань, выделяемая при жиловке одновременно с мышечной тканью и соединительная ткань, которая находится внутри мышечной ткани, то есть межмышечная. Содержание белка всех видов со- единительной ткани в пересчете на абсолютно сухое вещество устанавливается с учетом массовой доли ткани в туше. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch , where: 𝐶𝐶b protein conte At the fol To detect the extracted in t inside the m ) м соотношения и температуры owing stage, the connective tissue of slaughter animals w mass fraction of the connective tissue in a carcass, the con mming simultaneously with the muscle tissue and the con The protein content of all types of connective tissue on an absolutely dry matter basis was established taking into account the mass fraction of the connective tissue in a carcass.hfi РС; ПНЖКАСВ гов – С в пересчёте на uscle tissue (i.e., intramuscular connective tissue) we The coefficient of the consumer properties of the connective tissue was calculated using the ratio of protein in the connective tissue to the total protein in a carcass according to the equation: АСВ𝑡 пл (4) ффициент потребительских свойств жировой ткани мяса енасыщенных жирных кислот в жировой ткани мяса КР вещество, г на 100 г; БАСВ гов – содержание белка в мясе КР Коэффициент потребительских свойств соедини- тельной ткани, рассчитывается исходя из отношения белка в соединительной ткани к общему белку в туше по формуле: account. Cc = Pc ÷ Pcarcass, (3)fi (3) ; 9,0 и 4,0 – энергетическая ценно альная температура плавления для kc = Бс ÷ БТ , (3) енно для (3) ть жира и белка олного усвоения кота определены сравнительных where: Cc — the coefficient of the consumer properties of the connective tissue; Pc — the content of protein in the connective tissue, g/100 g; Pcarcass — the content of protein in a carcass, g/100 g. ть жира и белка олного усвоения кота определены сравнительных where: Cc — the coefficient of the consumer properties of the connective tissue; Pc — the content of protein in the connective tissue, g/100 g; Pcarcass — the content of protein in a carcass, g/100 g. кал/г; 37 С – максимальная температура плавления для ература плавления говяжьего жира [9]. ы потребительских свойств жировой ткани других видов коэффициенту на говяжью жировую ткань с учето где: kc — коэффициент потребительских свойств соедини- тельной ткани; Бс — содержание белка в соединительной ткани, г на 100 г; БТ — содержание белка в туше, г на 100 г. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости лено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch ество в костном уше. ельских свойств формуле: (2) ction of the ith bone in the skeleton, %; I𝑖𝑖 chem – indicator o position of the ith bone, g/100 g. rotein content on an absolutely dry matter basis in the bon Then, the protein content on an absolutely dry matter basis in the bony skeleton of cattle, sheep and goats, and pigs was established taking into account the bone yield in a carcass.fi е о о ро о о с о , с е с у е о од ос полученных показателей, рассчитан коэффициент потреби ходя из отношения белка в кости к общему белку в туше п фициент потребительских свойств костной ткани; Бк – со Далее установлено содержание белка в пересчете на абсолютно сухое вещество в костном скелете крупно- го и мелкого рогатого скота, свиней с учетом выхода кости в туше. F𝑖𝑖 – mass fr chemical com Then, the ержание белка в животных. Для соединительная rotein content on an absolutely dry matter basis in the bon nd goats, and pigs was established taking into account the On the basis of the obtained indicators, the coefficient of the consumer properties of the bone tissue was calculated using the ratio of protein in a bone to the total protein in a carcass according to the equation: т – содержание белка в туше, г на 100 г. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch этапе рассмотрена соединительная ткань мяса убойны овой доли соединительной ткани в туше учитывается при жиловке одновременно с мышечной тканью и соед С На основании полученных показателей, рассчитан коэффициент потребительских свойств костной тка- ни, исходя из отношения белка в кости к общему бел- ку в туше по формуле: cattle, sheep a carcass. On the b ndicators, the coefficient of the consum Cb = Pb ÷ Pcarcass, (2)fi onsum (2) кани, то есть межмышечная. Содер ете на абсолютно сухое вещество у kк = Бк ÷ БТ, (2) of the bone Содер тво у (2) bone ание белка всех танавливается с вается исходя из ssue was calculated using the ratio of protein in a bon rcass according to the equation: where: Cb — the coefficient of the consumer properties of bone tissue; Pb — the protein content in a bone, g/100 g; Pcarcass — the protein content in a carcass, g/100 g. ание белка всех танавливается с вается исходя из ssue was calculated using the ratio of protein in a bon rcass according to the equation: where: Cb — the coefficient of the consumer properties of bone tissue; Pb — the protein content in a bone, g/100 g; Pcarcass — the protein content in a carcass, g/100 g. ание белка всех танавливается с вается исходя из ssue was calculated using the ratio of protein in a bon rcass according to the equation: where: Cb — the coefficient of the consumer properties of bone tissue; Pb — the protein content in a bone, g/100 g; Pcarcass — the protein content in a carcass, g/100 g. ьной ткани в пересчете на абсолютно сухое вещество у доли ткани в туше. потребительских свойств соединительной ткани, рассчиты в соединительной ткани к общему белку в туше по формул где: kк — коэффициент потребительских свойств костной ткани; Бк — содержание белка в кости, г на 100 г; БТ — содержание белка в туше, г на 100 г. protein in a c ьной ткани в пересчете на абсолютно сухое вещество у оли ткани в туше. потребительских свойств соединительной ткани, рассчиты в соединительной ткани к общему белку в туше по формул где: kк — коэффициент потребительских свойств костной ткани; Бк — содержание белка в кости, г на 100 г; БТ — содержание белка в туше, г на 100 г. protein in a c : (3) Бс – содержание 100 г. как отношение ете на АСВ) к счете на АСВ). 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА To calculate the coef- ficients of the consumer properties of the bone tissue, the average weighted indicators of the protein and moisture content in the bony skeleton were detected on the basis of the chemical composition and yield of bones according to the equation: species of slaughter animals were carried out and the sp of bones was established [7,8]. To calculate the coeff perties of the bone tissue, the average weighted indicators ontent in the bony skeleton were detected on the basis of nd yield of bones according to the equation: 𝑛 представляет собой совокупность четырех видов тка- ней: мышечной, соединительной, жировой и костной. skeleton of a of each type На первом этапе были проведены расчеты по опре- делению структуры костного скелета всех видов убой- ных животных и установлены удельные веса каждого вида костей [7, 8]. Для расчета коэффициентов потре- бительских свойств костной ткани определены сред- невзвешенные показатели содержания белка и вла- ги в костном скелете на основе химического состава и выхода костей по формуле: е показатели содержания белка и влаги в костном ск ава и выхода костей по формуле: of each type consumer pr and moisture composition ве Iav chem = ෍(F𝑖𝑖 % 𝑛𝑛 𝑖𝑖=1 × I𝑖𝑖 chem) , (1) (1) ф р у Пср хим = ෍(Дi % n i=1 × Пi хим) , (1) (1) Дi % – массовая г на 100 г. 𝑖𝑖𝑖 (1) m – average weighted indicator of the chemical compositi where: Iav chem — average weighted indicator of the chemical composition, g/100 g; Fi % — mass fraction of the ith bone in the skeleton, %; Ii chem — indicator of the chemical composition of the ith bone, g/100 g. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch eef fatty tissue (𝑡𝑡melting beef ), 𝐶 l l i h ffi i k2 ср = tпл ÷ tпл гов.𝑘𝑘𝑘 = ННЖК ÷ ННЖ соотношения те ядины (𝑡𝑡пл гов), The summary equation for calculating the coefficient of the consumer properties of fatty tissues of meat from other examined animal species is as follows:𝑡𝑡𝐶𝑡𝑡 mary equation for calculating the coefficient of the consumer es of meat from other examined animal species is as follows: 𝐶𝐶𝑡 Сводная формула расчета коэффициента потреби- тельских свойств на жировую ткань мяса различных видов скота, за исключением КРС, выглядит следую- щим образом:𝑘𝑘𝑘 вида мяса (𝑡𝑡пл) и жировой ткани го 𝑘𝑘𝑘𝑘2 ср = 𝑡𝑡пл ÷ 𝑡𝑡пл гов ула расчета коэффициента потребительских свойств на дов скота за исключением КРС выглядит следующим об of fatty tis ю ткань 𝐶𝐶f = 𝐶𝐶f beef × UFA × 𝑡𝑡melting beef UFAbeef × 𝑡𝑡melting . (6) (6) исключением КРС, выглядит следующим об kж = kж гов × ННЖК × tпл гов ННЖКгов × tпл (6) . (6) . (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch ы потребительских свойств жировой ткани других видов коэффициенту на говяжью жировую ткань с учето тношения содержания ненасыщенных жирных кисло о ж ро о фор е:𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 где: kж гов — коэффициент потребительских свойств жировой ткани мяса КРС; — содержание полинена- сыщенных жирных кислот в жировой ткани мяса КРС в пересчёте на абсолютно сухое вещество, г на 100 г; — содержание белка в мясе КРС в пересчёте на аб- солютно сухое вещество, г на 100 г; 9,0 и 4,0 — энерге- тическая ценность жира и белка соответственно, ккал/г; 37 °С — максимальная температура плавления для пол- ного усвоения жиров; tпл гов — температура плавления го- вяжьего жира [9]. , where: 𝐶 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 matter (AD beef on an fat and pro f ll di ti , where: 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 matter (A b f𝑃 (ADM) basis). The calculatio beef fat (by correlation of the fats and the melting temperatu 𝐶𝐶f bee , h 𝐶𝐶beef h ffi𝑃𝑃𝑃𝑃 триваемого вида са КРС; 𝑘𝑘𝑘𝑘1 ср – рных кислот в ины (ННЖКгов), of fats; 𝑡𝑡melting beef – the melting temperature of beef fat [9]. icients of the consumer properties of the fatty tissue of meat imal species were determined with respect to the coefficient with consideration for the comparative indicators of the r absolutely dry matter (ADM) basis, g/100g; 9.0 and 4.0 – energy v ein, respectively, kcal/g; 370С – the maximum melting temperat n of fats; 𝑡𝑡melting beef – the melting temperature of beef fat [9]. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch Cf = Cf beef × C1 сomp ÷ C2 сomp, (5)fi𝐶 f = 𝐶𝐶f beef × 𝐶𝐶1 comp ÷ 𝐶𝐶2 comp kж = kж гов × k1 ср ÷ k2 ср, full diges (5) y tissu melting beef𝐶 ratio ators (5) где: kж — коэффициент потребительских свойств жировой ткани рассматриваемого вида мяса; kж гов — коэффици- ент потребительских свойств жировой ткани мяса КРС; k1 ср — сравнительный показатель соотношения содержа- ния ненасыщенных жирных кислот в жировой ткани рассматриваемого вида мяса (ННЖК) и жировой ткани говядины (ННЖКгов), 𝑡 The coefficients of the co examined animal species we fatty tissue with considerati , где: 𝐶𝐶f meat und beef fatty f tt id где: kж — коэффициент потребительских свойств жировой ткани рассматриваемого вида мяса; kж гов — коэффици- ент потребительских свойств жировой ткани мяса КРС; 𝑡 The coefficients of the con examined animal species wer , где: 𝐶𝐶f где: kж — коэффициент потребительских свойств жировой ткани рассматриваемого вида мяса; kж гов — коэффици- ент потребительских свойств жировой ткани мяса КРС; 𝑡 The coefficients of the con examined animal species wer , где: 𝐶𝐶f где: kж — коэффициент потребительских свойств жировой ткани рассматриваемого вида мяса; kж гов — коэффици- ент потребительских свойств жировой ткани мяса КРС; k1 ср — сравнительный показатель соотношения содержа- ния ненасыщенных жирных кислот в жировой ткани рассматриваемого вида мяса (ННЖК) и жировой ткани говядины (ННЖКгов), 𝑡 The coefficients of the co examined animal species wer fatty tissue with considerati , где: 𝐶𝐶f meat unde beef fatty f id umer properties of the fatty tissue of meat from other determined with respect to the coefficient for the beef for the comparative indicators of the ratio of the 11 где: Cf — the coefficient of the consumer properties of the fatty tissue of a kind of meat under consideration; Cf beef — the co- efficient of the consumer properties of the beef fatty tissue; C1 сomp — the comparative indicator of the ratio of the unsatu- rated fatty acid (UFA) content in the fatty tissue of a kind of meat under consideration and beef fatty tissue (UFAbeef),𝐶𝐶𝐶𝐶 (5) 𝐶 the coefficient of the consumer properties of the fatty tissue o consideration; 𝐶𝐶f beef – the coefficient of the consumer prope ssue; 𝐶𝐶1 comp – the comparative indicator of the ratio of the u umer properties of the fatty tissue of meat from other determined with respect to the coefficient for the beef for the comparative indicators of the ratio of the 11 где: Cf — the coefficient of the consumer properties of the fatty tissue of a kind of meat under consideration; Cf beef — the co- efficient of the consumer properties of the beef fatty tissue; C1 сomp — the comparative indicator of the ratio of the unsatu- rated fatty acid (UFA) content in the fatty tissue of a kind of meat under consideration and beef fatty tissue (UFAbeef),𝐶𝐶𝐶𝐶 (5) 𝐶 the coefficient of the consumer properties of the fatty tissue o consideration; 𝐶𝐶f beef – the coefficient of the consumer prope ssue; 𝐶𝐶1 comp – the comparative indicator of the ratio of the u р р k1 ср — сравнительный показатель соотношения содержа- ния ненасыщенных жирных кислот в жировой ткани рассматриваемого вида мяса (ННЖК) и жировой ткани говядины (ННЖКгов), fatty tissue with considerati 𝐶 meat unde beef fatty f tt id р р k1 ср — сравнительный показатель соотношения содержа- ния ненасыщенных жирных кислот в жировой ткани рассматриваемого вида мяса (ННЖК) и жировой ткани говядины (ННЖКгов), fatty tissue with considerati 𝐶 meat unde beef fatty f tt id C1 сomp = UFA ÷ UFAbeef; e fatty tissue of a kind of k1 ср = ННЖК ÷ ННЖКгов; C2 сomp — the comparative indicator of the melting temperature of the fatty tissue of a kind of meat under consideration (tmelting) and beef fatty tissue ( ),𝐶 issue (UFAbeef), 𝐶𝐶1 comp = UFA ÷ UFAbeef; 𝐶𝐶2 comp – the c the melting temperature of the fatty tissue of a kind of m k2 ср — сравнительный показатель соотношения температур плавления жировой ткани рассматриваемого вида мяса (tпл) и жировой ткани говядины (tпл гов), 𝑘𝑘𝑘𝑘ср ННЖК ННЖКгов beef fatty indicator id k2 ср — сравнительный показатель соотношения температур плавления жировой ткани рассматриваемого вида мяса (tпл) и жировой ткани говядины (tпл гов), 𝑘𝑘𝑘𝑘ср ННЖК ННЖКгов beef fatty indicator id C2 сomр = tmelting ÷ . (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch The calculation was carried out using the indicator of digestib energy value of polyunsaturated fatty acids (on an absolutely dry energy value of the total protein constituent of beef (on an ab (ADM) basis) The calculation was carried out using the indicato connective tissue, g/100 g; Pcarcass – the content of e beef fatty tissue was determined as a ratio of the жира (путем соотношения максимальной температу- ры плавления для полного усвоения жиров и темпера- туры плавления говяжьего жира): енности общей белковой составляющей говядины (в пе с учетом показателя усвояемости говяжьего жира (пу мпературы плавления для полного усвоения жиров его жира):𝑘𝑘𝑘𝑡 fats and the energy va energy va (ADM) b – the content of protein in th protein in a carcass, g/100 g. Then, the coefficient for 𝐶𝐶f beef = PUFAАDM beef × 9.0 PАDM beef × 4.0 × 370С 𝑡𝑡melting beef , (4) n of the maximum melting temperature for full digestibi mperature of beef fat): asis). The calculation was carried out using the indicato by correlation of the maximum melting temperature for he melting temperature of beef fat): y s (on an absolutely dry matter basis) to the uent of beef (on an absolutely dry matter 𝑘𝑘𝑘𝑘ж гов = ПНЖКАСВ гов × 9,0 БАСВ гов × 4,0 × 370С 𝑡𝑡пл гов ( ) , (4)𝐶 beef fat (b fats and th energy value of polyunsaturat energy value of the total pro (4) gestib re for (4) ( and th satura tal pr (4) ффициент потребительских свойств жировой ткани мяса ненасыщенных жирных кислот в жировой ткани мяса КР вещество, г на 100 г; БАСВ гов – содержание белка в мясе КР вещество, г на 100 г; 9,0 и 4,0 – энергетическая ценно кал/г; 370С – максимальная температура плавления для ература плавления говяжьего жира [9]. ы потребительских свойств жировой ткани других видов коэффициенту на говяжью жировую ткань с учето тношения содержания ненасыщенных жирных кисло дов жиров по формуле:𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 где: kж гов — коэффициент потребительских свойств жировой ткани мяса КРС; — содержание полинена- сыщенных жирных кислот в жировой ткани мяса КРС в пересчёте на абсолютно сухое вещество, г на 100 г; — содержание белка в мясе КРС в пересчёте на аб- солютно сухое вещество, г на 100 г; 9,0 и 4,0 — энерге- тическая ценность жира и белка соответственно, ккал/г; 37 °С — максимальная температура плавления для пол- ного усвоения жиров; tпл гов — температура плавления го- вяжьего жира [9]. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch и температур триваемого вида са КРС; 𝑘𝑘𝑘𝑘1 ср – рных кислот в Then, the coefficient for the beef fatty tissue was de- termined as a ratio of the energy value of polyunsaturated fatty acids (on an absolutely dry matter basis) to the en- ergy value of the total protein constituent of beef (on an absolutely dry matter (ADM) basis). The calculation was carried out using the indicator of digestibility of beef fat ношения содержания ненасыщенных жирных кисло дов жиров по формуле: 𝑘𝑘𝑘𝑘ж = 𝑘𝑘𝑘𝑘ж гов × 𝑘𝑘𝑘𝑘1 ср ÷ 𝑘𝑘𝑘𝑘2 ср (5) фициент потребительских свойств жировой ткани рассм ффициент потребительских свойств жировой ткани м оказатель соотношения содержания ненасыщенных ж Далее определен коэффициент на жировую ткань мяса КРС как отношение энергетической ценности полиненасыщенных жирных кислот (в пересчете на АСВ) к энергетической ценности общей белковой со- ставляющей говядины (в пересчете на АСВ). Расчет проведен с учетом показателя усвояемости говяжьего 88 вяди 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА е, г на 100 г. КРС как отношение M) basis). The calculation was carried out using the indicator of di ein in a carcass, g/100 g. hen, the coefficient for the beef fatty tissue was determined as a rati𝐶 protein in a carcass, g/100 g. (3) ficient of the consumer properties of the connective tissue; P жира (путем соотношения максимальной температу- ры плавления для полного усвоения жиров и темпера- туры плавления говяжьего жира): енности общей белковой составляющей говядины (в пе с учетом показателя усвояемости говяжьего жира (пу мпературы плавления для полного усвоения жиров его жира):𝑘𝑘𝑘𝑡 fats and th energy va energy v (ADM) b – the content of protein in t protein in a carcass, g/100 g. Then, the coefficient for есчете на АСВ). ем соотношения и температуры (by correlation of the maximum melting temperature for full digestibility of fats and the melting temperature of beef fat): melting temperature of beef fat): 𝐶𝑡 e of polyunsaturated fatty acids (on an absolutely dry matter basis) ue of the total protein constituent of beef (on an absolutely dry is). (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch , where: 𝐶 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 matter (AD beef on an fat and pro f ll di ti , where: 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏 matter (A b f𝑃 (ADM) basis). The calculatio beef fat (by correlation of th fats and the melting temperatu 𝐶𝐶f bee , h 𝐶𝐶beef th ffi𝑃𝑃𝑃𝑃 КРС; ПНЖКАСВ гов – С в пересчёте на С в пересчёте на ть жира и белка олного усвоения кота определены сравнительных 𝐶𝑡 (4) 𝐶 beef – the coefficient of the consumer properties of the beef – the content of the polyunsaturated fatty acids (on an abs M) basis) in the beef fatty tissue, g/100g; PАDM beef – the content o bsolutely dry matter (ADM) basis, g/100g; 9.0 and 4.0 – ener ein, respectively, kcal/g; 370С – the maximum melting tem where: Cf beef — the coefficient of the consumer properties of the beef fatty tissue; 𝐶𝐶f beef = PUFAАDM beef × 9.0 PАDM beef × 4.0 × 370С 𝑡𝑡melting beef (4) 𝐶𝐶f beef – the coefficient of the consumer properties of the beef fatty – the content of the polyunsaturated fatty acids (on an absolute M) basis) in the beef fatty tissue g/100g; PАDM beef – the content of pro — the content of the polyunsatu- rated fatty acids (on an absolutely dry matter (ADM) basis) in the beef fatty tissue, g/100g; 𝐶𝐶f beef = PUFAАDM beef × 9.0 PАDM beef × 4.0 × 370С 𝑡𝑡melting beef , (4) where: 𝐶𝐶f beef – the coefficient of the consumer properties of t 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 – the content of the polyunsaturated fatty acids (on — the content of pro- tein in beef on an absolutely dry matter (ADM) basis, g/100g; 9.0 and 4.0 — energy value of fat and protein, respectively, kcal/g; 37 °С — the maximum melting temperature for full digestion of fats; was carried out using the indicator of digestibility of maximum melting temperature for full digestibility of e of beef fat): 𝐶 = PUFAАDM beef × 9.0 PАDM beef × 4.0 × 370С 𝑡𝑡melting beef (4) — the melting temperature of beef fat [9]. (4) ффициент потребительских свойств жировой ткани мяса ненасыщенных жирных кислот в жировой ткани мяса КР вещество, г на 100 г; БАСВ гов – содержание белка в мясе КР вещество, г на 100 г; 9,0 и 4,0 – энергетическая ценно кал/г; 370С – максимальная температура плавления для ература плавления говяжьего жира [9]. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch ficients of the consumer properties of the fatty tissue of meat from nimal species were determined with respect to the coefficient for th i h id i f h i i di f h i𝑃𝑃𝑃𝑃 matter (ADM) basis) in the beef fatty tissue, g/100g; PАDM the c beef on an absolutely dry matter (ADM) basis, g/100g; 9.0 and 4 fat and protein, respectively, kcal/g; 370С – the maximum melt full digestion of fats; 𝑡𝑡melting beef – the melting temperature of beef fa The coefficients of the consumer properties of the fatty tissue examined animal species were determined with respect to the coe nt of the consumer properties of the beef fatty tissue; e polyunsaturated fatty acids (on an absolutely dry f fatty tissue, g/100g; PАDM beef – the content of protein in r (ADM) basis, g/100g; 9.0 and 4.0 – energy value of cal/g; 370С – the maximum melting temperature for The coefficients of the consumer properties of the fatty tissue of meat from other examined animal species were determined with respect to the coefficient for the beef fatty tissue with consideration for the comparative indi- cators of the ratio of the unsaturated fatty acid content and the melting temperature of these fats according to the equation: fatty acid content and the melting temperature of these fats ac : 𝐶𝐶𝐶𝐶 р ф р у 𝑘𝑘𝑘𝑘ж = 𝑘𝑘𝑘𝑘ж гов × 𝑘𝑘𝑘𝑘1 ср ÷ 𝑘𝑘𝑘𝑘2 ср (5) фициент потребительских свойств жировой ткани рассм ффициент потребительских свойств жировой ткани м оказатель соотношения содержания ненасыщенных ж ссматриваемого вида мяса (ННЖК) и жировой ткани говя Коэффициенты потребительских свойств жировой ткани других видов скота определены по отношению к коэффициенту на говяжью жировую ткань с учетом сравнительных показателей соотношения содержания ненасыщенных жирных кислот и температур плавле- ния этих видов жиров по формуле: g The coe examined a fatty tissue fat and p full diges The co examined f 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝐴𝐴𝐴𝐴𝐴𝐴 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 – the content of matter (ADM) basis) in the b beef on an absolutely dry ma fat and protein, respectively, unsaturate the equati р ф р у kж = kж гов × k1 ср ÷ k2 ср, (5) fatty tissu full digestion of fats; 𝑡𝑡melting beef𝐶 6 n for the comparative indicators of the ratio consideration for the comparative indicators rature of beef fat [9]. где: ki — коэффициент потребительной стоимости костной, соединительной, жировой тканей (индексы i = к, с, ж); Дi % — доля соответствующего вида ткани в общей массе жилованного мяса, %.𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖=к,с,ж ициент потребительной стоимости костной, соединит % , where: C Ctrimmed – coefficient o the consume tissue fibers th tis m – коэффициент ельских свойств вой ткани; Дс % – – массовая доля шечной ткани в х свойств всех (8) m – the coefficient of the consumer properties of the m he coefficient of the consumer properties of trimmed me the consumer properties of the connective tissue; Cf – the c r properties of the fatty tissue; Fс % – mass fraction of th in trimmed meat %; F% mass fraction of the fatty fiber where: Cm — the coefficient of the consumer properties of the muscle tissue; Ctrimmed — the coefficient of the consumer properties of trimmed meat; Cc — the coefficient of the consumer properties of the connective tissue; Cf — the coefficient of the consumer properties of the fatty tissue; Fс % — mass fraction of the connective tissue fibers in trimmed meat, %; Ff % — mass fraction of the fatty fibers in trimmed meat, %; Fm % — mass fraction of purely muscle tissue in trimmed meat, %. consumer properties of the fatty tissue; Fс % – mass fraction of the conn ue fibers in trimmed meat, %; Ff % – mass fraction of the fatty fibers in trim at, %; Fm % – mass fraction of purely muscle tissue in trimmed meat, %. The examined principles for calculating the coefficients of the con , ; f y – mass fraction of purely muscle tissue in trimmed meat, % ined principles for calculating the coefficients of th The examined principles for calculating the coefficients of the consumer properties for all tissues of slaughter animals are presented in a form of an algorithm in Fig. 2. erties for all tissues of slaughter animals are presented in a form of an algo g.2. Рассмотренные принципы расчетов коэффициентов потребительских свойств всех тканей убойных живот- ных представлены в форме алгоритма на рисунке 2. meat, %; Fm % The exa pr in 7 properties for all tissues of slaughter animals are presented in a form of in Fig.2. Figure 2. The algorithm for calculation of the coefficients of the consumer properties for all tissues of slaughter animals Рис. 2. где: ki — коэффициент потребительной стоимости костной, соединительной, жировой тканей (индексы i = к, с, ж); Дi % — доля соответствующего вида ткани в общей массе жилованного мяса, %.𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖=к,с,ж ициент потребительной стоимости костной, соединит % Figure 2 The algorithm for calculation of the coefficients of the consum A.2 – Алгоритм расчета коэффициентов потребительских свойств всех ткан Изучение морфологического состава мяса Изучение химического состава мяса Расчёт коэффициентов потребительских свойств костной ткани kк = Бк ÷ Бт Расчёт коэффициентов потребительских свойств соединительной ткани 𝑘𝑘𝑘𝑘с = Бс ÷ Бт Расчёт коэффициентов потребительских свойств жировой ткани Мясо крупного рогатого скота 𝑘𝑘𝑘𝑘жгов = ПНЖКАСВ ×9,0 БАСВ ×4,0 × 370С 𝑡𝑡𝑡𝑡пл гов Другие виды скота: 𝑘𝑘𝑘𝑘ж = 𝑘𝑘𝑘𝑘жгов× ННЖК× 𝑡𝑡𝑡𝑡пл гов ННЖКгов×𝑡𝑡𝑡𝑡пл Расчёт коэффициентов потребительских свойств мышечной ткани 𝑘𝑘𝑘𝑘жил = 1 −෍(𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %) 𝑖𝑖𝑖𝑖=к,с,ж Жилованное мясо: 𝑘𝑘𝑘𝑘м = 𝑘𝑘𝑘𝑘жил −(𝑘𝑘𝑘𝑘с × Дс% + 𝑘𝑘𝑘𝑘ж × Дж%) Дм % Чисто мышечная ткань: 7 all tissues of slaughter animals are presented in a form of f ll f l h l б й Study of meat morphological composition Study of meat chemical composition Calculation of coefficients of consumer properties for bone tissue 𝐶𝐶b = Pb ÷ Pcarcass 𝐶𝐶c = Pc ÷ Pcarcass Calculation of coefficients of consumer properties for connective tissue Calculation of coefficients of consumer properties for fatty tissue Cattle meat 𝐶𝐶f beef = PUFAАDM beef ×9.0 PАDM beef ×4.0 × 370С 𝑡𝑡melting beef Meat from other examined animal species: 𝐶𝐶f = 𝐶𝐶f beef× UFA× 𝑡𝑡melting beef UFAbeef×𝑡𝑡melting Calculation of coefficients of consumer properties for muscle tissue 𝐶𝐶𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡= 1 −෍(𝐶𝐶𝑖𝑖 × 𝐹𝐹𝑖𝑖 %) 𝑖𝑖=𝑏𝑏,с,𝑓𝑓 Trimmed meat: 𝐶𝐶𝑚𝑚= 𝐶𝐶𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡−(𝐶𝐶с × 𝑃𝑃с % + 𝐶𝐶𝑓𝑓× 𝐹𝐹𝑓𝑓 %) 𝐹𝐹𝑚𝑚 % Purely muscle tissue: Figure 2. The algorithm for calculation of the coefficients of the consumer properties for all tissues of slaughter animals Рис. 2. Алгоритм расчета коэффициентов потребительских свойств всех тканей убойных животных A.2 – Алгоритм расчета коэффициентов потребительских свойств всех тканей убойных ных 7C; LB0BK 𝑖𝑖𝑖𝑘𝑘𝑘 Fig.2 –The algorithm for calculation of the coeffici for all tissues of slaughter animals𝑖𝑏𝑓𝐶 где: ki — коэффициент потребительной стоимости костной, соединительной, жировой тканей (индексы i = к, с, ж); Дi % — доля соответствующего вида ткани в общей массе жилованного мяса, %.𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖=к,с,ж ициент потребительной стоимости костной, соединит % Кроме того, учитывается содержание в жилован- ном мясе внутримышечных волокон (соединительных и жировых) по формуле: i=к, с, ж); Д𝑖𝑖𝑖𝑖– доля соответствующего вида ткани , %. читывается содержание в жилованном мясе внутримы и жировых) по формуле:𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 In additio into account в общей массе шечных волокон In addition, the content of intramuscular fibers (con- nective tissue and fat) is taken into account in trimmed beef according to the equation: , the content of intramuscular fibers (connective tissue and n trimmed beef according to the equation: Cm = Ctrimmed −(Cс × Fс % + Cf × Ff %) Fm % , (8)fi ) ф р у 𝑘𝑘𝑘𝑘м = 𝑘𝑘𝑘𝑘жил −(𝑘𝑘𝑘𝑘с × Дс % + 𝑘𝑘𝑘𝑘ж × Дж %) Дм % 𝑘𝑘𝑘, (8) (8) (8) (8) ффициент потребительских свойств мышечной ткани; kжи свойств жилованного мяса; 𝑘𝑘𝑘𝑘с – коэффициент потреби ткани; kж – коэффициент потребительских свойств жиро единительнотканных волокон в жилованном мясе, %; Дж % н в жилованном мясе, %; Дм % – массовая доля чисто мы , %. е принципы расчетов коэффициентов потребительски животных представлены в форме алгоритма на рисунке 2. где: kм — коэффициент потребительских свойств мышечной ткани; kжил — коэффициент потребительских свойств жилованного мяса; kс — коэффициент потребительских свойств соединительной ткани; kж — коэффициент по- требительских свойств жировой ткани; Дс % — массовая доля соединительнотканных волокон в жилованном мясе, %; Дж % — массовая доля жировых волокон в жило- ванном мясе, %; Дм % — массовая доля чисто мышечной ткани в жилованном мясе, %. где: ki — коэффициент потребительной стоимости костной, соединительной, жировой тканей (индексы i = к, с, ж); Дi % — доля соответствующего вида ткани в общей массе жилованного мяса, %.𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖=к,с,ж ициент потребительной стоимости костной, соединит % Алгоритм расчета коэффициентов потребительских свойств всех тканей убойных животных A.2 – Алгоритм расчета коэффициентов потребительских свойств всех тканей убойных ных 7C; LB0BK Изучение морфологического состава мяса Изучение химического состава мяса Расчёт коэффициентов потребительских свойств костной ткани kк = Бк ÷ Бт Расчёт коэффициентов потребительских свойств соединительной ткани 𝑘𝑘𝑘𝑘с = Бс ÷ Бт Расчёт коэффициентов потребительских свойств жировой ткани Мясо крупного рогатого скота 𝑘𝑘𝑘𝑘жгов = ПНЖКАСВ ×9,0 БАСВ ×4,0 × 370С 𝑡𝑡𝑡𝑡пл гов Другие виды скота: 𝑘𝑘𝑘𝑘ж = 𝑘𝑘𝑘𝑘жгов× ННЖК× 𝑡𝑡𝑡𝑡пл гов ННЖКгов×𝑡𝑡𝑡𝑡пл Расчёт коэффициентов потребительских свойств мышечной ткани 𝑘𝑘𝑘𝑘жил = 1 −෍(𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %) 𝑖𝑖𝑖𝑖=к,с,ж Жилованное мясо: 𝑘𝑘𝑘𝑘м = 𝑘𝑘𝑘𝑘жил −(𝑘𝑘𝑘𝑘с × Дс% + 𝑘𝑘𝑘𝑘ж × Дж%) Дм % Чисто мышечная ткань: Fig.2 –The algorithm for calculation of the coefficients of the consumer prop for all tissues of slaughter animals Study of meat morphological composition Study of meat chemical composition Calculation of coefficients of consumer properties for bone tissue 𝐶𝐶b = Pb ÷ Pcarcass 𝐶𝐶c = Pc ÷ Pcarcass Calculation of coefficients of consumer properties for connective tissue Calculation of coefficients of consumer properties for fatty tissue Cattle meat 𝐶𝐶f beef = PUFAАDM beef ×9.0 PАDM beef ×4.0 × 370С 𝑡𝑡melting beef Meat from other examined animal species: 𝐶𝐶f = 𝐶𝐶f beef× UFA× 𝑡𝑡melting beef UFAbeef×𝑡𝑡melting Calculation of coefficients of consumer properties for muscle tissue 𝐶𝐶𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡= 1 −෍(𝐶𝐶𝑖𝑖 × 𝐹𝐹𝑖𝑖 %) 𝑖𝑖=𝑏𝑏,с,𝑓𝑓 Trimmed meat: 𝐶𝐶𝑚𝑚= 𝐶𝐶𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡−(𝐶𝐶с × 𝑃𝑃с % + 𝐶𝐶𝑓𝑓× 𝐹𝐹𝑓𝑓 %) 𝐹𝐹𝑚𝑚 % Purely muscle tissue: properties fo in Fig.2. (1) дневзвешенный показатель химического состава, г на 100 елета, %; Пi хим – показатель химического состава i-й кости ено содержание белка в пересчете на абсолютно сухое ве где: Пср хим — средневзвешенный показатель химического со- става, г на 100 г; Дi % — массовая доля i-й кости в скеле- та, %; — показатель химического состава i-й кости, г на 100 г. , where: Iav ch (6) еделяется как ых показателей тканей и долей ние показателя 𝐶𝑡 (6) on of the coefficients of the consumer properties of the musc as an indicator of the consumer properties of trimmed n for the obtained indicators of the consumer properties of the d connective (Cс) tissues and the fractions of the correspondin at; the value of the consumer properties of bone-in m Calculation of the coefficients of the consumer prop- erties of the muscle tissue is determined as an indicator of the consumer properties of trimmed meat with consider- ation for the obtained indicators of the consumer proper- ties of the bone (Cb), fatty (Cf) and connective (Cс) tissues and the fractions of the corresponding tissues in bone-in meat; the value of the consumer properties of bone-in meat of the corresponding species of farm animals was taken as a unit: (6) циентов потребительских свойств мышечной ткани оп бительских свойств жилованного мяса с учетом получен войств костной (kк), жировой (kж) и соединительной (k тканей в мясе на костях, за единицу принято знач оимости мяса на костях соответствующего вида скота: 𝑘𝑘𝑘𝑘жил = 1 −(𝑘𝑘𝑘𝑘к × Дк % + 𝑘𝑘𝑘𝑘ж × Дж % + 𝑘𝑘𝑘𝑘с × Дс %) 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖 Расчет коэффициентов потребительских свойств мышечной ткани определяется как показатель потре- бительских свойств жилованного мяса с учетом полу- ченных показателей потребительских свойств кост- ной (kк), жировой (kж) и соединительной (kс) тканей и долей соответствующих тканей в мясе на костях, за единицу принято значение показателя потребитель- ной стоимости мяса на костях соответствующего вида скота: Calcula determine considerat fatty (Cf) a bone-in m d s of farm animals was taken as a unit: Ctrimmed = 1 – (Cb × Fb % + Cf × Ff % + Cc × Fс %) жировой й массе s of farm animals was taken as a unit: (Cb × Fb % + Cf × Ff % + Cс × Fс %) or ∑ ൫Ci × Fi %൯ i=b,с,f , (7) Ctrimmed = 1 – (Cb × Fb % + Cf × Ff % + Cc × Fс %) Ctrimmed = 1 – Σ i = b, с, f (Ci × Fi %), (7) 𝑖𝑖𝑖𝑖=к,с,ж нт потребительной стоимости костной, со с, ж); Д𝑖𝑖𝑖𝑖 %– доля соответствующего вида kжил = 1 – (kк × Дк % + kж × Дж % + kc × Дс %) kжил = 1 – Σ i = к, с, ж (ki × Дi %), cor 𝑖𝑖𝑖𝑖=к,с,ж фициент потребительной стоимости костной, соединительной, i=к, с, ж); Д𝑖𝑖𝑖𝑖 %– доля соответствующего вида ткани в общ % kжил = 1 – (kк × Дк % + kж × Дж % + kc × Дс %) или kжил = 1 – Σ i = к, с, ж (ki × Дi %), (7) corresponding spec Ctrimmed = 1 Ctrimmed = 1 or (7) динит ткани (7) Ctrimm C фициен i=к, с или в об1 or 89 мыше e: Ci – THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 оимости мяса на костях соответствующего вида скота: 𝑘𝑘𝑘𝑘жил = 1 −(𝑘𝑘𝑘𝑘к × Дк % + 𝑘𝑘𝑘𝑘ж × Дж % + 𝑘𝑘𝑘𝑘с × Дс %) 𝑘𝑘𝑘𝑘𝑘𝑘 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 тоимости мяса на костях соответствующего вида скота: 𝑘𝑘𝑘𝑘жил = 1 −(𝑘𝑘𝑘𝑘к × Дк % + 𝑘𝑘𝑘𝑘ж × Дж % + 𝑘𝑘𝑘𝑘с × Дс %) 𝑘𝑘𝑘 ьной, жировой where: Ci — the coefficient of the consumer properties of the bone, connective and fatty tissues (indices i = b, с, f); Fi %— the proportion of the corresponding type of tissues in the total mass of trimmed meat, %. льной, жировой where: Ci — the coefficient of the consumer properties of the bone, connective and fatty tissues (indices i = b, с, f); Fi %— the proportion of the corresponding type of tissues in the total mass of trimmed meat, %. где: ki — коэффициент потребительной стоимости костной, 𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯𝑖𝑖𝑖 где: ki — коэффициент потребительной стоимости костной, соединительной, жировой тканей (индексы i = к, с, ж); Дi % — доля соответствующего вида ткани в общей массе жилованного мяса, %.𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘 𝑘𝑘𝑘𝑘жил = 1 − ෍൫𝑘𝑘𝑘𝑘𝑖𝑖𝑖𝑖 × Д𝑖𝑖𝑖𝑖 %൯ 𝑖𝑖𝑖𝑖=к,с,ж фициент потребительной стоимости костной, соединит % 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА На основании проведенных расчетов установлены коэффициенты потребительских свойств на все виды тканей, составляющих тушу крупного и мелкого рога- того скота, свиней, которые представлены в таблице 1. On the basis of the conducted calculations, the coefficients of the consumer properties for all types of tissues being constituents of a carcass of cattle, sheep and goats, and pigs were established (Table 1). Кроме основных составляющих: мышечной, сое- динительной, жировой и костных тканей в процессе разделки, обвалки и жиловки туш убойных животных образуются технические зачистки. В основном они состоят из лимфатических узлов и желез внутренней секреции, на долю которых приходится от 0,1% живой массы у свиней и мелкого рогатого скота и до 0,8% у крупного рогатого скота. Исходя из морфологиче- ского и химического состава технических зачисток, коэффициент потребительских свойств установлен на уровне 0,03, для трех видов убойных животных (круп- ного, мелкого рогатого скота и свиней). In addition to the main constituents (muscle, connective, fatty and bone tissues) technical scrapings are formed in the process of cutting, boning and trimming of slaughter animal carcasses. They are mainly composed of the lymphatic nodes and endocrine glands, which account for up to 0.8% of the live weight in cattle and 0.1% in sheep and goats, and pigs. Based on the morphological and chemical composition of the technical scrapings, the coefficient of the consumer properties was established at the level of 0.03 for three kinds of slaughter animals (cattle, sheep and goats, and pigs). Table 2 presents the summary data of yields and coefficients of the consumer properties by tissues of carcasses from cattle, sheep and goats, and pigs.hfi В таблице 2 представлены сводные данные по вы- ходам и коэффициентам потребительских свойств по тканям туш крупного и мелкого рогатого скота, свиней. The coefficients for the tissues of three kinds of slaugh- ter animals, which were established based on the per- formed experiments, will be used when determining the quality indicators of all meat products [10]. Установленные на основании проведенных иссле- дований коэффициенты для тканей туш трех видов убойного скота будут применяться при определении показателей качества всех мясных изделий [10]. Table 1. Coefficients of consumer properties \| Табл. 1. Результаты сновании разраб льских свойств по On the basis of the developed algorithm, the coefficients of the consumer properties by tissue composition of meat from cattle, sheep and goats, and pigs were calculated. The calculation of the indicators characterizing the consumer properties of all constituent parts of carcasses of the main species of slaughter animals is carried out taking into account the morphological composition. In the process of carcass cutting, bone-in meat, trimmed meat, muscle tissue, raw fat, connective tissue (cartilages) and bone tissue are extracted according to the technological instruction on meat boning and trimming. вляющих ического ышечную ветствии тельских а, свиней, Results On the basis of the developed algorithm, the coefficients of the cons erties by tissue composition of meat from cattle, sheep and goats, and pigs ulated. The calculation of the indicators characterizing the consumer propert На основании разработанного алгоритма проведе- ны расчеты коэффициентов потребительских свойств по тканевому составу мяса крупного и мелкого рогато- го скота, свиней. Расчет показателей, характеризующих потребительские свойства всех составляющих частей туш основных видов убойных животных, проводится с учетом морфологического состава. В процессе раз- делки туш выделяют мясо на костях, мясо жилованное, мышечную ткань, жир-сырец, соединительную ткань (хрящи) и костную ткань в соответствии технологиче- ской инструкцией по обвалке и жиловке мяса. й. Расчет показателей, характеризующих потребительские свойства всех сос й туш основных видов убойных животных, проводится с учетом морфол а. В процессе разделки туш выделяют мясо на костях, мясо жилованное, жир-сырец, соединительную ткань (хрящи) и костную ткань в со логической инструкцией по обвалке и жиловке мяса. основании проведенных расчетов установлены коэффициенты потре тв на все виды тканей, составляющих тушу крупного и мелкого рогатого ско ые представлены в таблице 1. F0 1. фициенты потребительских свойств Наименование Коэффициенты крупный рогатый скот свиньи мелкий рогатый ско OA> =0 : >ABOE, в том числе: 1,0 1,0 1,0 Мышечная ткань 1,60 1,77 1,61 Жировая ткань 0,27 0,38 0,24 С 0 2 0 2 0 2 pr ca 90 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 На основании морфологии мясных полуфабрика- тов и коэффициентов потребительских свойств от- дельных тканей проведены расчеты коэффициентов на все виды полуфабрикатов. Расчеты проводятся по формулам. Based on the morphology of the meat semi-prepared products and coefficients of the consumer properties of the individual tissues, the coefficients for all types of semi- prepared products were calculated. The calculations were carried out according to the equations. Коэффициент для бескостных полуфабрикатов (Кпб): The coefficient for the boneless semi-prepared products (Cblsp): (9) Кпб = Км × Ум + Кс ×Ус + Кж ×Уж; (9) Кпб = Км × Ум + Кс ×Ус + Кж ×Уж; (9) Коэффициент для полуфабрикатов на кости (Кпк): Коэффициент для полуфабрикатов на кости (Кпк): (9) Cblsp = Cm × Sm + Cc ×Sc + Cf × Sf; (9) The coefficient for the bone-in semi-prepared products (Cbsp): Кпк = Км × Ум + Кс ×Ус + Кж ×Уж+ Кк ×Ук; (10) где: Км — коэффициент потребительских свойств мышеч- ной ткани, Ум — удельный вес мышечной ткани в со- ставе продукта, Кс — коэффициент потребительских свойств соединительной ткани, Ус — удельный вес сое- динительной ткани в продукте, Кж — коэффициент по- требительских свойств жировой ткани, Уж — удельный вес жировой ткани в составе продукта, Кк — коэффи- циент потребительских свойств костной ткани, Ук — удельный вес костной ткани в составе продукта. (10) Cbsp = Cm × Sm + Cc ×Sc + Cf × Sf+ Cb × Sb; (10) where: Cm — the coefficient of the consumer properties of the muscle tissue, Sm — the specific weight of the muscle tissue in a product composition, Cс — the coefficient of the con- sumer properties of the connective tissue, Sс — the specific weight of the connective tissue in a product composition, Cf — the coefficient of the consumer properties of the fatty tissue, Sf — the specific weight of the fatty tissue in a product composition, Cb — the coefficient of the consumer proper- ties of the bone tissue, Sb — the specific weight of the bone tissue in a product composition. Таким образом, были определены коэффициенты на все наименования полуфабрикатов, которые вырабаты- ваются из говядины и свинины по техническим условиям (ТУ 9214–345–00419779–06 и ТУ 9214–456–00419779–03). Для крупнокусковых полуфабрикатов, установленный коэффициент потребительских свойств, колеблется: для говяжьих — от 1,51 до 1,27, для свиных — от 1,61 до 1,29, для бараньих — от 1,81 до 1,22. THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 The meat industry produces several main groups of meat products: semi-prepared products and culinary prod- ucts, sausage products, smoked meats and canned foods. The proportion of the meat constituent in each product group is different. В мясной промышленности вырабатывается несколь- ко основных групп мясных продуктов: полуфабрикаты и кулинарные изделия, колбасные изделия, продукты из мяса, консервы. Доля мясной составляющей в каждой группе продуктов значительно различается. Мясные полуфабрикаты содержат до 100% мясного сырья, поэтому определение мясной сырьевой состав- ляющей этих продуктов, практически дает возмож- ность устанавливать качественные характеристики собственно мясных продуктов. В мясных полуфабри- катах доля мышечной ткани зависит от содержания этой ткани в частях туш или конкретных мышц убой- ных животных, из которых они вырабатываются. Meat semi-prepared products contain up to 100% of meat raw material; thus, determination of meat raw mate- rial constituent in these products practically gives an op- portunity to establish the quality characteristics of meat products per se. In meat semi-prepared products, the pro- portion of muscle tissue depends on the content of this tis- sue in parts of carcasses or individual muscles of slaughter animals, from which they are produced.fi На основании коэффициентов потребительских свойств отдельных тканей и удельных весов этих тка- ней, входящих в состав конкретных продуктов, то есть, их морфологии проводится расчет показателей качества полуфабрикатов. On the basis of the coefficients of the consumer proper- ties of the individual tissues and specific weights of these tissues being constituents of particular products (i.e., their morphology), the calculation of the quality indicators of semi-prepared products was carried out.h The main constituent of the semi-prepared products is the muscle tissue. Its proportion in the beef semi-prepared products varies from 70 to 96%, in the pork semi-prepared products from 67 to 96.8%, and in the lamb semi-prepared products from 68 до 95%.h Основной составляющей полуфабрикатов являет- ся мышечная ткань. Ее доля в говяжьих полуфабри- катах колеблется от 70 до 96%, в свиных — от 67 до 96,8%, в бараньих — от 68 до 95%. Далее по значимо- сти в говяжьих и бараньих полуфабрикатах находится соединительная ткань от 5 до 25%, в свиных — жиро- вая ткань, доля которой в некоторых продуктах дохо- дит — до 50%. The next important tissue is the connective tissue (5 to 25%) in the semi-prepared products from beef and lamb and the fatty tissue in the pork semi-prepared products (up to 50% in several products). 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА Коэффициенты потребительских свойств Name \| Наименование Coefficients \| Коэффициенты cattle \| крупный рогатый скот pigs \| свиньи sheep and goats \| мелкий рогатый скот Bone-in meat, including: \| Мясо на костях, в том числе: 1.0 1.0 1.0 Muscle tissue \| Мышечная ткань 1.60 1.77 1.61 Fatty tissue \| Жировая ткань 0.27 0.38 0.24 Connective tissue, cartilages \| Соединительная ткань, хрящи 0.2 0.2 0.2 Bone \| Кость 0.1 0.1 0.19 s of consumer properties \| Табл. 1. Коэффициенты потребительских свойств Table 2. Summary table of yields and coefficients of the consumer properties by tissues Табл. 2. Сводная таблица по выходам и коэффициентам потребительских свойств п Table 2. Summary table of yields and coefficients of the consumer properties by tissues Табл. 2. Сводная таблица по выходам и коэффициентам потребительских свойств по тканям Table 2. Summary table of yields and coefficients of the consumer properties by tissues Табл. 2. Сводная таблица по выходам и коэффициентам потребительских свойств по тканям Cattle \| Крупный рогатый скот Pigs \| Свиньи Sheep and goats \| Мелкий рогатый скот yield, % \| выход, % coefficients \| коэффи- циенты yield, % \| выход, % coefficients \| коэффи- циенты yield, % \| выход, % coefficients \| коэффи- циенты Bone-in meat \| Мясо на костях 100.00 1.0 100.00 1.0 100.0 1.0 excluded: \| Исключается: bone \| кость 21.20 0.10 12.08 0.10 25.78 0.19 fat, back fat \| жир, шпик 2.75 0.27 15.66 0.38 1.43 0.24 connective tissue, cartilages \| соединительная ткань, хрящи 2.90 0.20 2.01 0.20 1.58 0.20 technical scrapings, losses \| технические зачистки, потери 0.90 — 0.20 — 0.20 — excluded, in total: \| Итого исключается 27.75 — 29.95 — 28.99 — Trimmed meat, including: \| Жилованное мясо, в том числе: 72.25 1.34 70.05 1.29 71.01 1.33 — fatty tissue \| — жировая ткань 4.22 0.27 21.44 0.38 7.31 0.24 — connective tissue \| — соединительная ткань 9.25 0.20 2.38 0.20 6.89 0.20 — purely muscle tissue \| — чисто мышечная ткань 58.78 1.60 46.23 1.77 56.81 1.61 91 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА For the portion-sized semi-prepared products and semi-prepared products in small pieces, the coefficients of the consumer properties are 1.51–1.29 for beef, 1.61–1.23 for pork, 1.5–0.83 for lamb. For the minced semi-prepared products and mincemeat, the respective coefficients are 1.19–0.83. 1,51–1,29 — для говяжьих, 1,61–1,23 — для свиных, от 1,5 до 0,83 — для бараньих, для рубленых полуфабрикатов и фаршевых — 1,19–0,83. Определение коэффициентов потребительной сто- имости на группу изделий, подвергнутых термической обработке, проведено только для мясной составляю- щей этих продуктов. Determination of the coefficients of the consumer properties for the group of products subjected to the ther- mal treatment was carried out only for the meat constitu- ent of these products. В общей сумме затрат на производство мясных продуктов, подвергнутых термической обработке (колбасные изделия, продукты из мяса и консервы), в среднем составляет порядка 70% стоимость сырья: мяса различных видов и основных мясных составля- ющих материалов. Количество сырья, направляемого на выработку конкретного продукта, четко регламен- тировано рецептурой и выходами продукции. Таким образом, для определения коэффициентов потре- бительских свойств конечного продукта достаточно установить коэффициенты для мясной составляющей и основных видов мясных материалов. In the total sum of the product costs for meat prod- ucts subjected to thermal treatment (sausage products, smoked meats and canned foods) the raw material costs are on average about 70%: meat of different kinds and the main meat constituent materials. An amount of raw ma- terial sent to production of a particular product is clearly regulated by a recipe and product yields. Therefore, to detect the coefficients of the consumer properties for the finished products, it is sufficient to establish the coeffi- cients of a meat constituent and the main types of the meat materials.hfi Расчеты коэффициентов потребительной стоимо- сти колбасных изделий проведены для всех наимено- ваний этой группы мясных продуктов на основании рецептур, т.е. соотношений сырья и основных матери- алов. Коэффициенты исчислены исходя из показате- лей потребительских свойств всех видов сырья и ма- териалов, входящих в рецептурный состав. The calculations of the coefficients of the value in use for sausage products were carried out for all items of this group of meat products on the basis of the recipes (i.e., the ratio of raw material and the main materials). (11) (11) Кмс = Км × Ум + Кс ×Ус + Кж ×Уж; (11) Кмс = Км × Ум + Кс ×Ус + Кж ×Уж; Cmс = Cm × Sm + Cс ×Sс + Cf ×Sf; (11)fi (11) Cmс = Cm × Sm + Cс ×Sс + Cf ×Sf; где: Км — коэффициент потребительских свойств мышеч- ной ткани, Ум — удельный вес мышечной ткани мясной составляющей продукта, Кс — коэффициент потреби- тельских свойств соединительной ткани, Ус — удельный вес соединительной ткани в мясной части продукта, Кж — коэффициент потребительских свойств жировой ткани, Уж — удельный вес жировой ткани мясной со- ставляющей в составе продукта. где: Км — коэффициент потребительских свойств мышеч- ной ткани, Ум — удельный вес мышечной ткани мясной составляющей продукта, Кс — коэффициент потреби- тельских свойств соединительной ткани, Ус — удельный вес соединительной ткани в мясной части продукта, Кж — коэффициент потребительских свойств жировой ткани, Уж — удельный вес жировой ткани мясной со- ставляющей в составе продукта. where: Cm — the coefficient of the consumer properties of the muscle tissue, Sm — the specific weight of the muscle tissue of the meat component of a product, Cс — the coefficient of the consumer properties of the connective tissue, Sс — the specific weight of the connective tissue in the meat part of a product, Cf — the coefficient of the consumer properties of the fatty tissue, Sf — the specific weight of the fatty tissue of the meat component of a product. where: Cm — the coefficient of the consumer properties of the muscle tissue, Sm — the specific weight of the muscle tissue of the meat component of a product, Cс — the coefficient of the consumer properties of the connective tissue, Sс — the specific weight of the connective tissue in the meat part of a product, Cf — the coefficient of the consumer properties of the fatty tissue, Sf — the specific weight of the fatty tissue of the meat component of a product. Для вареных колбасных изделий были установле- ны коэффициенты: для изделий высших сортов — 1,28–1,09, для первых сортов — 1,2–0,91, для вторых сортов — 1,04–0,94, для сосисок, сарделек, шпика- чек — 1,26–0,94. THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 Therefore, the coefficients were determined for all types of semi-prepared products that are produced from beef and pork under the technical specifications (TU 9214–345–00419779–06 and ТU 9214–456–00419779–03). For the semi-prepared products in chunks, the established coefficient of the consum- er properties varies: for beef from 1.51 to 1.27, for pork — from 1.61 tо 1.29, for lamb — from 1.81 to 1.22. Для порционных и мелкокусковых коэффициенты потребительских свойств составляют соответственно: 92 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА The coef- ficients were calculated based on the indicators of the con- sumer properties of all kinds of raw material and materials being a part of a recipe.hfi Коэффициенты потребительских свойств мясной составляющей продуктов (Кмс), подвергнутых тер- мической обработке на все виды мясных продуктов определялись на основе показателей качества основ- ных тканей: мышечной, соединительной и жировой по формуле: The coefficients of the consumer properties of the meat component of products (Cmc) subjected to thermal treat- ment for all kinds of meat products were determined based on the quality indicators of the main tissues: the muscle, connective and fatty tissues according to the equation: 1,51 × 50 + 1,37 × 20 + 0,85 × 25 + 0,68 × 5) : 100 = 1,28. 1,51 × 50 + 1,37 × 20 + 0,85 × 25 + 0,68 × 5) : 100 = 1,28. 1,51 × 50 + 1,37 × 20 + 0,85 × 25 + 0,68 × 5) : 100 = 1,28. With consideration of the sausage yield of 110%, the co- efficient for a finished product is 1.16 (1.28 : 110%). С учетом выхода колбасы — 110%, коэффициент на готовую продукцию составит 1,16 (1,28 : 110%). The coefficients for meat in carcasses, semi-carcasses, quarters by the fattiness categories, for bone-in meat by tissues, by grades, by cuts, by muscles and for all kinds of meat products (semi-prepared products, sausage products and smoked meats) are presented in «A Reference Book of the Indicators of the Consumer Properties of Meat and Meat Products», which was developed by the specialists of the Institute. The coefficients on meat of sheep and goats and the products produced from lamb are under develop- ment in 2016 and will be published. Коэффициенты на мясо в тушах, полутушах, чет- вертинах по категориям упитанности, на мясо на ко- стях по тканям, по сортам, по отрубам, по мышцам, а также на все виды мясных продуктов — полуфабри- каты, колбасные изделия и продукты из мяса пред- ставлены в «Справочнике показателей потребитель- ских свойств мяса и мясных продуктов», который разработали специалисты института. Коэффициенты на мясо мелкого рогатого скота и продукты, изготав- ливаемые из баранины, разрабатываются в 2016 году и будут изданы. (11) For cooked sausage products, the following coefficients were established: 1.28–1.09 for products of the top grades, 1.2–0.91 for the first grades, 1.04–0.94 for the second grades and 1.26–0.94 for small sausages and shpikachki (small sausages with fat).hfi В группе полукопченых колбас коэффициенты со- ставили 1,1–0,93, варено-копченых коэффициенты ко- лебались от 1,26 до 1,02, в группе сырокопченых кол- басных изделий коэффициенты — 1,26–0,86. The coefficients were 1.1–0.93 in the group of semi- smoked sausages, 1.26–1.02 in the group of cooked smoked sausages 1.26–1.02, and 1.26–0.86 in the group of the un- cooked smoked sausages. Следующая группа — это продукты из мяса, кото- рые до последнего времени назывались копчености. Для этих продуктов, которые выпускаются без шкур- ки, коэффициенты установлены от 1,51 до 0,57, выра- батываемых со шкуркой — от 1,46 до 0,53. The next group is the products from meat, which until recently were named smoked products. For this group, the following coefficients were established: 1.51 to 0.57 for the products without skin and 1.46 to 0.53 for products with skin. 93 THEORY AND PRACTICE OF MEAT PROCESSING №3 \| 2016 However, the established coefficients express the food value of meat products without taking into account the yields of finished products, which depend on the degree of thermal treatment. The coefficients should be adjusted to calculate the coefficients of consumer properties of such meat products according to a specific yield of finished products, which determine the degree of thermal treatment. For example, beef sausage is produced from beef with addition of mélange and the yield is 110%. Однако установленные коэффициенты выражают пищевую ценность мясных изделий без учета выхода готовой продукции, который зависит от степени тер- мической обработки. Для определения коэффициен- тов потребительских свойств таких мясных продук- тов в соответствии с конкретным выходом готовой продукции, который определяет степень термиче- ской обработки, коэффициенты должны корректи- роваться. Например, колбаса говяжья вырабаты- вается из говядины с добавлением меланжа, выход составляет 110%. In production of cooked beef sausage, the following ingredients are used: 50% beef of the top grade, coefficient: 1.51; 20% beef of the first grade, coefficient: 1.37; 25% fat beef, coefficient: 0.85; 5% mélange, coefficient: 0.68. На производство вареной говяжьей колбасы расхо- дуется: говядина высшего сорта 50%, коэффициент — 1,51; говядина первого сорта 20%, коэффициент — 1,37; говядина жирная 25%, коэффициент — 0,85; меланж 5%, коэффициент — 0,68. The calculation of the coefficient for cooked beef sausage: Расчет коэффициента на колбасу говяжью варе- ную: 1.51 × 50 + 1.37 × 20 + 0.85 × 25 + 0.68 × 5) : 100 = 1.28. Принадлежность к организации Небурчилова Нина Федоровна — кандидат экономических на- ук, доцент, ведущий научный сотрудник, руководитель направ- ления Экономических проблем мясной промышленности Центра «Экономико-аналитические исследования и информационные технологии, Всероссийский научно-исследовательский институт мясной промышленности имени В.М. Горбатова 109316, г. Москва, ул. Талалихина, 26. Тел.: 8–495–676–67–31 E-mail: econ@vniimp.ru Neburchilova Nina Fedorovna — candidate of economic sciences, docent, leading scientific worker, Head of the Direction of the Eco- nomic Problems in the Meat Industry of the Center of the Economic Analytical Investigations and Information Technologies, The V.M. Gorbatov All-Russian Meat Research Institute 109316, Talalikhina str. 26, Moscow, Russia Tel.: 8–495–676–67–31 E-mail: econ@vniimp.ru E-mail: econ@vniimp.ru E-mail: econ@vniimp.ru Петрунина Ирина Всеволодовна — старший научный сотруд- ник направления Экономических проблем мясной промыш- ленности Центра «Экономико-аналитические исследования и информационные технологии ФБГНУ «Всероссийский научно- исследовательский институт мясной промышленности имени В.М. Горбатова», 109316, г. Москва, ул. Талалихина, 26 Тел.: 8–495–676–67–31 E-mail: econ@vniimp.ru Критерии авторства Ответственность за работу и предоставленные сведения несут все авторы. Все авторы в равной степени участвовали в этой работе. Конфликт интересов Авторы заявляют об отсутствии конфликта интересов. Поступила 01.06.2016 Петрунина Ирина Всеволодовна — старший научный сотруд- ник направления Экономических проблем мясной промыш- ленности Центра «Экономико-аналитические исследования и информационные технологии ФБГНУ «Всероссийский научно- исследовательский институт мясной промышленности имени В.М. Горбатова», 109316, г. Москва, ул. Талалихина, 26 Тел.: 8–495–676–67–31 E-mail: econ@vniimp.ru Критерии авторства Ответственность за работу и предоставленные сведения несут все авторы. Все авторы в равной степени участвовали в этой работе. Конфликт интересов Авторы заявляют об отсутствии конфликта интересов. Поступила 01.06.2016 Petrunina Irina Vladimirovna — senior research worker, the Direc- tion of the Economic Problems in the Meat Industry of the Center of the Economic Analytical Investigations and Information Technologies of FBGNU “The V.M. Gorbatov All-Russian Meat Research Institute” Address: 109316, Talalikhina str. 26, Moscow, Russia Tel.: 8–495–676–67–31 E-mail: econ@vniimp.ru Петрунина Ирина Всеволодовна — старший научный сотруд- ник направления Экономических проблем мясной промыш- ленности Центра «Экономико-аналитические исследования и информационные технологии ФБГНУ «Всероссийский научно- исследовательский институт мясной промышленности имени В.М. Горбатова», 109316, г. Москва, ул. Талалихина, 26 Тел.: 8–495–676–67–31 E-mail: econ@vniimp.ru 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА 2016 \| №3 ТЕОРИЯ И ПРАКТИКА ПЕРЕРАБОТКИ МЯСА REFERENCES Антипова, Н.И. Дунченко. — М.: КолосС, 2007. — C. 853. 9. Файвишевский М.Л. Производство пищевых животных жиров. — М.: Антиква, 1995. — С. 6–14. 10 Лисицын А Б Price system and economic law conflicting / 9. Faivishevsky М.L. Production of food-grade animal fats. — М.: Antikva, 1995. — P. 6–14.l , у , 9. Файвишевский М.Л. Производство пищевых животных жиров. — М.: Антиква, 1995. — С. 6–14.l 10. Lisitsyn, A.B. Price system and economic law conflict- ing / A.B. Lisitsyn, N.F. Neburchilova, Е.А. Mishenina // Fleis- chwirtschaft International. — 2007. — Issue 5. — P. 52–53. р , 10. Лисицын А.Б. Price system and economic law conflicting / А.Б. Лисицын, Н.Ф. Небурчилова, Е.А. Мишенина // Fleischwirtschaft International. — 2007. — № 5. — P. 52–53. REFERENCES БИБЛИОГРАФИЧЕСКИЙ СПИСОК БИБЛИОГРАФИЧЕСКИЙ СПИСОК 1. Салимжанов И.К. Ценообразование: учебник / И.К. Са- лимжанов. — М.: КНОРУС, 2007. — С. 304. 1. Salimzhanov I.C. Price formation: textbook / I.C. Salimzha- nov. — М.: KNORUS, 2007. — P. 304. 2. Лисицын А.Б. Взаимосвязь потребительских свойств и цен на мясные продукты / А.Б. Лисицын, Н.Ф. Небурчилова, Е.А. Мишенина // Fleischwirtschaft International. Россия. — 2008. — № 1. — С. 77–78. 2. Lisitsyn А.B. Interrelation of consumer properties and prices on meat products / А.B. Lisitsyn, N.F. Neburchilova, Е.А. Mish- enina // Fleischwirtschaft International. Russia. — 2008. — Is- sue 1. — P.77–78. 3. Simmons N.J. Integrated technologies to enhance meat quality — An Australasian perspective / N.J. Simmons, C.C. Daly, C.R. Mudford, I. Richards, G. Jarvis, H. Pleiter. — Meat Science. — 2006. — Volume 74, Issue 1. — P. 172–179. 3. Simmons N.J. Integrated technologies to enhance meat quality — An Australasian perspective / N.J. Simmons, C.C. Daly, C.R. Mudford, I. Richards, G. Jarvis, H. Pleiter. — Meat Science. — 2006. — Volume 74, Issue 1. — P. 172–179. , 4. http://www.australian-meat.com/uploadedFiles/Foodser- vice/Resources/Publications/MSA-Beef-Fact-Sheet–2014.pdf. (дата обращения 04.08.2016). , 4. http://www.australian-meat.com/uploadedFiles/Foodser- vice/Resources/Publications/MSA-Beef-Fact-Sheet–2014.pdf. (дата обращения 04.08.2016). 5. http://www.mla.com.au/globalassets/mla-corporate/ blocks/marketing-beef-and-lamb/msa_beq-report14–15_web. pdf. (дата обращения 08.08.2016). 5. http://www.mla.com.au/globalassets/mla-corporate/ blocks/marketing-beef-and-lamb/msa_beq-report14–15_web. pdf. (дата обращения 08.08.2016). p (д р щ ) 6. Lisitsyn А.B. Aspects of determination of meat costs on the basis of the objective criteria / А.B. Lisitsyn, N.F. Neburchilova,, А.S. Chernova // All about meat. — 2015. Issue 4. — P. 6–9. 6. Лисицын А.Б. Аспекты определения стоимости мяса на основе объективных критериев качества / А.Б. Лисицын, Н.Ф. Небурчилова, А.С. Чернова // Все о мясе. — 2015. № 4. — С. 6–9. // 7. Chemical composition of the Russian food products / Refer- ence book. Under the editorship of the corresponding member of RAS, prof. I.M. Skurikhin and academician of RAMN V.А. Tutelyan. Reference book. -М.: DeLi print. — 2002, 236 pages. 7. Химический состав российских пищевых продуктов: Справочник / Под ред. член-корр. МАИ, проф. И.М. Скурихина и академика РАМН, проф. В.А. Тутельяна. — М.: ДеЛи принт, 2002. — C. 236. p , p g 8. Rogov I.А. Chemistry of food: textbook// I.А. Rogov, L.V. Anti- pova, N.I. Dunchenko. — М.: KolosS, 2007. — P. 853. 8. Рогов И.А. Химия пищи: учебное пособие // И.А. Рогов, Л.В. Антипова, Н.И. Дунченко. — М.: КолосС, 2007. — C. 853. 8. Рогов И.А. Химия пищи: учебное пособие // И.А. Рогов, Л.В. Выводы Установленные коэффициенты потребительских свойств позволят оперативно, в условиях постоянно меняющихся цен на мясо, определять стоимость сы- рья любого вида мясных изделий без предваритель- ных расчетов себестоимости при разделке, обвалке и жиловке. Кроме того, применяя эти коэффициенты, можно учесть и изменения в технологии производст- ва, которые влияют на выход продукции. The established coefficients of the consumer properties will allow operative assessment of raw material costs for all kinds of meat products in the conditions of constantly changing prices on meat without preliminary calculation of prime costs upon cutting, boning and trimming. More- over, using these coefficients, it is possible to account for changes in a production technology, which influence the product yield. Использование установленных показателей обес- печит менеджменту предприятия возможности опе- ративного принятия управленческих решений для определения эффективных путей развития произ- водства с целью повышения конкурентоспособности продукции и создания таких условий, при которых потребность в конкретном товаре будет удовлетво- ряться с минимальными затратами и получением по- ложительных финансовых результатов. The use of the established indicators will provide enter- prise management with an opportunity to promptly make managerial decisions for determining effective ways for production development in order to strengthen product competitiveness and create conditions, under which a de- mand for a particular good will be satisfied with minimal expenses and positive financial results. 94 Критерии авторства All authors bear responsibility for the work and presented data. All authors made an equal contribution to the work. Conflict of interest The authors declares no conflict of interest. Received 01.06.2016 Конфликт интересов Received 01.06.2016 95
https://openalex.org/W4220853200	https://zenodo.org/records/6390345/files/IRS4020.pdf	Latin	null	INFLUENCE OF EXTERNAL BARRIER STRUCTURES ON THERMAL PHYSICAL PROPERTIES.	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	2,215	INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference meter of the Earth's surface is 160 W / m, but these figures vary for different latitudes. Humidity, cloudy weather, atmospheric dust, altitude, seasons, daily temperature and more. The current issue is how much of the world's solar energy is used for human consumption. This can be achieved by using different construction methods using new energy sources when building a new building or reconstructing an existing building. In the 60s and 70s, when the first steps were taken to use non-conventional forms of energy in the CIS countries, autonomous power-supply photovoltaics appeared and justified itself in space. By the end of the 1980s, solar panels were installed to provide hot water to an area of 150,000 m2. The production of solar collectors amounted to 80 thousand m 2 per year. As a result of the economic difficulties of the 1990s, the development of non-conventional energy in our country has been halted. But today, the use of non-conventional energy sources is becoming more widespread around the world and in our country. In order to receive solar energy, the receiving surface must be on the south side, which means that the latitude of residential buildings is considered to be efficient. Passive use of solar radiation by receiving direct light from glass windows is used indirectly through walls, roofs, barriers of winter gardens. meter of the Earth's surface is 160 W / m, but these figures vary for different latitudes. Humidity, cloudy weather, atmospheric dust, altitude, seasons, daily temperature and more. The current issue is how much of the world's solar energy is used for human consumption. This can be achieved by using different construction methods using new energy sources when building a new building or reconstructing an existing building. In the 60s and 70s, when the first steps were taken to use non-conventional forms of energy in the CIS countries, autonomous power-supply photovoltaics appeared and justified itself in space. By the end of the 1980s, solar panels were installed to provide hot water to an area of 150,000 m2. The production of solar collectors amounted to 80 thousand m 2 per year. As a result of the economic difficulties of the 1990s, the development of non-conventional energy in our country has been halted. But today, the use of non-conventional energy sources is becoming more widespread around the world and in our country. Abstract: The article analyzes the current state of the problems encountered in the design of buildings. To this end, various engineered discussed concept upon which the design of roofs formed. As one of the elements of the mechanism of implementation of these plans are specific examples of the construction of energy-efficient roof. Key words: energy efficient, roof constructions, buildings From the beginning of time, humans have used sunlight. The beginnings of architecture in Central Asia, especially in Uzbekistan, date back to the III century BC. Many of the buildings and structures that were built in the IXth Century and have survived to the present day are rightly considered to be the pinnacle of the art of construction. Buildings and structures built in Samarkand, Bukhara, Khorezm, Tashkent and other cities in the XVI-XVII centuries can be considered as a high school of construction. clearly visible. 90% of primary energy is lost in mining, production, transportation, storage and consumption of energy resources. This is primarily due to the fact that the raw material undergoes many technological processes before it reaches the consumer, and the traditional energy supply becomes more expensive. Therefore, architects and builders in the development of concepts for the development of architectural design of the XXI century, the city structure and individual buildings in working projects to preserve natural resources and as much as possible new solutions, especially solar energy. llash is required. New sources include solar energy, wind energy, hydropower, and deep-seated energy. The country's heat balance accounts for 90% of non-renewable energy sources, of which 30% is oil, 40% is gas and 20% is coal. All fossil fuels are a form of solar energy that has gone through various stages, regenerated and reached us millions of years later, and is in danger of running out and becoming more expensive. The energy of the light that the sun sends to the earth is really huge, 47% of the 100% of the energy coming to the earth falls to the earth, the rest of the energy is distributed to the universe and provides the planet's heat balance. Solar energy per square 87 INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference In order to receive solar energy, the receiving surface must be on the south side, which means that the latitude of residential buildings is considered to be efficient. Passive use of solar radiation by receiving direct light from glass windows is used indirectly through walls, roofs, barriers of winter gardens. The disadvantage of solar energy is that it is unique to all alternative energy sources. For example, solar radiation activity varies from 2.2 MWh / year depending on the latitude, and the daily oscillations can be considered even higher. Today, Uzbekistan has 3,800 boiler and 1,136 thermal power plants, thousands of kilometers of communication pipelines, which emit harmful substances, combustion products and require new investments. Uzbekistan is a republic with more than 300 sunny days a year. The total capacity of solar energy is estimated at 95 billion tons of conventional fuel, 1% of which is consumed by wind turbines, which is comparable to the consumption of energy in Uzbekistan as a whole. In the design and construction of modern residential buildings and structures, first of all, the climatic characteristics of the construction site are taken into account. As for the climate of Uzbekistan, it is located in the northern hemisphere, in the central part of Central Asia. In addition to its geographical location, the climate of Uzbekistan is influenced by the altitude and topography of its territory. 88 INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference A quarter of the country's territory is mountainous, the rest is 100-200 m above sea level. The plain joins the hills from west to southeast and the hills to the mountains. The territory of the Republic of Uzbekistan is 447.4 thousand km2 and borders more than 5300 km, mainly between the Amudarya and Syrdarya rivers. The area of plains (deserts) is 75% of the territory of the republic. The deserts are located at an altitude of 300-400 m above sea level, and the climate is sharply continental. The average temperature in July is 30o-32o C, and in January -2o-3o C. Annual precipitation is around 100-300 mm. The part of the territory of the republic at an altitude of 400-1200 meters above sea level is the Adir region. The climate of the hills is more temperate than the desert. The rainfall is higher (300-450 mm) than in the desert, and the summer lasts longer. INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference floor of a building was left open to allow wind to blow. Otherwise, the heat from the building will melt the ice and cause the building to collapse. floor of a building was left open to allow wind to blow. Otherwise, the heat from the building will melt the ice and cause the building to collapse. The duration of the hot climate in Tashkent is more than 5-6 months. Therefore, natural ventilation is used in the buildings to keep the room temperature from overheating. In addition, the walls and windows of buildings should be designed to protect from solar radiation (screens - blinds), and natural ventilation measures should be taken on the roofs of solid attics. In sharply continental climates, natural ventilation of buildings using windows at night and closing windows during the day is effective in protecting them from high temperatures. In industrial buildings, where the production process is subject to certain requirements, artificial cooling, ie air conditioning, creates a temperate climate in the rooms. Currently, many one- and two-storey industrial buildings are being built in villages and towns on the basis of standard designs. The convenience of this is that if the upper part of the two-storey buildings is protected from solar radiation by the method of natural ventilation, the high temperature at the bottom will be absorbed into the ground. One of the most effective ways to protect against solar radiation is to plant shady, cool trees around the industrial building. In areas where strong winds and rainfall are observed, the outer surface of the structures is covered with a waterproof ceramic and moisture-proof layer. In areas with low rainfall, the walls of buildings are plastered with a 2-4 cm thick cement-sand mixture. It is known that the thermal conductivity and moderate humidity of external barrier structures depend on the climate and indoor environment. In order to create a temperate climate inside the rooms of the building and to choose the rational external barrier structures, their thermal- physical and humidity state engineering calculations are performed. We will look at the main climatic parameters used to design exterior barriers. Использованная литература: 1. Sagatov B. U. About transfer of effort through cracks in ferro-concrete elements //European science review. – 2016. – №. 7-8. – С. 220-221. Использованная литература: 1. Sagatov B. U. INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference The mountainous region is located at an altitude of 1000-2800 m above sea level. Summers in the mountains are short and cool, with lots of rain and long winters. Strong winds blow in many parts of Uzbekistan, especially in the plains. According to the requirements of hygienists, builders and QMQ 2.01.01-97 *, QMQ 2.01.01-94, the size-historical and design solution of buildings under construction in each climate zone and their use in these climatic conditions should be different. The design of buildings and their external barriers will focus primarily on creating a temperate climate for people to live and work. For humans to live, the average temperature should be 18-24o C. If the room temperature is below + 8 ° C, it is cold, +8 ° +15 ° is cool, +16° + 28° is warm, and above 28° C is hot. Heating buildings in winter and protecting them from solar radiation in summer depends on the climate of the region. For example, the average July temperature in Tashkent is + 28.6o C and the maximum absolute temperature is +44.5o C, the minimum absolute temperature is -31.7o C, and the maximum daily amplitude of temperature fluctuations is +24.9o C. In Termez, these values are +30.4 o C and 27.6 o C. The coldest daily temperature for the winter in Samarkand is -18 o C with an annual temperature of 0.98, in Nukus it is -27o C and in Jizzakh -22o C. However, in the future it is recommended to take these indicators as a result of practical research. In very cold climates, the design of public and residential buildings does not include balconies, loggias and balconies. Residential buildings have the same height, and light lanterns are rarely used. For centuries, in glacial areas, the first 89 INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference About transfer of effort through cracks in ferro-concrete elements //European science review. – 2016. – №. 7-8. – С. 220-221. 2. Ablayeva U., Normatova N. Energy saving issues in the design of modern social buildings //Problems of Architecture and Construction. – 2019. – Т. 2. – №. 1. – С. 59-62. elements //European science review. – 2016. – №. 7-8. – С. 220-221. 2. Ablayeva U., Normatova N. Energy saving issues in the design of modern social buildings //Problems of Architecture and Construction. – 2019. – Т. 2. – №. 1. – С. 59-62. 3. Норматова Н. А. Проектирование энергосберегающих зданий в условиях узбекистана //Academy. – 2020. – №. 11 (62). 3. Норматова Н. А. Проектирование энергосберегающих зданий в условиях узбекистана //Academy. – 2020. – №. 11 (62). 90 INNOVATIVE RESEARCH IN SCIENCE International scientific-online conference 4. Asatov N. A., Sagatov B. U., Maxmudov B. I. O. G. L. Тashqi to’siq konstruksiyalarini issiqlik fizik xususiyatlariga ta’siri //Science and Education. – 2021. – Т. 2. – №. 5. – С. 182-192. 4. Asatov N. A., Sagatov B. U., Maxmudov B. I. O. G. L. Тashqi to’siq konstruksiyalarini issiqlik fizik xususiyatlariga ta’siri //Science and Education. – 2021. – Т. 2. – №. 5. – С. 182-192. 5. Аблаева Ў. Ш., Норматова Н. А. Тошкент: лойиҳалашнинг анъанавийликдан хозирги кунигача //Science and Education. – 2021. – Т. 2. – №. 5. – С. 206-216. 5. Аблаева Ў. Ш., Норматова Н. А. Тошкент: лойиҳалашнинг анъанавийликдан хозирги кунигача //Science and Education. – 2021. – Т. 2. – №. 5. – С. 206-216. 6. Aблаева Ў. Ш., Норматова Н. А. Ўзбекистондаги мавжуд биноларнинг энергия тежамкор шамоллатиладиган тизимлари асосий системалари //Science and Education. – 2021. – Т. 2. – №. 5. – С. 193-205. 6. Aблаева Ў. Ш., Норматова Н. А. Ўзбекистондаги мавжуд биноларнинг энергия тежамкор шамоллатиладиган тизимлари асосий системалари //Science and Education. – 2021. – Т. 2. – №. 5. – С. 193-205. 7. Sagatov B. U. O’zbekistonda energiya tejamkor binolar qurilishining ahvoli //Science and Education. – 2022. – Т. 3. – №. 1. – С. 261-265. 7. Sagatov B. U. O’zbekistonda energiya tejamkor binolar qurilishining ahvoli //Science and Education. – 2022. – Т. 3. – №. 1. – С. 261-265. 91
https://openalex.org/W4387130979	https://herpetologia.fciencias.unam.mx/index.php/revista/article/download/751/484	English	null	FIRST RECORD OF ECTOMELY IN Phrynosoma orbiculare (SQUAMATA: PHRYONOSOMATIDAE): PREDATION INJURY OR CONGENITAL ANOMALY?	Revista latinoamericana de herpetología	2,023	cc-by-sa	2,060	Palabras clave.– Especie endémica, lagartija, México, anomalía morfológica, pérdida parcial de extremidad. Abstract.– We report a case of a forelimb anomaly in an adult female of Phrynosoma orbiculare found in Abstract.– We report a case of a forelimb anomaly in an adult female of Phrynosoma orbiculare found in Aguascalientes, Mexico. Keywords.– Endemic species, lizard, Mexico, morphological anomaly, partial limb loss. In lizards, limb morphology is directly related to uses of habitats and microhabitats and represents the outcome of phylogenetic conservatism and the genetic pool of species (Vitt et al., 1997; Melville & Swain, 2000; Herrel et al., 2002). Considering it is also correlated positively with locomotor performance (Melville & Swain, 2000), and this in turn with dominance (Robson & Miles, 2000), it is expected that male limb morphology might be under stronger selective pressures than female limb morphology (Herrel et al., 2002). Therefore, any kind of limb anomaly might constitute a potential cost to the overall fitness of individuals (Gkourtsouli-Antoniadou et al., 2017). However, limb anomalies are quite common in different lizard species (Gleed-Owen, 2012; Kolenda et al., 2017; Gkourtsouli-Antoniadou et al., 2017; Christopoulos & Pafilis, 2020; Mora et al., 2020) and are regarded as having genetic and environmental origins (Rothschild et al., 2012; Kolenda et al., 2017; Christopoulos & Pafilis, 2020). Here we report the first case of a forelimb anomaly in an individual of Mountain Horned Lizard Phrynosoma orbiculare. Durango, Tamaulipas, Nuevo León, Zacatecas, Aguascalientes, and San Luis Potosi), center (Guanajuato, Querétaro, Hidalgo, Morelos, Tlaxcala, Ciudad de México, Estado de México, and Puebla), west (Jalisco and Michoacán) and south of the country (Veracruz, Sherbrooke, 2003; Moreno-Barajas et al., 2013; Ramírez-Bautista et al., 2014; Payan-Cazares et al., 2023). In the state of Aguascalientes, P. orbiculare is found in the municipalities of Calvillo, San José de Gracia, Jésus María, Tepezalá, and Aguascalientes (Vázquez-Díaz & Quintero-Díaz, 2005). Durango, Tamaulipas, Nuevo León, Zacatecas, Aguascalientes, and San Luis Potosi), center (Guanajuato, Querétaro, Hidalgo, Morelos, Tlaxcala, Ciudad de México, Estado de México, and Puebla), west (Jalisco and Michoacán) and south of the country (Veracruz, Sherbrooke, 2003; Moreno-Barajas et al., 2013; Ramírez-Bautista et al., 2014; Payan-Cazares et al., 2023). In the state of Aguascalientes, P. orbiculare is found in the municipalities of Calvillo, San José de Gracia, Jésus María, Tepezalá, and Aguascalientes (Vázquez-Díaz & Quintero-Díaz, 2005). On August 20th, 1982 at 14:45 h, Zeferino Uribe-Peña collected alive an adult female of P. NOTA CIENTÍFICA Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare —e751— 172-175 NOTA CIENTÍFICA Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare —e751— 172-175 NOTA CIENTÍFICA Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare —e751— 172-175 FIRST RECORD OF ECTOMELY IN PHRYNOSOMA ORBICULARE (SQUAMATA: PHRYNOSOMATIDAE) PRIMER REGISTRO DE ECTOMELIA EN PHRYNOSOMA ORBICULARE (SQUAMATA: PHRYNOSOMATIDAE) César A. Díaz-Marín1, Tonantzin Carmona-Zamora1, Cinthya Mendoza-Almeralla1 & Aurelio Ramírez-Bautista1 1Laboratorio de Ecología de Poblaciones, Centro de Investigaciones Biológicas, Instituto de Ciencias Básicas e Ingeniería, Universidad Autónoma del Estado de Hidalgo, km. 4.5 carretera Pachuca-Tulancingo, 42184, Mineral de La Reforma, Hidalgo, México Corresponding author e-mail: cesaardm@hotmail com Received: 2023-07-10. Accepted: 2023-08-26. Published: 2023-09-05. Editor: Ernesto Raya-García, México. Resumen.– Documentamos un caso de una anomalía en una extremidad anterior de una hembra adulta de Phrynosoma orbiculare encontrada en Aguascalientes, México. Palabras clave.– Especie endémica, lagartija, México, anomalía morfológica, pérdida parcial de extrem Palabras clave.– Especie endémica, lagartija, México, anomalía morfológica, pérdida parcial de extremidad. Detalles de la extremidad anterior en vista lateral (C) y ventral (D). Note el color y la forma distintivos de las escamas regeneradas en la base de la extremidad anterior y la cicatrización en el brazo (flechas rojas). Figura 1. Vista dorsal (A) y ventral (B) de hembra de Phrynosoma orbiculare con anomalía en la extremidad anterior del municipio de San José Gracia en el estado de Aguascalientes en México. Detalles de la extremidad anterior en vista lateral (C) y ventral (D). Note el color y la forma distintivos de las escamas regeneradas en la base de la extremidad anterior y la cicatrización en el brazo (flechas rojas). Figure 1. Dorsal (A) and ventral view (B) of the female of Phrynosoma orbiculare with forelimb anomaly from San José de Gracia municipality in the state of Aguascalientes in Mexico. Details of the forelimb in lateral (C) and ventral view (D). Note the distinct color and shape of regenerated scales at the base of the forelimb and scarring on the arm (red arrows). Figure 1. Dorsal (A) and ventral view (B) of the female of Phrynosoma orbiculare with forelimb anomaly from San José de Gracia municipality in the state of Aguascalientes in Mexico. Details of the forelimb in lateral (C) and ventral view (D). Note the distinct color and shape of regenerated scales at the base of the forelimb and scarring on the arm (red arrows). REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 Palabras clave.– Especie endémica, lagartija, México, anomalía morfológica, pérdida parcial de extremidad. orbiculare (77.63 mm snout-vent length) at Barranca Los Pilares, Sierra de San Blas de Pabellón, San José de Gracia, Aguascalientes (22.214219° N, 102.483921° W, 2,420 m a.s.l.). The lizard was found on the leaf litter of an oak-juniper forest. It was humanely euthanized, fixed in 10% formalin, preserved in 70% ethanol, and stored by ARB at Laboratorio de Ecología de Poblaciones, Centro de Investigaciones Biológicas (CIB), Universidad Autónoma del Estado de Hidalgo (UAEH). Subsequently, we deposited this specimen in the herpetological collection of the CIB, UAEH, under the voucher number CIB- 6469. This adult female of P. orbiculare had an incomplete left forelimb, from which the arm was complete and scarred, the forearm was stunted, and the hand and toes were lacking (Figure 1A, B). The scalation was normal, except at the base, where it had two tough and large scales of different color (Figure 1C, D). The The Mountain Horned Lizard Phrynosoma orbiculare is an endemic species of Mexico, inhabiting an elevation range between 1,371-3,450 m a.s.l. across the Sierra Madre Occidental, Sierra Madre Oriental, Transmexican Volcanic Belt, and Chihuahuan Desert (Sherbrooke, 2003; Ramírez-Bautista et al., 2014). This species occupies xeric scrublands and pine-oak forests in states of the north (Sonora, Sinaloa, Chihuahua, REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 - 172 - Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare TINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 Díaz-Marín et al.- Ectomely in Phrynosoma or ista dorsal (A) y ventral (B) de hembra de Phrynosoma orbiculare con anomalía en la extremidad anterior del municipio de San José Gracia en el estado de Aguascalie alles de la extremidad anterior en vista lateral (C) y ventral (D). Note el color y la forma distintivos de las escamas regeneradas en la base de la extremidad anterior y n el brazo (flechas rojas). orsal (A) and ventral view (B) of the female of Phrynosoma orbiculare with forelimb anomaly from San José de Gracia municipality in the state of Aguascalientes in M orelimb in lateral (C) and ventral view (D). Note the distinct color and shape of regenerated scales at the base of the forelimb and scarring on the arm (red arrows). Figura 1. Vista dorsal (A) y ventral (B) de hembra de Phrynosoma orbiculare con anomalía en la extremidad anterior del municipio de San José Gracia en el estado de Aguascalientes en México. Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare Gkourtsouli-Antoniadou, I., A. Deimezis-Tsikoutas, K. Vassaki, A. Vezyrakis & P. Pafilis. 2017. A tail where it shouldn’t be: a morphological anomaly in Podarcis erhardii. Herpetology Notes 10:233-234. change in coloration probably occurred either because of the fixation process, scarring, or both. Given that the lower portion was missing, this forelimb anomaly could be classified as an ectomely, according to Rothschild et al. (2012). Herrel, A., J.J. Meyers & B. Vanhooydonck. 2002. Relationships between microhabitat use and limb shape in phrynosomatid lizards. Biological Journal of the Linnean Society 77:149-163. Previous studies in different lizard species have considered two potential causes for limb anomalies: injuries from predation attempts (Mora et al., 2020) or malformation during embryo development (Raynaud, 1990). Predators of Phrynosoma lizards are quite diverse, including mammals such as canids (e.g., Vulpes spp. and Canis latrans) and rodents (e.g., Onychomys torridus); birds such as falcons (Falco spp.) and hawks (e.g., Buteo lineatus); and snakes (e.g., Crotalus atrox and Masticophis spp.; Sherbrooke, 2003, 2022). Snakes generally attack and swallow their prey (including Phrynosoma lizards) from the head, while mammals and birds might grab them from any part of the body, including limbs (Sherbrooke, 2003). Given the available evidence of the P. orbiculare reported here, this limb anomaly probably would be related to a vertebrate predation attempt (e.g., mammal or bird), but we cannot rule out the possibility that it could result from a malformation of embryo development. Kolenda, K., M. Wieczorek, A. Najbar, B. Najbar & T. Skawiński. 2017. Limb malformation and tail bifurcation in sand lizards (Lacerta agilis) and common lizards (Zootoca vivipara) from Poland. Herpetology Notes 10:713-716. Melville, J. & R. Swain. 2000. Evolutionary relationships between morphology, performance and habitat openness in the lizard genus Niveoscincus (Scincidae: Lygosominae). Biological Journal of the Linnean Society 70:667-683. Mora, J.M., L.I. López & N. Campos-Loría. 2020. An adult male Basiliscus plumifrons Cope, 1875 without an arm survives in the tropical rain forest of Costa Rica. Revista Latinoamericana de Herpetología 3:123-125. A previous record in P. cornutum has shown that horned lizards can survive predator attacks with visible and healed wounds (Sherbrooke, 2022); however, to our knowledge, there are no records of Phrynosoma lizards that had survived a predation attempt with partial limb loss. Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare Records of other lizard species suggest that individuals with ectomely do not decrease their mobility and flee capacity (Gleed-Owen, 2012; Kolenda et al., 2017; Christopoulos & Pafilis 2020; Mora et al., 2020), but they probably would not be able to climb. Taking all into account, this is the first case of ectomely in a Phrynosoma lizard and further studies could help to elucidate if this kind of anomaly is environmentally or genetically determined. Moreno-Barajas, R., F. Rodríguez-Romero, A. Velázquez-Rodríguez & A. Aragón-Martínez. 2013. Variación geográfica en Phrynosoma orbiculare (Sauria: Phrynosomatidae): análisis de las subespecies. Acta Zoológica Mexicana 29:129-143. Payan-Cazares, E., E. A. Gamez-Duarte, J. D. Jacobo- González & A. Castro-Bastidas. 2023. Phrynosoma orbiculare (Squamata: Phrynosomatidae). Revista Latinoamericana de Herpetología 6:162-163. Ramírez-Bautista, A., U. Hernández-Salinas, R. Cruz-Elizalde, C. Berriozabal-Islas, D. Lara-Tufiño, I. Goyenechea Mayer- Goyenechea, & J.M. Castillo-Cerón. 2014. Los Anfibios y Reptiles de Hidalgo, México: Diversidad, Biogeografía, y Conservación. Sociedad Herpetológica Mexicana, A. C. Ciudad de México, México. Acknowledgements.– We thank Zeferino Uribe-Peña for donating the specimen of Phrynosoma orbiculare to Laboratorio de Ecología de Poblaciones, Centro de Investigaciones Biológicas, Universidad Autónoma del Estado de Hidalgo. REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 - 173 - CITED LITERATURE Raynaud, A. 1990. Developmental mechanism involved in the embryonic reduction of limbs in reptiles. International Journal of Developmental Biology 34:233-243. Christopoulos, A. & P. Pafilis. 2020. Hindlimb malformation in a widely distributed skink, Chalcides ocellatus. Herpetology Notes 13:16-17. Christopoulos, A. & P. Pafilis. 2020. Hindlimb malformation in a widely distributed skink, Chalcides ocellatus. Herpetology Notes 13:16-17. Robson, M.A. & D.B. Miles. 2000. Locomotor performance and dominance in male tree lizards, Urosaurus ornatus. Functional Ecology 14:338-344. Gleed-Owen, C. 2012. Zootoca vivipara (common or viviparous lizard): injury or pathology? Herpetological Bulletin 120:35-36. Gleed-Owen, C. 2012. Zootoca vivipara (common or viviparous lizard): injury or pathology? Herpetological Bulletin 120:35-36. REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 - 174 - Díaz-Marín et al.- Ectomely in Phrynosoma orbiculare Rothschild, B.M., H.P. Schultze & R. Pellegrini. 2012. Herpetological osteopathology: annotated bibliography of amphibians and reptiles. Springer Science and Business Media, New York, USA. Vázquez-Díaz, J. & G.E. Quintero-Díaz. 2005. Anfibios y reptiles de Aguascalientes. Comisión Nacional para el Conocimiento y Uso de la Biodiversidad, Ciudad de México, México. Sherbrooke, W.C. 2003. Introduction to Horned Lizards of North America. University of California Press, Berkeley, California, USA. Vitt, L.J., J P. Caldwell, P.A. Zani & T.A. Titus. 1997. The role of habitat shift in the evolution of lizard morphology: evidence from tropical Tropidurus. Proceedings of the National Academy of Sciences 94:3828-3832. Sherbrooke, W.C. 2022. Phrynosoma cornutum (Texas Horned Lizard). Predator attack survival. Herpetological Review 53: 135-146. - 175 - REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 REVISTA LATINOAMERICANA DE HERPETOLOGÍA Vol.06 No.03 / Julio-Septiembre 2023 - 175 -
https://openalex.org/W4212862632	https://bmcvetres.biomedcentral.com/track/pdf/10.1186/s12917-022-03165-z	English	null	Antemortem diagnosis of anomalous origin of the left coronary artery from the pulmonary artery in a dog	BMC veterinary research	2,022	cc-by	4,459	© The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom- mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Takamura et al. BMC Veterinary Research (2022) 18:74 https://doi.org/10.1186/s12917-022-03165-z Open Access Abstract Background: In both humans and animals, anomalous origin of the left coronary artery from the pulmonary artery (ALCAPA) is a rare congenital coronary artery anomaly. In veterinary medicine, ALCAPA is reported to be discovered only during autopsy or necropsy, and diagnostic methods and prognosis remain poorly understood in dogs. Case presentation: A 6-month-old Kaninchen Dachshund was diagnosed with functional mitral valve regurgitation and ALCAPA. Echocardiography identified anomalous vessels in the left ventricular wall and abnormal origin of the left coronary artery from the pulmonary artery. Further evaluation with coronary computed tomographic angiography demonstrated the left coronary artery arising from the posterior aspect of the main pulmonary artery together with the characteristic findings of ALCAPA. The right coronary artery was found to be dilated and tortuous. Furthermore, dilated coronary collateral arteries within the ventricular septum and along the epicardial surface were observed. The dog underwent surgery, but the origin of the anomalous artery could not be ligated, and it died from pulmonary edema 5 months after surgery. Conclusion: Anomalous origin of the left coronary artery from the pulmonary artery is overlooked in clinical practice due to its rarity. Coronary computed tomographic angiography was useful to definitively diagnose ALCAPA in a low- invasive manner. Antemortem diagnosis of ALCAPA was shown to be possible in dogs for the first time, and presence of unexplained mitral valve regurgitation should raise concern to this anomaly. Keywords: Canine, Congenital heart disease, Coronary artery anomaly stage of life if left untreated [3, 4]. Until the neonatal phase, ALCAPA does not interfere with the hemodynam- ics of the coronary arteries, and no difference in blood pressure occurs between the systemic and pulmonary arterial circulation through the presence of ductus arte- riosus. When the ductus arteriosus closes after birth, pulmonary arterial circulation pressure decreases physi- ologically, consequently lowering the blood flow rate through the LCA [1, 3]. Antemortem diagnosis of anomalous origin of the left coronary artery from the pulmonary artery in a dog Kazuki Takamura1, Ayaka Chen1, Shin Ono2 and Masami Uechi1* Correspondence: uechi.masami@jasmine-vet.co.jp 1 JASMINE Veterinary Cardiovascular Medical Center, 1‑8‑37 Nakagawa, Tsuzuki, Yokohama, Kanagawa 224‑0001, Japan Full list of author information is available at the end of the article Background Anomalous left coronary artery from the pulmonary artery (ALCAPA) is a rare congenital heart disease in humans that usually presents as an isolated heart defect [1–3]. As its name suggests, the left coronary artery (LCA) arising from the main pulmonary artery (PA) is the diagnostic hallmark of the disease. Of the two types of ALCAPA, infantile or adult, the adult variant is much rarer, as this malformation induces death in the initial In veterinary medicine, ALCAPA is also a rare disor- der discovered during autopsy or necropsy [5–8], and valid diagnostic methods or prognosis remain unclear in dogs. This report demonstrates the validity of coronary Correspondence: uechi.masami@jasmine-vet.co.jp 1 JASMINE Veterinary Cardiovascular Medical Center, 1‑8‑37 Nakagawa, Tsuzuki, Yokohama, Kanagawa 224‑0001, Japan Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom- mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Takamura et al. BMC Veterinary Research (2022) 18:74 Page 2 of 6 Fig. 1 A Right lateral and B dorsoventral thoracic radiograph demonstrating severe cardiomegaly. The pulmonary parenchyma appears normal Fig. 1 A Right lateral and B dorsoventral thoracic radiograph demonstrating severe cardiomegaly. The pulmonary parenchyma appears normal ig. 1 A Right lateral and B dorsoventral thoracic radiograph demonstrating severe cardiomegaly. The pulmonary parenchym and B dorsoventral thoracic radiograph demonstrating severe cardiomegaly. The pulmonary parenchyma appears normal computed tomography angiography (CCTA) aided with echocardiography in diagnosing ALCAPA in a small breed dog. This will support clinicians in diagnosing the disease while the dogs are alive and in making proposals for treatments. evaluation were conducted. Abnormalities detected included elevated NT-proBNP (4887 pmol/L; reference range ≤ 900 pmol/L) and ANP (259.9 pg/mL; reference range 8.6–105.8 pg/mL). Cardiac troponin I was normal (0.016 ng/mL; reference range 0.006–0.129 pg/mL). Systemic blood pressure (oscillometry) was 118/50/77 mmHg (systolic/diastolic/mean). A six-lead electrocardiography performed in the right lateral recumbency demonstrated sinus rhythm with one crite- rion for left ventricular enlargement (R wave amplitude of 3.1 mV in lead II). However, occasional premature ven- tricular contraction was observed on monitor electrocar- diography during transthoracic echocardiography (TTE). The right lateral and dorsoventral thoracic radiographs (Fig. 1A and B) revealed severe cardiomegaly (vertebral heart size, 11.6 v) [9] and left atrial (LA) enlargement (vertebral left atrium size, 2.8 v) [10], with no evidence of pulmonary edema.h 1 Vetmedin® chewable 1.25 mg, Boehringer Ingelheim Animal Health Japan Co., Ltd., Tokyo, Japan 2 Apinac® 25 mg, DS Pharma Animal Health Co., Ltd., Osaka, Japan 3 Luprac® 4 mg, Mitsubishi Tanabe Pharma Corporation,Osaka, Japan 4 Nitrol® 5 mg, Eisai Co., Ltd., Tokyo, Japan Case presentation A 6-month-old female Kaninchen Dachshund with a tentative diagnosis of mitral valve dysplasia was referred to our center for further evaluation. The dog was smaller than its littermates and had initially presented to a veterinarian at 4 months of age due to a history of tachypnea. A cardiac murmur was auscultated, and echocardiography revealed severe mitral valve regurgi- tation (MR) with left ventricular and atrial enlargement. Pimobendan1 (0.58 mg/kg PO q12h), alacepril2 (3.9 mg/ kg PO q12h), torasemide3 (0.1 mg/kg PO q12h), and isosorbide dinitrate4 (2.0 mg/kg PO q12h) were pre- scribed as treatment for MR and pulmonary edema. The patient was referred to the university animal hospital for further evaluation, where mitral dysplasia with abnormal coronary vessels was diagnosed. The transthoracic echocardiogram revealed severe LA dilation (2D short axis left atrial to aortic ratio 2.12) [11]. The left ventricle (LV) was enlarged (M-mode left ven- tricular internal dimension in end-diastole normalized to bodyweight 2.33) [11] with mildly reduced systolic function (M-mode left ventricular internal dimension in end-systole normalized to bodyweight 1.55 [12] and fractional shortening 31.5% [12]). However, no apparent abnormalities were observed in LV wall movement, sub- jectively. Color Doppler analysis was obtained through a right parasternal long-axis four-chamber view, and the mitral valve showed a moderate regurgitation signal (Fig. 2A). No structural abnormalities were found in the mitral valve complex, which consists of the mitral valve, mitral valve annulus, and papillary muscles; however, the leaflet coaptation area was decreased. The blood flow of multiple aberrant vessels was observed within the inter- ventricular septum in the right parasternal long-axis On presentation to JASMINE, the patient was lean with a body condition score of 4/9 and weighed 3.36 kg. A grade 3/6 left apical systolic murmur was auscultated. The femoral pulses were normal and synchronous, and lung sounds were normal in all areas. The patient was alert with a heart rate of 140 bpm, a respiratory rate of 48 breaths/min, and a temperature of 38.9 °C. Complete blood count, plasma biochemistry, and cardiac biomarker Takamura et al. BMC Veterinary Research (2022) 18:74 Page 3 of 6 Fig. 2 A Color Doppler image of right parasternal long-axis view demonstrating moderate mitral valve regurgitation and color flow through coronary collaterals in the interventricular septum (white arrow). B Right parasternal short-axis view demonstrating the anomalous origin of the left coronary artery from the pulmonary artery (yellow arrow). Case presentation C Color Doppler image of right parasternal short-axis view at the mitral valve level, demonstrating the retrograde flow in the LCA to the PA (white arrowhead). D Continuous wave Doppler flow tracing demonstrating continuous, mostly systolic retrograde flow, in the dilated coronary artery. LA: left atrium; LV: left ventricle; PA: pulmonary artery; RVOT: right ventricular outflow tract Fig. 2 A Color Doppler image of right parasternal long-axis view demonstrating moderate mitral valve regurgitation and color flow through coronary collaterals in the interventricular septum (white arrow). B Right parasternal short-axis view demonstrating the anomalous origin of the left coronary artery from the pulmonary artery (yellow arrow). C Color Doppler image of right parasternal short-axis view at the mitral valve level, demonstrating the retrograde flow in the LCA to the PA (white arrowhead). D Continuous wave Doppler flow tracing demonstrating continuous, mostly systolic retrograde flow, in the dilated coronary artery. LA: left atrium; LV: left ventricle; PA: pulmonary artery; RVOT: right ventricular outflow tract aorta. Images were acquired at 10 and 90% of the R–R interval. Acquisition parameters of the scanning were as follows; spiral pitch factor: 0.175, tube voltage: 120 kVp, reference tube current: 250 mV, slice thickness: 0.5 mm, and a low frequency reconstruction algorithm. Contrast medium of 600 mg iodine/kg (2 mL/kg) was delivered via a power injector system6 into the cephalic vein at a rate of 0.7 mL/s to obtain an enhanced levophase. The examina- tion revealed LCA and RCA originating from the PA and aorta, respectively (Fig. 3A, B and C, 4B and C). The dila- tion of RCA, LCA, and their branches; the left circumflex artery; and the left anterior descending artery (Fig. 4A, B and C) were also noted. The anomalous origin of the main LCA from the PA, and prominence of the right cor- onary circulation with an extensive network of collaterals four-chamber view (Fig. 2A). Large aberrant vessels were most prominent within the LV wall and were best visual- ized with color Doppler (Fig. 2C). The continuous wave Doppler analysis of the aberrant vessels revealed continu- ous systodiastolic flow with a maximal velocity diastole (velocity, 1.20 m/s; pressure gradient, 5.76 mmHg) and a minimal velocity systole (velocity, 0.86 m/s; pressure gradient, 2.94 mmHg) (Fig. 2D). The left coronary artery (LCA) optimum was not detected in the left coronary sinus, and the trunk of the coronary artery was directly connected to the PA (Fig. 5 Aquilion PRIME, Canon Medical Systems Corporation,Tochigi, Japan 6 Oypalomin® 300 injection, Fuji Pharma Co., Ltd.,Tokyo, Japan Case presentation 2B). After these examinations, torasemide3 dosage was increased from 0.1 mg/kg PO q12h to 0.2 mg/kg PO q12h in order to reduce the volume overload. An ECG-gated CCTA of the thorax was performed under general anesthesia using a multidetector 80-row computed tomography scanner5 in the helical mode to obtain a detailed anatomic image of the abnormal vessels. The scanner was positioned to encompass the ascending Takamura et al. BMC Veterinary Research (2022) 18:74 Page 4 of 6 Fig. 3 A Axial and B sagittal coronary computed tomographic angiographic images demonstrating the origin of the left coronary artery (arrow) from the pulmonary artery. C Axial coronary computed tomographic angiographic images demonstrating the origin of the right coronary artery (arrowhead) from the aorta. The entire right coronary artery appears enlarged. Ao: aorta; LAD: left anterior descending artery; LCA: left coronary artery; LCX: left circumflex artery; PA: pulmonary artery; RCA: right coronary artery Fig. 3 A Axial and B sagittal coronary computed tomographic angiographic images demonstrating the origin of the left coronary artery (arrow) from the pulmonary artery. C Axial coronary computed tomographic angiographic images demonstrating the origin of the right coronary artery (arrowhead) from the aorta. The entire right coronary artery appears enlarged. Ao: aorta; LAD: left anterior descending artery; LCA: left coronary artery; LCX: left circumflex artery; PA: pulmonary artery; RCA: right coronary artery Fig. 4 Three-dimensional volume-rendered computed tomography angiography images. A, B The left coronary artery anomaly originates from the pulmonary artery. C The right coronary artery is tortuous and dilated. D Prominent collateral vessels between RCA, LCX, and LAD. Ao: aorta; LAA: left atrial appendage; LAD: left anterior descending artery; LCX: left circumflex artery; PA: main pulmonary artery; RCA: right coronary artery Fig. 4 Three-dimensional volume-rendered computed tomography angiography images. A, B The left coronary artery anomaly originates from the pulmonary artery. C The right coronary artery is tortuous and dilated. D Prominent collateral vessels between RCA, LCX, and LAD. Ao: aorta; LAA: left atrial appendage; LAD: left anterior descending artery; LCX: left circumflex artery; PA: main pulmonary artery; RCA: right coronary artery not be confirmed and ligated, retrograde blood flow via LCA was recognized in the PA on post-operative echocar- diography. Tachypnea resolved, but the same cardiac med- ications as prior to surgery was needed throughout the period after discharge since cardiomegaly did not improve. Discussion and conclusionsh This paper presents a very rare finding of MR which the principal cause lies in the coronary artery anomaly in a dog. To the author’s knowledge, this is the first report of an antemortem diagnosis of ALCAPA in a dog. Echocardiography enables visualization of the abnor- mal origin of the LCA in humans [3, 13]. Dilated RCA, retrograde blood flow in the anomalous coronary artery, diastolic flow in the PA, and abnormal septal or epicar- dial color flow signals from the collateral vessels are com- mon findings. Left ventricular dilation, left ventricular or global hypokinesis, and mitral insufficiency can also be identified. The adult variant of ALCAPA is characterized by compensatory collaterals between the RCA and LCA [4, 14, 15]. The echocardiographic findings in the pre- sented case were similar to those of the human adult var- iant of ALCAPA and the collateral vessels explained the dog’s survival until 6 months of age. However, the RCA could not be visualized and none of the findings strongly support left ventricular hypokinesis. Of note, increased fractional shortening is generally observed under mod- erate to severe MR and volume overload, but as it was not observed in this case, a decrease in contractility was suggested. To conclude, this study reports the first canine case of antemortem ALCAPA diagnosis, which can help veteri- narians to arrive at this diagnosis in living dogs. CCTA is a low-invasive and common method to visualize coronary artery anomaly in humans and has proven to be applicable in small breed dogs. In young dogs show- ing cardiomegaly with mitral insufficiency, ALCAPA should be considered as a differential diagnosis, and a combination of TTE and CCTA is recommended for definitive antemortem diagnosis. For diagnosing coronary artery anomalies in human medicine, the performance of CCTA has proven to be highly viable [4, 15], for example, allowing visualization of the LCA arising from the main PA in ALCAPA. In the presented case, dilation of both the LCA and RCA with the presence of prominent collateral vessels was observed by CCTA, comparable with that in the human adult variant of ALCAPA. The origin of the LCA was located at the left inferolateral aspect of the main PA close to the pulmonary valve, which is representative of the human adult variant. Case presentation On 3 months follow-up, continuous systodiastolic flow was still observed in the aberrant vessels with no promi- nent changes in velocity compared to the preoperative echocardiographic values. The dog died from pulmonary edema 5 months after the surgery. Necropsy could not be performed due to lack of consent from the owner. that communicate and supply the LCA territory were found (Fig. 4D). At the owner’s request, surgical interventions were per- formed 78 days after the first visit. The initial surgical plan was to use the internal thoracic artery as a bypass graft, but it was not sufficiently long to reach the left anterior descending artery (LAD). An alternative plan was applied using the left carotid artery to supply arterial blood to the LAD; the left carotid artery was dissected from the neck to the thoracic cavity and anastomosed to LAD under cardio- pulmonary bypass. As the origin of LCA at the PA could Takamura et al. BMC Veterinary Research (2022) 18:74 Page 5 of 6 Page 5 of 6 Takamura et al. BMC Veterinary Research (2022) 18:74 (or coronary arterial translocation), Takeuchi repair, and use of a bypass graft with anomalous artery liga- tion have been established. In the case presented, since the distance between the anomalous coronary artery and aortic root was long, a coronary artery bypass graft of the left carotid artery and LAD was performed with cardiopulmonary bypass. However, the proximal anomalous artery was not ligated. Consequently, it is possible that the steal phenomenon was accelerated, with increased pulmonary blood flow, causing a further volume overload on the left heart. Therefore, in simi- lar cases, ligation of the proximal anomalous artery is strongly recommended. Persistent cardiac enlargement observed during the period after surgery was explaina- ble since the ligation was not achieved in the presented case. However, even if the dog had not taken surgery, it would eventually have faced the risk of death from cardiac failure as described in one dog with ALCAPA [5]. Improvement in tachypnea after surgery is likely the result of arterial blood supply to the LAD and there remains a possibility that the dog would have survived if ligation of the anomalous artery had succeeded. Funding Not applicable. Acknowledgements The authors would like to thank the Editage (www.Editage.com) for English language assistance. Abbreviations ALCAPA: Anomalous origin of the left coronary artery from the pulmonary artery; CCTA: Coronary computed tomography angiography; LA: Left atrial; LAD: Left anterior descending artery; LCA: Left coronary artery; LV: Left ventri- cle; MR: Mitral valve regurgitation; PA: Pulmonary artery; RCA: Right coronary artery; TTE: Transthoracic echocardiogram. Authors’ contributions All authors read and approved the final manuscript. KT was responsible for the clinical and surgical management of the patient, conceptualized the manuscript, collected the images, wrote and reviewed the manuscript. AC collected CT images. SO collected CT images and reviewed the manuscript. MU was responsible for the surgical management of the patient and reviewed the manuscript. Discussion and conclusionsh Retrograde blood flow from the LCA to the PA was clearly observed that resembled a “steal phenomenon” described as the stealing of blood supply from the left coronary circulation provided by collateral vessels from the dilated right coronary sys- tem. The degree of collateral formation relates to the degree of successive myocardial ischemia [15], so it was hypothesized that the collateral vessels maintained oxy- genated blood flow to the LCA territory for the dog in this case. The size of the dog in this case was smaller compared to recent reports using CCTA for visualiza- tion of coronary arteries [16, 17], but it was possible to clearly visualize the coronary artery. This case report supports the utility of CCTA as a diagnostic method for ALCAPA in dogs, as it depicts collaterals that cannot be clearly recognized by echocardiography alone. Competing interests The author declares that they have no competing interests. Declarations 17. Laborda-Vidal P, Pedro B, Baker M, Gelzer AR, Dukes-McEwan J, Maddox TW. Use of ECG-gated computed tomography, echocardiography and selective angiography in five dogs with pulmonic stenosis and one dog with pulmonic stenosis and aberrant coronary arteries. J Vet Cardiol. 2016;18:418–26. Received: 25 August 2021 Accepted: 27 January 2022 Received: 25 August 2021 Accepted: 27 January 2022 Received: 25 August 2021 Accepted: 27 January 2022 Availability of data and materials As ALCAPA is an anatomical anomaly, surgery is required for definitive treatment [1–3, 15, 18, 19]. Operative strategies such as coronary button transfer Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Takamura et al. BMC Veterinary Research (2022) 18:74 Page 6 of 6 Ethics approval and consent to participate The study was approved by the Ethical Committee of the JASMINE Veterinary Cardiovascular Medical Center. The dog’s owner supplied written consent for understanding and approval of this study. 18. Agarwal PP, Dennie C, Pena E, Nguyen E, Labounty T, Yang B, et al. Anomalous coronary arteries that need intervention: review of pre- and postoperative imaging appearances. Radiographics. 2017;37:740–57. 18. Agarwal PP, Dennie C, Pena E, Nguyen E, Labounty T, Yang B, et al. Anomalous coronary arteries that need intervention: review of pre- and postoperative imaging appearances. Radiographics. 2017;37:740–57. Consent for publication 19. Mongé MC, Eltayeb O, Costello JM, Sarwark AE, Carr MR, Backer CL. Aortic implantation of anomalous origin of the left coronary artery from the pul- monary artery: long-term outcomes. Ann Thorac Surg. 2015;100:154–61. 19. Mongé MC, Eltayeb O, Costello JM, Sarwark AE, Carr MR, Backer CL. Aortic implantation of anomalous origin of the left coronary artery from the pul- monary artery: long-term outcomes. Ann Thorac Surg. 2015;100:154–61. The dog’s owner supplied a written consent for publication. Author details 1 Springer Nature remains neutral with regard to jurisdictional claims in pub- lished maps and institutional affiliations. 1 JASMINE Veterinary Cardiovascular Medical Center, 1‑8‑37 Nakagawa, Tsuzuki, Yokohama, Kanagawa 224‑0001, Japan. 2 SKYVETS, Inc., 2958‑4 Yanokuchi, Inagi, Tokyo 206‑0812, Japan. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: References 1. Cowles RA, Berdon WE. Bland-white-Garland syndrome of anomalous left coronary artery arising from the pulmonary artery (ALCAPA): a historical review. Pediatr Radiol. 2007;37:890–5. 2. Cherian KM, Bharati S, Rao SG. Surgical correction of anomalous origin of the left coronary artery from the pulmonary artery. J Card Surg. 1994;9:386–91. 3. Dodge-Khatami A, Mavroudis C, Backer CL. Anomalous origin of the left coronary artery from the pulmonary artery: collective review of surgical therapy. Ann Thorac Surg. 2002;74:946–55. 3. Dodge-Khatami A, Mavroudis C, Backer CL. Anomalous origin of the left coronary artery from the pulmonary artery: collective review of surgical therapy. Ann Thorac Surg. 2002;74:946–55. 4. Heermann P, Heindel W, Schülke C. Coronary artery anomalies: diagnosis and classification based on cardiac CT and MRI (CMR) – from ALCAPA to anomalies of termination. RoFo. 2017;189:29–38. 5. Turk MA, Turk JR, Hopkins MG, Wagner JA. Unexpected death in an adult dog with anomalous origin of the left coronary artery from the pulmo- nary trunk. Cornell Vet. 1984;74:344–8. 6. Sandusky GE, Smith CW. Anomalous left coronary artery in a calf. J Am Vet Med Assoc. 1978;173:475–7. 7. Tanaka A, Yasuda M, Murakami T. Aberrant origin of coronary arteries from pulmonary trunk in cattle. Adv Anim Cardiol. 2009;42:1–7 (in Japanese). 8. Shank AM, Bryant UK, Jackson CB, Williams NM, Janes JG. Anomalous origin of the left coronary artery from the pulmonary artery (ALCAPA) in four calves. Vet Pathol. 2008;45:634–9. 9. Buchanan J, Bucheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc. 1995;206:194–9. 9. Buchanan J, Bucheler J. Vertebral scale system to measure canine heart size in radiographs. J Am Vet Med Assoc. 1995;206:194–9. 10. Malcolm EL, Visser LC, Phillips KL, Johnson LR. Diagnostic value of verte- bral left atrial size as determined from thoracic radiographs for assess- ment of left atrial size in dogs with myxomatous mitral valve disease. J Am Vet Med Assoc. 2018;253:1038–45. 11. Keene BW, Atkins CE, Bonagura JD, Fox PR, Häggström J, Fuentes VL, et al. ACVIM consensus guidelines for the diagnosis and treatment of myxoma- tous mitral valve disease in dogs. J Vet Intern Med. 2019;33:1127–40. 12. Cornell CC, Kittleson MD, Della Torre P, Häggström J, Lombard CW, Ped- ersen HD, et al. Allometric scaling of M-mode cardiac measurements in normal adult dogs. J Vet Intern Med. 2004;18:311–21. References • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: 13. Kothari J, Lakhia K, Solanki P, Parmar D, Boraniya H, Patel S. Anomalous origin of the left coronary artery from the pulmonary artery in adult- hood: challenges and outcomes. Korean J Thorac Cardiovasc Surg. 2016;49:383–6. 14. Shriki JE, Shinbane JS, Rashid MA, Hindoyan A, Withey JG, DeFrance A, et al. Identifying, characterizing, and classifying congenital anomalies of the coronary arteries. Radiographics. 2012;32:453–68. 15. Peña E, Nguyen ET, Merchant N, Dennie C. ALCAPA syndrome: not just a pediatric disease. Radiographics. 2009;29:553–65. 16. Blake RR, Longo M, Santarelli G, Liuti T, Martinez-Pereira LVY. Coronary arteriovenous malformation in a dog with a complex arrhythmia and hypothyroidism. J Vet Cardiol. 2019;23:38–44. 16. Blake RR, Longo M, Santarelli G, Liuti T, Martinez-Pereira LVY. Coronary arteriovenous malformation in a dog with a complex arrhythmia and hypothyroidism. J Vet Cardiol. 2019;23:38–44.
https://openalex.org/W2913027883	https://revistas.unphu.edu.do/index.php/aula/article/download/83/83, https://dialnet.unirioja.es/descarga/articulo/8931190.pdf	es		Gestión de Riesgos (Sendai) y Objetivos Desarrollo Sostenible en Puerto Plata	Aula Revista de Humanidades y Ciencias Sociales	2,018	cc-by	4,267	REVISTA DE HUMANIDADES Y CIENCIAS SOCIALES DE LA UNIVERSIDAD NACIONAL PEDRO HENRÍQUEZ UREÑA 2 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO Gestión de Riesgos (Sendai) y Objetivos Desarrollo Sostenible en Puerto Plata Juan Roberto Suriel Guzmán Director del Departamento de Medio Ambiente y Recursos Naturales, UNPHU. Santo Domingo, República Dominicana r.suriel@unphu.edu.do Fecha de recepción: 9 de enero de 2018 Fecha de aceptación: 9 de abril de 2018 Favor de citar este artículo de la siguiente forma: Suriel Guzmán, J. (2018). Gestión de Riesgos (Sendai) y Objetivos Desarrollo Sostenibles en Puerto Plata. Revista AULA. Vol.62, Número 1,enero – junio 2018. Santo Domingo : Amigo del Hogar. RESUMEN Este artículo trata sobre los desastres recurrentes del Municipio de San Felipe de Puerto Plata y cómo el Acuerdo Marco para la Reducción de Desastres 2015-2030 de Sendai, junto a los Objetivos de Desarrollo Sostenible (ODS), se interrelacionan y de cómo pueden ser utilizados para la Gestión de riesgos ante desastres. En él se muestran trabajos previos sobre gestión de riesgos de desastres realizados en este territorio. Esto es importante porque parece que no han sido comprendidos en toda su extensión por las autoridades y las comunidades. Prueba de ello son los efectos negativos, repetitivos, de amenazas naturales y antrópicas convertidas en realidad como ocurrió con las pérdidas ocurridas durante las lluvias de los meses de octubre y noviembre de 2016, las vaguadas de inicio del año 2017 y las ocasionadas durante el paso por las cercanías de República Dominicana de los huracanes Irma y María en el mes de septiembre de 2017. Los hallazgos encontrados mostraron que existe insuficiencia de entendimiento y sensibilidad de parte de los actores claves del Municipio de San Felipe de Puerto Plata al trabajo coordinado de la gestión de riesgos a desastres naturales y antrópicos. Se recomienda desarrollar proyectos a nivel local, del gobierno central y organismos internacionales tomando en cuenta el Marco de Sendai y los Objetivos de Desarrollo Sostenible (ODS) y la Estrategia Nacional de Desarrollo, de manera que los trabajos, además de aumentar la sostenibilidad ambiental, logren el futuro que queremos, aumente la resiliencia y se reflejen en los indicadores de ODS y las metas e indicadores del Marco de Sendai. Palabras clave: cambio climático, gestión de riesgos, objetivos de desarrollo sostenible Puerto Plata, Sendai Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 3 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO ABSTRACT This article refers to the recurring disasters affecting the Municipality of San Felipe de Puerto Plata and how the Framework Agreement for Disaster Reduction 2015-2030 of Sendai, together with the Sustainable Development Goals (SDG), are interrelated and can be used for a Disaster Risk Management. It shows previous work on disaster risk management in this territory. This is important for it seems it have been poorly understood by authorities and communities. Prove of this, is the repetitive negative effects of natural threats and anthropic occurrences; as it occurred with the losses during the rains of October and November 2016, the beginning of 2017, and those caused during the passage through the vicinity of the Dominican Republic of Hurricanes Irma and María in September 2017. The findings showed that there is insufficient understanding and sensitivity, from the key stakeholders of the Municipality of San Felipe de Puerto Plata, to the coordinated work of risk management of natural and anthropic disasters. The development of projects at the local level by the central government and international organizations, taking into account the Sendai Framework, the Sustainable Development Goals and the National Development Strategy. The recommended works, in addition to increasing Environmental Sustainability, would achieve the future we want, increase resilience and reflect it in the SDG indicators of the Sendai framework. Palabras clave: climate change, risk management, Sustainable Development Goals (SDG), Puerto Plata, Sendai Gestión de Riesgos (Sendai) y Objetivos Desarrollo Sostenible en Puerto Plata Introducción Los resultados recurrentes de inundaciones asociadas a las precipitaciones del mes de noviembre de 2016, las vaguadas del inicio del año 2017 y las ocasionadas durante el paso por las cercanías de República Dominicana (RD), de los huracanes Irma y María en el mes de septiembre de 2017, llevaron a las autoridades a declarar la provincia de Puerto Plata con todos sus municipios en estado de emergencia en cada uno de estos fenómenos naturales. El centro de Operaciones de Emergencias (COE), en el “Informe final por vaguada del 8 de diciembre de 2016” (COE Vaguada, 2017), el “Informe final del huracán Irma del 15 de septiembre de 2017” (COE, 2017) y el “Informe final del huracán María del 02 de octubre de 2017” (COE, 2017), así como visitas al área de estudio, muestran cómo Puerto Plata fue afectada por estos fenómenos naturales. La motivación de escribir este artículo se origina a partir mi experiencia como director del “Proyecto para el Fortalecimiento de las Estructuras Organizativo-Funcionales de la Gestión de Riesgos Ante Desastres en la RD”, de la Presidencia de la República Dominicana y la Unión Europea durante el año 2015. Esto me permitió conocer la situación presente en muchos municipios del país; que afecta el desarrollo de los mismos mientras no se intervenga adecuadamente. Por mi interés de cooperar con el país, una motivación más, pude observar bien de cerca el sufrimiento de esos dominicanos afectados por los desastres que necesitan de un trabajo como nación en beneficio de todos. Además, este articulo incluye parte de Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 4 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO las conclusiones del proyecto de investigación que realicé en el curso “Reducción de Riesgo de Desastres y Desarrollo Local Sostenible”, del Centro Internacional de Formación de la Organización Internacional del Trabajo (OIT) y la oficina de la Naciones Unidas (UN) para la Reducción de Riesgos de Desastres (UNISDR) en Turín, Italia. La República Dominicana es miembro de Naciones Unidas, que el día 25 de septiembre de 2015 adoptó de conjunto, 17 Objetivos de Desarrollo Sostenible. También, la R.D. firmó el 18 de marzo de 2015 un Acuerdo sobre Gestión de Riesgos a Desastres, el Marco de Sendai, que es el instrumento sucesor del Marco de Acción de Hyogo, con un principio fundamental de aumento de la resiliencia de las naciones y las comunidades ante los desastres. Este artículo se enmarca en la aplicación del Acuerdo Marco para la Reducción de Desastres 2015-2030 de Sendai y los Objetivos de Desarrollo Sostenible (ODS) en el Municipio de San Felipe de Puerto Plata. Desarrollo En el Informe de la vaguada del 8 de diciembre de 2016 del COE (COE Vaguada, 2017), Puerto Plata encabezó la lista de comunidades afectadas con cierre de vías por deslizamientos de tierra, crecidas de ríos, cañadas, afectaciones de puentes y pérdidas de vidas humanas. Durante la visita realizada el 26 de noviembre de 2016 al área del Municipio de San Felipe de Puerto Plata, el Director de la Defensa Civil, mediante comunicación oral, reportó que en las precipitaciones de ese mes resultaron unas 5,714 casas inundadas, 28 casas destruidas, junto a una cantidad de desplazados de 28,710 personas. En las fotos se muestra el desarrollo de la investigación y daños a vivienda producida por la erosión provocada por el río San Marcos. Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 Imagen I. Reunión con el director de la Defensa Civil de Puerto Plata Imagen II. Erosión cimientos vivienda provocada por el río San Marcos En el Informe del Huracán Irma del 15 de septiembre de 2017 del COE, se plasma que Puerto Plata estuvo de nuevo dentro de los municipios más afectados con personas movilizadas, viviendas afectadas y destruidas, además de infraestructura vial con daños parciales o totales, servicios básicos afectados, pérdidas de medios de vida, daños severos a la agropecuaria y un muelle afectado (COE Irma, 2017). Con el huracán María del 02 de octubre de 2017, el COE reportó a Puerto Plata en Alerta Roja, entre los más afectados, con daños severos a la infraestructura vial, personas movilizadas, viviendas afectadas y destruidas, servicios básicos afectados, pérdidas de medios de vida y daños severos a la agropecuaria; también fue afectado parcialmente un (1) muelle y suspendido el arribo de un crucero (COE María, 2017). AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO El Municipio de Puerto Plata ha sido intervenido en los aspectos de Gestión de Riesgos en varias ocasiones. Ya en el año 2015 fue clasificado como uno de los diez municipios más vulnerables de la República Dominicana por el estudio de Selección de Municipios del Proyecto “Fortalecimiento de las Estructuras Organizativo-Funcionales de la Gestión de Riesgo ante Desastres en RD” No. FED/2014/353-229, tomando en cuenta su grado de susceptibilidad. La provincia de Puerto Plata forma parte de la región del Cibao Norte, está ubicada en la costa, en el sector septentrional de la República Dominicana. Tiene nueve municipios y doce distritos municipales. La capital de la provincia es San Felipe de Puerto Plata, donde se ubica el 49% de la población de la provincia. Tiene dos distritos municipales, Maimón y Yásica Arriba; se encuentra limitada al norte, por el Océano Atlántico, al sur por el municipio de Altamira, al oeste por los municipios de Imbert y Luperón, y al este por el municipio de Villa de Montellano. Fue fundada por Nicolás de Ovando en los inicios de la colonización hispánica, hacia 1502. Se ubica en las coordenadas geográficas 19°48’N -70°41’W, ocupando una superficie de 459.71 km², clima tropical, población de San Felipe de Puerto Plata y sus dos distritos municipales, Yásica Arriba (8,791 habitantes), Maimón (21,795 habitantes) con una población total de 158,756 habitantes según el IX Censo de Población y Vivienda 2010; altura sobre el nivel del mar de 8 metros (ONE, 2010). Los desastres por fenómenos naturales son muy recurrentes en el Municipio de San Felipe de Puerto Plata, lo que nos motivó a desarrollar un trabajo sobre “Gestión de Riesgos y Sostenibilidad” para el cumplimiento final del Curso “Reducción de Riesgo de Desastres y Desarrollo Local Sostenible” en el Centro Internacional de Formación, CIF, de la Organización Internacional del Trabajo, OIT, de la Naciones Unidas (UN). Junio-diciembre 2016, Turín, Italia. 5 El trabajo se inició con la participación en un proyecto de Gestión de Riesgos en la República Dominicana; en el que se realizó un diagnóstico de riesgo de los municipios vulnerables partiendo de una lista de 45, para seleccionar los 10 más vulnerables. Para ello se tomaron en consideración análisis previos, estudios y reportes de intervenciones realizadas por diversos organismos nacionales e internacionales, entre los cuales están el Documento País 2014 sobre la Gestión de Riesgos en República Dominicana y el Plan Nacional de Gestión Integral del Riesgo de Desastres. También se usó el sistema de indicadores de Riesgo de Desastre y Gestión de Riesgos del Programa para América Latina y el Caribe del Banco Interamericano de Desarrollo, BID, junto con los Indicadores del Progreso para medir la Reducción del Riesgo de Desastres y la Implementación del Marco de Acción de Hyogo de la Estrategia Internacional para la Reducción de Riesgos de la Oficina de Naciones Unidas para la Reducción del Riego de Desastres, (UNISDR). Este estudio confirmó al municipio de San Felipe de Puerto Plata como uno de los 10 más vulnerables. En la investigación también se realizaron consultas con actores claves del municipio para conocer la operatividad del Comité Municipal de Prevención, Mitigación y Respuesta (PMR), así como una evaluación del riesgo basado en estudios previamente realizados, que evidenció la situación peligrosa en que se encuentran los diversos grupos sociales. De tal manera, el estudio debió servir para promover acciones a nivel comunitario y local. Como resultados de la investigación para el desarrollo del trabajo se pueden mencionar los siguientes hallazgos, como las causas que contribuyen a generar desastres en el municipio de San Felipe de Puerto Plata: Las características físico-geográficas del municipio de San Felipe de Puerto Plata, y los Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 6 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO factores antrópicos presentes en ella, lo exponen a diferentes amenazas, con vulnerabilidades que representan riesgos potenciales. Las principales amenazas del municipio son, geológicas sísmicas y asociadas, como los tsunamis; climáticas por ciclones tropicales y asociadas como deslizamientos e inundaciones, y antrópicas, como la contaminación del río San Marcos y del sistema de cañadas que nacen en la Loma Isabel de Torres y cruzan diferentes puntos de la ciudad (Martínez, 2013). Las vulnerabilidades física, social, ambiental, económica e institucional, no pueden ser peores y son originadas por una alta concentración de personas en comunidades desfavorecidas; los altos niveles de pobreza en barrios periféricos; un alto nivel de analfabetismo en el municipio correspondiente a un 25% (ONE, 2010); el deficiente servicio de agua potable en gran parte de la población; la existencia limitada de planes de respuesta ante las diferentes amenazas: inundaciones, deslizamientos, huracanes, tsunamis y contaminación; la no existencia de programas ni de iniciativas en el ayuntamiento local para la reducción de riesgos de desastres; la falta de capacitación de los actores locales en estos temas; la falta de Plan de Ordenamiento Territorial; la falta de coordinación en el Comité Municipal de Prevención, Mitigación y Respuestas (PMR); razones de institucionalidad, política y legal por permisibilidad de las Autoridades Municipales de asentamientos de familias en área que son vulnerables a las amenazas naturales; razones sociales y económicas como la falta de educación y conciencia de la ciudadanía para disponer adecuadamente los desechos sólidos; el vertido de desechos industriales y razones ambientales (Giraldo, 2013). Objetivos de Desarrollo Sostenible Representantes de la mayoría de los países del mundo, reunidos en las Naciones Unidas el Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 25 de septiembre de 2015, adoptaron conjuntamente 17 Objetivos de Desarrollo Sostenible, centrados en erradicar la pobreza, proteger el planeta y asegurar la prosperidad para todas las personas. Estos objetivos tienen 169 metas y 231 indicadores para lograr el futuro que queremos, dando paso a la Agenda 2030 para el Desarrollo Sostenible, mejor conocida como Agenda 2030. La República Dominicana está comprometida con el cumplimiento de estos objetivos. Los ODS (ONU, 2015) son: Objetivo 1: Poner fin a la pobreza en todas sus formas en todo el mundo. Objetivo 2: Poner fin al hambre, lograr la seguridad alimentaria y la mejora de la nutrición y promover la agricultura sostenible. Objetivo 3: Garantizar una vida sana y promover el bienestar en todas las edades. Objetivo 4: Garantizar una educación inclusiva, equitativa y de calidad y promover oportunidades de aprendizaje durante toda la vida para todos. Objetivo 5: Lograr la igualdad entre los géneros y empoderar a todas las mujeres y las niñas. Objetivo 6: Garantizar la disponibilidad de agua y su gestión sostenible y el saneamiento para todos. Objetivo 7: Garantizar el acceso a una energía asequible, segura, sostenible y moderna para todos. Objetivo 8: Promover el crecimiento económico sostenido, inclusivo y sostenido, el empleo pleno y productivo y el trabajo decente para todos. AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO Objetivo 9: Construir infraestructuras resilientes, promover la industrialización inclusiva y sostenible y fomentar la innovación. Objetivo 10: Reducir la desigualdad en y entre los países. Objetivo 11: Lograr que las ciudades y los asentamientos humanos sean inclusivos, seguros, resilientes y sostenibles. Objetivo 12: Garantizar modalidades de consumo y producción sostenibles. Objetivo 13: Adoptar medidas urgentes para combatir el cambio climático y sus efectos. Objetivo 14: Conservar y utilizar en forma sostenible los océanos, los mares y los recursos marinos para el desarrollo sostenible. Objetivo 15: Proteger, restablecer y promover el uso sostenible de los ecosistemas terrestres, gestionar los bosques de forma sostenible, luchar contra la desertificación, detener e invertir la degradación de las tierras y poner freno a la pérdida de la diversidad biológica. Objetivo 16: Promover sociedades pacíficas e inclusivas para el desarrollo sostenible, facilitar el acceso a la justicia para todos y crear instituciones eficaces, responsables e inclusivas a todos los niveles. Objetivo 17: Fortalecer los medios de ejecución y revitalizar la Alianza Mundial para el Desarrollo Sostenible. Acuerdo Marco para la Reducción de Desastres 2015-2030 de Sendai. 7 La Gestión de Riesgos a nivel mundial es liderada por “El Marco de Sendai para la Reducción del Riesgo de Desastres 2015-2030”, el cual se adoptó en la tercera Conferencia Mundial de las Naciones Unidas celebrada en Sendai (Japón) el 18 de marzo de 2015 (Naciones Unidas, 2015). El Marco de Sendai es el instrumento sucesor del Marco de Acción de Hyogo para 20052015: su principio es el aumento de la resiliencia de las naciones y las comunidades ante los desastres. Sendai tiene como objetivo lograr la reducción sustancial del riesgo de desastres y de las pérdidas ocasionadas tanto en vidas, medios de subsistencia y salud como en bienes económicos, físicos, sociales, culturales y ambientales de las personas, las empresas, las comunidades y los países. Para alcanzar el resultado anterior es vital, según Sendai, trabajar para prevenir la aparición de nuevos riesgos de desastres y reducir los existentes implementando medidas integradas e inclusivas de índole económica, estructural, jurídica, social, sanitaria, cultural, educativa, ambiental, tecnológica, política e institucional que prevengan y reduzcan el grado de exposición a las amenazas y la vulnerabilidad a los desastres, aumenten la preparación para la respuesta y la recuperación y refuercen de ese modo la resiliencia. En el Marco de Sendai se mencionan por primera vez siete metas que incluyen: reducción sustancial de la mortalidad mundial en desastres, reducción sustancial del número de personas afectadas, reducción de las pérdidas económicas en relación al PIB mundial, reducción sustancial de los daños del desastre a la infraestructura y la interrupción de los servicios básicos, incluidos los servicios de salud y educación, aumento en el número de países con estrategias nacionales y locales para la reducción Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 8 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO del riesgo de desastres para el año 2020, una mayor cooperación internacional, un mayor acceso a peligros múltiples, sistemas de alerta temprana e información del riesgo de desastre y evacuaciones. Sendai, basada en la experiencia de Hyogo recomienda a los Estados adoptar medidas específicas en todos los planos: local, nacional, regional y mundial, y trabajar en cuatro prioridades que son: 1: Comprender el riesgo de desastres. 2: Fortalecer la gobernanza del riesgo de desastres para gestionar dicho riesgo. 3: Invertir en la reducción del riesgo de desastres para la resiliencia. 4: Aumentar la preparación para casos de desastre a fin de dar una respuesta eficaz y para “reconstruir mejor” en los ámbitos de la recuperación, la rehabilitación y la reconstrucción. Como se muestra, la República Dominicana tiene compromisos internacionales que cumplir con El Marco de Sendai para la Reducción del Riesgo de Desastres 2015-2030 y los Objetivos de Desarrollo Sostenible. Por cuanto es una responsabilidad del Estado Dominicano promover a nivel nacional y local estos objetivos y metas. Conclusión Debido a las diferentes intervenciones en la gestión de riesgo en el Municipio de San Felipe de Puerto Plata y a los repetidos efectos negativos asociados a los eventos naturales y actividades humanas, se llega a la conclusión, entre otras cosas, de que existe insuficiencia de entendimiento y sensibilidad de los actores cla- Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 ve del Municipio de San Felipe de Puerto Plata respecto al trabajo coordinado de la gestión de riesgos a desastres naturales y antrópicos. Los daños materiales, el impacto a ecosistemas y las pérdidas de vidas humanas recurrentes por las inundaciones asociadas a las precipitaciones de los meses de noviembre-diciembre de 2016 y las ocasionadas por los Huracanes Irma y María en el 2017; así lo demuestran. Visto lo anterior, se recomienda inicialmente apoyar el fortalecimiento de los principales actores del Municipio de San Felipe de Puerto Plata de la República Dominicana en el trabajo coordinado de la gestión y reducción de los riesgos a desastres y la adaptación al cambio climático con los siguientes propósitos u objetivos: 1. Motivar a los actores clave del municipio al trabajo coordinado de la gestión y reducción de los riesgos. 2. Promover el conocimiento y la capacitación entre los actores clave del Municipio de San Felipe de Puerto Plata, sobre los riesgos naturales y antrópicos identificados en su territorio. 3. Motivar a las autoridades municipales a incorporar la gestión de riesgos y el cambio climático en sus planes de desarrollo municipal y comunitario de una manera sostenible. Las líneas de acción para lograr estos objetivos pueden ser: Definir espacios de coordinación entre todos los actores relacionados con la gestión y reducción de riesgos a desastres. Elaborar un diagnóstico participativo de riesgos a desastres (naturales y antrópicos) de los dos Distritos Municipales (DM), Maimón y Yásica Arriba del municipio de San Felipe de Puerto Plata. AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO Lograr que los actores clave del municipio identifiquen y reconozcan los riesgos naturales y antrópicos presentes en su territorio. Establecer centros de documentación e información sobre gestión de riesgos de desastres, para consulta de los actores clave y la comunidad en general. Desarrollar un plan de capacitación comunitaria en gestión integral de riesgos a desastres. Preparar y realizar un taller municipal sobre gestión de riesgo y cambio climático en los Planes de Desarrollo Municipal en la Provincia de San Felipe de Puerto Plata. Al mismo tiempo es recomendable desarrollar otros proyectos a nivel local, del gobierno central y organismos internacionales tomando en cuenta el Marco de Sendai, los Objetivos de Desarrollo Sostenible y la Estrategia Nacional de Desarrollo para lograr un desarrollo sostenible (Suriel, 2016). Los proyectos a desarrollarse dentro del marco del aumento de la resiliencia y de la gestión de riesgos a desastres, ayudarán a cumplir con los objetivos de Desarrollo Sostenible de Naciones Unidas mencionados previamente. Los trabajos a ejecutarse con estos lineamientos, además de aumentar la sostenibilidad medio ambiental para lograr el futuro que queremos y la reducción de riesgos a desastres; contribuirán con los objetivos y se reflejarán en los indicadores de ODS y con las metas e indicadores del Marco de Sendai para el aumento de la resiliencia de la nación y las comunidades ante los desastres. Específicamente en el caso del Municipio de San Felipe de Puerto Plata; estas medidas son de gran urgencia para que la región sea menos vulnerable y tenga mayores posibilidades de desarrollo sostenible. 9 Referencias Centro de Operaciones de Emergencia COE (2017). Informe Final por Vaguada 8/12/2016. Santo Domingo, República Dominicana. Centro de Operaciones de Emergencia COE (2017). Informe Final Huracán Irma 15/09/2017. Santo Domingo, República Dominicana. Centro de Operaciones de Emergencia COE (2017). Informe Final Huracán María 02/10/2017. Santo Domingo, República Dominicana. Dirección General de Ordenamiento y Desarrollo Territorial –DGODT. (2012). Amenazas y Riesgos Naturales en República Dominicana – Compendio de Mapas. Programa de Prevención de Desastres y Gestión de Riesgos. Santo Domingo. Cocco Quezada, A. (s.f.). Temporada de Huracanes en la República Dominicana. Recuperado de http://www.acqweather.com/HURACANES.htm Dirección General de Ordenamiento y Desarrollo Territorial –DGODT. (2013). Plan Municipal de Gestión de Riesgos San Felipe de Puerto Plata de la Dirección General de Ordenamiento y Desarrollo Territorial (DGODT), a través del Programa de Prevención de Desastres y Gestión de Riesgos (1708/ OC-DR). Financiamiento del Banco Interamericano de Desarrollo (BID). Santo Domingo. Gaceta Oficial. Ley 147-02. (2002) Sistema Nacional de Prevención, Mitigación y Respuesta. Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597 10 AULA EL MEDIO AMBIENTE SOSTENIBLE DE LA MANO DEL DESARROLLO ECONÓMICO Giraldo, M. (2013). Propuestas integrales y estrategia para la reducción de riesgos. Programa de Prevención de Desastres y Gestión de Riesgos. DGODT – BID. Santo Domingo. Programa DELNET – Organización Internacional del Trabajo-OIT. (2014). El riesgo de desastres: origen, evaluación, reducción y prevención en el marco del desarrollo local sostenible. Martínez, I. (2013). Caracterización Amenazas Naturales por Deslizamientos e Inundación. Municipio San Felipe de Puerto. Programa de Prevención de Desastres y Gestión de Riesgos. DGODT – BID. Santo Domingo. Programa DELNET – Organización Internacional del Trabajo-OIT. (2016). Curso: Reducción Del Riesgo De Desastres y Desarrollo Local Sostenible, Programa Delnet CIF/OIT. (2016). Materiales de la Unidad 1 a la 6; Guía, Marco Teórico y Glosario. Naciones Unidas. (2015). Marco de Sendai para la Reducción del Riesgo de Desastres 2015-2030. Oficina de las Naciones Unidas para la Reducción del Riesgo de Desastres (UNISDR). Recuperado de www.unisdr.org. Oficina Nacional de Estadísticas (ONE). (2010). IX CENSO NACIONAL DE POBLACIÓN Y VIVIENDA. Santo Domingo, República Dominicana. Suriel, R. (2016). Fortalecimiento de capacidades de los actores clave del municipio de San Felipe de Puerto Plata al trabajo coordinado de la Gestión de Riesgos de Desastres. Turín, Italia: Programa DELNET – Organización Internacional del Trabajo-OIT. PNUD/ UNESCO/ OIM/ Comisión Europea/ ANESVAD. (2012). Mapa de inundación, rutas de evacuación y zonas seguras en caso de tsunami. Ciudad de Puerto Plata. Juan Roberto Suriel Guzmán Bore nus rerum et veri ut intionsequi officil lantibus, quis es idi rem aliqui ium sunt. Voluptiae sunt hari te quisi dolorerum earions erepedi assi corio doluptassi consequ amendita is quia nonsed quosa quis enis ped maio qui odi alit facim excesedi volore, officipsusae maiorero doluptibus mil miliatem cor aut eium ut laccumque cum fugitiatur? Soluptas maximus, cum quam, comnimu sdandel essunt maximporerci beatur? As magnimintum sequatem rehendelent harchil int am, velitium voluptat. Volupta temquam, sam nulland itatquam, eumenet quod et, sitionsequat minihitio optatiu sandenis elendant ernatur andae. Itae ne sum et erunt, se num elit que comnit, sequae. Ipitatatis aut re volorum quidus aut ulpa dolor as et laci optatio. Ut ulpa quasper spicidu stiur? Ectem ut que laut omnihit estrum autenih illabor aliquas eatium, il eaquist Vol. 61, número 2, ene.-jun. 2018 ISSN:0254-7597
https://openalex.org/W4383999446	https://www.scielo.br/j/bpsr/a/FmvrZcZtWHffJHmmfMBRypc/?lang=en&format=pdf	English	null	Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn	DOAJ (DOAJ: Directory of Open Access Journals)	2,019	cc-by	16,133	brazilianpoliticalsciencereview After a decade dominated by policies that prioritized pro-market economic reforms and rejected state-led redistribution, beginning at the end of the 1990s a left turn swept through the region, with the electoral victory and subsequent implementation of policy agendas emphasizing social spending and redistribution. Consequently, income inequality fell substantially in prac- tically all the countries governed by the left (CORNIA, 2014; LOƵ PEZ-CALVA and LUSTIG, 2010). M Leftist policies produced social impacts via different paths, however. According to a signiϐicant part of the contemporary literature on Latin America, democracies recently governed by the left can be divided into two groups, according to those paths (CASTANǂ EDA, 2006; WEYLAND, 2009). The ϐirst group is formed of countries such as Brazil, Chile, and Uruguay, in which presidents were elected from traditional left-wing parties, defending al- ternative policies without antagonizing the prevailing economic order. Such governments enacted moderate redistributive policies, mainly guaranteeing the provision of a minimum income to the poorest sectors of society, without advancing major changes to the struc- tures of taxation and government transfers (HUNTER and SUGIYAMA, 2009; PRIBBLE and HUBER, 2013). In such cases, the left’s success among voters and the subsequent imple- mentation of redistributive policies came about without affecting political stability. This path is similar to the one taken by various social democratic parties in consolidated democ- racies, which, for many analysts, constitutes the expected pattern of evolution. The second group includes countries such as Bolivia, Ecuador, and Venezuela, where a new left emerged in a context of crisis of the traditional party system and explosive growth of social movements, subsequently electing presidents who emphasized a break with the economic order. These leftist governments implemented much more intense re- distributive policies, substantially increasing both taxation of the richest sectors of society and public transfers to the poorest (CORRALES, 2010; KENNEMORE and WEEKS, 2011). In such cases, the emergence of the left is associated with a radicalization of the politi- cal process and of government, accompanied by political instability in the form of violent protests, secession attempts, and even coups (RITTINGER and CLEARY, 2013). According to a substantial number of studies on the subject, such a path does not ϐit with either the theoretical models or empirical patterns associated with the experiences of consolidated democracies. brazilianpoliticalsciencereview Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn∗ Fabiano Santos∗∗ https://orcid.org/0000-0002-6661-7666 ∗∗Universidade Estadual do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro, Brazil Acir Almeida∗∗∗ https://orcid.org/0000-0002-7402-4510 ∗∗∗Instituto de Pesquisa Econômica Aplicada, Rio de Janeiro, Rio de Janeiro, Brazil Thiago Moreira da Silva∗∗ https://orcid.org/0000-0001-8451-8621 ∗∗Universidade Estadual do Rio de Janeiro, Rio de Janeiro, Rio de Janeiro, Brazil This article offers an explanation for the intensity of re- distributive policies from Latin American leftist governments dur- ing the so-called ‘left turn’. It challenges the idea that the recent radical leftist governments are the product of region-speciϐic char- acteristics. Based on theoretical models that analyze the implica- tions of inequality in democracies and the moderating effects of different economic and institutional structures, it argues instead that the left’s redistributive policies were more intense in countries where capital mobility is low and there was no pro-elite legislative veto player. To evaluate this explanation, we offer original evidence from time-series -cross-section regression models of social spend- ing, based on data from thirteen Latin American countries over the years 2003 to 2015. The results show partial support for the ex- planation: there is strong evidence that social spending decreased with the ideological distance between the president and the pro- elite veto player, but only weak evidence that capital mobility at- tenuated this effect. Keywords: Redistribution; social spending; capital mobility; veto player; Latin America; left turn. () http://dx.doi.org/10.1590/1981-3821201900010005 This publication is registered under a CC-BY Licence. For data replication, see https://doi.org/10.7910/DVN/H6NGHG. Earlier versions of this paper were presented at the Berlin Social Science Center (WZB, 2014), the annual meeting of the American Political Science Association (2014), the VIII Congresso Latinoamer- icano de Ciencia Polı́tica (2015), and the Institute of Applied Economic Research (IPEA, 2016). The authors gratefully acknowledge Ricardo Ceneviva, Miguel Foguel, Luciana Jaccoud, Wolfgang Merkel, Dieter Nohlen, Thamy Pogrebinschi, Salvador Werneck Vianna, Bernard Webbels, and the anonymous reviewers for their comments and suggestions. (2019) 13 (1) e0005 - 1/31 Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn M ost Latin American democracies experienced signiϐicant political and social transformations at the dawn of the twenty-ϐirst century. brazilianpoliticalsciencereview However, we must emphasize that our aim is not to offer a new explanation for what are considered Latin American political idiosyncrasies, but rather to demonstrate that well-established theories about the interaction between the political and the economic realms in consolidated democracies remain useful, with only minor adaptations, for explaining policy differences among recent leftist governments in Latin America. context marked by, on one side, acute inequality and, on the other, an economic and insti- tutional structure conducive to intense redistributive policies. Our claim is that although Latin American countries entered this century with high levels of inequality, and practi- cally all of them experienced favorable macroeconomic conditions during most of its ϐirst ϐifteen years or so, leftist governments implemented radical redistributive policies only in countries where there was relatively low capital mobility and the legislature was not a veto player. The theoretical foundations for this claim come from theories about how inequal- ity affects the intensity of redistributive policies in democracies (MELTZER and RICHARD, 1981), its implications for economic growth (ALESINA and RODRIK, 1994), and how this effect is moderated by different economic and institutional structures (BATES and LIEN, 1985; PERSSON and TABELLINI, 1994). However, we must emphasize that our aim is not to offer a new explanation for what are considered Latin American political idiosyncrasies, but rather to demonstrate that well-established theories about the interaction between the political and the economic realms in consolidated democracies remain useful, with only minor adaptations, for explaining policy differences among recent leftist governments in Latin America. We assess our claim by testing its implications for levels of government social ex- penditure, using time-series-cross-section (TSCS) data from the thirteen countries which comprise the so-called left turn in Latin America, over the years 2003 to 2015. This period roughly coincides with the so-called commodities price boom, during which the region en- joyed above average economic growth rates, thus facilitating redistribution. Since our goal is to explain why the left adopted radical redistributive policies in some countries but not in others, our sample only includes the countries where at least one leftist president was elected. We use the level of social expenditure (data from the Economic Commission for Latin American and the Caribbean (ECLAC)) as proxy for the government’s redistributive effort, as have several previous studies (e.g.: AVELINO, BROWN and HUNTER, 2005), ac- cording to McCarty and Pontusson (2009, p. 666). brazilianpoliticalsciencereview The theoretical foundations for this claim come from theories about how inequal- ity affects the intensity of redistributive policies in democracies (MELTZER and RICHARD, 1981) it i li ti f i th (ALESINA d RODRIK 1994) d h thi context marked by, on one side, acute inequality and, on the other, an economic and insti- tutional structure conducive to intense redistributive policies. Our claim is that although Latin American countries entered this century with high levels of inequality, and practi- cally all of them experienced favorable macroeconomic conditions during most of its ϐirst ϐifteen years or so, leftist governments implemented radical redistributive policies only in countries where there was relatively low capital mobility and the legislature was not a veto player. The theoretical foundations for this claim come from theories about how inequal- ity affects the intensity of redistributive policies in democracies (MELTZER and RICHARD, 1981), its implications for economic growth (ALESINA and RODRIK, 1994), and how this effect is moderated by different economic and institutional structures (BATES and LIEN, 1985; PERSSON and TABELLINI, 1994). However, we must emphasize that our aim is not to offer a new explanation for what are considered Latin American political idiosyncrasies, but rather to demonstrate that well-established theories about the interaction between the political and the economic realms in consolidated democracies remain useful, with only minor adaptations, for explaining policy differences among recent leftist governments in Latin America. context marked by, on one side, acute inequality and, on the other, an economic and insti- tutional structure conducive to intense redistributive policies. Our claim is that although Latin American countries entered this century with high levels of inequality, and practi- cally all of them experienced favorable macroeconomic conditions during most of its ϐirst ϐifteen years or so, leftist governments implemented radical redistributive policies only in countries where there was relatively low capital mobility and the legislature was not a veto player. The theoretical foundations for this claim come from theories about how inequal- ity affects the intensity of redistributive policies in democracies (MELTZER and RICHARD, 1981), its implications for economic growth (ALESINA and RODRIK, 1994), and how this effect is moderated by different economic and institutional structures (BATES and LIEN, 1985; PERSSON and TABELLINI, 1994). brazilianpoliticalsciencereview Rather, it constitutes a speciϐic characteristic of the region, related, for ex- ample, to its tradition of populism or to transformations undergone by its party systems. In this article, we question the idea that Latin America’s recent radical leftist gov- ernments1 reϐlect characteristics speciϐic to the region, arguing instead that they consti- tute empirical manifestations of more general processes of democratic consolidation in a 1We use the terms ‘moderate’ and ‘radical’ to characterize the intensity of redistributive policy, and by ex- tension, of the government that implemented it. (2019) 13 (1) e0005 - 2/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva context marked by, on one side, acute inequality and, on the other, an economic and insti- tutional structure conducive to intense redistributive policies. Our claim is that although Latin American countries entered this century with high levels of inequality, and practi- cally all of them experienced favorable macroeconomic conditions during most of its ϐirst ϐifteen years or so, leftist governments implemented radical redistributive policies only in countries where there was relatively low capital mobility and the legislature was not a veto player. The theoretical foundations for this claim come from theories about how inequal- ity affects the intensity of redistributive policies in democracies (MELTZER and RICHARD, 1981), its implications for economic growth (ALESINA and RODRIK, 1994), and how this effect is moderated by different economic and institutional structures (BATES and LIEN, 1985; PERSSON and TABELLINI, 1994). However, we must emphasize that our aim is not to offer a new explanation for what are considered Latin American political idiosyncrasies, but rather to demonstrate that well-established theories about the interaction between the political and the economic realms in consolidated democracies remain useful, with only minor adaptations, for explaining policy differences among recent leftist governments in Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Fabiano Santos, Acir Almeida & Thiago Moreira da Silva context marked by, on one side, acute inequality and, on the other, an economic and insti- tutional structure conducive to intense redistributive policies. Our claim is that although Latin American countries entered this century with high levels of inequality, and practi- cally all of them experienced favorable macroeconomic conditions during most of its ϐirst ϐifteen years or so, leftist governments implemented radical redistributive policies only in countries where there was relatively low capital mobility and the legislature was not a veto player. Brief literature review Latin America’s left turn has inspired many studies addressing different aspects of the phenomenon. One notable characteristic of these studies is the way in which they depart from the traditional view in the comparative literature, which supposes that ac- curate accounts of political phenomena in Latin America require the formulation of new, regional-speciϐic theories2. It is our purpose to advance and improve upon this scientiϐic program, embodied in the institutionalist and the political economy paradigms, which an- alyzes the Latin American case using an integrated theoretical lens capable of accounting for as many cases as possible. Our focus is on the policies implemented by the left. Here, studies vary in terms of their emphasis on similarities and differences. Those pointing to common elements of the Latin American left highlight a common agenda centered on equality, redistribution, and social inclusion (ARDITI, 2008). Studies emphasizing differences, on the other hand, tend to draw attention to the actual economic and social policies, usually classiϐied as either moderate or radical (CASTANǂ EDA, 2006; WEYLAND, 2009). The moderate left preserves the pillars of the market economy and does not interfere with prices and contracts, tack- ling poverty in a gradual fashion, with ϐiscal prudence and respect to private property. The radical left, on the other hand, intervenes in the economy, controlling prices and nation- alizing companies, and seeks to reduce poverty by drastically raising public spending and intensely redistributing income. The literature offers various explanations for such differences among leftist gov- ernments, some being of a political and others of a more economic nature. Among the former, studies initially focused on the evolution of the party system. According to this perspective, left radicalism in government results from the party system’s low degree of institutionalization (FLORES-MACIƵAS, 2010) or from the electoral context in which left- wing parties emerged and consolidated (MADRID, 2010). Regarding the ϐirst explanation, however, Weyland (2009, p. 150) argues convincingly that the causal relationship is in fact the reverse of that postulated, with the emergence of radical leftist parties contributing to the destabilization of the party system. Concerning parties’ electoral trajectories, although this may eventually explain ideological differences, it is not sufϐicient to account for varia- tion among leftist policies, as Campello (2015), Murillo, Oliveros and Vaishnav (2011) and Weyland (2009) have shown. brazilianpoliticalsciencereview Evidence from TSCS linear models with country-speciϐic random effects and Driscoll-Kraay standard errors lends partial support to our explanation: we ϐind strong evidence that social spending decreased with the ideo- logical distance between the president and the (pro-elite) legislative veto player, but only weak evidence that the degree of capital mobility attenuated this effect. The remainder of this article is organized into four sections. The next section pro- vides a brief review of the recent literature on the Latin America’s left turn, outlining differ- ent explanations for the variation in policies among leftist governments. The third section selectively discusses the theoretical literature on inequality and redistribution, speciϐically those studies that analyze the redistributive implications of democracy and the moderat- ing effects of economic and institutional structures. Based on the main contributions to this literature, we offer an explanation for the variation found in redistributive policies implemented by recent leftist governments in the region. The fourth section evaluates the (2019) 13 (1) e0005 - 3/31 e0005 - 3/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn empirical validity of the explanation. The ϐifth and ϐinal section brieϐly summarizes the main ϐindings and offers some ϐinal considerations. Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn empirical validity of the explanation. The ϐifth and ϐinal section brieϐly summarizes the main ϐindings and offers some ϐinal considerations. 2The origin of this view can be traced to Lipset’s (1981) classical study, in which Latin American cases are treated as a cluster, separate from the other experiences of mass representative democracy. Brief literature review Adopting a more institutionalist approach, Nunes (2014) argues that the radical leftist governments emerged due to the absence of at least one of two moderating political mechanisms: the occurrence of a runoff in the presidential election and the existence of an effective veto player in the legislature. While the ϐirst generates policy moderation by requiring that candidates win the support of the median voter, the second does so by re- quiring minority presidents to obtain support from opposition parties. Nevertheless, the ϐirst mechanism does not seem to operate as the author suggests. The theoretical results presented by Bordignon, Tommaso and Tabellini (2016) suggest that it is not the occur- rence of a runoff that generates moderation, but rather the two-round rule. Considering that there are only two cases of leftist presidents elected by a simple majority under one round of balloting (Paraguay’s Lugo and Nicaragua’s Ortega, in 2006), the empirical rele- vance of Nunes’ (2014) ϐirst mechanism appears highly doubtful. Among the economic explanations, there is a virtual consensus that the macroe- conomic climate in Latin America during the ϐirst decade of the twenty-ϐirst century fa- vored the expansion of social policies, mainly due to the sharp income growth spurred by a signiϐicant and persistent increase in commodity prices. Weyland (2009) identiϐies this phenomenon as the main cause of the different intensities of leftist governments’ redis- tributive policies, arguing that radical policies were adopted only in countries in which the commodities sector carried greater weight (as in Bolivia, Ecuador, and Venezuela). Addi- tionally, Murillo, Oliveros and Vaishnav (2011) hypothesize that the higher capital inϐlow due to the increasing value of exports granted leftist governments the latitude to redis- tribute. On the other hand, Campello (2015, pp. 19-20) argues that integration into inter- national ϐinancial market mitigated the left’s redistributive policies. What does the accumulated empirical evidence say? Among the political explana- tory variables, the most consistent result is a negative correlation between the level of institutionalization of party systems and the intensity of leftist policies (FLORES-MACIƵAS, 2010, p. 423; NUNES, 2014, p. 51). As previously mentioned, however, it is highly prob- able that the causal direction underlying this relationship is inverse to that postulated. With respect to the effect of the legislature, the ϐindings to date cannot be considered con- clusive, due to their contamination by measurement problems. Brief literature review Another explanation in line with low institutionalization arguments is found in Biglaiser (2016), according to whom far-left policies were adopted where the electorate 2The origin of this view can be traced to Lipset’s (1981) classical study, in which Latin American cases are treated as a cluster, separate from the other experiences of mass representative democracy. (2019) 13 (1) e0005 - 4/31 e0005 - 4/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva granted the president a clear mandate, in the form of a wide margin of votes and a leg- islative majority. The argument is that this type of electoral result is a strong sign that the electorate will reward legislative support for the presidential agenda, thus increas- ing/decreasing defections among the opposition/government’s base. Despite a lack of solid theoretical foundation, this explanation is consistent with the idea that democracies in the region are ‘delegative’ in nature (O’DONNELL, 1994), in the sense that presidents enjoying the support of a large majority of voters can govern virtually without political or institutional constraints. Fabiano Santos, Acir Almeida & Thiago Moreira da Silva granted the president a clear mandate, in the form of a wide margin of votes and a leg- islative majority. The argument is that this type of electoral result is a strong sign that the electorate will reward legislative support for the presidential agenda, thus increas- ing/decreasing defections among the opposition/government’s base. Despite a lack of solid theoretical foundation, this explanation is consistent with the idea that democracies in the region are ‘delegative’ in nature (O’DONNELL, 1994), in the sense that presidents enjoying the support of a large majority of voters can govern virtually without political or institutional constraints. granted the president a clear mandate, in the form of a wide margin of votes and a leg- islative majority. The argument is that this type of electoral result is a strong sign that the electorate will reward legislative support for the presidential agenda, thus increas- ing/decreasing defections among the opposition/government’s base. Despite a lack of solid theoretical foundation, this explanation is consistent with the idea that democracies in the region are ‘delegative’ in nature (O’DONNELL, 1994), in the sense that presidents enjoying the support of a large majority of voters can govern virtually without political or institutional constraints. Brief literature review Biglaiser (2016), in turn, combines the president’s legislative status with her margin of victory at the polls to gauge the nature of her mandate, which impedes an evaluation of the contribution of each of those variables to the strong positive correlation that the author detects between the electoral mandate and the intensity of leftist policies. Among the economic explana- tions, the only strong and consistent evidence is the positive correlation between the im- portance of rents from commodities and how far to the left the government’s policies are (CAMPELLO, 2015; MURILLO, OLIVEROS and VAISHNAV, 2011; WEYLAND, 2009). structured, without considering their ideological location. Murillo, Oliveros and Vaishnav (2011) takes into account the ideology of the opposition in the legislature, but not of the legislative median, which is the crucial actor in the veto player theory. Biglaiser (2016), in turn, combines the president’s legislative status with her margin of victory at the polls to gauge the nature of her mandate, which impedes an evaluation of the contribution of each of those variables to the strong positive correlation that the author detects between the electoral mandate and the intensity of leftist policies. Among the economic explana- tions, the only strong and consistent evidence is the positive correlation between the im- portance of rents from commodities and how far to the left the government’s policies are (CAMPELLO, 2015; MURILLO, OLIVEROS and VAISHNAV, 2011; WEYLAND, 2009). We consider the institutional explanation based on the veto player argument the- oretically sound but empirically fragile, given the current limitations of the evidence. On the other hand, notwithstanding the strong (positive) association between the size of the commodities sector and the intensity of leftist policies, we ϐind the economic explanation unsatisfactory since it cannot account for the political conϐlicts that accompanied the im- plementation of intense redistributive policies. For these reasons, in the next two sections we offer a modest contribution to the debate. First, we identify the causal connections pro- posed in the literature between the intensity of redistributive policies and the institutional and economic structures, and, based on these, build our explanation for the policy differ- ences between Latin American leftist governments during the recent commodities boom. Then, we evaluate these relationships empirically. Brief literature review Nunes (2014) measures the existence of a legislative veto player by means of an indicator that the parties are well (2019) 13 (1) e0005 - 5/31 e0005 - 5/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn structured, without considering their ideological location. Murillo, Oliveros and Vaishnav (2011) takes into account the ideology of the opposition in the legislature, but not of the legislative median, which is the crucial actor in the veto player theory. Biglaiser (2016), in turn, combines the president’s legislative status with her margin of victory at the polls to gauge the nature of her mandate, which impedes an evaluation of the contribution of each of those variables to the strong positive correlation that the author detects between the electoral mandate and the intensity of leftist policies. Among the economic explana- tions, the only strong and consistent evidence is the positive correlation between the im- portance of rents from commodities and how far to the left the government’s policies are (CAMPELLO, 2015; MURILLO, OLIVEROS and VAISHNAV, 2011; WEYLAND, 2009). Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn structured, without considering their ideological location. Murillo, Oliveros and Vaishnav (2011) takes into account the ideology of the opposition in the legislature, but not of the legislative median, which is the crucial actor in the veto player theory. Biglaiser (2016), in turn, combines the president’s legislative status with her margin of victory at the polls to gauge the nature of her mandate, which impedes an evaluation of the contribution of each of those variables to the strong positive correlation that the author detects between the electoral mandate and the intensity of leftist policies. Among the economic explana- tions, the only strong and consistent evidence is the positive correlation between the im- portance of rents from commodities and how far to the left the government’s policies are (CAMPELLO, 2015; MURILLO, OLIVEROS and VAISHNAV, 2011; WEYLAND, 2009). structured, without considering their ideological location. Murillo, Oliveros and Vaishnav (2011) takes into account the ideology of the opposition in the legislature, but not of the legislative median, which is the crucial actor in the veto player theory. Inequality and redistribution In this section, we brieϐly present the micro foundations of the relationship be- tweeninequalityandredistributionindemocracies, tounderstandwhyleftistgovernments seeking to reduce inequality adopt redistributive policies with very different levels of in- tensity. Despite the large literature on the subject, for our purposes it will sufϐice to make use of the main insights from the seminal works of Meltzer and Richard (1981), Alesina and Rodrik (1994), and Persson and Tabellini (1994). Imagine a perfectly competitive closed economy with no uncertainty, formed of individuals that only differ in terms of their relative endowments of capital (either physi- cal or human) and (unqualiϐied) labor, and whose income is an increasing function of the former. All labor income is consumed, but not necessarily that derived from capital, which may be invested. Any increase in production occurs via an increase in labor productivity, which in turn requires investment. Suppose also that there is a government elected by ma- jority rule and according to universal suffrage, which directly provides basic infrastructure and services that beneϐit all citizens and are ϐinanced by a ϐlat tax levied on capital income only. The government has the option of redistributing income, through a mechanism that consists of charging a surplus tax on capital income and transferring the corresponding (2019) 13 (1) e0005 - 6/31 e0005 - 6/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva revenue to all individuals, equally. The level of the surplus tax - i.e., the intensity of the redistributive policy - is the government’s only choice variable and the only issue guiding voters’ decision. The economy and politics operate in two periods: in the ϐirst (t), capital- ists make decisions on investment and individuals elect a government; and in the second (t+1) individuals receive their earnings, the government deϐines and collects the surplus tax on capital income, and then distributes the corresponding revenue. Fabiano Santos, Acir Almeida & Thiago Moreira da Silva revenue to all individuals, equally. The level of the surplus tax - i.e., the intensity of the redistributive policy - is the government’s only choice variable and the only issue guiding voters’ decision. The economy and politics operate in two periods: in the ϐirst (t), capital- ists make decisions on investment and individuals elect a government; and in the second (t+1) individuals receive their earnings, the government deϐines and collects the surplus tax on capital income, and then distributes the corresponding revenue. Inequality and redistribution First, consider the decisions made by individual voters when electing a govern- ment. As the tax is only levied on capital gains and income is increasing on such yields, voters with lower income prefer more intense redistribution. Considering that elections are held under majority rule, the candidate elected is the one who promises to implement the redistributive policy preferred by the voter with median income. How intense is this policy? Imagine the usual scenario in which income is concentrated among the richest, meaning that the median income is lower than the average. In such a case, the median voter pays surplus tax in the amount of τyM and receives transfers of ´y, where τ, yM and ´y represent the surplus tax, the median income, and the average income, respectively. Since ´y yM (by assumption), the median voter will favor redistribution, though only up to the level in which her personal income equals the average income - beyond that, she would receive less in transfers than she pays in taxes (MELTZER and RICHARD, 1981). The expectation of redistribution negatively affects investment decisions, as the surplus tax reduces the expected net income from capital. A reduction in investment implies a lower increase in labor productivity, negatively affecting the median voter’s ϐinal income, because both labor and capital incomes (and therefore transfers) increase less. Taking these effects into account, the median voter prefers a more moderate redistributive policy, below that necessary for equalizing the median and average incomes post transfers (ALESINA and RODRIK, 1994). Notice, however, that the median voter (and the government elected) is subject to a time consistency problem. Suppose that capitalists make their investment de- cisions (in t) based on the winning policy platform, which consists of moderate redistribu- tion. Once the investments have been made, however, the median voter beneϐits from the opportunistic implementation of the more intense redistributive policy (in t+1), i.e. the one that equalizes the median and average incomes. Given that the government seeks to please the median voter (for reelection purposes), it implements the more intense redistributive policy. Anticipating the government’s opportunistic behavior, capitalists make investment decisions not according to the redistributive policy the median voter prefers in t, but rather the one which maximizes her utility in t+1, which is more intense. Inequality and redistribution Despite the lower invest- ment in t producing a less favorable result for the median voter, there is no way of avoiding it because the government cannot credibly commit to the moderate redistributive policy. The equilibrium policy is therefore more intense and less efϐicient. In sum, higher (initial) income inequality generates (the expectation of) more in- tenseredistributivepolicy, whichleadstolessinvestment, andϐinally, lesseconomicgrowth. (2019) 13 (1) e0005 - 7/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn However, as we discuss next, the ϐirst nexus of this causal chain is conditioned by economic and institutional structures. Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn bpsr However, as we discuss next, the ϐirst nexus of this causal chain is conditioned by economic and institutional structures. Capital mobility The capitalists’ strategic reduction in investment in anticipation of an excessive sur- plustaxisexclusivelybasedonthepremisethatcapitalmaybeconsumedinsteadofinvested. Actually, the theoretical literature that analyzes the relationship between taxation and in- vestment tends to work with models including multiple politically independent economies and the premise that capitalists can shift their assets between them. In this scenario, the lower the cost paid by capitalists to transfer their assets, the greater the government incen- tives for increasing the net rate of return on investment – in other words, for minimizing the surplus tax on capital (BATES and LIEN, 1985; BRETSCHER and HETTICH, 2002). The cost of shifting or reallocating capital deϐines its mobility. Economies with low capital mobility are those mostly based on ϐixed assets or real estate, particularly in- vestments in activities linked to the exploitation of natural resources, such as agriculture and mining. Economies based on manufacturing industries with intensive human capital are said to have intermediate capital mobility. Capital is more mobile when it assumes the form of a ϐinancial asset, allowing it to easily migrate to another country, especially in the current context of strong international integration. Given that capital ϐlight implies subsequent lower economic growth, negatively impacting the income of the median voter, governments from countries whose economies have a high degree of capital mobility are expected to adopt more moderate redistributive policies (BOIX, 2003, Chap. 02). However, this effect may not obtain when redistribu- tive policies aim at improving health or education, given “the importance of human capital for employers needing to be competitive in the world economy” (HUBER, MUSTILLO and STEPHENS, 2008, p. 425). bpsr One way to distinguish these two models analytically is by making use of the con- cept of ‘veto players’ (TSEBELIS, 1995). A veto player is any individual or collective agent whose agreement is necessary to the approval of a policy. Every party that is a member of a majority coalition is a veto player in the deϐinition of the government’s agenda, for ex- ample. Equally, when the median party in parliament is not the same which controls the executive the legislature becomes a veto player. Pure majoritarian democracies feature a single actor serving as the veto player: the majority party in the legislature, which also controls the government. Consensual democracies, on the other hand, can have multiple veto players. The main general result from this approach is that the production of new and relevant policies decreases when there are veto players with divergent preferences (TSE- BELIS, 1995, p. 298). One implication of this result is that the institutional structure can serve as a mechanism for the government to credibly commit to a moderate redistribu- tive policy (ACEMOGLU and ROBINSON, 2001; NORTH and WEINGAST, 1989; PERSSON and TABELLINI, 1994). The existence of a veto player representing voters whose income is higher than that of the median voter guarantees that the government will not be able to implement a more intense redistributive policy. In summary, based on the theoretical literature reviewed in this section, we can conclude that the demand for redistribution increases with income inequality, and the ex- tent to which the former is effectively translated into redistributive policy decreases with capital mobility or the existence of a pro-elite veto player. A variety of empirical studies have been conducted to test these relations. However, the results have proven contro- versial, mainly due to methodological problems. Examples of studies offering favorable evidence are: Borge and Rattsø (2004) and Milanovic (2000), on the positive effect of in- equality on redistributive policy; and Carlsen, Langset and Rattsø (2005), Hallerberg and Basinger (1998) and Hays (2003), on the attenuating effects of capital mobility and veto players. For the speciϐic case of Latin American countries, there is consistent evidence that capital mobility negatively affects the government’s redistributive effort (KAUFMAN and SEGURA-UBIERGO, 2001; RUDRA, 2002; WIBBELS, 2006) and suggestive evidence that the existence of veto players restricts the government’s ability to change redistributive policies (HUBER and STEPHENS, 2012)3. Veto players The idea that the government cannot credibly commit to more moderate redis- tribution strongly depends on the assumption that the decision-making process operates according to the majoritarian model of democracy, according to which the institutional structure favors the election of a single-party majority and the implementation of its pol- icy agenda. In this scenario, the median preference in the electorate is converted into pub- lic policy almost automatically. This is not the case in the model of democracy known as ‘consensual’, which seeks to ‘share, disperse and limit power’ (LIJPHART, 1999). Consen- sual democracies promote the formation of multi-party government coalitions and afford a more relevant role to the legislature in policy-making. This leads to decision-making processes that contemplate minority preferences, meaning there is no guarantee that the electorate’s median preference will be converted into public policy. e0005 - 8/31 (2019) 13 (1) e0005 - 8/31 Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr bpsr At ϐirst glance, these relations seem to offer a promising basis for explaining the in- tensityofthe redistributivepolicies implementedbythe recentleftistgovernmentsinLatin America. Figure 01 illustrates how the capital mobility and pro-elite veto player variables relate to the exemplary cases of radical (Bolivia, Ecuador, and Venezuela) and moderate left (Brazil, Chile, and Uruguay) governments. We observe radical redistributive policies only in countries with economies that heavily depend on the commodities sector and in 3The literature on the redistributive effects of capital mobility and veto players is extensive. Since a compre- hensive evaluation is beyond the purposes of this article, we simply refer to the main works that establish empirical support for these. (2019) 13 (1) e0005 - 9/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn which the left has won an absolute majority in the legislature – in other words, where capital mobility is low and there was no pro-elite veto player4. On the other hand, we ob- serve moderate policies in countries with more complex economies (Brazil and Uruguay) or where the median party in the legislature was not left-leaning (Brazil and Chile)5. which the left has won an absolute majority in the legislature – in other words, where capital mobility is low and there was no pro-elite veto player4. On the other hand, we ob- serve moderate policies in countries with more complex economies (Brazil and Uruguay) or where the median party in the legislature was not left-leaning (Brazil and Chile)5. which the left has won an absolute majority in the legislature – in other words, where capital mobility is low and there was no pro-elite veto player4. On the other hand, we ob- serve moderate policies in countries with more complex economies (Brazil and Uruguay) or where the median party in the legislature was not left-leaning (Brazil and Chile)5. Figure 1: Capitalmobility,pro-elitevetoplayers, andredistributivepolicyinselectedcountries Note: Moderate; *Radical. Source: Elaborated by the authors. 1: Capitalmobility,pro-elitevetoplayers, andredistributivepolicyinselectedcountries Figure 1: Capitalmobility,pro-elitevetoplayers, andredistributivepolicyinselectedcountries Note: Moderate; *Radical. Source: Elaborated by the authors. Source: Elaborated by the authors. This explanation is partly in line with that provided by Campello (2015, Chap. 02) who uses the same theoretical approach to highlight the attenuating effect of inter- national ϐinancial integration on redistributive policies. 4In Bolivia, although the 2005 election did not produce a majority in the senate, the support declared by two opposing senators for the president and his party changed this situation in January 2006 (SANDI, 2006). In Ecuador, although the left won the majority vote in September 2007 exclusively to rewrite the constitution, it suspended the legislature and assumed its powers in November of that year (CONAGHAN, 2008, p. 57). 5Although they are part of the government coalition led by the left, the median parties in the Brazilian and Chilean legislatures – the Brazilian Democratic Movement Party (PMDB) and the Christian Democratic Party (PDC), respectively – are centrally-oriented. In Chile, non-directly elected Senators with conservative lean- ings were decisive from 1999 to 2005, due to the legislative stalemate between the government and the opposition. bpsr It also chimes with those pro- posed by Murillo, Oliveros and Vaishnav (2011), Nunes (2014) and Weyland (2009), in the sense that it emphasizes the attenuating effect of the legislative veto (as in the ϐirst two studies) and the dependence on commodities (like in the third). However, our ex- planation stands out for its emphasis on the effects of both economic ‘and’ institutional structures, according to which moderate redistributive policies are obtained due to suf- ϐiciently high capital mobility or the presence of a pro-elite legislative veto player. It is also important to point out how our interpretation of the effect of commod- ity economies on redistributive policies differs from that of rentier-states’ (WEYLAND, 2009). According to the former, radical redistributive policies were viable in countries in which the commodities sector carried more weight because capital mobility was lower and, therefore, majority governments (i.e., without a pro-elite veto player) could levy a higher tax on capital. According to the latter, those policies were viable not because the government levied a higher tax, but because it experienced a larger inϐlow of rev- enue stemming from the commodities sector, either by means of direct production or of participation in that sector (JENSEN and WANTCHEKON, 2004, p. 817; SMITH, 2004, pp. 233-234). According to the rentier-state perspective, therefore, governments from 4In Bolivia, although the 2005 election did not produce a majority in the senate, the support declared by two opposing senators for the president and his party changed this situation in January 2006 (SANDI, 2006). In Ecuador, although the left won the majority vote in September 2007 exclusively to rewrite the constitution, it suspended the legislature and assumed its powers in November of that year (CONAGHAN, 2008, p. 57). 5Although they are part of the government coalition led by the left, the median parties in the Brazilian and Chilean legislatures – the Brazilian Democratic Movement Party (PMDB) and the Christian Democratic Party (PDC), respectively – are centrally-oriented. In Chile, non-directly elected Senators with conservative lean- ings were decisive from 1999 to 2005, due to the legislative stalemate between the government and the opposition. (2019) 13 (1) e0005 - 10/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva commodity-based countries could adopt more intense redistributive policies without soaring redistributive conϐlict (MORRISON, 2009; SMITH, 2004). bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Fabiano Santos, Acir Almeida & Thiago Moreira da Silva commodity-based countries could adopt more intense redistributive policies without soaring redistributive conϐlict (MORRISON, 2009; SMITH, 2004). Studies on how leftist governments in commodity-based countries ϐinanced their social policies reveal mechanisms consistent with the redistributive approach. Accord- ing to Hogenboom (2012), Bolivian president Evo Morales increased the public sector’s share in the proϐits from two of the country’s largest natural gas ϐields, as well as the taxes and royalties from private companies producing this commodity, from 18 to 82 percent; Ecuadorian president Rafael Correa practically doubled the tax on windfall proϐits from transnational oil companies; and Venezuela under Chavez increased royalties and taxes on foreign companies from 01 to 30 percent and 34 to 50 percent, respectively. On the other hand, Letelier and Dávila (2015) show how the Chilean legislature blocked the socialist’s government aim to charge royalties on the copper industry right from the be- ginning of its term. It was only in 2005, ϐive years after its inauguration, that the Chilean government could get legislative approval for the tax change, but only after substituting the royalties for the more moderate alternative of a speciϐic tax on mining proϐits. Empirical analysis In this section, we perform statistical regressions to more systematically evalu- ate the empirical support for our explanation of the intensity of the redistributive poli- cies implemented by the Latin American leftist governments of the 2000s. To do so, we ϐirst need to identify the countries that were part of the left turn. The movement is widely regarded to have begun in 1998, with the ϐirst election of Hugo Chávez to the presidency of Venezuela, after which followed the electoral victo- ries of leftist presidents in several other countries in the region, some of them more than once. However, political events in 2015 and 2016 indicate that the movement may be waning (CASTANǂ EDA, 2016). Following Levitsky and Roberts (2011, p. 05), we consider only countries that elected at least one president who: 01. ran for ofϐice with an elec- toral platform centered on social policies, particularly the reduction of inequality; and 02. once elected, preserved a signiϐicant part of her platform in government. According to these criteria, we identiϐied 19 leftist presidents in 13 countries, in ofϐice between 1999 and 2015. Table 01 lists the countries and their respective presidents, per year of election and period in ofϐice. (2019) 13 (1) e0005 - 11/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Table 1: Leftist presidents, Latin America, 1999-2016 Country President Party Year of election Years in government Argentina N. Kirchner FPV-PJ 2003 5/2003-12/2007 C. Kirchner FPV-PJ 2007 12/2007-12/2011 2011 12/2011-12/2015 Bolivia Morales MAS 2005 1/2006-1/2010 2009 1/2010-1/2015 2014 1/2015-present Brazil Lula PT 2002 1/2003-12/2006 2006 1/2007-12/2010 Roussef PT 2010 1/2011-12/2014 2014 1/2015-5/2016a Chile Lagos PPD 2000 3/2000-3/2006 Bachelet PS 2006 3/2006-3/2010 2013 3/2014-present Dominican Rep. Data and methods Measuring the intensity of redistributive policy (henceforth IRP) is an extraordi- nary challenge because of the different tools and strategies for implementing such policy. The former includes taxation, social expenditures and regulations - for example, on mini- mum wages or land reform. The strategies are related to the deϐinition of the pace (either gradual or precipitous), the target population, tax progressivity and compensations. The proper evaluation of different redistributive policies requires considering these various aspects. To the best of our knowledge, however, there is no valid measure of overall IRP. To use actual income redistribution as proxy for IRP is not without problems when one considers the distinction between policies and outcomes. Since redistribution also de- pends on the operation of market forces, which are often difϐicult to identify and isolate, IRPs cannot be inferred directly from observed redistribution. In the case of the recent Latin American leftist governments, although there is reasonable agreement that their re- distributive policies have contributed signiϐicantly to the reduction of inequality, the mag- nitude of this contribution is still under debate (CORNIA, 2014; LOƵ PEZ-CALVA et al., 2014). Alternatively, several studies use social spending as a proxy for IRP, as pointed out in Mc- Carty and Pontusson (2011, p. 666). Although the former is only one aspect of the latter, there are compelling arguments and evidence that social spending strongly reϐlects the government’s redistributive effort (AVELINO, BROWN and HUNTER, 2005; CASTLES and OBINGER, 2007; HUBER and STEPHENS, 2012, Chap. 03). For these reasons, to test our explanation about the left’s IRP we estimate time-series-cross-section (TSCS) regression models of overall social spending as percentage of the Gross Domestic Product (GDP). Our sample only includes the thirteen countries listed in Table 01, and the thir- teen years from the period 2003-2015. This is sufϐicient for the purpose of our empirical investigation, which is testing the effects of the economic (capital mobility) and institu- tional (veto player) structures on the intensity of the redistributive policies implemented by Latin American leftist governments during the recent commodities boom. It is impor- tant to bear in mind that our goal is simply to evaluate whether these two variables account for differences between the policies enacted by these speciϐic governments, and not to test their relevance for redistributive policies in the region (much less in general). Empirical analysis Mejı́a PRD 2000 8/2000-8/2004 Ecuador Correa MPAIS 2006 1/2007-8/2009 2009 8/2009-5/2013 2013 5/2013-5/2017 El Salvador Funes FMLN 2009 6/2009-5/2014 Cerén FMLN 2014 6/2014-present Guatemala Colom UNE 2007 1/2008-1/2012 Nicaragua Ortega FSLN 2006 1/2007-1/2012 2011 1/2012-1/2017 2016 1/2017-present Paraguay Lugo APC 2008 8/2008-6/2012a Peru Humala PNP 2011 7/2011-7/2016 Uruguay Vázquez FA 2004 3/2005-2/2010 Mujica FA 2009 3/2010-2/2015 Vázquez FA 2014 3/2015-present Venezuela Chávez MVR 1998 2/1999-1/2001 2000 1/2001-1/2007 2006 1/2007-1/2013 Chávez/Maduro PSUV 2012 1/2013-presentb Notes: APC= Alianza Patriótica para el Cambio; FA= Frente Amplio; FMLN= Frente Farabundo Martı́ para la Liberación Nacional; FPV-PJ= Frente para la Victoria-Partido Justicialista; FSLN= Frente Sandinista de Liberación Nacional; MAS= Movimiento al Socialismo; MPAIS= Movimiento Patria Altiva i Soberana; MVR= Movimiento V [Fifth] República; PNP= Partido Nacionalista Peruano; PPD= Partido por la Democracia; PRD= Partido Revolucionario Dominicano; PS= Partido Socialista; PSUV= Partido Socialista Unido de Venezuela; PT= Partido dos Trabalhadores; UNE= Unidad Nacional de la Esperanza; a Impeached; b Maduro took ofϐice following Chávez’s death, on March 2013. Table 1: Leftist presidents, Latin America, 1999-2016 Table 1: Leftist presidents, Latin America, 1999-2016 Notes: APC= Alianza Patriótica para el Cambio; FA= Frente Amplio; FMLN= Frente Farabundo Martı́ para la Liberación Nacional; FPV-PJ= Frente para la Victoria-Partido Justicialista; FSLN= Frente Sandinista de Liberación Nacional; MAS= Movimiento al Socialismo; MPAIS= Movimiento Patria Altiva i Soberana; MVR= Movimiento V [Fifth] República; PNP= Partido Nacionalista Peruano; PPD= Partido por la Democracia; PRD= Partido Revolucionario Dominicano; PS= Partido Socialista; PSUV= Partido Socialista Unido de Venezuela; PT= Partido dos Trabalhadores; UNE= Unidad Nacional de la Esperanza; a Impeached; b Maduro took ofϐice following Chávez’s death, on March 2013. Notes: APC= Alianza Patriótica para el Cambio; FA= Frente Amplio; FMLN= Frente Farabundo Martı́ para la Liberación Nacional; FPV-PJ= Frente para la Victoria-Partido Justicialista; FSLN= Frente Sandinista de Liberación Nacional; MAS= Movimiento al Socialismo; MPAIS= Movimiento Patria Altiva i Soberana; MVR= Movimiento V [Fifth] República; PNP= Partido Nacionalista Peruano; PPD= Partido por la Democracia; PRD= Partido Revolucionario Dominicano; PS= Partido Socialista; PSUV= Partido Socialista Unido de Venezuela; PT= Partido dos Trabalhadores; UNE= Unidad Nacional de la Esperanza; a Impeached; b Maduro took ofϐice following Chávez’s death, on March 2013. Sources: Political Database of the Americas; Murillo, Oliveros and Vaishnav (2011). (2019) 13 (1) e0005 - 12/31 (2019) 13 (1) e0005 - 12/31 Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr p g 8According to the results from the following tests: the Breusch-Pagan test of cross-sectional heterogene- ity, likelihood-ratio test of panel heteroskedasticity, Pesaran’s test of contemporaneous correlation (EBER- HARDT, 2011), Wooldridge’s test of panel serial correlation, and joint F test of no year-speciϐic-effect. We performed all tests and estimations in Stata 11. 7We used the statistical package Amelia II, for the R software. Data and methods This lim- ited objective derives directly from the scholarly debate about the differences between the radical and the moderate left, as discussed in the ϐirst and second sections of this article. Besides, there is no reason to expect any bias from the exclusion of countries where no left- ist president was elected, since they cannot add relevant information for the evaluation of our explanation as to why the left adopted radical policies in some countries, but not in others6. 6The sample does include center and right-of-center presidents, since only a few countries were governed by the left during the whole 2003-2015 period. Keeping those observations was necessary to preserve the time-series-cross-section structure of the data. (2019) 13 (1) e0005 - 13/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Graph 01 shows the evolution of social spending in each of the thirteen countries of the left turn, during the 2003-2015 period. The data is from the Economic Commission for Latin American and the Caribbean (ECLAC). Although there are some gaps in the se- ries - Peru (2003-2004), Paraguay (2003-2006), and especially in Uruguay (2003-2010) and Venezuela (2010-2015) - we can see that social spending increased in almost all coun- tries, most notably in Argentina, Chile, Ecuador and Paraguay. Graph 02 shows the average yearly change (in percentage points) in total social spending by country, during the leftist governments (as identiϐied in Table 01). There are substantial differences: the left sub- stantially increased social spending in most of the countries, though much more intensely in Paraguay (under president Lugo) and Argentina (under Néstor and Cristina Kirchner). h 1: Annualsocialspendingas%ofGDP,perselectedcountries(LatinAmerica, 2003-2015 Graph 1: Annualsocialspendingas%ofGDP,perselectedcountries(LatinAmerica, 2003-20 Graph 1: Annualsocialspendingas%ofGDP,perselectedcountries(LatinAmerica, 2003-2015) Source: Economic Commission for Latin American and the Caribbean (ECLAC) at http://estadisticas.cepal. org/cepalstat/web_cepalstat/estadisticasindicadores.asp?idioma=i. Source: Economic Commission for Latin American and the Caribbean (ECLAC) at http://estadisticas.cepal. org/cepalstat/web_cepalstat/estadisticasindicadores.asp?idioma=i. (2019) 13 (1) e0005 - 14/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Graph 2: Average yearly change in social spending (as % of GDP) during leftist govern- ments, by country (Latin America, 2003-2015) Source: Economic Commission for Latin American and the Caribbean (ECLAC) at http://estadisticas.cepal. org/cepalstat/web_cepalstat/estadisticasindicadores.asp?idioma=i. Data and methods (2019) 13 (1) e0005 - 15/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn The main purpose of the ϐixed-effects speciϐication is to avoid bias in the coefϐicient es- timates due to correlation between independent variables and omitted (non-observed) factors that affect the dependent variable at the group-level. This strategy may come with great costs though. By accounting for group-level variation, it reduces the model’s capacity to estimate the effects of time-invariant independent variables. Besides, the estimation of country-speciϐic effects may substantially inϐlate the standard errors, thus reducing the ef- ϐiciency of coefϐicient estimates from variables that explain within-group variation. It is the ϐirst of these costs that concerns us the most, since capital mobility, one of our explanatory variables, shows low longitudinal variation. To evaluate whether a model required ϐixed effects, we performed Mundlak’s (1978) procedure, which is robust to heteroskedasticity and serial correlation, unlike the usual Hausman test (WOOLDRIDGE, 2010, p. 332)9. The results indicate that, for most speciϐications, there is no signiϐicant difference between the estimates from ϐixed and random effects models. In such cases, we use the latter strategy, for efϐiciency reasons. bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn The main purpose of the ϐixed-effects speciϐication is to avoid bias in the coefϐicient es- timates due to correlation between independent variables and omitted (non-observed) factors that affect the dependent variable at the group-level. This strategy may come with great costs though. By accounting for group-level variation, it reduces the model’s capacity to estimate the effects of time-invariant independent variables. Besides, the estimation of country-speciϐic effects may substantially inϐlate the standard errors, thus reducing the ef- ϐiciency of coefϐicient estimates from variables that explain within-group variation. It is the ϐirst of these costs that concerns us the most, since capital mobility, one of our explanatory variables, shows low longitudinal variation. To evaluate whether a model required ϐixed effects, we performed Mundlak’s (1978) procedure, which is robust to heteroskedasticity and serial correlation, unlike the usual Hausman test (WOOLDRIDGE, 2010, p. 332)9. The results indicate that, for most speciϐications, there is no signiϐicant difference between the estimates from ϐixed and random effects models. Data and methods Graph 2: Average yearly change in social spending (as % of GDP) during leftist govern- ments, by country (Latin America, 2003-2015) Graph 2: Average yearly change in social spending (as % of GDP) during leftist govern- ments, by country (Latin America, 2003-2015) Source: Economic Commission for Latin American and the Caribbean (ECLAC) at http://estadisticas.cepal. org/cepalstat/web_cepalstat/estadisticasindicadores.asp?idioma=i. Source: Economic Commission for Latin American and the Caribbean (ECLAC) at http://estadisticas.cepal. org/cepalstat/web_cepalstat/estadisticasindicadores.asp?idioma=i. As mentioned above, out of the total 169 country-year dyads, there are 20 missing values in the social spending series, most of them for Uruguay (eight observations) and Venezuela (six observations). We substituted these missing observations, as well as a few others in the independent variables, using Honaker and King’s (2010) method of multiple imputation for TSCS data. Multiple imputation uses the information from the observed portions of the dataset, including many potentially relevant variables, to generate multi- ple values for the missing cells. Honaker and King’s (2010) algorithm considers the usual characteristics of TSCS data, which are heterogeneity and correlation on the longitudinal or the cross-sectional dimensions (or both), thus generating more ‘reasonable’ estimates for the missing values. In our application, we generated ϐive complete imputed datasets using all the variables included in the regression analyses as well as their ϐirst-order lags7. To proper specify a regression model for TSCS data, it is ϐirst necessary to eval- uate the presence of unit heterogeneity, heteroskedasticity and error dependency. Using the complete dataset resulting from replacing the missing values with the average of the (ϐive) corresponding multiple-imputation values, we found strong evidence of country het- erogeneity, heteroskedasticity, contemporaneous and serial correlation, but no evidence of time heterogeneity8. The modelling of country heterogeneity, whether as ϐixed or random effects, is a controversial issue (CLARK and LINZER, 2015; PLUǆ MPER, TROEGER and MANOW, 2005). p g , 8According to the results from the following tests: the Breusch-Pagan test of cross-sectional heterogene- ity, likelihood-ratio test of panel heteroskedasticity, Pesaran’s test of contemporaneous correlation (EBER- HARDT, 2011), Wooldridge’s test of panel serial correlation, and joint F test of no year-speciϐic-effect. We performed all tests and estimations in Stata 11. 9Using the Stata module ‘mundlak’, from Perales (2013). g , ( ) 10Brooks and Manza (2006) show that persistence in social spending is due to mass policy preferences, which have a high degree of temporal stability. According to the authors, “this characteristic does not disqual- ify [other temporally-stable variables] as causal factors behind social policymaking” (Brooks and Manza, 2006, p. 822). Data and methods In such cases, we use the latter strategy, for efϐiciency reasons. How to treat serial correlation in TSCS data is another controversial issue. There are two basic strategies: adjusting the standard errors or including the lagged dependent variable as a correlate. The ϐirst approach treats time dependence as a nuisance, whereas the second, as substantive information. Beck and Katz’s (1996) defend the second ap- proach, pointing that it ‘causes researchers to think about the dynamics of their model’. On the other hand, Achen (2000) calls attention to the fact that this strategy makes the co- efϐicient estimates of theoretically relevant variables vanish, since the correlation between the dependent variable and its lag is usually very high, even when there is no substantive reason to expect the ϐirst to be a function of the latter. For the speciϐic case of social spend- ing, we agree with Achen (2000, p. 02) that there are no theoretically substantive grounds for including the lagged term, and thus estimate (static) models that simply adjust the stan- dard errors for serial correlation10. In any case, we do check the robustness of the results by also including the (ϐirst) lag of the dependent variable. Another time dependence issue is non-stationarity, a process through which the statistical properties of a time series are functions of (as opposed to ϐixed in) time. Re- gressing a nonstationary variable on another one may produce statistically signiϐicant re- lationship between them even if they are not causally related, for they are both functions of time. Since the dependent variable shows a clear longitudinal pattern (Graph 01), we tested for panel data unit roots (a feature of non-stationarity) and were able to conϐidently reject it11. Using the Stata module mundlak, from Perales (2013). 10Brooks and Manza (2006) show that persistence in social spending is due to mass policy preferences, which have a high degree of temporal stability. According to the authors, “this characteristic does not disqual- ify [other temporally-stable variables] as causal factors behind social policymaking” (Brooks and Manza, 2006, p. 822). g ( ) 10Brooks and Manza (2006) show that persistence in social spending is due to mass policy preferences, which have a high degree of temporal stability. According to the authors, “this characteristic does not disqual- ify [other temporally-stable variables] as causal factors behind social policymaking” (Brooks and Manza, 2006, p. 822). 12It should be noted that when the number of clusters is small (as in our sample), standard errors may be biased downwards. Esarey and Menger (2018) provide a method for solving this potential bias, but it does not work with random effects or time series, which characterize our data. However, it is somewhat comforting that their simulations show that random effects perform better than any cluster adjustment method when the model is correctly speciϐied (Esarey and Menger, 2018, p. 02). 15Available at: http://psephos.adam-carr.net/. Data and methods p ) ing to Levin, Lin and Chu’s (2002) test. By rejecting non-stationarity, we do not need to consider mic) models based on ϐirst differences or error correction. (2019) 13 (1) e0005 - 16/31 Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Finally, to adjust for the error structure we use Driscoll and Kraay’s (1998) stan- dard errors (DKSE) with a correction for ϐirst-order serial correlation. DKSE is an adapta- tion of the (heteroskedasticity and autocorrelation consistent) Newey-West estimator that is also consistent in the presence of contemporaneous correlation. DKSE’s method is very much like the usual panel corrected standard errors (PCSE) used by Beck and Katz (1995), in the sense that it adjusts the covariance matrix for the same data characteristics but pre- serves the least squares estimator of the coefϐicients. However, unlike PCSE, DKSE also allows for the estimation of random (country-speciϐic) effects, which is the main reason why we decided to use it12. 14Available at: http://pdba.georgetown.edu/. 13Available at http://spot.colorado.edu/ bakerab/data.html. y p ( y g p ) 13Available at http://spot.colorado.edu/ bakerab/data.html. Variables Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn the absolute difference between the median’s ideology and the president’s whenever the for- mer is located to the right of the latter, and zero otherwise. The sample mean and standard deviation are, respectively, 2.0 and 3.7, and the range goes from 0 to 15.1. Based on the hypothesis that the existence of a pro-elite veto player attenuates the IRP, we expect social spending to decrease with the ideological distance between the president and the pro-elite legislative veto player. the absolute difference between the median’s ideology and the president’s whenever the for- mer is located to the right of the latter, and zero otherwise. The sample mean and standard deviation are, respectively, 2.0 and 3.7, and the range goes from 0 to 15.1. Based on the hypothesis that the existence of a pro-elite veto player attenuates the IRP, we expect social spending to decrease with the ideological distance between the president and the pro-elite legislative veto player. We use two variables to test the effect of capital mobility: ‘Financial openness’ and ‘Commodity economy’. The ϐirst is the Chinn-Into yearly index, which measures a country’s degree of capital account openness, based on restrictions on cross-border ϐi- nancial transactions (CHINN and ITO, 2006)16. The sample mean, standard deviation and range are, respectively, 71, 31 and [0,100]. The second variable is the weight of the com- modities sector in the economy, measured as the percentage share of the agricultural and mineral sectors in the GDP, based on data from ECLAC17. For this variable the re- spective summary statistics are: 16.7, 8.0, and [6.3, 48.8]. By assumption, capital mobil- ity is increasing on ϐinancial openness and decreasing on commodity economy. For the sake of clarity of interpretation, we thus use the complement of the second variable (Non- commodity economy = 100-commodity economy), so that, from the hypothesis that the IRP is decreasing on capital mobility, we expect ϐinancial openness or non-commodity economy (or both) to negatively affect social spending. Our explanation also posits that either capital mobility or the existence of a pro-elite veto player is sufϐicient to reduce the left’s IRP. In econometric terms, it means that the expected (negative) effect of each of those factors is decreasing on the values of the other. Variables We include independent variables that account for our main explanatory factors - capital mobility and the ideological distance between the president and the pro-elite leg- islative veto player -, alternative explanations, and some controls emphasized by previous analyses of social expenditures. ‘President’sideology’istheestimateofthepresident’slocationontheWiesehomeier- Benoit’s left-right scale (WIESEHOMEIER and BENOIT, 2009), which ranges from 01 (ex- treme left) to 20 (extreme right). The estimates are based on perception data from expert surveys about parties and presidents’ positions on policy dimensions, including taxes and spending. For our purposes, it is important to note that respondents ranked this dimension about average in overall importance, and it ranks last in terms of party divergence, charac- teristics that led the authors to conclude that “political actors do not differentiate their po- sitions much on this economic policy issue” (WIESEHOMEIER and BENOIT, 2009. p. 1438). Therefore, we can safely assume that our measure of ideology reϐlects neither the president’s nor political parties’ propensities to spend. The data are from Andy Baker’s dataset of ‘Latin American Election Results with Party Ideology Scores 2.0’ (BAKER and GREENE, 2011)13. The sample mean and standard deviation are, respectively, 8.8 and 4.9, and the range goes from 2.0 to 18.5. We thus expect the president’s ideology (in this case, her degree of con- servatism) to have a negative effect on social spending. However, since the sample consists mostly of leftist presidents, there may not be enough variation to capture this effect. Usingthesamedataset, complementedwithinformationfromthe‘PoliticalDatabase of the Americas’14 and Adam Carr’s election archive,15 we identiϐied the median party in the lower chamber of the legislature. The variable ‘Distance from the pro-elite veto player’ equals e0005 - 17/31 (2019) 13 (1) bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn the absolute difference between the median’s ideology and the president’s whenever the for- mer is located to the right of the latter, and zero otherwise. The sample mean and standard deviation are, respectively, 2.0 and 3.7, and the range goes from 0 to 15.1. Based on the hypothesis that the existence of a pro-elite veto player attenuates the IRP, we expect social spending to decrease with the ideological distance between the president and the pro-elite legislative veto player. 16Available at http://web.pdx.edu/ ito/Chinn-Ito_website.htm. 17Available at http://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/Portada.asp. Available at http://web.pdx.edu/ ito/Chinn-Ito_website.htm. 17Available at http://estadisticas.cepal.org/cepalstat/WEB_CEPALSTAT/Portada.asp. 16Available at http://web.pdx.edu/ ito/Chinn-Ito_website.htm. Variables Therefore, we also test the inter- actions between the veto player distance and the capital mobility variables, for which we expect positive effects. In what relates to the (alternative) explanations discussed in Section 02, the in- clusion of non-commodity economy allows for an indirect test of Weyland’s (2009) hy- pothesis that the left’s redistributive policies were more intense in countries that ben- eϐited more from the commodities boom, due to the larger inϐlow of nontax revenues. However, as opposed to ours, this explanation implies a null effect for the interaction be- tween that variable and the veto player distance. Following Murillo, Oliveros and Vaish- nav (2011), to test the positive effect of capital inϐlow from exports we also include the variable ‘Current account’, which is the sum of net exports of goods and services as per- centage of GDP, based on data from the World Bank. The variable ‘Popular vote’, which is the percentage of votes received by the president in the ϐirst round of the election, cap- tures Biglaiser’s (2016) hypothesis that IRP increased on the size of the popular support for (leftist) presidents. On the other hand, to account more speciϐically for the potential negative effect of electoral competition on the left’s IRP (BARRILLEAUX, HOLBROOK and e0005 - 18/31 (2019) 13 (1) bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva LANGER 2002; GARAY 2016) we also include ‘Margin of victory’ which is the difference bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva LANGER, 2002; GARAY, 2016), we also include ‘Margin of victory’, which is the difference between the vote shares received by the president and the runner-up in the ϐirst round of the election. Fabiano Santos, Acir Almeida & Thiago Moreira da Silva LANGER, 2002; GARAY, 2016), we also include ‘Margin of victory’, which is the difference between the vote shares received by the president and the runner-up in the ϐirst round of the election. LANGER, 2002; GARAY, 2016), we also include ‘Margin of victory’, which is the difference between the vote shares received by the president and the runner-up in the ϐirst round of the election. Finally, as controls, we include economic and socio-demographic variables that previous studies indicate as potentially relevant (AVELINO, BROWN and HUNTER, 2005; HUBER and STEPHENS, 2012; KAUFMAN and SEGURA-UBIERGO, 2001). Variables The ϐirst and most important is ‘trade openness’ (measured as total imports and exports as percent- age of GDP), which is common place in studies that evaluate the impact of globalization on governmental spending, even though the direction of the effect is the object of much con- troversy, with some authors arguing that it should decrease social spending due to market pressures, and others that social spending in fact increases to compensate for higher expo- sure to external shocks (GARRETT, 1998; RODRIK, 1998). ‘GDP per capita’ (in purchasing power parity, per $1,000) controls for the level of economic development and its expected positive impact. This variable enters the model lagged by one year to avoid endogeneity. The level of unemployment, the percentage of the population aged 65 or older, and the per- centage of the population living in urban areas are usual controls for the demand side of social spending. However, we include only the ϐirst of these variables, since the other two are extremely correlated between them and with GDP per capita (all pairwise correlations are above 0.80). Lati 50] 50]* 11] 67] 35] 15] 17]* 74]* 90] error ced Results Table 02 presents the results from ϐive TSCS models of social spending. Models 01 through 03 treat country heterogeneity as random effects, in accordance with the test re- sults reported at the bottom of the table. The ϐirst and second models estimate the effects of the president’s share of the popular vote and her margin of victory separately, due to the strong correlation (r= 0.74) between these variables. We keep popular vote in the sub- sequent models since its inclusion contributes slightly more to the adjusted R2. Models 03 and 04 test separately the (partial and interactive) effects of ϐinancial openness and non- commodity economy, the variables that capture the concept of capital mobility. Again, the reason for not testing them simultaneously is to avoid collinearity issues. The fourth spec- iϐication is the one that best ϐits the data, accounting for one third of the variance. Since this speciϐication is the only one for which country ϐixed effects cannot be rejected (with 90% conϐidence), we report the results from both the random (4A) and the ϐixed effects (4B) versions. (2019) 13 (1) e0005 - 19/31 Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Latin American countries (2003-2015) Random effects Random effects Fixed effects (3) (4A) (4B) 50]* -0.087 [0.046]* -0.073 [0.045] -0.104 [0.049] 50]* -0.117 [0.053] -0.141 [0.052]* -0.171 [0.053]*** 11] -0.013 [0.013] — — 67] — -0.027 [0.052] 0.017 [0.092] -0.001 [0.002] — — — 0.014 [0.008]* 0.013 [0.007]* 35] 0.014 [0.037] -0.004 [0.038] 0.007 [0.035] 15] 0.023 [0.016] 0.028 [0.014]* 0.026 [0.013]* — — — 17] -0.036 [0.016] -0.028 [0.016]* -0.043 [0.017] 74] 0.256 [0.067]* 0.301 [0.072]* 0.289 [0.071]*** 90] -0.011 [0.086] 0.003 [0.092] 0.029 [0.067] 9.35 7.86 0.223 0.076 0.076 0.288 0.315 0.437 169 169 169 errors with AR1. Pooled estimates based on ϐive multiple-imputed datasets.Notes: ocedure; c. Overall R2 for Models 01 through 4A, and within R2 for model 4B; p < e0005 - 20/31 : * : ] ]* ] ]* ] ]* ] ]* ] tes: p < * : * * : 74]** 84] (20 Fabiano Santos, Acir Almeida & Thiago Moreira da Silva Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr The coefϐicient estimates, standard errors and signiϐicance levels are based on the point estimates generated from the ϐive multiple-imputed (MI) datasets, according to the rules spec- iϐied in Rubin and Schenker (1986, p. Results 367). The (pooled) coefϐicient estimates are the average of the MI coefϐicient estimates, their standard errors are an increasing function of the MI vari- ance estimates (within-imputation variance) and the variance of the MI coefϐicient estimates (between-imputation variance), and the signiϐicance tests are based on an approximation of the t statistic with degrees of freedom that depend on the relative sizes of the within and between MI variances. The evidence provides weak to moderate support for the partisan hypothesis that so- cial spending decreased as the president’s ideological position moved to the right. Increasing the president’s location on the Wiesehomeier-Benoit scale by one unit decreased annual social spending by about 0.10 percentage points (p.p.). More importantly, there is strong support for the hypothesis that social spending decreased as the ideological distance between the president and the pro-elite legislative veto player increased, and weak support for the hypothesis that this effect was attenuated by capital mobility, measured as the weight of the non-commodity sector in the economy. On the other hand, there is no support for the hypothesis that social spending was lower where capital mobility is higher. Graph 03 illustrates the estimate of the total effect of the distance from the pro-elite veto player, evaluated at different levels of non-commodity economy. When the latter variable equals 50% (near its sample minimum), the estimated effect is -0.61, signiϐicant at the 05% level (standard error= 0.251). The magnitude of the estimate decreases as non-commodity economy increases, such that when the latter is at its sample average (83%) the former equals -0.17, sig- niϐicant at the 01% level (the partial coefϐicient estimate from model 4B). The estimated effect is not signiϐicant only for the sample values of non-commodity economy greater than 88%. Graph 3: Estimated effect of Distance from pro-elite veto player, at values of non- commodity economy 3: Estimated effect of Distance from pro-elite veto player, at values of non- Graph 3: Estimated effect of Distance from pro-elite veto player, at values of non- commodity economy y y Obs.: Based on the estimates from model 4B. Dotted lines represent the 95% conϐidence interval. Source: Elaborated by the authors. (2019) 13 (1) e0005 - 21/31 Obs.: Based on the estimates from model 4B. Dotted lines represent the 95% conϐidence interval. Source: Elaborated by the authors. Results (2019) 13 (1) Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn To assess the substantive relevance of these effects we calculated the expected change in social spending from meaningful changes in the independent variables. Thus, in- creasing the ideology of leftist presidents (i.e., with values below ten on the Wiesehomeier- Benoit scale) from its sample minimum, 2.0 (El Salvador since 2009), to its maximum, 8.4 (Argentina), we estimate that yearly social spending increases by an average of 0.64 p.p., which corresponds to a fraction of .22 of its sample standard deviation (s.d.). Using the same criteria, by increasing the distance from the pro-elite veto player from zero to 15 (El Salvador, 2009-2011) we estimate yearly social spending to decrease on average by 9.15 p.p. (3.14 s.d.) when non-commodity economy is near its sample minimum, and by 2.55 p.p. (0.88 s.d.) when it is at its sample average. Thus, the estimated impact of the president’s ideology can be regarded as small to moderate, whereas the impact of her di- vergence with the pro-elite veto player, moderate to high. Among the alternative explanations, there is only weak support for Biglaiser’s (2016) electoral mandate hypothesis that the IRP increased with the (leftist) president’s share of the popular vote. The evidence does not lend support to the hypotheses that the IRP increased with the capital inϐlow from (net) exports (MURILLO, OLIVEROS and VAISH- NAV, 2011), with the nontax revenue from the commodities sector (WEYLAND, 2009) or that it was higher where the left faced weaker electoral competition (i.e., a higher margin of victory). Regarding the control variables, only trade openness and GDP per capita are statistically signiϐicant. The ϐirst had a negative effect on social spending, which contra- dicts the compensation hypothesis (GARRETT, 1998; RODRIK, 1998), whereas the second had an (expected) positive effect. Unemployment, meanwhile, had no signiϐicant effect. In sum, the econometric evidence lends partial support to our explanation of the intensity of redistributive policies implemented by the Latin American left during the re- centcommoditiesboom. Thereismoderateevidencethatsocialspendingwaslowerwhere the president’s ideological preference was more to the right, and strong evidence that it decreased with the ideological distance between the president and the (pro-elite) legisla- tive veto player. There is weak evidence that the degree of capital mobility, measured as the share of the non-commodity sector in the economy, attenuated the veto-player effect. Results Given the behavior of the R2, this loss of signiϐicance is more likely to reϐlect collinearity issues. Finally, the only consequence of excluding the observations from Venezuela and Uruguay is to strengthen the signiϐicance of the coefϐicient estimate for that interaction, which is a strong indication that the results are not driven by the imputations. to expect current social spending to be a function of past social spending. Therefore, we interpret the sensitivity of the results to this speciϐication as a sample characteristic. Con- trolling for either urban or aged populations, instead of GDP per capita, substantially de- creases the ϐitness of the models (for example, the adjusted R2 for model 4A reduces from 0.32 to 0.14 and 0.19, respectively), but only the inclusion of the second variable affects our results, making the interaction in model 4A no longer statistically signiϐicant. Given the behavior of the R2, this loss of signiϐicance is more likely to reϐlect collinearity issues. Finally, the only consequence of excluding the observations from Venezuela and Uruguay is to strengthen the signiϐicance of the coefϐicient estimate for that interaction, which is a strong indication that the results are not driven by the imputations. Results However, there is no evidence that social spending decreased with capital mobility. To evaluate the robustness of these results, we performed additional analyses, considering each of the following: 01. including the ϐirst lag of the dependent variable; 02. substituting GDP per capita either for the percentage of the population living in ur- ban areas or the percentage aged at least 65, two variables that other studies include but which are highly correlated with income per capita in our sample; and 03. excluding from the sample Venezuela and Uruguay, the two countries with the highest number of missing values in the dependent variable. The inclusion of the lag of the dependent variable causes the statistical signiϐicance of the coefϐicient estimates to vanish, with the only notable ex- ception being that of GDP per capita. As we argued before, there is no theoretical reason (2019) 13 (1) e0005 - 22/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva to expect current social spending to be a function of past social spending. Therefore, we interpret the sensitivity of the results to this speciϐication as a sample characteristic. Con- trolling for either urban or aged populations, instead of GDP per capita, substantially de- creases the ϐitness of the models (for example, the adjusted R2 for model 4A reduces from 0.32 to 0.14 and 0.19, respectively), but only the inclusion of the second variable affects our results, making the interaction in model 4A no longer statistically signiϐicant. Given the behavior of the R2, this loss of signiϐicance is more likely to reϐlect collinearity issues. Finally, the only consequence of excluding the observations from Venezuela and Uruguay is to strengthen the signiϐicance of the coefϐicient estimate for that interaction, which is a strong indication that the results are not driven by the imputations. Fabiano Santos, Acir Almeida & Thiago Moreira da Silva to expect current social spending to be a function of past social spending. Therefore, we interpret the sensitivity of the results to this speciϐication as a sample characteristic. Con- trolling for either urban or aged populations, instead of GDP per capita, substantially de- creases the ϐitness of the models (for example, the adjusted R2 for model 4A reduces from 0.32 to 0.14 and 0.19, respectively), but only the inclusion of the second variable affects our results, making the interaction in model 4A no longer statistically signiϐicant. Summary and concluding remarks Besides, the evidence that legislative veto players contributed to attenuating redistributive policy con- (2019) 13 (1) e0005 - 23/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn tradicts arguments that cast doubt on the relevance of formal political institutions in Lat America, particularly those designed to constrain the chief executive (O’DONNELL, 1994 Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn tradicts arguments that cast doubt on the relevance of formal political institutions in Latin America, particularly those designed to constrain the chief executive (O’DONNELL, 1994). Regarding the political debate, our ϐindings shed light on the recent conϐlicts and transformations that Latin American democracies have experienced. Regarding the for- mer, we offer an interpretation that emphasizes their redistributive nature, locating the conϐlicts among the different elements that comprise the political dynamics of very un- equal democracies, instead of blaming particularities of the region. This more general per- spective may be useful to reformers who look for remedies based on the international ex- perience. The ϐindings also allow us to reinterpret some of the political phenomena charac- teristic of the Latin American left turn, such as, for example, institutional transformations aimed at broadening and deepening direct popular participation in politics (AVRITZER, 2009; POGREBINSCHI, 2013). Mechanisms such as referenda and plebiscites, to which several radical leftist governments resorted once elected, can be understood as a resource to circumvent pro-elite veto players in the legislature. Participatory institutions, by their turn, can be thought of as a mechanism for reducing the costs to the poor of collective action, increasing their capacity to preserve leftist policies under future right-of-center governments. Finally, our explanation needs to be applied with the utmost caution to the recent reshufϐling of leftist governments in Latin America and the emergence of an aggressive right-wing agenda in countries like Brazil, Argentina and Chile. In this article, we identi- ϐied the reasons why radical redistributive policies were adopted in countries like Ecuador and Venezuela, instead of in Brazil or Chile. The same reasons need not apply to the reverse movement - i.e., the extent to which the new right-wing governments will be able to roll back earlier welfare-based policies. Summary and concluding remarks Based on theoretical models that analyze the implications of redistribution in highly unequal democracies, this article offered an explanation for the differences in the intensity of redistributive policies implemented by Latin American leftist governments during the 2000s and 2010s. In contrast with explanations that focus on region-speciϐic factors, we claimed that more radical policies were adopted by the left only in countries where capital mobility was sufϐiciently low and there was no legislative pro-elite veto player. To test our explanation, we estimated TSCS models of social spending, using a sample of thirteen Latin American countries over the thirteen years between 2003 and 2015. The countries are all those where at least one leftist president was elected (the only ones relevant for testing differences among leftist governments), and where the time pe- riod roughly coincides with the commodities boom, when conditions for redistribution were favorable throughout the region. The results provide partial support for our hy- potheses: there is strong evidence that social spending decreased with the ideological distance between the president and the (pro-elite) legislative veto player, but only weak evidence that the degree of capital mobility attenuated this effect. Moreover, the redis- tributivetheory of democracy seems toϐitthe facts better thanthe mainalternativeoffered by the literature, namely the rentier state theory (WEYLAND, 2009), according to which the intensity of redistributive policy is conditioned not by the existence of economic and political constraints on the taxation of the wealthy elite, but by the growth of the govern- ment’s non-tax revenue, due to higher commodity prices. Our ϐindings provide potential contributions to both the academic and the politi- cal debates. With respect to the former, we add new empirical evidence that strongly sup- ports the veto players theory and partially supports the redistributive theory of democ- racy. Although testing these theories was not the purpose of this study, the results en- courage their application to the recent context of Latin American politics. Summary and concluding remarks Following Pierson’s seminal work on policy retrench- ment (PIERSON, 1996), we believe that the factors that explain variations in the intensity of redistributive policies are not the same as those that help us ϐigure out why the reversal of these policies may occur in a more profound way in some Latin American countries than in others. Revised by Matthew Richmond Submitted on October 06, 2017 Accepted on January 19, 2019 References ACEMOGLU, Daron and ROBINSON, James A. (2001), A theory of political transitions. American Economic Review. Vol. 91, Nº 04, pp. 938-963. ACEMOGLU, Daron and ROBINSON, James A. (2001), A theory of political transitions. American Economic Review. Vol. 91, Nº 04, pp. 938-963. ACHEN, Chirstopher H. (2000), Why lagged dependent variables can suppress the explana- tory power of other independent variables. Paper presented at the Annual Meeting of the Political Methodology Section of the American Political Science Association, UCLA. e0005 - 24/31 bpsr Department of Political Science and Institute for Social Research. University of Michi- gan. ALESINA, Alberto and RODRIK, Dani (1994), Distributive politics and economic growth. Quarterly Journal of Economics. Vol. 109, Nº 02, pp. 465-490. ALESINA, Alberto and RODRIK, Dani (1994), Distributive politics and economic growth. Quarterly Journal of Economics. Vol. 109, Nº 02, pp. 465-490. ARDITI, Benjamin (2008), Arguments about the left turns in Latin America: a post-liberal politics? Latin American Research Review. Vol. 43, Nº 03, pp. 59-81. AVELINO, George; BROWN, David S., and HUNTER, Wendy (2005), The effects of capital mobility, trade openness, and democracy on social spending in Latin America, 1980– 1999. American Journal of Political Science. Vol. 49, Nº 03, pp. 625-641. AVRITZER, Leonardo (2009), La nueva izquierda, la crisis de representación y la partici- pación social en América Latina. In: La ’nueva izquierda’ en América Latina: derechos humanos, participación polı́tica, y sociedad civil. Edited by ARNSON, Cynthia J.; AR- MONY, Ariel C.; SMULOVITZ, Catalina; CHILLIER, Gastón; PERUZZOTTI, Enrique, and COHEN, Giselle. Washington: The Woodrow Wilson Center Press. pp. 177-192. BAKER, Andy and GREENE, Kenneth F. (2011), The Latin American left’s mandate: free- market policies and issue voting in new democracies. World Politics. Vol. 63, Nº 01, pp. 43-77. BARRILLEAUX, Charles; HOLBROOK, Thomas, and LANGER, Laura (2002) Electoral com- petition, legislative balance, and American state welfare policy. American Journal of Political Science. Vol. 46, Nº 02, pp. 415-427. BATES, Robert H. and LIEN, Da-Hsiang Donald (1985), A note on taxation, development and representative government. Politics & Society. Vol. 14, Nº 01, pp. 53-70. BECK, Nathaniel and KATZ, Jonathan N. (1996), Nuisance vs. substance: specifying and estimating time-series-cross-section models. Political Analysis. Vol. 06, pp. 01-36. BECK, Nathaniel and KATZ, Jonathan N. (1995), What to do (and not to do) with time-series cross-section data. American Political Science Review. Vol. 89, Nº 03, pp. 634-647. References BIGLAISER, Glen (2016), Mandate and the market: policy outcomes under the left in Latin America. Comparative Politics. Vol. 48, Nº 02, pp. 185-204. BOIX,Carles(2003), Democracyandredistribution.NewYork: CambridgeUniversityPress. 284 pp.. BORDIGNON, Massimo; TOMMASO, Nannicini, and TABELLINI, Guido (2016), Moderating political extremism: single round versus runoff elections under plurality rule. Ameri- can Economic Review. Vol. 106, Nº 08, pp. 2349-2370. (2019) 13 (1) e0005 - 25/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn BORGE, Lars-Erik and RATTSØ, Jørn (2004), Income distribution and tax structure: em- pirical test of the Meltzer–Richard hypothesis. European Economic Review. Vol. 48, Nº 04, pp. 805-826. bpsr bpsr BORGE, Lars-Erik and RATTSØ, Jørn (2004), Income distribution and tax structure: em- pirical test of the Meltzer–Richard hypothesis. European Economic Review. Vol. 48, Nº 04, pp. 805-826. BORGE, Lars-Erik and RATTSØ, Jørn (2004), Income distribution and tax structure: em- pirical test of the Meltzer–Richard hypothesis. European Economic Review. Vol. 48, Nº 04, pp. 805-826. BRETSCHGER, Lucas and HETTICH, Frank (2002), Globalization, capital mobility and tax competition: theory and evidence for OECD countries. European Journal of Political Economy. Vol. 18, Nº 04, pp. 695-716. BROOKS, Clem and MANZA, Jeff (2006), Why do welfare states persist? The Journal of Politics. Vol. 68, Nº 04, pp. 816-827. CAMPELLO, D. (2015), The politics of market discipline in Latin America: globalization and democracy. New York: Cambridge University Press. 254 pp.. CARLSEN, Fredrik; LANGSET, Bjørg, and RATTSØ, Jørn (2005), The relationship between ϐirm mobility and tax level: empirical evidence of ϐiscal competition between local gov- ernments. Journal of Urban Economics. Vol. 58, Nº 02, pp. 273-288. CASTANǂ EDA, Jorge G. (2016), The death of the Latin American left. The New York Times The Opinion Pages. Available at ˂https://www.nytimes.com/2016/03/23/opinion/the- death-of-the-latin-american-left.html˃. Accessed on March 15, 2018. CASTANǂ EDA, Jorge G. (2006), Latin America’s left turn: a tale of two lefts. Foreign Affairs. Vol. 85, Nº 28, pp. 28-43. CASTLES, Francis G. and OBINGER, Herbert (2007), Social expenditure and the politics of redistribution. Journal of European Social Policy. Vol. 17, Nº 03, pp. 206-222. CHINN, Menzie D. and ITO, Hiro (2006), What matters for ϐinancial development? Capital controls, institutions, and interactions. Journal of Development Economics. Vol. 81, Nº 01, pp. 163-192. CLARK, Tom S. and LINZER, Drew A. (2015), Should I use ϐixed or random effects? Political Science Research and Methods. Vol. 03, Nº 02, pp. 399-408. CONAGHAN, Catherine M. References e0005 - 27/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn bpsr HUNTER, Wendy and SUGIYAMA, Natasha Borges (2009), Democracy and social policy in Brazil: advancing basic needs, preserving privileged interests. Latin American Politics and Society. Vol. 51, Nº 02, pp. 29-58. JENSEN, Nathan and WANTCHEKON, Leonard (2004), Resource wealth and political regimes in Africa. Comparative Political Studies. Vol. 37, Nº 07, pp. 816-841. KAUFMAN, Robert R. and SEGURA-UBIERGO, Alex (2001), Globalization, domestic politics, and social spending in Latin America: a time-series cross-section analysis, 1973–97. World Politics. Vol. 53, Nº 04, pp. 553-587. KENNEMORE, Amy and WEEKS, Gregory (2011), Twenty-ϐirst century socialism? The elu- sive search for a post-neoliberal development model in Bolivia and Ecuador. Bulletin of Latin American Research. Vol. 30, Nº 03, pp. 267-281. LETELIER S., Leonard. E. and DAƵVILA A., Mireya (2015), The political economics of tax reform in Chile. New Political Economy. Vol. 20, Nº 06, pp. 832-850. LEVIN, Andrew; LIN, Chien-Fu, and CHU, Chia-Shang James (2002), Unit root tests in panel data: asymptotic and ϐinite-sample properties. Journal of econometrics. Vol. 108, Nº 01, pp. 01-24. LEVITSKY, Steven and ROBERTS, Kenneth M. (eds)(2011), Latin America’s ’left turn’: a framework for analysis. In: The resurgence of the Latin American left. Baltimore: The Johns Hopkins University Press. pp. 01-30. LIJPHART, Arend (1999), Patterns of democracy: government forms and performance in thirty-six democracies. New Haven: Yale University Press. 368 pp.. LIPSET, Seymour Martin (1981), Political man: the social bases of politics. Baltimore: The John Hopkins University Press. 608 pp.. LOƵ PEZ-CALVA, Luis F. and LUSTIG, Nora (eds)(2010), Explaining the decline in inequality in Latin America: technological change, educational upgrading and democracy. In: De- clining inequality in Latin America: a decade of progress? New York and Washington: United Nations Development Program and Brookings Institution Press. pp. 03-24. LOƵ PEZ-CALVA, Luis F.; LUSTIG, Nora; PESSINO, Carola, and SCOTT, John (2014), The impact of taxes and social spending on inequality and poverty in Argentina, Bolivia, Brazil, Mexico, Peru, and Uruguay: introduction to the special issue. Public Finance Review. Vol. 42, Nº 03, pp. 287-303. MADRID, Raúl L. (2010), The origins of the two lefts in Latin America. Political Science Quarterly. Vol. 125, Nº 04, pp. 587-609. References (2008), Ecuador: Correa’s plebiscitary presidency. Journal of Democracy. Vol. 19, Nº 02, pp. 46-60. CORNIA, Giovanni Andrea (ed)(2014), Inequality trends and their determinants: Latin America over the period 1990-2010. In: Falling inequality in Latin America: policy changes and lessons. Oxford: Oxford University Press. pp. 23 – 48. CORRALES, Javier (2010), The repeating revolution: Chávez’s new politics and old eco- nomics. In: Leftist governments in Latin America: successes and shortcomings. Edited Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr bpsr by WEYLAND, Kurt; MADRID, Raúl L., and HUNTER, Wendy. New York: Cambridge University Press. pp. 28-56. DRISCOLL, John C. and KRAAY, Aart C. (1998), Consistent covariance matrix estimation with spatially dependent panel data. Review of economics and statistics. Vol. 80, Nº 04, pp. 549-560. EBERHARDT, Marcus (2011), XTCD: Stata module to investigate variable/residual cross- section dependence. Statistical Software Components S457237, Boston College Depart- ment of Economics. ESAREY, Justin and MENGER, Andrew (2018), Practical and effective approaches to dealing with clustered data. Political Science Research and Methods. pp. 01-19. doi: 10.1017/psrm.2017.42. FLORES-MACIƵAS, Gustavo A. (2010), Statist vs. pro-market: explaining leftist govern- ments’ economic policies in Latin America. Comparative Politics. Vol. 42, Nº 04, pp. 413-433. GARAY, Candelaria (2016), Social policy expansion in Latin America. New York: Cambridge University Press. 406 pp.. GARRETT, Geoffrey (1998), Global markets and national politics: collision course or vir- tuous circle? International organization. Vol. 52, Nº 04, pp. 787-824. HALLERBERG, Mark and BASINGER, Scott (1998), Internationalization and changes in tax policy in OECD countries: the importance of domestic veto players. Comparative Polit- ical Studies. Vol. 31, Nº 03, pp. 321-352. HAYS, Jude C. (2003), Globalization and capital taxation in consensus and majoritarian democracies. World Politics. Vol. 56, Nº 01, pp. 79-113. HOGENBOOM, Barbara (2012), Depoliticized and repoliticized minerals in Latin America. Journal of Developing Societies. Vol. 28, Nº 02, pp. 133-158. HONAKER, James and KING, Gary (2010), What to do about missing values in time-series cross-section data. American Journal of Political Science. Vol. 54, Nº 02, pp. 561-581. HUBER, Evelyne and STEPHENS, John D. (2012), Democracy and the left: social policy and inequality in Latin America. Chicago: University of Chicago Press. 368 pp.. HUBER, Evelyne; MUSTILLO, Thomas, and STEPHENS, John D. (2008), Politics and social spending in Latin America. The Journal of Politics. Vol. 70, Nº 02, pp. 420-436. References (2019) 13 (1) (2019) 13 (1) e0005 - 28/31 bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva bpsr bpsr McCARTY, Nolan and PONTUSSON, H. Jonas (2011), The political economy of inequality and redistribution. In: The Oxford handbook of economic inequality. Edited by NOLAN, Brian; SALVERDA, Wiemer, and SMEEDING, Timothy M.. London: Oxford University Press. pp. 665-692. MELTZER, Allan H. and RICHARD, Scott F. (1981), A rational theory of the size of the gov- ernment. Journal of Political Economy. Vol. 89, Nº 05, pp. 914-927. MILANOVIC, Branko (2000), The median-voter hypothesis, income inequality, and income redistribution: an empirical test with the required data. European Journal of Political Economy. Vol. 16, Nº 03, pp. 367-410. MORRISON, Kevin M. (2009), Oil, nontax revenue, and the redistributional foundations of regime stability. International Organization. Vol. 63, Nº 01, pp. 107-138. MUNDLAK, Yair (1978), On the pooling of time series and cross section data. Economet- rica: Journal of the Econometric Society. Vol. 46, Nº 01, pp. 69-85. MURILLO, Maria Victoria; OLIVEROS, Virginia, and VAISHNAV, Milan (2011), Economic constraints and presidential agency. In: The resurgence of the Latin American left. Edited by LEVITSKY, Steven and ROBERTS, Kenneth M.. Baltimore: The Johns Hopkins University Press. pp. 52-70. NORTH, Douglass C. and WEINGAST, Barry R. (1989), Constitutions and commitment: the evolution of institutions governing public choice in seventeenth-century England. The Journal of Economic History. Vol. 49, Nº 04, pp. 803-832. NUNES, Felipe (2014), The left variance: how electoral mechanisms shape leftist presiden- tial behavior in Latin America. Revista Ibero-Americana de Estudos Legislativos. Vol. 01, pp. 25-34. O’DONNELL, Guilhermo A. (1994), Delegative democracy. Journal of Democracy. Vol. 05, Nº 01, pp. 55-69. PERALES, Francisco (2013), MUNDLAK: Stata module to estimate random-effects regres- sions adding group-means of independent variables to the model. Statistical Software Components S457601. Boston College Department of Economics. PERSSON, Torsten and TABELLINI, Guido (1994), Representative democracy and capital taxation. Journal of Public Economics. Vol. 55, Nº 01, pp. 53-70. PIERSON, Paul (1996), The new politics of the welfare state. World Politics. Vol. 48, Nº 02, pp. 143-179. (2019) 13 (1) e0005 - 29/31 bpsr Capital Mobility, Veto Players, and Redistribution in Latin America During the Left Turn PLUǆ MPER, Thomas; TROEGER, Vera E., and MANOW, Philip (2005), Panel data analysis in comparative politics: Linking method to theory. European Journal of Political Research. Vol. 44, Nº 02, pp. 327-354. References bpsr bpsr PLUǆ MPER, Thomas; TROEGER, Vera E., and MANOW, Philip (2005), Panel data analysis in comparative politics: Linking method to theory. European Journal of Political Research. Vol. 44, Nº 02, pp. 327-354. POGREBINSCHI, Thamy (2013), The pragmatic turn of democracy in Latin America. Paper presented in the American Political Science Association Annual Meeting. Chicago, IL. PRIBBLE, Jennifer and HUBER, Evelyne (2013), Social policy and redistribution: Chile and Uruguay. In: The resurgence of the Latin American left. Edited by LEVITSKY, Steven and ROBERTS, Kenneth M.. Baltimore: The Johns Hopkins University Press. pp. 117-138. RITTINGER, Eric R. and CLEARY, Mattew (2013), Confronting coup risk in the Latin Ameri- can left. Studies in Comparative International Development. Vol. 48, Nº 04, pp. 403-431. RODRIK, Dani (1998), Why do more open economies have bigger governments? Journal of Political Economy. Vol. 106, Nº 05, pp. 997-1032. RUBIN, Donald B. and SCHENKER, Nathaniel (1986), Multiple imputation for interval esti- mation from simple random samples with ignorable nonresponse. Journal of the Amer- ican statistical Association. Vol. 81, Nº 394, pp. 366-374. RUDRA, Nita (2002), Globalization and the decline of the welfare state in less-developed countries. International Organization. Vol. 56, Nº 02, pp. 411-445. SANDI, R. (2006) Partido de Evo Morales se asegura control del Congreso boliviano. Nacion.com. Available at ˂www.nacion.com/ln_ee/2006/enero/18/ultima-la5.html˃. Accessed on January 18, 2006. SMITH, Benjamin (2004), Oil wealth and regime survival in the developing world, 1960– 1999. American Journal of Political Science. Vol. 48, Nº 02, pp. 232-246. TSEBELIS, George (1995), Decision making in political systems: veto players in presiden- tialism, parliamentarism, multicameralism and multipartyism. British Journal of Polit- ical Science. Vol. 25, Nº 03, pp. 289-325. WEYLAND, Kurt G. (2009), The rise of Latin America’s two lefts: insights from rentier state theory. Comparative Politics. Vol. 41, Nº 02, pp. 145-164. WIBBELS, Erik (2006), Dependency revisited: international markets, business cycles, and social spending in the developing world. International Organization. Vol. 60, Nº 02, pp. 433-468. WIESEHOMEIER, Nina and BENOIT, Kenneth (2009), Presidents, parties, and policy com- petition. The Journal of Politics. Vol. 71, Nº 04, pp. 1435-1447. (2019) 13 (1) bpsr Fabiano Santos, Acir Almeida & Thiago Moreira da Silva WOOLDRIDGE, Jeffrey M. (2010), Econometric analysis of cross section and panel data. Cambridge: The MIT Press. 776 pp.. bpsr WOOLDRIDGE, Jeffrey M. (2010), Econometric analysis of cross section and panel data. Cambridge: The MIT Press. 776 pp.. (2019) 13 (1) e0005 - 31/31
https://openalex.org/W3157696138	https://periodicos.ufsc.br/index.php/geosul/article/download/69947/45838	Portuguese	null	O Programa “Produtor de Água” como proposta de fortalecimento socioeconômico e de recuperação dos recursos naturais	Geosul	2,021	cc-by	8,834	1 Universidade Federal de Goiás, Jataí, Goiás, Brasil, mainaracosta@gmail.com, 0000-0002-4709- 0897 2 Universidade Federal de Goiás, Jataí, Goiás, Brasil, evandrospfc@gmail.com, 0000-0003-1499- 8388 THE “WATER PRODUCER” PROGRAM AS A PROPOSAL FOR SOCIOECONOMIC STRENGTHENING AND FOR THE RECOVERY OF NATURAL RESOURCES THE “WATER PRODUCER” PROGRAM AS A PROPOSAL FOR SOCIOECONOMIC STRENGTHENING AND FOR THE RECOVERY OF NATURAL RESOURCES Abstract: This article aims to analyze the socio environmental effects of the territorialization of capitalist agriculture in the municipality of Rio Verde, in State of Goiás, in order to understand its reflexes in social, environmental and economic spheres. To this end, a theoretical survey was first conducted on the Brazilian agrarian question and later the acquisition of secondary data from IBGE and primary data through field collections, and interviews with farmers and secretaries of the municipality. The territorialization of agriculture in the municipality of Rio Verde is understood today as an extremely homogenizing and concentrating process of wealth, leading to the marginalization of family farmers and, at the same time, to an excessive use of natural resources, jeopardizing biodiversity and the quality of local water resources. We identified that the “modernization of agricultural activities” proposal was based solely on economic growth, leading to social and environmental unsustainability, thus, intensifying socio-spatial inequalities and different local territorial disputes. p Keywords: Agrarian Question. Capitalist Agriculture. Cerrado Biome. p eywords: Agrarian Question. Capitalist Agriculture. Cerrado Biome. Geosul, Florianópolis, v. 36, n. 78, p. 356-380, jan./abr. 2021. https://doi.org/10.5007/2177-5230.2021.e69947 O PROGRAMA “PRODUTOR DE ÁGUA” COMO PROPOSTA DE FORTALECIMENTO SOCIOECONÔMICO E DE RECUPERAÇÃO DOS RECURSOS NATURAIS 1 Mainara da Costa Benincá1 Evandro César Clemente2 Resumo: O presente artigo tem como objetivo analisar o programa “Produtor de Água” - PPA no município de Rio Verde-Goiás, de forma a averiguar em que medida este tem contribuído para o fortalecimento socioeconômico dos envolvidos, bem como na recuperação dos recursos naturais locais. Como estratégia metodológica, realizou-se um levantamento teórico com base na compreensão de conceitos em torno da temática e posteriormente um levantamento a campo para a obtenção de dados de fonte primária. Com base nas informações coletadas, pode-se dizer que o PPA está consolidado em uma área de monoculturas, onde o agronegócio está efetivado, sendo um grande desafio assegurar o desenvolvimento socioeconômico e ambiental, fortalecer o agricultor, dar suporte para a melhoria de vida da população e promover desse modo um uso consciente dos recursos ambientais. Palavras-chave: Agronegócio. Recursos Ambientais. Sustentabilidade. EL PROGRAMA “AGUA” COMO PROPUESTA DE FORTALECIMIENTO SOCIOECONÓMICO Y RECUPERACIÓN DE RECURSOS NATURALES Resumen: Este artículo tiene como objetivo analizar los efectos socioambientales de la territorialización de la agricultura capitalista en Río Verde-GO, en la comprensión de los reflejos en las esferas sociales, ambientales y económicas. Para esto, primero se realizó una encuesta teórica sobre la cuestión agraria brasileña y, más tarde, 357 BENINCÁ, CLEMENTE sobre la adquisición de datos secundarios del IBGE y datos primarios a través de la recogida de datos en campo, con entrevistas con agricultores y secretarios de la municipalidad. La territorialización de la agricultura en el municipio de Río Verde se entiende hoy como un proceso extremadamente homogeneizador y concentrador de riquezas, que conduce a una marginación de los agricultores familiares y al mismo tiempo a un uso excesivo de los recursos naturales, poniendo en riesgo la biodiversidad y la calidad de los recursos hídricos locales. Se identifica que la propuesta de "modernización de las actividades agrícolas" se basó únicamente en el crecimiento económico, lo que condujo a la insostenibilidad social y ambiental, intensificando las desigualdades socio-espaciales y las diferentes disputas territoriales locales. Palabras clave: Cuestión Agraria. Agricultura Capitalista. Bioma Cerrado INTRODUÇÃO O estudo de uma bacia hidrográfica faz-se necessário para compreender de forma sistêmica o uso e ocupação de uma determinada área. De acordo com Tucci (2009) uma bacia hidrográfica pode ser entendida como uma área definida topograficamente, drenada por um curso de água ou um sistema conectado de cursos de água, de forma que toda a vazão efluente seja descarregada em uma única saída. A compreensão de uma bacia hidrográfica deve partir de uma visão integrada e não compartimentada, sendo os elementos pertencentes totalmente conectados entre si. De acordo com estudos de Tundisi e Matsumura-Tundisi (2011) a concepção de bacia hidrográfica é a unidade mais correta no que diz respeito ao gerenciamento e otimização de usos múltiplos, principalmente no que diz respeito aos recursos hídricos para a sua conservação. O planejamento e o gerenciamento de uma bacia hidrográfica devem partir de uma visão integrada dos diversos elementos que a compõe e que inevitavelmente se interconectam. A proposta do presente artigo é compreender o Programa Produtor de Água, instalado na Bacia Hidrográfica Ribeirão Abóbora - BHRA, nas suas contribuições socioeconômicas e ambientais. Para tal em um primeiro momento realizou-se a caracterização da BHRA nos seus diferentes usos da terra, destacando a presença de extensas áreas de lavouras temporárias, de áreas com predominância de pastagens, áreas de vegetação remanescente de Cerrado e outras. Após caracterização da bacia, a proposta foi abordar o Programa “Produtor de Água” - PPA, executado na mesma localizada no município de Rio Verde-GO a partir de 2011, após consecutivas tentativas da população local para recuperação das áreas degradadas. 358 BENINCÁ, CLEMENTE Em um segundo momento abordar-se-á o PPA no Município de Rio Verde- GO, partindo do seu processo de implantação para compreender a execução e as etapas desenvolvidas do mesmo. Serão apresentadas de forma detalhada as três etapas do programa, bem como a localização e a caracterização da área de estudo no Município de Rio Verde-GO. Por fim, apresentar-se-á uma avaliação dos proprietários do PPA sobre as ações do programa, no que tange aos aspectos econômicos, sociais e ambientais, tendo em vista a efetivação do mesmo há oito anos no local. Com base nos dados coletados, busca-se uma análise do PPA Rio Verde como uma proposta de desenvolvimento rural sustentável para o município. A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) A bacia hidrográfica pode ser considerada um sistema físico onde a entrada é o volume de água precipitado e a saída é o volume de água escoado pelo exutório, considerando-se como perdas intermediárias os volumes evaporados e transpirados e também os infiltrados profundamente. Em evento isolado pode-se desconsiderar estas perdas e analisar a transformação de chuva em vazão feita pela bacia (TUCCI, 2001, p. 41). Esses e outros indicadores podem ser utilizados para analisar as condições de uma determinada bacia hidrográfica. Tundisi e Matsumura-Tundisi (2011) enumeram alguns dos indicadores das condições que podem fornecer o índice de qualidade de uma bacia hidrográfica: Qualidade da água dos rios e riachos; espécies de peixes e vida selvagem (fauna terrestre) presentes; taxa de preservação ou de perda de áreas alagadas; taxa de preservação ou de perda das florestas nativas; taxa de preservação ou contaminação das fontes de abastecimento de água; taxa de urbanização (% de área da bacia hidrográfica); relação – população urbana/população rural (TUNDISI; MATSUMURA-TUNDISI, 2011, p. 155). Além dos indicadores de qualidade, Tundisi e Matsumura-Tindisi (2011), consideram também os indicadores de vulnerabilidade nas interferências em uma bacia hidrográfica, sendo: os poluentes tóxicos; as cargas de poluentes; a descarga urbana; a descarga agrícola; alterações na população; efeitos gerais das atividades humanas e os potenciais de eutrofização. Desse modo, identifica-se a necessidade de compreender a dinâmica nas bacias hidrográficas, tendo em vista que o crescimento populacional, juntamente com as diferentes atividades econômicas, intensifica o uso dos recursos hídricos e também os diferentes usos dos solos, comprometendo a qualidade de vida em um local. 359 BENINCÁ, CLEMENTE Visando manter um equilíbrio ambiental em uma bacia hidrográfica, os aspectos humanos e físicos precisam ser pensando de modo integrado. O alto processo de degradação dos recursos naturais atualmente no Brasil reflete as condições de desequilíbrio ambiental presentes nessas bacias. Nesse trabalho, o recorte espacial de estudo é a Bacia Hidrográfica do Ribeirão Abóbora (BHRA), localizada na área central do Município de Rio Verde-GO, nas coordenadas geográficas de Latitude: 8028500m N; 8039000m N Longitude: 496000n E; 504500m E, com uma área de 199,05 km², um perímetro de 78,25 km e um comprimento axial 29,27 km, conforme apresentado no mapa 3 (ALVES, 2016). Mapa 1 – Localização da Bacia Hidrográfica Ribeirão Abóbora – Rio Verde (GO) Fonte: Sistema de coordenadas UTM. Fuso 22 Sul, SIEGAS2000. SIEG, 2019, IBGE 2018. Org.: QUEIROZ JÚNIOR, V. S.; BENINCÁ, M.C. (2018). A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) Fonte: Sistema de coordenadas UTM. Fuso 22 Sul, SIEGAS2000. SIEG, 2019, IBGE 2018. Org.: QUEIROZ JÚNIOR, V. S.; BENINCÁ, M.C. (2018). A BHRA possui forte influência antrópica, sendo cortada por importantes rodovias, anel viário, plantas agroindustriais e também cercada por loteamentos urbanos. O relevo é uma das características que favorecem a mecanização da agricultura, possuindo uma formação suavemente ondulada, com uma altitude média de 777,88 m e uma amplitude altimétrica de 240 m (ALVES, 2016). O clima da região é marcado por duas estações bem definidas, uma seca (de maio a outubro) e outra chuvosa (novembro a abril), com índices pluviométricos elevados, sendo que no período chuvoso, ocorre uma precipitação média mensal 360 BENINCÁ, CLEMENTE acima de 200 mm e temperatura média de 24ºC. Já no período seco essa precipitação mensal é menor que 50 mm, com temperatura média de 22ºC (INMET, 2018). acima de 200 mm e temperatura média de 24ºC. Já no período seco essa precipitação mensal é menor que 50 mm, com temperatura média de 22ºC (INMET, 2018). A declividade da Bacia do Ribeirão Abóbora é predominantemente suave ondulada com 106,86 km2, seguida de uma declividade plana com 56.53 km2, ondulado de 34,48 km² e, com menos representatividade, forte ondulada 1,21 km² e montanhosa de apenas 0,02 km² (ALVES, 2016). Os três tipos de solos predominantes na bacia do Ribeirão das Abóboras são os: Latossolos, Argissolos e Neossolos, os quais, dentro de suas limitações, são utilizados para atividades agrícolas (ALVES, 2016). No decorrer do trabalho de campo, os proprietários da BHRA relataram a necessidade de utilizarem correções para os solos, com o uso de calcário (corrigir a acidez) e adubos químicos (corrigir a fertilidade). A BHRA era coberta por uma vegetação típica de Cerrado, formada por um mosaico de paisagens naturais com diferentes fitofisionomias de savanas estacionais (DIAS, 1996). A área de vegetação remanescente é de apenas 41,39 km2, que corresponde a 20,8% da área da BHRA, localizada geralmente em torno dos cursos fluviais e, de acordo com Alves (2016) diversos trechos do Ribeirão Abóbora não estão em conformidade quanto aos limites legais do Código Ambiental Municipal de Rio Verde (GO), Lei 5.090/2005. O alto da BHRA possui uma vegetação mais densa, com a predominância de espécies arbóreo arbustiva e o restante da bacia identifica-se uma vegetação mais arbustiva. A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) Essas características físicas (relevo, vegetação e solos) da bacia determinam parcialmente as atividades desenvolvidas em cada área, sendo a maior concentração de pecuária extensiva na parte superior e a de agricultura nas demais áreas. A disponibilidade hídrica do Ribeirão Abóbora é significativa, sendo o responsável pela captação de 70% da água para o abastecimento da população do Município de Rio Verde (BRICEÑO; SOUZA, 2013). Características essas que favoreceram a implantação de atividades diversas na área da BHRA, como a mineração de cascalho, a pecuária, a agricultura, atividades agroindustriais, dentre outros, impactando de forma negativa os recursos naturais da mesma. O mapeamento da Bacia do Ribeirão Abóbora foi obtido por meio do software ArcGIS 10.1, licenciado para o Laboratório de Geoinformação, com uma classificação não supervisionada por segmentação e classificação por máxima 361 BENINCÁ, CLEMENTE verossimilhança, com posterior correção visual com auxílio do Google Earth Pro. Utilizaram-se de imagens Landsat 8 (canais 6R5G4B), com resolução de 30 metros, para junho de 2017. verossimilhança, com posterior correção visual com auxílio do Google Earth Pro. Utilizaram-se de imagens Landsat 8 (canais 6R5G4B), com resolução de 30 metros, para junho de 2017. O Mapa 2 apresenta o uso e cobertura da terra na Bacia Hidrográfica do Ribeirão Abóbora, caracterizando-a com uma extensa área destinada à agricultura, uma significativa área de pecuária e outras de vegetação, silvicultura e de granjas. Mapa 2 – Uso e cobertura da terra da BHRA em Rio Verde (GO) – 2017 p ( ) Fonte: Sistema de Coordenadas UTM, imagem Ladsat8. Org.: MARTINS, A.P. (2019). Fonte: Sistema de Coordenadas UTM, imagem Ladsat8. Org.: MARTINS, A.P. (2019). A Bacia Hidrográfica do Ribeirão Abóbora é cortada pela BR-060 e pelo anel viário que circunda a cidade e liga a BR-060 à GO-174, importante acesso aos demais municípios que integram a cadeia de grãos do estado de Goiás. Quando analisados os dados do uso e cobertura da terra da bacia, em porcentagem, evidenciam-se as principais áreas, sendo em 1ª lugar para a agricultura (54%), em 2º lugar a vegetação remanescente (23,7%) em 3º lugar a pastagem (13,8 %) e em 4º lugar a área construída (5%), de acordo com a Tabela 1. A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) Tabela 1 – Uso e cobertura da terra da Bacia Hidrográfica Ribeirão Abóbora em Rio Verde (GO) - 2017 2017 Área total da Bacia 5.248,6ha 100% Água 6,4 0,1% Tabela 1 – Uso e cobertura da terra da Bacia Hidrográfica Ribeirão Abóbora em Rio Verde (GO) - 2017 2017 Área total da Bacia 5.248,6ha 100% Água 6,4 0,1% Tabela 1 – Uso e cobertura da terra da Bacia Hidrográfica Ribeirão Abóbora em Rio Verde (GO) - 2017 2017 Área total da Bacia 5.248,6ha 100% Água 6,4 0,1% 362 BENINCÁ, CLEMENTE Vegetação remanescente 1236, 6 23,7% Agricultura 2831, 8 54% Pastagem 726, 5 13,8% Silvicultura 119,0 2,3% Solo descoberto 59,3 1,1% Área construída 258,4 5% Fonte: Sistema de Coordenadas UTM, imagem Landsat8 – 2017. Org.: BENINCÁ, M. C. (2019). Por meio do levantamento a campo, identificou-se que os principais usos da terra na Bacia Hidrográfica Ribeirão Abóbora são: a pecuária leiteira nas áreas com maior declividade e a agricultura mecanizada nas áreas com relevo plano a suavemente ondulado, havendo o cultivo de soja no período de safra e o de milho no período da “safrinha”, conforme apresentado nas Fotografias 1 e 2. Fotografia 1 - Pecuária leiteira em propriedade da Bacia Hidrográfica Ribeirão Abóbora g p p g Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017). Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017). A pecuária leiteira é uma atividade típica dos proprietários da região, inclusive sendo anterior à implantação das culturas de grãos. Atualmente está presente nas propriedades onde o relevo é de ondulado a suavemente ondulado, que dificulta a mecanização agrícola, impedindo a implantação da monocultura de grãos. A pecuária leiteira é praticada por meio de fortes investimentos em tecnologias, com a presença de assistência técnica e complementação da alimentação com rações minerais. Consequência desses altos investimentos é a 363 BENINCÁ, CLEMENTE produção diária de leite, chegando em alguns casos a 3.000 L/dia, com algumas propriedades tendo 100 vacas leiteiras, em média 30 l/dia por animal. A maior área da Bacia Hidrográfica Ribeirão Abóbora é atualmente utilizada para o cultivo de grãos, sendo a soja na principal safra (novembro a março) e o milho na safrinha (março a julho). Como já apresentado anteriormente, aproximadamente 122,8 km2 da bacia são destinados à agricultura, que anualmente expande sua área e produtividade. A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) Na Fotografia 2 é possível visualizar a produção de soja, em parte já colhida, em uma área de relevo plano, propícia à mecanização agrícola. Fotografia 2 - Soja (safra normal) atividade predominante nas áreas com terreno relativamente plano Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017) Fotografia 2 - Soja (safra normal) atividade predominante nas áreas com terreno relativamente plano Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017) Por meio de entrevistas com os proprietários da Bacia Ribeirão Abóbora, identificou-se que a produtividade média da soja é de 58 sacas/hectares e a produção média de milho é de aproximadamente 90 sacas/hectares na região, valores próximos da média de produtividade do Município de Jataí, 62 sacas/hectares de soja e 106 sacas/hectares de milho no ano de 2018. O cultivo desses grãos é baseado em equipamentos tais como: tratores, plantadeiras, pulverizadores, colheitadeiras e outros, sendo a produção comercializada em grande parte com a Caramuru e a Comigo. Na Fotografia 3 pode-se observar o cultivo do milho no período safrinha, sendo o segundo maior em produção e área plantada entre as lavouras temporárias. O mesmo é cultivado em grande parte após a colheita da soja e é responsável por 364 BENINCÁ, CLEMENTE uma parte significativa da renda dos proprietários da Bacia Ribeirão Abóbora, tendo em vista a quantidade produzida. uma parte significativa da renda dos proprietários da Bacia Ribeirão Abóbora, tendo em vista a quantidade produzida. ografia 3 – Milho (safrinha) plantado nas áreas antes ocupadas pela soja (safra norma Fotografia 3 – Milho (safrinha) plantado nas áreas antes ocupadas pela soja (safra normal) Fotografia 3 – Milho (safrinha) plantado nas áreas antes ocupadas pela soja (safra normal) Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017) Fonte: Trabalho de Campo, BENINCÁ, M.C. (2017) Além da agricultura e da pecuária leiteira, o uso da terra na Bacia Ribeirão Abóbora também compreende as granjas de aves e suínos, em função da instalação da Perdigão (atual BRFFOODS) no ano de 1999. Fruto do Projeto Buriti3 a empresa se instalou no município de Rio Verde com o objetivo de promover o deslocamento da agroindústria do Sul do país para o Centro-Oeste brasileiro, estimulada pelas produções de soja e milho regional (CARMO et al., 2002). De acordo com Carmo et al. sse p ojeto o p a tado po u a g a de e p esa do seg e to da p odução a co a, a e d gão S. A. no ano de 1998, e consistia em incentivar os pequenos produtores de frango a se tornarem integradores (a empresa fornece tecnologia necessária para produção e em contrapartida os produtores vendem os animais terminados exclusivamente para esses fornecedores) dessa empresa com a promessa da ascensão do negócio que muitas vezes era familiar (PRIA, 2001). 3 Esse projeto foi implantado por uma grande empresa do segmento da produção avícola, a P A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) (2002), a agroindústria da Perdigão do Município de Rio Verde é composta por cinco unidades, sendo: uma unidade de abate de aves; uma unidade de abate de suínos; uma unidade de fabricação de alimentos de origem, suína e de aves; uma fábrica de ração e um incubatório. Outras empresas foram sendo atraídas pela Perdigão, constituindo um importante gerador de emprego e renda para Rio Verde, porém, ocasionando problemas de ordens ambientais, sociais e mesmo de renda (desigualdade). Devido à fatores como a proximidade e as vantagens econômicas apresentadas, grande parte dos proprietários rurais da Bacia Ribeirão Abóbora se integrou à Perdigão, construindo os aviários e módulos de engorda de suínos. 365 BENINCÁ, CLEMENTE Segundo Alves (2016), 26 granjas estão instaladas na bacia, sendo voltadas tanto para a criação de aves e de suínos, sendo eles de porte médio ou grande para fornecimento de matéria-prima exclusivamente para a BRFFOODS. Segundo Alves (2016), 26 granjas estão instaladas na bacia, sendo voltadas tanto para a criação de aves e de suínos, sendo eles de porte médio ou grande para fornecimento de matéria-prima exclusivamente para a BRFFOODS. A silvicultura também é uma atividade presente na Bacia Ribeirão Abóbora, utilizada como barreiras protetoras no entorno das granjas de aves e suínos e como produtora de bioenergia para atender a COMIGO e a BRFFOODS (ALVES, 2016). Parte da área urbana de Rio Verde também está situada sobre a Bacia, contribuindo para comprometer a qualidade dos recursos naturais, tendo em vista o elevado crescimento populacional das últimas décadas. Os novos loteamentos representam uma nova “ameaça” à qualidade da água da BHRA e os mesmos tendem a se expandir frente ao crescimento populacional e a ação de grupos especuladores urbanos nos últimos anos. De acordo a Secretária Municipal de Meio Ambiente de Rio Verde, a abertura de novos loteamentos pode comprometer a qualidade dos recursos da BHRA, principalmente a redução das áreas de vegetação, de nascentes e possíveis contaminações dos solos e da água. Alguns loteamentos se encontram em fase de planejamento, mas, segundo a Secretária, não devem ser liberados para que o abastecimento de água da população Rio Verdense no futuro não seja comprometido. Devido a essas barreiras para com a abertura de novos loteamentos, algumas propriedades rurais da Bacia, por se localizarem próximas a zona urbana de Rio Verde, se tornaram chácaras para locações de eventos nos finais de semana. A caracterização da Bacia Hidrográfica Ribeirão Abóbora (BHRA) A área de vegetação remanescente totaliza 20,8% do total da BHRA, sendo constantemente substituída por lavouras temporárias, ou seja, a retirada parcial ou completa da vegetação tem avançado com o tempo na bacia. Identifica-se que a vegetação remanescente encontra-se principalmente no entorno dos cursos fluviais e que, devido à implantação do Programa Produtor de Água da Agência Nacional das Águas tem passado por um processo de preservação e/ou recuperação mais resistente. Desse modo, a ocupação da BHRA caracteriza-se por extensas áreas de lavouras temporárias, áreas de pastagens destinadas à pecuária em grande parte leiteira e de corte, áreas de vegetação remanescente de Cerrado e demais áreas com a silvicultura. Na sequência apresentar-se-á caracterização do Programa “Produtor de Água”, instalado na bacia do Ribeirão Abóbora. 366 BENINCÁ, CLEMENTE A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida O excessivo e inadequado uso dos recursos hídricos tem comprometido a sua qualidade e a quantidade disponível, levando a sociedade geral 367 BENINCÁ, CLEMENTE a questionamentos sobre maneiras de promover uma conservação desse bem, ou seja, um uso racional do mesmo. a questionamentos sobre maneiras de promover uma conservação desse bem, ou seja, um uso racional do mesmo. As reflexões a este respeito surgiram no município de Rio Verde ainda na década de 1990, mais precisamente no ano de 1995, quando proprietários rurais, juntamente com o Poder Público local, analisaram as possibilidades de adotarem um programa de recuperação de nascentes, contenção de erosões e assoreamentos, em prol da conservação do alto da Bacia Ribeirão Abóbora. A preocupação partiu pelo fato de a bacia ser responsável por mais de 70% o abastecimento de água do município e, pelo fato das mesmas e encontrarem-se em um elevado processo de degradação. Juntamente com essa Bacia, encontra-se a Microbacia Ribeirão da Laje, que integram a captação para o abastecimento urbano de Rio Verde, das plantas industriais e das propriedades rurais da região. Partindo dessa necessidade, proprietários do alto da Bacia Ribeirão Abóbora se uniram e buscaram auxílio junto com ao Poder Público local em prol da recuperação dessas áreas degradadas. A iniciativa partiu primeiramente de uma das proprietárias do alto da BHRA, diante da elevada degradação de sua área, devido à intensa exploração de cascalho. O desenvolvimento rural sustentável deve engendrar profundas transformações no âmbito político, social, econômico e ambiental e o que se identifica pelo programa de Rio Verde são medidas na ordem apenas ambiental e de forma pontual na esfera social e econômica. De acordo com Sorrentino (2011), faz- se necessário despertar em cada indivíduo o sentimento de pertencimento, participação e responsabilidade. Esse sentimento pode ser identificado dentre os proprietários, tendo em vista que o programa foi implantado no Município de Rio Verde em 2011, mas a mobilização dos proprietários ocorreu desde o ano de 1995 em busca da recuperação de áreas degradadas, diante da impossibilidade de uso pelo intenso processo de assoreamento, erosões dentre outros. A elaboração do mesmo ocorreu por meio de uma metodologia participativa e partindo de uma necessidade dos moradores. Sorrentino (2001) afirma que a questão ambiental se revela como uma construção histórica, não se trata de trabalhar um conhecimento fechado, mas sim em andamento, de forma participativa. A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida O Programa Produtor de Água (PPA) foi projetado tendo como origem a iniciativa da Agência Nacional de Águas (ANA), cujo objetivo era de reversão da situação de alta degradação ambiental, por meio de medidas mitigadoras para os problemas ambientais em diversas regiões brasileiras. O PPA é uma iniciativa da ANA voltada à conservação de recursos hídricos e se fundamenta na Política Nacional de Recursos Hídricos (Lei 9433/97) que dispõe, dentre suas diretrizes, “articulação da gestão de recursos hídricos com a do uso do solo” (ANA, 2018). Segundo a ANA (2012), o programa é um instrumento pelo qual a União apoia a melhoria, a recuperação e a proteção de recursos hídricos em bacias hidrográficas estratégicas, tendo como base ações executadas no meio rural voltadas à redução da erosão e do assoreamento de mananciais, de forma a proporcionar o aumento da qualidade da água e a tornar mais regular sua oferta. O programa efetiva-se por meio de articulações e parcerias entre instituições da esfera pública em suas três escalas municipal, estadual, federal e também privada, visando ao desenvolvimento da política de Pagamento por Serviços Ambientais (PSA) como um instrumento de incentivo aos proprietários rurais que adotarem, de forma voluntária, práticas voltadas à conservação de água e solo (ANA, 2012). A proposta contida no PPA é de promover o desenvolvimento rural sustentável, que, de acordo com Sachs (2008), não pode se restringir apenas à sustentabilidade ambiental, pois há outras dimensões não menos importantes, tais como a social, a econômica e a cultural, sendo ambas necessárias para que a sustentabilidade ambiental seja realizada. Nesse sentido, além das práticas preservacionistas, compreendemos que devem haver ações e políticas visando garantir a reprodução social e a permanência no campo dos agricultores familiares, bem como garantir a soberania alimentar e a disponibilidade de água com qualidade, sob risco de que as ações da sustentabilidade ambiental não serem tão eficazes sem ações nas demais esferas mencionadas. A água se apresenta como um bem fundamental para a vida, sendo condição necessária para todos os seres vivos, inclusive os humanos, em praticamente todas as atividades. A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida Desse modo, a construção participativa do mesmo pode ser elencada como um aspecto positivo, tendo em vista que as necessidades dos proprietários foi o ponto de partida para a elaboração do programa. 368 BENINCÁ, CLEMENTE De acordo com a entrevistada a relação de amizade entre os proprietários intensificou-se, contribuindo desse modo para o sucesso de programa. Loureiro (2011) considera que a construção de conceitos, atitudes e habilidades podem possibilitar o entendimento da realidade de vida e a atuação lúcida de atores sociais individuais e coletivos no ambiente. A mobilização é uma das formas de promover essas ações e, consequentemente, benfeitorias para toda a comunidade envolvida. De acordo com a Entrevistada 2, os primeiros passos foram de entrar em contato com a Prefeitura Municipal de Rio Verde-GO no ano de 1995, para que a mesma, juntamente com órgãos competentes pudessem articular ações em prol dessas áreas degradadas. Um grupo de moradores do alto da BHRA, liderados pela Entrevistada 2, realizaram uma reunião com representantes do Poder Público e com a imprensa local, para que juntos elaborassem um projeto de recuperação de nascentes e contenção de processos erosivos para o local (ENTREVISTADA 2). A partir dessa reunião, algumas medidas começaram a ser tomadas em prol da recuperação dessas áreas, iniciando nesse mesmo ano os trabalhos de contenção dos processos erosivos, o plantio de mudas e algumas outras ações pontuais. Diante da situação de calamidade e, por ser a principal fonte de abastecimento de água do Município, a Prefeitura aprovou um projeto de aproximadamente R$ 80.000,00 destinado para a recuperação das nascentes, cobrindo todas as voçorocas. Segundo a Entrevistada 2, essas medidas foram fundamentais para a contenção dos processos erosivos e para começar um processo lento de recuperação dessas áreas, sendo o próximo passo o reflorestamento. O reflorestamento iniciou-se no ano de 1999, com o plantio de mudas nativas, em parceria com instituições de ensino do Município. Foi criado o Movimento Águas no Rio (AMAR), para melhor articular as etapas de recuperações das nascentes entre os participantes, que são proprietários rurais, secretários da Prefeitura do município, professores da Universidade de Rio Verde (UNIRV), do Instituto Federal Goiano (IF-Goiano), funcionários da EMATER, entre outros. A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida De acordo coma Entrevistada 2, as ações foram ocorrendo de forma lenta, quando em 14 de setembro de 2007, devido à uma falha no sistema de bombeamento da unidade industrial da antiga Perdigão S/A, levou a contaminação das águas do Ribeirão Abóbora por dejetos de animais. A empresa foi notificada pela Promotoria de Defesa do Meio Ambiente da Comarca de Rio Verde, nos autos 369 BENINCÁ, CLEMENTE do Inquérito Civil nº 06/07. O Ministério Público do Estado de Goiás, com um termo de compromisso e ajustamento de conduta, sendo uma das obrigações da empresa, a elaboração e execução de projeto de recuperação das nascentes do Ribeirão Abóbora (MINISTERIO PÚBLICO DO ESTADO DE GOIÁS, 2007). Entre as exigências do Ministério Público, estava a recuperação das nascentes do Ribeirão Abóbora, com o cercamento das 54 nascentes catalogadas e o plantio de mudas nativas nas mesmas. Esse trabalho marcou o início de um processo de recuperação da Bacia, tendo em vista, que a partir desse momento, as ações se reforçaram. Ainda mobilizados em prol da recuperação do Ribeirão Abóbora, a união dos moradores, juntamente com professores de instituições públicas e privadas, bem como de representantes do Poder Público deu origem à ideia de implantar o programa da Agências Nacional das Águas (ANA), denominado de Produtor de Água (PPA). De acordo com a Entrevistada 2, a referência foi o programa de Extrema-MG, local da instalação do primeiro PPA, se tornando um exemplo a ser seguido. A parceria com a Secretaria de Desenvolvimento Econômico Sustentável de Rio Verde, da época, juntamente com os parceiros, foi fundamental para a implantação do programa Produtor de Água no município, partindo da iniciativa de um grupo de proprietários da Bacia Hidrográfica Ribeirão Abóbora. Com o apoio dessa e de outras entidades e servidores públicos do município, o que era visto como um sonho pelos moradores da Ribeirão Abóbora, passou a ser realidade. Partindo dessa ação, o município foi contemplado com o Programa em 28 de outubro de 2011, sendo o mesmo implementado e regulamentado pela Lei Municipal número 6.033/2011, promulgada com amparo no artigo 30, incisos I e II da Constituição Federal (MERIDA, 2014). A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida Segundo o Artigo 10, fica criado no município de Rio Verde o Programa Produtor de Águas, sob a coordenação do órgão municipal que tem por atribuição o desenvolvimento da política ambiental, visando promover a recuperação das nascentes e dos corpos d’água que abastecem o Município, a fim de garantir a qualidade e a quantidade de água, incentivando os proprietários rurais a se envolverem no processo, mediante compensação pelos serviços ambientais por eles prestados (LEI MUNICIPAL 6.033/2011) As 54 nascentes catalogadas já estavam cercadas, conforme o estabelecido por lei, então as próximas ações começaram a ser planejadas, para a execução 370 BENINCÁ, CLEMENTE completa do PPA. Após a aprovação do Projeto e da criação da Lei Municipal em 2011, reuniões começaram a ser realizadas para a apresentação do mesmo junto à comunidade local. completa do PPA. Após a aprovação do Projeto e da criação da Lei Municipal em 2011, reuniões começaram a ser realizadas para a apresentação do mesmo junto à comunidade local. Após aprovação do projeto, os 32 proprietários das 54 nascentes da BHRA foram convocados para uma reunião em que foram expostos os objetivos gerais do PPA e, ainda as possíveis etapas futuras do mesmo. O projeto tinha como objetivo geral incentivar a recuperação e conservação das nascentes do Município de Rio Verde, afim de garantir a qualidade e a quantidade de água para abastecimento urbano, compensando os proprietários rurais pelos serviços ambientais prestados. O pagamento por serviços ambientais é visto pela ANA como uma forma de motivação para esses proprietários, de modo a agregá-los por uma causa coletiva. No caso de Rio Verde-GO, o pagamento é feito com os recursos do Fundo Municipal do Meio Ambiente (SECRETARIA MUNICIPAL DO MEIO AMBIENTE, 2017). Para que o mesmo fosse realizado, traçaram-se objetivos específicos: aumentar a cobertura vegetal no entorno das nascentes; reduzir os processos de assoreamento e eutrofização dos corpos d’água; buscar a melhoria qualiquantitativa da água nos mananciais; aumentar o grau de proteção das áreas conservadas e recuperar as áreas degradadas; despertar a conscientização ambiental, mostrando que novas concepções e hábitos podem melhorar a qualidade de vida e do meio ambiente; garantir a sustentabilidade socioeconômica e ambiental das atividades desenvolvidas na microbacia, por meio de incentivos financeiros aos proprietários rurais; condicionar o diagnóstico das nascente, à necessidade de realização de subprojetos para aumentar o quantitativo de informações sobre a microbacia (RIO VERDE, 2012). A caracterização e breve histórico do programa “Produtor de Água”: partindo da necessidade dos proprietários da área envolvida Além da apresentação dos objetivos do projeto, nessa mesma reunião foi definido que a primeira etapa seria desenvolvida no Ribeirão Abóbora e, posteriormente no Ribeirão Laje e no Córrego Marimbondo. De acordo com a Engenheira Ambiental da Secretaria Municipal do Meio Ambiente de Rio Verde, os proprietários da BHRA apoiaram de forma efetiva e se organizaram para a formação de uma Associação, tendo em vista que o Pagamento por Serviços Ambientais (PSA) só seria feito, mediante o cumprimento deste requisito. Constituíram assim a Associação de Produtores de Água da Microbacia do Ribeirão, tendo como presidente a Engenheira Ambiental da Prefeitura de Rio Verde, filha de um dos proprietários de área do Ribeirão Abóbora. 371 BENINCÁ, CLEMENTE Na sequência serão apresentados a localização e caracterização da área de estudo no Município de Rio Verde-GO, bem como, de forma detalhada, as três etapas do PPA. As etapas desenvolvidas Durante a implantação do Programa Produtor de Água no Município de Rio Verde-GO, foram estabelecidos três locais para a instalação do mesmo: nas bacias hidrográficas do Ribeirão Abóbora, Córrego Marimbondo e do Ribeirão Laje, porém, as ações estão restritas e concentradas até o presente ano somente no Ribeirão Abóbora. O início das atividades foi definido pelo intenso processo de degradação do Ribeirão Abóbora e, posteriormente, serão direcionadas para o Ribeirão Lage (em 2018), sendo que até o momento não foram direcionadas ao Córrego Marimbondo. O Mapa 3 apresenta todas as etapas do Programa, bem como a drenagem dos cursos fluviais. a 3 – As três áreas de atuação do Programa Produtor de Água do Município de Rio a 3 – As três áreas de atuação do Programa Produtor de Água do Município de Rio Verde-GO apa 3 – As três áreas de atuação do Programa Produtor de Água do Município de R Verde-GO Fonte: Sistema Estadual de Geoinformação (SIEG), 2018. Org.: PRADO, F. S. (2018) Verde-GO Verde-GO Fonte: Sistema Estadual de Geoinformação (SIEG), 2018. Org.: PRADO, F. S. (2018) Conforme apresentado no Mapa 3, em vermelho a área do Córrego Marimbondo com 86,32 km2, em amarelo o ponto acima da captação do Ribeirão Abóbora com 54,9 km2 e verde, a área acima do ponto de captação do Ribeirão da Laje com 17,54 km2 (PREFEITURA DE RIO VERDE, 2012). 372 BENINCÁ, CLEMENTE As captações de água para o abastecimento de Rio Verde estão localizadas em ambas as microbacias, representando assim forte justificativa para a instalação do PPA no local. As águas do Ribeirão Lage e Abóbora são as responsáveis pelo abastecimento urbano da população de Rio Verde-GO, bem como para o processo produtivo das indústrias e das propriedades rurais (PREFEITURA DE RIO VERDE, 2012). Desse modo, a primeira etapa do Programa consistiu em determinar o grau de conservação das APP’s para avaliar a área de cobertura vegetal e a presença/ausência de fatores de perturbação, para posteriormente constatar a situação das nascentes (RIO VERDE, 2012). A classificação dessas nascentes fez- se necessária para mensurar a degradação total da área e como base de cálculo para o pagamento por serviços ambientais (PSA). Em um primeiro momento utilizou-se do estudo de Furtado (2008), que identificou 54 nascentes que abastecem o Ribeirão Abóbora em 29 propriedades rurais. As etapas desenvolvidas Porém, de acordo com a Secretaria Municipal do Meio ambiente de Rio Verde (2018), hoje já foram identificadas mais 9 nascentes, totalizando 63 em 32 propriedades. Identificou-se que o PPA, no período de 2011 até 2017, realizou atividades apenas no Ribeirão Abóbora, tendo em vista as várias dificuldades encontradas, conforme apresentadas pela engenheira ambiental, responsável pelo PPA: ausência de uma equipe técnica com maior número de profissionais; recurso financeiro insuficiente para o acompanhamento de campo; excessiva burocracia para a consolidação da associação e, consequentemente para o pagamento pelos serviços ambientais, dentre outros. Em uma segunda entrevista com a atual Secretária do Meio Ambiente (2018), ela relatou que a partir desse ano, as atividades serão direcionadas para o Córrego Marimbondo. O georreferenciamento da área foi iniciado, porém, não possuem previsão para o trabalho de recuperação de nascentes e o PSA nessa área. Localização e caracterização da área de estudo Com base nos levantamentos realizados, busca-se caracterizar a área onde concretamente está implementado o PPA de Rio Verde, o Ribeirão Abóbora. Atualmente 32 propriedades da microbacia estão inseridas no Programa, porém, no Mapa 4, apenas vinte e nove foram representadas, por falta das coordenadas de 373 BENINCÁ, CLEMENTE três propriedades, que não estão recebendo o pagamento pelos serviços ambientais. Mapa 4 – Propriedades da Bacia Hidrográfica Ribeirão Abóbora participantes do Program Produtor de Água Fonte: Sistema Estadual de Geoinformação (SIEG), 2015 Org.: JUNIOR QUEIROZ, V.S. (2015) – Propriedades da Bacia Hidrográfica Ribeirão Abóbora participantes do Programa Produtor de Água Fonte: Sistema Estadual de Geoinformação (SIEG), 2015 Org.: JUNIOR QUEIROZ, V.S. (2015 A BHRA constitui um importante manancial de abastecimento do município, é a área objeto deste projeto desde 2011 até o presente ano (ANA, 2018). O local passava por um processo acelerado de degradação e se trata da principal fonte de abastecimento de água de Rio Verde, sendo assim, apresentava uma forte necessidade de recuperação, já que a intensificação da degradação poderia comprometer seriamente o abastecimento doméstico da população local. As atividades no local iniciaram-se no ano de 2011 e, estão ativas até o momento, sendo que a continuidade do programa está condicionada a avaliações de 7 em 7 anos, que serão feitas pela Câmara Técnica do programa. O Ribeirão Abóbora, de acordo com o Termo de Fomento firmado com a Associação dos Produtores de Água da Microbacia do Ribeirão Abóbora, possui 63 nascentes catalogadas em 32 propriedades rurais, sendo 18 preservadas, 41 em regeneração e 4 degradadas (RIO VERDE, 2018). De acordo com a Secretaria Municipal de Meio Ambiente de Rio Verde, das 32 propriedades ativas no PPA, 29 estão recebendo o Pagamento por Serviços Ambientais, as demais, não entregaram a documentação exigida e, dessa forma, não podem receber o valor destinado por não cumprirem os requisitos exigidos em contrapartida. Os dados refletem a necessidade da implantação de programa como o PPA para promover a recuperação ambiental, tendo em vista o intenso 374 BENINCÁ, CLEMENTE desmatamento e uso excessivo dos solos, o elevado uso de produtos químicos, que geralmente são tóxicos e assim prejudiciais a vegetação, solos e água. O PPA representa assim, uma das alternativas para minimizar os impactos das atividades desenvolvidas na BHRA e de forma ainda pontual, podem apresentar alguns resultados. Localização e caracterização da área de estudo Um dos exemplos é quando analisa-se a área de vegetação remanescente, do ano de 2011 para 2017, um valor significativo de aumento de 1075,5 ha para 1236,6 ha, perante o aumento de área de agricultura. O PPA contribuiu inegavelmente para que as áreas de vegetação se expandissem, diante do cercamento das nascentes e consequentemente da preservação das APP’s, contribuindo para a melhoria ambiental na bacia e, consequentemente, da água. Ao longo da pesquisa de campo realizada na área de estudo, foram visualizados alguns aspectos do PPA, permitindo um conhecimento mais aprofundado da realidade das propriedades do Ribeirão Abóbora e dos reflexos do programa. Em um dos acessos às propriedades, foi possível ver a placa com as informações referentes ao PPA, como forma de conservação dos solos e reflorestamento das áreas de APP’s do Ribeirão. Apresenta-se um desafio um programa como o PPA, que consiste na recuperação de nascentes, de controle de assoreamentos e de aumento da disponibilidade de água, em uma área onde a agricultura capitalista se apresenta de forma tão intensa e estabelece um padrão produtivo que não respeita os ciclos naturais, tendendo a comprometer os mesmos. A fotografia 16 representa de forma muito clara essa contradição entre os objetivos do programa e a realidade local, que para atender a proposta de desenvolvimento rural sustentável precisa ser repensado o padrão produtivo pautado no intenso uso de tecnologia e mecanização por modelos agroecológicos e/ou sintrópicos. O estudo de uma bacia hidrográfica evidencia a conexão existente entre os elementos naturais e humanos e, nesse caso, as atividades estão sendo executadas em apenas uma parte da BHRA. Promover o desenvolvimento rural sustentável é algo muito mais complexo, mesmo sendo o PPA um avanço para a bacia, as ações para um padrão sustentável devem ser realizadas de forma mais ampla com mudanças estruturais, reorientando os padrões produtivos principalmente. Porém, não deve-se desconsiderar os benefícios que o mesmo tem possibilitado aos proprietários participantes com e sem nascentes. Conclui-se dessa forma que os traços da agricultura capitalista, se apresenta de forma intensa nas propriedades pesquisadas, representado a principal fonte de renda dos proprietários. 375 BENINCÁ, CLEMENTE Quando questionados sobre a questão da renda, 15 dos 16 proprietários, disseram ter outras fontes de renda, como aposentadoria, aluguel de imóveis, comércio e outros. Localização e caracterização da área de estudo Quando questionados sobre a questão da renda, 15 dos 16 proprietários, disseram ter outras fontes de renda, como aposentadoria, aluguel de imóveis, comércio e outros. O programa Produtor de Água na visão dos proprietários da Bacia Hidrográfica Ribeirão Abóbora O Programa Produtor de Água (PPA) foi elaborado para implementar medidas e ações para contemplar os âmbitos ambientais, sociais e econômicos, como uma proposta de desenvolvimento rural sustentável, desse modo, em entrevistas com os proprietários envolvidos, buscou-se a avaliação dos mesmos. Os questionamentos foram feitos em torno da eficácia ou não do programa, bem como das possibilidades de melhorias do mesmo. Primeiramente utilizou-se de questões objetivas, para identificar as concepções gerais dos proprietários rurais participantes e, em um segundo momento, de questões discursivas para um maior detalhamento dessa avaliação. A primeira questão levantada foi em relação à avaliação feita pelos mesmos a respeito do programa. Cerca de 12 proprietários avaliam o PPA como bom, 2 proprietários como regular e 2, como ótimo. Nenhum dos pesquisados consideram o programa como péssimo. A efetivação do programa era um sonho idealizado pela população local, dessa forma, mesmo apresentando falhas, que serão apresentadas em seguida, é considerado algo extremamente inovador pela população local. Os proprietários que caracterizaram o programa como regular, consideram principalmente a ausência de um maior acompanhamento técnico, um valor baixo pago pelos serviços ambientais e a falta das cercas. A pesquisa de campo nos possibilitou um contato direto com os sujeitos em estudo, proporcionando desse modo, uma análise crítica fatos. Quando questionados sobre a importância da participação em reuniões semestrais, todos foram unânimes em dizer que consideram sim importante essa participação. Alguns casos relatam ser um valor suficiente, outros por se tratarem de proprietários com um padrão de vida relativamente alto, acabam não atualizando seus cadastros e desse modo ficando impossibilitados de receber o pagamento pelo serviço ambiental realizado. Esses proprietários acabam se desligando do programa e não mais participam das ações em prol de um melhor uso da Microbacia. 376 BENINCÁ, CLEMENTE Em relação a satisfação desses proprietários com o programa, a grande maioria reclama da falta de acompanhamento, de reuniões e de atividades desempenhadas pela associação dos proprietários da Microbacia Ribeirão Abóbora. Muitos conhecem o programa como um meio apenas de receber por um serviço ambiental, tendo em vista a ausência de reuniões periódicas, de oficinas sobre questões ambientais e até mesmo de visitas periódicas por parte dos técnicos. O programa Produtor de Água na visão dos proprietários da Bacia Hidrográfica Ribeirão Abóbora Neste caso, as ações do programa, que já são pontuais por ocorrerem somente no Ribeirão Abóbora, ficam ainda mais restritas, por conta de que parte dos proprietários rurais não aderiram ao programa e não adotaram manejos menos predatórios, terminam por influenciar indiretamente negativamente nas ações e medidas efetuadas nas propriedades vizinhas, já que numa bacia hidrográfica os elementos naturais encontram-se fortemente integrados. Assim, a divisão das propriedades rurais e a falta de consciência de parte dos proprietários rurais, tende a comprometer os resultados das ações e medidas, que poderiam ser evitadas por uma maior orientação por parte da equipe do PPA, de modo a promover uma maior divulgação e consequentemente atividades de recuperação de áreas degradas mais efetivas não só no Ribeirão Abóbora, mas nas duas outras áreas, o Ribeirão Laje e Córrego Marimbondo, incluídas no projeto inicial. Considerações Finais Com base nas informações coletadas, pode-se dizer que o PPA está consolidado em uma área de monoculturas, onde o agronegócio está efetivado, sendo um grande desafio assegurar o desenvolvimento socioeconômico e ambiental, fortalecer o agricultor, dar suporte para a melhoria de vida da população e promover desse modo um uso consciente dos recursos ambientais. O programa Produtor de Água da ANA á apresentado como uma proposta inovadora por ter uma preocupação ambiental, apresentar ações de recuperação de nascentes, conservar os recursos hídricos e por fortalecer economicamente os envolvidos, por meio do Pagamento Por Serviços Ambientais. Partindo dos objetivos propostos embasou-se na seguinte hipótese: o programa Produtor de Água na prática apresenta limitações, como a sua atuação restrita e pontual, em apenas uma bacia, que apesar de ser estratégica por ser um dos principais mananciais do município, torna suas ações insuficientes na promoção do desenvolvimento rural 377 BENINCÁ, CLEMENTE sustentável, não incluindo sequer, de forma efetiva, todos os proprietários rurais presentes na Bacia Hidrográfica Ribeirão Abóbora. No que se refere o município de Rio Verde, esse conceito se apresenta de forma utópica, quando analisa-se as atividades econômicas desenvolvidas e as prioridades que as mesmas possuem. A agropecuária e a indústria são as principais fontes de renda, com um modelo altamente tecnológico visando atender o mercado externo, nada inclusivo, promovendo assim, uma maior desigualdade. A territorialização da agricultura capitalista em Rio Verde foi responsável por uma exclusão em relação ao acesso a condições econômicas, técnicas e socias, intensificando as desigualdades e marginalizando os agricultores familiares. Dessa forma, quanto à manutenção desses agricultores no campo, identifica-se a proposta de desenvolvimento rural sustentável falha, pois os mesmos não possuem as condições básicas, que são políticas públicas e investimentos de modo geral, para desenvolverem suas atividades. Apesar de ter sido implementado há aproximadamente oito anos, muitas ações ainda deverão ser efetuadas, sendo necessário uma análise mais aprofundada e detalhada no decorrer das ações. Desse modo, mesmo em caráter de execução apresenta-se várias alternativas para que o mesmo possa atuar de forma mais ampla, contemplando um maior número de proprietários, envolvendo um maior número de instituições e com um maior respaldo do poder público, principalmente no que tange as políticas públicas territoriais com ações para um padrão sustentável devem ser realizadas de forma mais ampla com mudanças estruturais, reorientando os padrões produtivos principalmente. REFERÊNCIAS ALVES, W. S. Aspectos físicos e qualidade da água da bacia hidrográfica do Ribeirão das Abóboras, no município de Rio Verde, Sudoeste de Goiás. 2016. 171 f. Dissertação (Mestrado em Geografia) - Universidade Federal de Goiás, Jataí, GO, 2016. BRASIL. Produtor de água. Disponível em: <https://www.ana.gov.br/textos-das- paginas-do-portal/programa-produtor-de-agua>. Acesso em: 10 abr. 2017. BRASIL. Conservador das Águas de Extrema (MG) é referência nacional. 2018. Disponível em: <http://legado.brasil.gov.br/noticias/meio- ambiente/2015/03/conservador-das-aguas-de-extrema-mg-e-referencia-nacional>. Acesso em: 10 fev. 2019. BRASIL. Conservador das Águas de Extrema (MG) é referência nacional. 2018. Disponível em: <http://legado.brasil.gov.br/noticias/meio- ambiente/2015/03/conservador-das-aguas-de-extrema-mg-e-referencia-nacional>. Acesso em: 10 fev. 2019. 378 BENINCÁ, CLEMENTE BRASIL. Programa Produtor de Água. Disponível em: <http://www3.ana.gov.br/portal/ANA/programas-e-projetos/programa-produtor-de- agua>. Acesso em: 26 mar. 2017. BRASIL. Lei Federal nº 12.651/12, de 25 de maio de 2012. Dispõe sobre a proteção da vegetação nativa; altera as Leis nº 6.938, de 31 de agosto de 1981, 9.393, de 19 de dezembro de 1996, e 11.428, de 22 de dezembro de 2006; revoga as Leis nº 4.771, de 15 de setembro de 1965, e 7.754, de 14 de abril de 1989, e a Medida Provisória nº 2.166- 67, de 24 de agosto de 2001; e dá outras providências. Disponível em: <http://www.planalto.gov.br/ccivil_03/_Ato2011- 2014/2012/Lei/L12651.htm>. Acesso em: 15 jun. 2017. BRASIL. Lei Federal nº 12.651/12, de 25 de maio de 2012. Dispõe sobre a proteção da vegetação nativa; altera as Leis nº 6.938, de 31 de agosto de 1981, 9.393, de 19 de dezembro de 1996, e 11.428, de 22 de dezembro de 2006; revoga as Leis nº 4.771, de 15 de setembro de 1965, e 7.754, de 14 de abril de 1989, e a Medida Provisória nº 2.166- 67, de 24 de agosto de 2001; e dá outras providências. Disponível em: <http://www.planalto.gov.br/ccivil_03/_Ato2011- 2014/2012/Lei/L12651.htm>. Acesso em: 15 jun. 2017. BRASIL. Ministério do Planejamento, Orçamento e Gestão. Programa desenvolvimento dos Cerrados. 2018. Disponível em: <http://www.abrasil.gov.br/ nivel3/index.asp?id=164&cod=CREND>. Acesso em: 10 out. 2018. BRASIL. Ministério do Meio Ambiente. Secretaria de Biodiversidade e Florestas. PROBIO. Projeto de conservação e utilização sustentável da diversidade biológica brasileira: relatório de atividades. Brasília, DF: MMA, 2002b. BRASIL. Ministério do Meio Ambiente. Programa de Monitoramento Ambiental dos Biomas Brasileiros (PMABB). Terraclass Cerrado. Disponível em: <https://www.mma.gov.br/biomas/cerrado/projeto-terraclass.html>. Acesso em: 20 maio 2018. BRASIL. Ministério do Desenvolvimento Agrário. A questão agrária no Brasil. Disponível em: <http://www.mda.gov.br/>. Acesso em: 10 abr. 2016. BRICEÑO, A. E. E.; SOUZA, V. L. REFERÊNCIAS Pagamento por serviços ambientais pela conservação de nascentes da microbacia do Ribeirão Abóbora, Rio Verde, Goiás, Brasil. In: CONGRESSO BRASILEIRO DE ENGENHARIA SANITÁRIA E AMBIENTAL (ABES), 27., 2013, Goiânia. Anais... Goiânia, GO: ABES, 2013. p. CARMO L. R.; GUIMARÃES, E. N.; AZEVEDO A. M. M. Agroindústria, População e ambiente no sudoeste de Goiás. In: _____. Migrações e ambiente no Centro- Oeste. Campinas, SP: PRONEX, 2002. p.1-23. DIAS, B. F. S. Cerrados: uma Caracterização. In: ______ (coord.). Alternativas de desenvolvimento dos Cerrados: manejo e conservação dos recursos naturais renováveis. Brasília, DF: Fundação Pró-Natureza, 1996. p. 11-25. GOIÁS. Instituto Mauro Borges de Estatística e Estudos Socioeconômicos. Banco de dados estatísticos de Goiás. Disponível em: <http://www.imb.go.gov.br/bde>. Acesso em: 15 fev. 2018. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE). Censo agropecuário brasileiro de 2006. Disponível em: <https://biblioteca.ibge.gov.br/visualizacao/periodicos/51/agro_2006.pdf.>. Acesso em: 15 jun. 2017. INSTITUTO BRASILEIRO DE GEOGRAFIA E ESTATÍSTICA (IBGE). Censo agropecuário brasileiro de 2017. Disponível em: 379 BENINCÁ, CLEMENTE <https://sidra.ibge.gov.br/pesquisa/censo-agropecuario/censo-agropecuario-2017>. Acesso em: 12 mar. 2019. <https://sidra.ibge.gov.br/pesquisa/censo-agropecuario/censo-agropecuario-2017>. Acesso em: 12 mar. 2019. <https://sidra.ibge.gov.br/pesquisa/censo-agropecuario/censo-agropecuario-2017>. Acesso em: 12 mar. 2019. LOUREIRO, C. F. B. et al. (Orgs). Educação ambiental: repensando o espaço da cidadania. 5. ed. São Paulo, SP: Cortez, 2011. LOUREIRO, C. F. B. et al. (Orgs). Educação ambiental: repensando o espaço da cidadania. 5. ed. São Paulo, SP: Cortez, 2011. MERIDA, C. O pagamento por serviços ambientais como instrumento de efetividade do desenvolvimento sustentável em Rio Verde, Goiás. 2014. 125f. Dissertação (Mestrado em Direito) - Programa de Pós-Graduação em Direito, Relações Internacionais e Desenvolvimento, Pontifícia Universidade Católica de Goiás, Goiânia, GO, 2014. MINISTÉRIO PÚBLICO DE GOIAS. Disponível em: <http://www.mpgo.mp.br/portal/principal>. Acesso em: 07 fev. 2019. em: RIO VERDE. Lei Municipal nº 6.033, de 28 de outubro de 2011.Cria o programa produtor de água. Disponível em: <http://www.rioverde.go.gov.br/docsoscs/bacia_justificativa4.pdf>. Acesso em: 12 fev. 2017. RIO VERDE. Diagnóstico da implantação do programa produtores de água na cidade de Rio Verde. Rio Verde, GO, 2012. Disponível em: <http://www.rioverde.go.gov.br>. Acesso em: 12 set. 2017 SACHS, I. Ecodesenvolvimento: crescer sem destruir. São Paulo, SP: Editora Vértice, 1985. SACHS, I. Desenvolvimento: includente, sustentável, sustentado. Rio de Janeiro, RJ: Garamond, 2008. 152 p. SACHS, I. Caminhos para o desenvolvimento sustentável. Rio de Janeiro, RJ: Garamond, 2009. SORRENTINO, M. Ambientalismo e participação na contemporaneidade. São Paulo, SP: Educ/Fapesp, 2001. TUCCI, C. E. M. Hidrologia: Ciência e aplicação. 4. ed. Porto Alegre, RS: Editora da UFRGS/ABRH, 2009. 943p. TUNDISI, J G. REFERÊNCIAS Água no século XXI: enfrentando a escassez. 2. Ed. São Carlos, SP: RiMa, 2005. 248p. TUNDISI, J. G.; MATSUMURA-TUNDISI, T. Recursos hídricos no século XXI. São Paulo, SP: Oficina de textos, 2011. 328 p. 380 BENINCÁ, CLEMENTE LICENÇA DE USO Ç Este artigo está licenciado sob a Licença Creative Commons CC-BY. Com essa licença você pode compartilhar, adaptar, criar para qualquer fim, desde que atribua a autoria da obra. HISTÓRICO Recebido em: 09-11-2019 Aprovado em: 30-07-2020 NOTAS DE AUTOR Mainara da Costa Benincá - Concepção. Coleta de dados, Análise de dados, Elaboração do manuscrito, revisão e aprovação da versão final do trabalho p ç vandro César Clemente – Participação ativa da discussão dos resultados; Revisão e aprovação da versão al do trabalho. FINANCIAMENTO Bolsa Capes CNPQ - Conselho Nacional de Desenvolvimento Científico e Tecnológico Bolsa Capes CNPQ - Conselho Nacional de Desenvolvimento Científico e Tecnológico p CNPQ - Conselho Nacional de Desenvolvimento Científico e Tecnológico
https://openalex.org/W2073710803	http://eprints.whiterose.ac.uk/81046/1/WRRO_81046.pdf	English	null	Dis/ability and austerity: beyond work and slow death	Disability & society	2,014	cc-by	3,348	promoting access to White Rose research papers Universities of Leeds, Sheffield and York http://eprints.whiterose.ac.uk/ This is a copy of the final published version of a paper published via gold open access in Disability and Society. This open access article is distributed under the terms of the Creative Commons Attribution Licence (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/81046 promoting access to White Rose research papers This is a copy of the final published version of a paper published via gold open access in Disability and Society. This open access article is distributed under the terms of the Creative Commons Attribution Licence (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/81046 White Rose Research Online URL for this paper: http://eprints.whiterose.ac.uk/81046 Published paper Goodley, D., Lawthom, R. and Runswick-Cole, K. (2014) Dis/ability and austerity: Beyond work and slow death. Disability and Society, 29 (6). 980 - 984. Doi: 10.1080/09687599.2014.920125 Dis/ability and austerity: beyond work and slow death To cite this article: Dan Goodley, Rebecca Lawthom & Katherine Runswick-Cole (2014) Dis/ ability and austerity: beyond work and slow death, Disability & Society, 29:6, 980-984, DOI: 10.1080/09687599.2014.920125 To link to this article: http://dx.doi.org/10.1080/09687599.2014.920125 To cite this article: Dan Goodley, Rebecca Lawthom & Katherine Runswick-Cole (2014) Dis/ ability and austerity: beyond work and slow death, Disability & Society, 29:6, 980-984, DOI: 10.1080/09687599.2014.920125 To cite this article: Dan Goodley, Rebecca Lawthom & Katherine Runswick-Cole (2014) Dis/ ability and austerity: beyond work and slow death, Disability & Society, 29:6, 980-984, DOI: 10.1080/09687599.2014.920125 To link to this article: http://dx.doi.org/10.1080/09687599.2014.920125 To link to this article: http://dx.doi.org/10.1080/09687599.2014.920125 White Rose Research Online eprints@whiterose.ac.uk This article was downloaded by: [University of Sheffield] On: 26 February 2015, At: 07:04 On: 26 February 2015, At: 07:04 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Disability & Society Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/cdso20 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Versions of published Taylor & Francis and Routledge Open articles and Taylor & Francis and Routledge Open Select articles posted to institutional or subject repositories or any other third-party website are without warranty from Taylor & Francis of any kind, either expressed or implied, including, but not limited to, warranties of merchantability, fitness for a particular purpose, or non-infringement. Any opinions and views expressed in this article are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor & Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions It is essential that you check the license status of any given Open and Open Select article to confirm conditions of access and use. This article may be used for research, teaching, and private study purposes. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms- and-conditions It is essential that you check the license status of any given Open and Open Select article to confirm conditions of access and use. It is essential that you check the license status of any given Open and Open Select article to confirm conditions of access and use. Disability & Society, 2014 Disability & Society, 2014 Vol. 29, No. 6, 980–984, http://dx.doi.org/10.1080/09687599.2014.920125 Vol. 29, No. 6, 980–984, http://dx.doi.org/10.1080/09687599.2014.920125 Dis/ability and austerity: beyond work and slow death D G dl a* R b L th b d K th i R i k C l b aSchool of Education, University of Shefﬁeld, Shefﬁeld, UK; bDepartment of Psychology, Manchester Metropolitan University, Manchester, UK aSchool of Education, University of Shefﬁeld, Shefﬁeld, UK; bDepartment of Psychology, Manchester Metropolitan University, Manchester, UK (Received 15 April 2014; ﬁnal version received 17 April 2014) (Received 15 April 2014; ﬁnal version received 17 April 2014) The forthcoming book Dis/ability Studies: Theorising Disablism and Ableism argues that we are living in an historical epoch which might be described as neoliberal-ableism, in which we are all subjected to slow death, increased preca- rity and growing debility. In this paper we apply this analysis to a consideration of austerity with further reference to disability studies and politics. ownloaded by [University of Sheffield] at 07:04 26 February 2015 Keywords: disability; austerity; slow death; neoliberal-ableism © 2014 The Author(s). Published by Taylor & Francis. This is an Open Access article distributed under the terms of the Creative Commons Attribution License http://creativecom mons.org/licenses/by/3.0/, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The moral rights of the named author(s) have been asserted. © 2014 The Author(s). Published by Taylor & Francis. CURRENT ISSUES Dis/ability and austerity: beyond work and slow death Dan Goodleya, Rebecca Lawthomb and Katherine Runswick-Coleb Austerity By 2008 the world knew: the reckless nature of capitalism had led to economic meltdown. We had simply spent and borrowed too much. To overcome this we were, and still are, expected to ‘make do and mend’ (Tosh 2013). Rather than contest the system (capitalism), we should, well, work it more responsibly – be more prudent in our spending and more committed in our labour. We are, of course, ‘all in this together’, so the British Chancellor told us, and his government remains committed to ‘the makers, doers and savers’.1 Indeed, it would appear that we are all equal, although some are more equal than others. This is evidenced by the impact of auster- ity measures in Britain since the 2010 emergency budget. In 2012 the Disability in Austerity Study led by Demos estimated losses of £2000–3000 (in families’ house- hold income) over the course of the current parliament, with disabled people losing £9 billion in welfare support and one-third losing their Disability Living Allowance (Wood 2012). This had resulted in disabled people and their families experiencing: An ever-diminishing level of civic and social engagement – our households are becoming socially more isolated, and reducing the amount of activities they engage in – from essen- tials such as work and medical appointments to ‘luxuries’ such as volunteering and train- ing. This is at odds with the Government’s vision of stronger and active communities. Retrenchment of services – both statutory services and third sector services are being cut, leaving disabled people with nowhere else to turn. The concept of the safety net no longer resonates with people experiencing serious crises before help is provided. Declining mental health – our households are increasingly experiencing anxiety, depres- sion and fear for the future, with some relying on increased medication. (Wood 2012: 7) This experience is nowhere more evident than today in the midst of neoliberal capitalism. Soldatic and Chapman (2010) depict this as a time of ﬂexibility, Corresponding author. Email: d.goodley@shefﬁeld.ac.uk *Corresponding author. Email: d.goodley@shefﬁeld.ac.uk 981 Disability & Society casualisation, work readiness and productivity. The rationality of market rule, they say, demands an adaptable worker; albeit one caught up in the insecurities associated with the global economy and the winding back of welfare beneﬁts and entitlements. Austerity For Cooley (2011), while at the beginnings of neoliberalism there was talk of free enterprise, the ‘American way’ and working for one’s family, latter iterations of neo- liberal discourse would suggest we are entering a cultural epoch where such a vision of human development automatically inducts each and everyone into what Jakobsen (2009, 224) describes as ‘a relational structure that provides for privatised resource- provision’. The way we would like to think of this relational structure is as follows: neoliberalism provides an ecosystem for the nourishment of ableism, which we can deﬁne as neoliberal-ableism. We are all expected to overcome economic downturn and respond to austerity through adhering to ableism’s ideals, its narrow conceptions of personhood, its arrogance and its propensity to buddy-up with other fascistic ide- ologies that celebrate the minority over the majority means that it is a world-view, an ontology and a methodology for the making human-kinds that will eventually, if left unchallenged, bulldoze the disparate variegated nature of human kind (Goodley 2014). A key principle of ‘body management within neoliberalism’, as Mitchell observes, is that: ownloaded by [University of Sheffield] at 07:04 26 February 2015 those who don’t adequately maintain their bodies are held personally responsible for their descent into the chaos of ill health and non-well-being proﬁting from the misfortunes of another; a parasitism of privilege allowed only to those who embody the normative capacities of neoliberal identities. (Mitchell 2014, 3) ‘The truth is that all this work is debilitating’ (Puar 2012). Crip alternatives (slowing the death down) The politics of disability permit us to think again (and differently) about living through (and being recognised by) neoliberal-ableism. Here are two thoughts. Firstly, why all this talk of austerity when we should be talking about poverty? In Britain the ‘bedroom tax’, the recent reduction of Disabled Student Allowance, and the reduction of numbers judged to be disabled (and therefore eligible to welfare beneﬁts) put disabled people yet again amongst the poorest of the poor. If labour is slow death, living as a disabled person in 2014 might mean a quick one: ownloaded by [University of Sheffield] at 07:04 26 February 2015 A disabled man died penniless when he lost his beneﬁts after being judged ﬁt to work – despite being so ill doctors said he needed a new heart. Robert Barlow died last November aged 47 while suffering from a heart defect and brain tumour. The classi- cally trained pianist – who worked for years as a microbiologist in Cambridge – was deemed ﬁt to work by beneﬁts assessors Atos, even though doctors at the time urged him to have a heart transplant. He passed away less than two years later. Now Mr Barlow’s Cambridge family and Labour MP Luciana Berger want the Govern- ment to learn lessons from this tragic case. His aunt Joan Westland, 85, of Cherry Hinton, said: ‘I don’t know how they expected him to work. Nobody would have loved to work more than him, but he simply couldn’t.’2 It is absolutely essential that we consider the ways in which poverty and disability are once again being cast together as inseparable categories. Disabled People Against the Cuts3remind us that freedom can rarely be found in capitalism. Disability politics lead the way, yet again, in thinking about alternatives to neoliberal-ableist capitalism. It is absolutely essential that we consider the ways in which poverty and disability are once again being cast together as inseparable categories. Disabled People Against the Cuts3remind us that freedom can rarely be found in capitalism. Disability politics lead the way, yet again, in thinking about alternatives to neoliberal-ableist capitalism. Secondly, what alternatives does disability offer to the slow death of neoliberal- ableism and false promises of austerity? For David Mitchell (2014, 1–2): ‘disability subjectivities create new forms of embodied knowledge and collective conscious- ness. Labour/slow death Work, for Lauren Berlant (2007, 2010, 2011), is best epitomised as a practice of slow death, a concept that refers to the physical wearing out of a population and the deterioration of people in that population, which is very nearly a deﬁning condition of their experience and historical existence. We are, she suggests, exhausted by neo- liberal capitalism: we are all in slow death, but for some this is more apparent. This includes: the bodies of US waged workers will be more fatigued, in more pain, less capable of ordinary breathing and working, and die earlier than the average for higher-income workers, who are also getting fatter, but at a slower rate and with relatively more opportunity for exercise. (Berlant 2007, 775) To this collective we would add disabled people and many of their close allies. Neoliberal-ableism displays, to borrow the words of Robert McRuer (2012c), a ‘spe- cial genius’ at making lopsided growth, wealth for a few, and immiseration for many more, seem sexy, and – well – progressive and modern. Disabled people have to become, in the words of David Mitchell (2014), ‘able-disabled’; or, as Goodley (2014) notes, disabled people have to embrace able- ism to overcome their disabling conditions; or as Goodley and Runswick-Cole (in press) have argued, individuals need to embolden the ability side of the dis/ability complex in order to survive, hopefully thrive, but deﬁnitely make do and mend. This is what Mitchell (2014) and McRuer (2012a, 2012b) have in mind around neoliberal tolerance: 982 D. Goodley et al. (1) if you are prepared to work hard then you are in; (1) if you are prepared to work hard then you are in; ( ) y p p y (2) don’t forget now: we are all in this together as workers and consumers; and ) y p p y 2) don’t forget now: we are all in this together as workers and consumers; and (2) don’t forget now: we are all in this together as workers and consumers; and (3) in these austere times we will get ourselves ‘back on our feet’ – work will set us free. (3) in these austere times we will get ourselves ‘back on our feet’ – work will set us free. (3) in these austere times we will get ourselves ‘back on our feet’ – work will set us free. An account from our Inclusive Living colleagues: Pete and Wendy Crane, along with colleagues including person-centred planning coordinator Max Neil, have developed Crip alternatives (slowing the death down) Queer and disabled people’s interdependencies provide alternative ethical maps for living together outside of, even in opposition to, the dictates of normalcy’. This resonates with stories emerging from our current ESRC project ‘Big Society? Dis- abled People with Learning Disabilities and Civil Society’ (grant reference: ES/K004883/1). The research team, from The University of Shefﬁeld, Manchester Metropolitan University, Northumbria University and The University of Bristol, are working with organisations of/for disabled people, activists and allies to discover how disabled people with learning disabilities are participating in their communities, in public services and in social action. The team is exploring disabled people with learning disabilities access to social capital and networks of interdependence as well as their social emotional well-being in a context of austerity. Here is a story from our project: An account from our Inclusive Living colleagues: Pete and Wendy Crane, along with colleagues including person-centred planning coordinator Max Neil, have developed 983 Disability & Society Disability & Society the Circles of Support approach to community living. This involves individuals such as Matt (a pseudonym) using their welfare beneﬁts to recruit a network of advocates and support staff to support him 24/7 in his own home which is located a couple of streets down from his parents. (Goodley 2014, 128) the Circles of Support approach to community living. This involves individuals such as Matt (a pseudonym) using their welfare beneﬁts to recruit a network of advocates and support staff to support him 24/7 in his own home which is located a couple of streets down from his parents. (Goodley 2014, 128) Disability, then, offers opportunities for reconsidering our relationships with life, labour and slow death. Could care, rather than work, be a place to ﬁnd identity and recognition? Why wear yourself out? Disability provides a moment to intervene in slow death: why work yourself to death? Why (just) work? How do we support one another in a time of austerity? Why sweat to improve one’s embodied and cognitive lot? How else might we live together to support, care and enable one another? What do we gain when we fail to meet neoliberalism’s normative labouring standards? For Kolárová : cripness is rich with failure; cripness is infused with negativity; yet we do not always see it as such. Crip alternatives (slowing the death down) The rich archive of the labour of crip failure is here and at hand … but do we, as the crip community celebrate those as crip failures that can sustain our visions of utopia and whose negative energies move us towards the crip horizon?’ (Kolárová in press, n.p.) ownloaded by [University of Sheffield] at 07:04 26 February 2015 These crip horizons might be found as disability is, simultaneously for and against slow death. And this appears to be a worn out place that all occupy and can, via our collectivities and assemblages, recuperate our possibilities. What alternatives do we see for our worn-out bodies? Ones, we would assume, far beyond work and slow death. Downloaded by [University of Sheffield] at 07:04 Notes 1. See http://disabilitynow.org.uk/blog/budget-bleaker-future 1. See http://disabilitynow.org.uk/blog/budget-bleaker-future 2. See http://www.cambridge-news.co.uk/News/Cambridge-microbiologist-Robert-Barlow- dies-penniless-in-Liverpool-after-his-beneﬁts-were-withdrawn-despite-him-needing-a-new- heart-20140409063051.htm 3. See http://dpac.uk.net/ References Berlant, L. 2007. “Slow Death: Sovereignty, Obesity, Lateral Agency.” Critical Inquiry 33: 754–780. Berlant, L. 2007. “Slow Death: Sovereignty, Obesity, Lateral Agency.” Critical Inquiry 33: 754–780. Berlant, L. 2007. “Slow Death: Sovereignty, Obesity, Lateral Agency.” Critical Inquiry 33: 754–780. Berlant, L 2010. “Cruel Optimism.” In The Affect Theory Reader, edited by M. Gregg and G. J. Seigworth, 93–117. Durham, NC: Duke University Press. Berlant, L. 2011. Lauren Berlant, “Austerity, Precarity, Awkwardness”. Retrieved on 14th May 2013 from http://supervalentthought.ﬁles.wordpress.com/2011/12/berlant-aaa-2011ﬁnal.pdf. Cooley, R. 2011. “Disabling Spectacles: Representations of Trig Palin and Cognitive Disability.” Journal of Literary & Cultural Disability Studies 5 (3): 303–320. dl / b l d h bl d bl d Goodley, D. 2014. Dis/Ability Studies: Theorising Disablism and Ableism. Londo Routledge. g Goodley, D, and K. Runswick-Cole. In press. “Dishumanism: Thinking about the Human through Disability.” Discourse: Cultural Politics of Education. g y f Jakobsen, J. 2009. “The Economics of Fear, the Politics of Hope, and the Perversity of Happiness.” Women & Performance: A Journal of Feminist Theory 19 (2): 219–226. pp f f y ( ) Kolárová, K. In press. “Cruising for a Crip(Topia) in the Context of Neoliberal Transforma- tions of the Czech Republic.” Journal of Literary and Cultural Disability Studies. McRuer, R. 2012a. “Disability and the Globalisation of Austerity Politics.” Keynote Paper presented at the Contact Zone: Disability, Culture, Theory Conference, 25–27 October, University of Cologne. D. Goodley et al. 984 McRuer, R. 2012b. “Afterword: The End of Contested Corporealities.” Review of Educatio Pedagogy, and Cultural Studies 34 (3–4): 208–211. g gy ( ) McRuer, R. 2012c. “Afterword.” Bioethical Inquiry 9: 357–358. Mitchell, D. 2014. “Gay Pasts and Disability Future(s) Tense. Heteronormative Trauma and Parasitism in Midnight Cowboy.” Journal of Literary & Cultural Disability Studies 8 (1), 1–16. Puar, J. K. 2012. “Precarity Talk: A Virtual Roundtable with Lauren Berlant.” Judith Butler, Bojana Cvejic, Isabell Lorey, Jasbir Puar, and Ana Vujanovic, TDR: The Drama Review 56 (4): 163–177. ( ) oldatic, K., and A. Chapman. 2010. “Surviving the Assault? The Australian Disability Movement and the Neoliberal Workfare State.” Social Movement Studies 9 (2): 139–154. Soldatic, K., and A. Chapman. 2010. Surviving the Assault? The Australian Disability Movement and the Neoliberal Workfare State.” Social Movement Studies 9 (2): 139–154. Tosh, J. 2013. “Women and Austerity: Beyond ‘Make Do and Mend’.” Psychology of Women Section Review 15 (1): 46–52. ( ) Tosh, J. 2013. References “Women and Austerity: Beyond ‘Make Do and Mend’.” Psychology of Women Section Review 15 (1): 46–52. ( ) Wood, C. 2012. For Disabled People the Worst is Yet to Come…Destination Unknown: Summer 2012. London: Demos. ownloaded by [University of Sheffield] at 07:04 26 February 2015
https://openalex.org/W2899507676	https://digital.csic.es/bitstream/10261/179201/1/L%c3%b3pezDeCastroJA_HowERAP1andERAP2.pdf	English	null	How ERAP1 and ERAP2 Shape the Peptidomes of Disease-Associated MHC-I Proteins	Frontiers in immunology	2,018	cc-by	15,253	How ERAP1 and ERAP2 Shape the Peptidomes of Disease-Associated MHC-I Proteins José A. López de Castro* Centro de Biología Molecular Severo Ochoa (CSIC-UAM), Madrid, Spain Four inﬂammatory diseases are strongly associated with Major Histocompatibility Complex class I (MHC-I) molecules: birdshot chorioretinopathy (HLA-A∗29:02), ankylosing spondylitis (HLA-B∗27), Behçet’s disease (HLA-B∗51), and psoriasis (HLA-C∗06:02). The endoplasmic reticulum aminopeptidases (ERAP) 1 and 2 are also risk factors for these diseases. Since both enzymes are involved in the ﬁnal processing steps of MHC-I ligands it is reasonable to assume that MHC-I-bound peptides play a signiﬁcant pathogenetic role. This review will mainly focus on recent studies concerning the effects of ERAP1 and ERAP2 polymorphism and expression on shaping the peptidome of disease-associated MHC-I molecules in live cells. These studies will be discussed in the context of the distinct mechanisms and substrate preferences of both enzymes, their different patterns of genetic association with various diseases, the role of polymorphisms determining changes in enzymatic activity or expression levels, and the distinct peptidomes of disease-associated MHC-I allotypes. ERAP1 and ERAP2 polymorphism and expression induce signiﬁcant changes in multiple MHC-I-bound peptidomes. These changes are MHC allotype-speciﬁc and, without excluding a degree of functional inter-dependence between both enzymes, reﬂect largely separate roles in their processing of MHC-I ligands. The studies reviewed here provide a molecular basis for the distinct patterns of genetic association of ERAP1 and ERAP2 with disease and for the pathogenetic role of peptides. The allotype-dependent alterations induced on distinct peptidomes may explain that the joint association of both enzymes and unrelated MHC-I alleles inﬂuence different pathological outcomes. Edited by: Maria Teresa Fiorillo, Università degli Studi di Roma La Sapienza, Italy Reviewed by: Peter M. Van Endert, Institut National de la Santé et de la Recherche Médicale (INSERM), France Marlene Bouvier, University of Illinois at Chicago, United States Edited by: Maria Teresa Fiorillo, Università degli Studi di Roma La Sapienza, Italy Reviewed by: Peter M. Van Endert, Institut National de la Santé et de la Recherche Médicale (INSERM), France Marlene Bouvier Reviewed by: Peter M. Van Endert, Institut National de la Santé et de la Recherche Médicale (INSERM), France Marlene Bouvier, University of Illinois at Chicago, United States Correspondence: José A. López de Castro aldecastro@cbm.csic.es Specialty section: This article was submitted to Inﬂammation, a section of the journal Frontiers in Immunology Specialty section: This article was submitted to Inﬂammation, a section of the journal Frontiers in Immunology Received: 21 August 2018 Accepted: 04 October 2018 Published: 30 October 2018 INTRODUCTION The role of the Major Histocompatibility Complex (MHC) as a risk factor for numerous diseases is well established (1). Although usually involving MHC-II alleles a more restricted set of inﬂammatory disorders is strongly associated with MHC-I. The most prominent of these associations concern birdshot chorioretinopathy (BSCR), ankylosing spondylitis (AS), Behçet’s disease (BD) and psoriasis, which involve HLA-A∗29:02, HLA-B∗27, HLA-B∗51, and HLA-C∗06:02, respectively (2–10). The associations of BSCR and AS with MHC-I are extremely strong, with about 100 and 90% of patients being A∗29:02 and B∗27 positive, respectively. The associations of HLA-B∗51 and C∗06:02 with BD and psoriasis are lower, yet quite signiﬁcant, with odds ratios around 5–6. Received: 21 August 2018 Accepted: 04 October 2018 Published: 30 October 2018 Keywords: MHC, inﬂammatory diseases, ERAP, antigen processing, ankylosing spondylitis, uveitis, Behçet’s disease, psoriasis REVIEW REVIEW published: 30 October 2018 doi: 10.3389/ﬁmmu.2018.02463 published: 30 October 2018 doi: 10.3389/ﬁmmu.2018.02463 Citation: López de Castro JA (2018) How ERAP1 and ERAP2 Shape the Peptidomes of Disease-Associated MHC-I Proteins. F t I l 9 2463 In spite of the strength of these associations, both the pathogenetic mechanisms and the role of the MHC molecules in the four diseases remain unknown. Whereas, BSCR is an eye-speciﬁc Front. Immunol. 9:2463. doi: 10.3389/ﬁmmu.2018.02463 October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro inﬂammatory disorder, typically aﬀecting the choroid and retina (11), the three other diseases show distinct systemic manifestations, involving a variety of tissues, cell populations, and immunological pathways, a complexity that poses considerable diﬃculty to the characterization of pathogenetic factors (12–14). shaping the constitutive peptidomes of disease-associated MHC- I proteins from live cells, and will discuss the contribution of these studies to our understanding of ERAP1/ERAP2 function and pathogenetic role. ERAP1 AND ERAP2 HAVE COMPLEMENTARY SUBSTRATE PREFERENCES The development of genome wide association studies and other genetic strategies lead, among many other ﬁndings, to a discovery that promised to shed new light into the pathogenetic role of MHC-I in disease. Namely, the involvement of the endoplasmic reticulum aminopeptidases (ERAP)1 and 2 as prominent risk factors for MHC-I-associated diseases (12). These enzymes are key players in the antigen processing pathway of MHC-I molecules because they trim peptides to the optimal size for MHC-I binding (15, 16) and can also over-trim and destroy MHC-I ligands (17–20). ERAP1 and ERAP2 are structurally related Zn-metallopeptidases that share about 50% amino acid sequence identity and many similarities in their three-dimensional structure (49–52), reviewed in Stratikos and Stern (53). Both enzymes consist in four globular domains. Their catalytic sites are located in domain II and a cavity, encompassing areas of domains II and IV, is the substrate-binding site. Conformational changes involving the reorientation of amino acid side chains in the catalytic site and domain rearrangements leading to open (less active) and closed (more active) conformations govern the enzymatic activity of ERAP1 and, probably, of ERAP2. Yet, the distinct structure of both the active site and the peptide binding site determine critical diﬀerences in enzymatic speciﬁcity and substrate handling. For the purpose of this review, the most important of these diﬀerences are those concerning N-terminal residue speciﬁcity and substrate length preferences. ERAP1 can cleave all peptide bonds except those involving Pro, but shows a wide range of eﬃciency depending on the N-terminal side chain of the substrate: it shows preference for nonpolar residues and is much less eﬃcient with polar and charged ones (54). ERAP2 can cleave very few residues, being most eﬃcient with Arg (12, 16, 45). In addition, ERAP1 preferentially cleaves peptides longer than 9-mers and becomes virtually inactive with 8-mers and shorter peptides. This remarkable feature, which is seemingly unique among aminopeptidases, is known as the molecular ruler mechanism (55) and tunes the length preferences of the enzyme to the optimal length of MHC-I ligands. In contrast, ERAP2 cleaves best 9-mers and shorter peptides, becoming progressively less eﬃcient with longer ones (52). ERAP1 is associated with AS, psoriasis and BD in epistasis with the risk HLA-B∗27 (21) and -B∗40:01 (22), C∗06:02, (23), and HLA-B∗51 alleles (24), respectively. ERAP2 is associated with AS and psoriasis (25–28), but not in epistasis with MHC-I. ERAP1 AND ERAP2 HAVE COMPLEMENTARY SUBSTRATE PREFERENCES ERAP2 and ERAP1 are also risk factors for BSCR (2, 29). The joint association of ERAP1/2 and MHC-I with multiple inﬂammatory disorders underlines the relevance of MHC-I- bound peptides in these diseases. Since peptides presented by MHC-I determine both the recognition by T lymphocytes (30, 31) and NK cells (32), ERAP1 and ERAP2 can inﬂuence both innate and adaptive immune pathways in these diseases. Indeed, a role of ERAP1 in modulating innate immunity has been demonstrated (33, 34), and its involvement in cross-presentation has also been claimed (35–37). ERAP1, and perhaps also ERAP2, may have additional functions besides antigen processing (38, 39). For instance, ERAP1 can be secreted from macrophages in response to lipopolyssacharide and interferon-gamma by a TLR-dependent mechanism (40), which enhances their phagocytic activity (41). This might be relevant to the modulation of innate immunity. A potentially important issue is the involvement of both ERAP1 and ERAP2 in the blood pressure homeostasis, as revealed by genetic analyses (42, 43). The relevance of secreted ERAP1 in this process is supported by a study showing that the redox state of the ER-resident thioredoxin ERp44 controls blood pressure. This protein is involved in the retention of ERAP1 and other proteins in the ER by a thiol-dependent mechanism (44). ERAP2 secretion might be promoted in an analogous way but, to my knowledge, this has not been demonstrated. The mechanism by which secreted ERAP1 and ERAP2 may modulate hypertension is not yet resolved. These enzymes may inﬂuence the processing of relevant peptide hormones regulating blood pressure (44–46). In addition, cleavage of Arg residues by secreted enzymes, may promote the formation of nitric oxide, as shown for ERAP1 (47), which may in turn inﬂuence blood pressure and inﬂammation (48). The putative relevance of these alternative functions should not be underestimated. Yet, it seems that the interplay between ERAP1/ERAP2 and MHC-I molecules should hide the mechanism underlying their joint association with disease. Frontiers in Immunology \| www.frontiersin.org Ankylosing Spondylitis Although a number of non-synonymous ERAP1 polymorphisms were initially found to inﬂuence risk of AS (60, 68), the strong LD within the ERAP1 gene required further reﬁnement to dissect the actual disease-promoting polymorphisms. The association of ERAP1 with AS was most consistent with a two-mutation model involving a primary eﬀect of the high activity K528 variant and a secondary eﬀect determined by D575/R725 (21). Both polymorphisms are jointly present in the AS-associated Hap1 to Hap3 haplotypes (56). Accordingly, Hap10, a low activity haplotype including the two protective changes R528 and N575/Q725, signiﬁcantly inﬂuenced AS protection. The Q730E polymorphism showed a strong association with AS, Q730 being the risk allotype, but ﬁne mapping studies narrowed down the AS-promoting region of ERAP1 to a portion of the gene that excludes codon 730 (69), raising the possibility that Q730E might not directly inﬂuence AS risk. In addition, recent analyses suggest that the association of D575/R725, tagged by rs10050860 at codon 575, is in tight LD with the splice altering SNP rs7063, which is the one that seems to account for the contribution to AS susceptibility (61). The risk allele of this polymorphism leads to increased expression of the shortest ERAP1 isoform, 19E, and higher ERAP1 protein levels. Thus, our current view is that AS risk is favored by high ERAP1 activity, mainly due to the direct eﬀect of K528, without excluding some additional eﬀect of D575/R725, and to linked polymorphisms increasing ERAP1 protein expression of the high activity variants. Other ERAP1 polymorphisms aﬀect the expression level, rather than the structure, of ERAP1, as well as the relative levels of the two isoforms of this enzyme commonly expressed in all individuals (59–61). Whereas, diﬀerences in ERAP1 expression have an obvious eﬀect on peptide cleavage, the eﬀect of altered isoform levels is less straightforward. It seems that diﬀerences in the posttranscriptional dynamics between the two isoforms results in an indirect inﬂuence of the polymorphism aﬀecting isoform proportions on the overall expression of the ERAP1 protein (61). Due to the complexity and variety of eﬀects induced by ERAP1 polymorphism, the functional spectrum of this enzyme across individuals is potentially very diverse. Yet, tight linkage disequilibrium (LD) among many of these polymorphisms probably limits this diversity at the population level. LD can also increase the eﬀect of single SNPs on enzymatic activity by promoting higher expression of the more active allotype. ERAP1 POLYMORPHISM AND EXPRESSION ERAP1 is expressed in all individuals and displays a signiﬁcant degree of polymorphism, much of which aﬀects the enzymatic activity, the expression level or both. Natural ERAP1 variants are complex allotypes including multiple non-synonymous single nucleotide polymorphisms (SNPs), known as haplotypes. Ten ERAP1 haplotypes, designated as Hap1 to Hap10 account for over 99% of the natural ERAP1 variants in human populations (56). Polymorphic amino acids are frequently located near the catalytic site (residues 346, 349), in the peptide binding site (residues 725 and 730) or in interdomain regions or other locations that can aﬀect the conformational rearrangements associated with the acquisition of the enzymatic activity (residues 528 and 575). Therefore, these polymorphisms may inﬂuence ERAP1 activity in multiple ways. Several in vitro studies have established that the K528R polymorphism has large eﬀects This review will focus on recent studies dealing with the eﬀects of the polymorphism and expression of ERAP1 and ERAP2 on October 2018 \| Volume 9 \| Article 2463 2 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro (19–21, 46, 57), due to its inﬂuence on the kinetics of the conformational transition between the active and inactive states of the enzyme (58). In contrast, the Q730E change inﬂuences the enzymatic activity in a way that depends on substrate length, by increasing the preference of ERAP1 for shorter substrates (58). This is likely related to the involvement of this residue in substrate binding. (19–21, 46, 57), due to its inﬂuence on the kinetics of the conformational transition between the active and inactive states of the enzyme (58). In contrast, the Q730E change inﬂuences the enzymatic activity in a way that depends on substrate length, by increasing the preference of ERAP1 for shorter substrates (58). This is likely related to the involvement of this residue in substrate binding. In a recent study a polymorphism located in the ERAP2 promoter region, rs7586269, was associated with opposite changes in the expression of ERAP1 and ERAP2, so that decreased expression of ERAP2 correlated with increased ERAP1 expression, suggesting a concerted regulation of the expression of both genes by this SNP (67). Yet, unlike rs2248374 and rs10044354, the minor allele frequency of rs7586269 is rather low in most populations. ERAP1 POLYMORPHISM AND EXPRESSION The activity of natural ERAP1 variants depends on the speciﬁc combination of polymorphic residues (19, 20) and, due to the diverse alterations induced by diﬀerent mutations, the activity of a given variant cannot be easily predicted from the eﬀects of single changes. Yet, natural ERAP1 variants with K528 (Hap1– Hap3) are more active than those with R528, and the less active haplotype, Hap10, is the one that accumulates the largest number of amino changes that negatively aﬀect ERAP1 activity. Ankylosing Spondylitis A paradigmatic case is the correlation between the highly active variant K528 with SNPs determining increased ERAP1 expression (60–62). ERAP2 POLYMORPHISM AND EXPRESSION In contrast to ERAP1, non-synonymous changes aﬀecting the amino acid sequence of ERAP2 seem to be very limited. A polymorphism coding for the K392N change aﬀects ERAP2 activity (63). Although both variants are expressed in some populations (64), the N392 allele is in strong LD with SNP rs2248374, a frequent polymorphism that promotes nonsense- mediated RNA decay and impairs protein expression (65). As a result, only the K392 variant is expressed in most populations. Since both rs2248374 alleles show similar frequencies due to balancing selection about 25% of individuals fail to express ERAP2. Other polymorphisms, such as rs10044354, or SNPs in strong LD with it, also have a large impact on ERAP2 protein expression (2, 66). Heterozygosity at these loci, and probably additional polymorphisms inﬂuencing quantitative expression of the enzyme (61), introduce a further level of heterogeneity in the activity of ERAP2 across individuals due to diﬀerent enzyme levels. ERAP2 expression, as determined by the rs2248374 SNP, is associated with AS risk, both in HLA-B∗27-positive and -negative individuals (26). One or more additional polymorphisms in tight LD additionally inﬂuence ERAP2 expression in a quantitative way. The corresponding risk alleles are those determining increased ERAP2 levels (61). Thus, as for ERAP1, AS risk is favored by ERAP2 activity. Psoriasis The association of ERAP1 with psoriasis shows a similar pattern as in AS: multiple studies, reviewed in Lopez de Castro et al. (12) and Ombrello et al. (56), suggest that K528 and Q730 in ERAP1 predispose to disease in epistasis with C∗06:02. The major risk haplotype is Hap2, which includes these two residues, and the protective haplotype is Hap10, which includes residues R528 and E730. The epistasis between ERAP1 and C∗06:02 was challenged by one study upon stratiﬁcation of patients by age at onset (71). Behçet’s Disease The initial report detecting the epistatic association of ERAP1 and HLA-B∗51 with BD showed that the polymorphisms encoding the D575N/R725Q changes were associated with disease risk, ﬁtting a recessive model, so that risk of BD is conferred by homozygosity at this locus (24). Further studies determined that Hap10 is the risk haplotype for this disease (70). Thus, the same ERAP1 variant that is protective for AS October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 3 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro Conditional analyses controlling for the contribution of ERAP1 suggested an association of ERAP2 with psoriasis independent of ERAP1 (27) and also that the risk allele of ERAP1 and the protective allele of ERAP2, the latter tagged by the intronic SNP rs2910686, frequently segregated together. Since the risk allele of rs2910686 (C) was the same as in AS (25) this result suggests that, like in AS, ERAP2 expression may predispose to psoriasis, although the eﬀect may often be masked by ERAP1. In these studies no subdivision of disease by association with C∗06:02 or age at onset was considered. In a recent report from Poland (72) such subdivision was carried out. Although somewhat limited by the size of the cohorts, this study, which tagged ERAP2 using rs2248374, a polymorphism directly determining ERAP2 protein expression (65), found that presence of ERAP2 was protective from psoriasis, although this association showed up only among C∗06:02-positive individuals with very early disease onset. In spite of the limitations imposed by sample size, this and other studies suggest that the association of both ERAP1 and ERAP2 with psoriasis is probably variable in this heterogeneous pathology, and a precise deﬁnition of the role of these enzymes, and their functional interaction with C∗06:02, should beneﬁt from large scale studies allowing robust statistical assessments within disease subsets. increases risk of BD, acting in both cases in epistasis with the susceptibility MHC allele. This ﬁnding strongly suggests that the pathogenetic role of ERAP1 in the two diseases is MHC-I allotype-dependent, presumably related to eﬀects on the processing of their corresponding ligands. To my knowledge there is no evidence for an association of ERAP2 with BD, making this disease the only one of the four major MHC-I associated disorders for which a role of this enzyme has not been demonstrated. Birdshot Chorioretinopathy Evidence for an involvement of ERAP1 in BSCR was suspected from the signiﬁcant eﬀects of ERAP1 polymorphism on the A∗29:02 peptidome (66) and has been demonstrated in a recent study (29) that identiﬁed the low activity Hap10 haplotype as a risk factor for this disease. Its contribution was further enhanced by co-occurrence with the splice-altering variant rs7063, inﬂuencing the relative expression of the ERAP1 isoforms, whose risk allele (T) determines lower protein levels. Unlike the situation in BD, the association of Hap10 with BSCR does not appear to follow a recessive model. Also unlike BD, ERAP2 presence and higher expression level are risk factors for BSCR (2, 29). The inﬂuence of ERAP1 and ERAP2 in MHC-I-associated diseases is summarized in Table 1. THE PEPTIDOMES OF DISEASE ASSOCIATED MHC-I MOLECULES As virtually all other human MHC-I molecules, the peptidomes of those associated with disease share a number of fundamental features, the most important of which are the following (73– 78): (1) MHC-I ligands show strong length restrictions, with an optimal size of nine amino acid residues, and frequencies falling abruptly with shorter or larger length, (2) peptides are bound to the MHC-I molecule into speciﬁc pockets of its peptide binding site, termed A–F; the main interactions involve the peptidic amino- and carboxyl-termini, in pockets Deﬁning the association of ERAP2 with psoriasis (27, 28) has been complicated due to the seemingly non-epistatic nature of this association, the heterogeneity of the disease and the apparently opposite eﬀects of ERAP1 and ERAP2 in the same haplotype, leading to a masking of the latter by ERAP1. TABLE 1 \| ERAP1 and ERAP2 in MHC-I-associated diseasesa. Disease ERAP1 riskb ERAP2 risk Genetics Enzymatic activity Genetics Enzymatic activity Ankylosing spondylitis Hap1-Hap3 Higher expression High ERAP2 presence and higher expression level High Behçet’s disease Hap10 (recessive) Low Not reported Not reported Birdshot chorioretinopathy Hap10 (non-recessive) Lower expression Low ERAP2 presence and higher expression level High Psoriasis Hap2 Disease subset dependent (?) High Disease subset dependent (?) Disease subset dependent (?) Disease subset dependent (?) aSee text for references. bERAP1 is in epistasis with the susceptibility MHC-I allele at least in AS, BD, and psoriasis, although in the latter might be differences among disease subsets. As for BSCR, the association with A29:02 is virtually 100% (3). For the polymorphic nucleotide and amino acid changes of the Hap1-Hap3 and Hap10 variants of ERAP1 see (56). TABLE 1 \| ERAP1 and ERAP2 in MHC-I-associated diseasesa. Disease ERAP1 riskb ERAP2 risk Genetics Enzymatic activity Genetics Enzymatic activity Ankylosing spondylitis Hap1-Hap3 Higher expression High ERAP2 presence and higher expression level High Behçet’s disease Hap10 (recessive) Low Not reported Not reported Birdshot chorioretinopathy Hap10 (non-recessive) Lower expression Low ERAP2 presence and higher expression level High Psoriasis Hap2 Disease subset dependent (?) High Disease subset dependent (?) Disease subset dependent (?) Disease subset dependent (?) aSee text for references. bERAP1 is in epistasis with the susceptibility MHC-I allele at least in AS, BD, and psoriasis, although in the latter might be differences among disease subsets. As for BSCR, the association with A29:02 is virtually 100% (3). For the polymorphic nucleotide and amino acid changes of the Hap1-Hap3 and Hap10 variants of ERAP1 see (56). THE PEPTIDOMES OF DISEASE ASSOCIATED MHC-I MOLECULES MHC-I polymorphism can dramatically alter the size, shape, and polarity of side chain binding pockets. The main features of the ligands of major disease-associated MHC-I molecules are summarized in Figure 1. THE PEPTIDOMES OF DISEASE ASSOCIATED MHC-I MOLECULES TABLE 1 \| ERAP1 and ERAP2 in MHC-I-associated diseasesa. aSee text for references. bERAP1 is in epistasis with the susceptibility MHC-I allele at least in AS, BD, and psoriasis, although in the latter might be differences among disease subsets. As for BSCR, the association with A29:02 is virtually 100% (3). For the polymorphic nucleotide and amino acid changes of the Hap1-Hap3 and Hap10 variants of ERAP1 see (56). aSee text for references. bERAP1 is in epistasis with the susceptibility MHC-I allele at least in AS, BD, and psoriasis, although in the latter might be differences among disease subsets. As for BSCR, the association with A29:02 is virtually 100% (3). For the polymorphic nucleotide and amino acid changes of the Hap1-Hap3 and Hap10 variants of ERAP1 see (56). October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 4 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro identiﬁcation of high numbers of HLA-B∗27 ligands allowed the reﬁnement of these motifs and the characterization of new ones (79, 84–86), providing a much more accurate picture of the HLA- B∗27 peptidome that, nevertheless did not alter the essential features deﬁned by earlier studies. One of these features, which is particularly relevant to this review, is the remarkable preference of HLA-B27 for basic (R, K) and small (G, A, S) N-terminal residues, which account for about 20 and 48% of the peptidome, respectively. Along with the Arg2 motif, the high frequency of HLA-B∗27 ligands with dibasic N-terminal sequences, which are particularly resistant to digestion by cytosolic amino-peptidases, might endow HLA-B∗27 with the capacity to present peptides generated in particularly low amounts (87). In addition, the HLA- B∗27 binding motif is particularly well suited to the TAP binding preferences (88–90), which should favor the import into the ER of HLA-B∗27 ligands directly generated in the cytosol. A and F, respectively, as well as a few main anchor residues, which in humans are generally located at position (P)2 and C-terminal (PC) and bind in pockets B and F, respectively, (3) other interactions involve the peptidic main chain and a number of additional amino acid side chains of secondary anchor residues, which are generally those at P1, P3, and PC-2, without excluding others. Diﬀerent HLA-A, -B, and -C allelic products typically diﬀer in the nature of the main anchor residues, which provide the optimal binding to each particular allotype. The HLA-B∗27 Peptidome HLA-B∗27 was one of the earliest MHC molecules whose structure (76, 77) and peptidome (82) were investigated, and one of the most intensively studied, due to its association with AS. The initial studies already determined that the main anchor residues of the B∗27:05 prototype were Arg2 and basic, aliphatic and aromatic residues at PC. Whereas, Arg2 was a hallmark of most HLA-B∗27 ligands, the polymorphism of the major subtypes, including some diﬀerentially associated with AS, frequently aﬀected C-terminal residues, drastically decreasing the tolerance for basic residues (i.e., B∗27:02, B∗27:04), or restricting this position almost only to nonpolar ones (i.e., B∗27:06) (83). With the arrival of mass spectrometry-based immunopeptidomics, the The HLA-B∗51 Peptidome cell line. An 8-fold increase of extended ligands (from 3 to 24 peptides) was observed in ERAP1-inhibited cells compared to the ERAP1 proﬁcient counterpart, of which most were C-terminally extended peptides carrying the Arg2 motif. These observations illustrate the relevance of ERAP1 in optimizing the length of MHC-I ligands, a well-established function of this enzyme. Some ﬂuctuation in P1 residue frequencies, including a decrease of Arg1, was also observed in ERAP1-deﬁcient cells. However, the moderate number of identiﬁed peptides in this study and the low frequency of many P1 residues limited the statistical signiﬁcance of these ﬂuctuations and precluded a precise assessment of the inﬂuence of ERAP1 on P1 residue usage. Several distinctive features, compared to other disease-associated MHC-I molecules, characterize the HLA-B∗51 peptidome (80, 92). It shows a particularly high percentage of octamers and low allowance for peptides >9-mers. Most important, it has a dual motif of Ala2 and Pro2, which deﬁnes two very diﬀerent subpeptidomes, both of which share a same, very restrictive, C- terminal motif of aliphatic residues, including Val and Ile, and a lower percentage of Leu. The Ala2 and Pro2 subpeptidomes drastically diﬀer in the usage of P1 residues. Whereas, Leu and other nonpolar residues are favored at P1 among peptides with Pro2, in the Ala2 subpeptidome Asp1 is largely predominant. The Ala2 subpeptidome shows lower aﬃnity (80), which is consistent with experimental binding studies (94, 95). Uniquely among de major disease-associated MHC-I molecules, basic residues at P1 are virtually absent among HLA-B∗51 ligands, a feature that might be related to the lack of association of ERAP2 with BD. A more recent study reported a huge number (about 15,000) of HLA-B∗27 ligands from ERAP1-proﬁcient transgenic rats and the counterparts carrying a homozygous deletion of the ERAP1 gene (98). An interesting outcome of this study, also observed with functional ERAP1 variants (see below) is that only about one-third of the HLA-B∗27 peptidome was altered upon ERAP1 depletion, suggesting that most of the peptides are not subjected to ERAP1 processing or that their generation/destruction balance results in unchanged peptide amounts, relative to absence of the enzyme. A signiﬁcant eﬀect on peptide length, resulting in longer ligands in the absence of ERAP1 was also observed, as well as changes in P1 residue usage, in particular among peptides showing quantitative expression diﬀerences. ERAP1 AND THE PEPTIDOMES OF DISEASE-ASSOCIATED MHC-I MOLECULES In various studies from our laboratory (99–102) we addressed the eﬀects of natural ERAP1 polymorphism on the HLA- B∗27 peptidome. The focus was on the quantitative changes in peptide amounts, rather than on peptides detected only in a speciﬁc ERAP1 context, and the strategy used was to compare the HLA-B∗27 peptidomes from human cell lines expressing distinct ERAP1 variants. Already in our initial study (99), we observed that the HLA-B∗27 peptidome expressed in the context of an AS-associated ERAP1 haplotype (Hap2) diﬀered substantially from the peptidome expressed in an AS-protective context (Hap10) in showing a higher percentage of 9-mers and increased frequencies of ERAP1-resistant ﬂanking and P1 residues. A second study (100) further determined that of the two polymorphisms accounting for the association of ERAP1 with AS, namely at residues 528 and 575/725, the former one dominated the alterations of the HLA-B∗27 peptidome. Two types of studies have addressed the role of ERAP1 on shaping disease-associated MHC-I peptidomes: those involving the depletion of ERAP1 and those addressing the role of ERAP1 polymorphism. ERAP1 is expressed in the normal cells of all individuals, although its expression can be largely diminished or absent altogether in transformed cells (96). The eﬀects of ERAP1 downregulation on the peptidome have been examined so far in HLA-B∗27 (97, 98) and A∗29:02 (66).The eﬀects of ERAP1 polymorphism on the peptidomes expressed in live cells have been examined in HLA-B∗27 (99–102), HLA-A∗29 (66), and HLA-B∗51 (103). Considering the complexity of ERAP1 haplotypes, the diverse inﬂuence of individual mutations (58), and the combined eﬀects of co-occurring polymorphisms (19, 20), we recently attempted to rank the eﬀects of individual changes in ERAP1 on the HLA-B∗27 peptidome through a multilevel comparison between cell lines with appropriate ERAP1 backgrounds (102). This study conﬁrmed the dominant eﬀect of K528R on decreasing the enzymatic activity of ERAP1, as reﬂected mainly in the The HLA-C∗06:02 Peptidome By far the best characterization of the C∗06:02 peptidome has been reported by Mobbs and colleagues (81). Based on their study this molecule binds mainly 9-mers, is less tolerant than HLA- B27 for 10-mers and 11-mers, and somewhat more for longer peptides. The predominant residue at P2 is Arg, but its frequency is much smaller than in HLA-B∗27 (about 30 vs. >95%). C- terminal residues are predominantly aliphatic/aromatic, which is frequent among MHC-I molecules, including HLA-B∗27. A major distinctive feature of C∗06:02 is a prominent motif of basic residues at PC-2 (R+K: ∼41%), suggesting that this may be a main anchor position of C∗06:02 ligands. For the purpose of this review it is important to note the diversity at P1, which is comparable to that in the A∗29:02 peptidome, so that no single residue has a frequency higher than ≈10%. The HLA-B∗51 Peptidome For instance, Gly1 was more frequent among peptides predominant in the presence of ERAP1 and Ala1, Ser1, and Lys1 were increased in the absence of the enzyme. Another interesting observation is that the main anchor residues that most contribute to binding aﬃnity, Arg2 and C-terminal aliphatic/aromatic residues, were more frequent among peptides longer tan 9-mers, which suggests a compensatory eﬀect for their suboptimal length. The HLA-A∗29:02 Peptidome Beyond earlier studies (91), the most extensive characterization of the A∗29:02 peptidome is found in three recent reports (66, 92, 93). Unlike other disease-associated MHC-I molecules A∗29:02 shows only one dominant, highly restrictive, main anchor motif of Tyr at PC. Some looser restriction for aliphatic/aromatic residues is observed at P2 and PC-2. There are no clear preferences at P1, although aromatic residues (Y, F) jointly account for about 20% of the A∗29:02 ligands. FIGURE 1 \| Major features of the peptidomes of disease-associated MHC-I molecules. Length distribution and % residue frequencies at P1, P2, and PC of B27:05, A29:02, B51:01, and C06:02 ligands (N: 1,117, 4,931, 1,620, and 1,287 peptides, respectively). The data are from the following references: B27:05 (79), A29:02 (66), B51:01 (80), and C06:02 (81). FIGURE 1 \| Major features of the peptidomes of disease-associated MHC-I molecules. Length distribution and % residue frequencies at P1, P2, and PC of B27:05, A29:02, B51:01, and C06:02 ligands (N: 1,117, 4,931, 1,620, and 1,287 peptides, respectively). The data are from the following references: B27:05 (79), A29:02 (66), B51:01 (80), and C06:02 (81). October 2018 \| Volume 9 \| Article 2463 5 Frontiers in Immunology \| www.frontiersin.org Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro Frontiers in Immunology \| www.frontiersin.org ERAP1 and the HLA-B∗51 Peptidome ERAP1 and the HLA-B 51 Peptidome Our initial characterization of the HLA-B∗51 peptidome (80) concerned B∗51:01 in an ERAP2-positive cell line (721.221) expressing the Hap1/Hap8 haplotypes (104), of which Hap1 is of high activity. In another study we determined the features of the B∗51:08 peptidome from an ERAP2-positive cell line expressing the low activity ERAP1 variant Hap10 (103). B51:08 is, like B∗51:01, associated with BD and diﬀers from the latter only by the E152V and L156D changes (103, 105). Hap10 is the ERAP1 haplotype associated with risk of BD (70). Therefore, it was of interest to assess the eﬀect of this low activity haplotype on the HLA-B∗51 peptidome. The location of the polymorphic residues between B∗51:01 and B∗51:08 at positions far away from the interaction site of the peptidic N-terminus facilitated the distinction of the eﬀects due to subtype polymorphism from those depending on the ERAP1 context when comparing both HLA-B∗51 peptidomes (103). y Further complexity is introduced by polymorphisms that aﬀect ERAP1 protein levels, either by altering gene expression or through alternative splicing (61). For instance, the risk K528 variant of ERAP1 is consistently associated with higher ERAP1 protein because the SNP rs30187, coding for the K528R change, is in LD with one or more polymorphisms altering gene expression (60–62). In addition, as already mentioned, the splicing altering variant rs7603 determines the relative amounts of two ERAP1 isoforms diﬀering in Exon 20, which codes for a short C-terminal portion of the molecule. Because of their distinct transcriptional dynamics, alterations in the relative expression of both isoforms result in distinct protein levels. Conditional analyses showed that rs7603 accounted for the association of D575N/R725Q with AS, suggesting that the contribution of this protein polymorphism to disease risk is mainly through the eﬀect of the linked rs7603 allele on increasing protein levels (61). This may explain the signiﬁcant association of D575N/R725Q with AS in spite of the relatively minor eﬀect of these changes on the B∗27:05 peptidome (100, 102). ERAP1 and the HLA-A∗29:02 Peptidome ERAP1 and the HLA-A∗29:02 Peptidome The eﬀects of ERAP1 on the A∗29:02 peptidome were analyzed through comparisons involving three ERAP2-negative LCL, two of which expressed ERAP1 variants with distinct enzymatic activity, and the third one expressing very low ERAP1 levels (66). In a high activity ERAP1 context (Hap2/3) the A∗29:02 peptidome showed the following diﬀerences compared to either a peptidome expressed in the context of lower activity variants (Hap6/8) or low ERAP1 protein levels (about 16- fold less): (1) increased amounts of 9-mers relative to longer peptides, (2) bulkier and more hydrophobic P1 and P2 residues, (3) higher global aﬃnity for A∗29:02, (4) higher global hydrophobicity (Figure 3). These eﬀects were similar in both comparisons, indicating rather small diﬀerences between the A∗29:02 peptidomes generated with either low activity ERAP1 Frontiers in Immunology \| www.frontiersin.org ERAP1 and the HLA-B∗27 Peptidome Effects of ERAP1 Depletion Depletion of ERAP1 in human cells in one study (97) resulted in longer peptides, a signiﬁcant increase of C-terminally extended ligands, and some alteration in P1 residue frequencies. The percentage of 9-mers relative to longer peptides decreased from about 50% in the wild type to about 35% in the ERAP1-inhibited October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 6 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro increased frequency of 9-mers, relative to longer peptides, the skewing of P1 frequencies toward ERAP1-resistant residues, and the increased aﬃnity of the peptidome in the K528 context (Figure 2A). The eﬀect of D575N/R725Q was much smaller, but added to that of polymorphism at residue 528 in decreasing ERAP1 trimming, thus accounting for the low activity of the protective Hap10 haplotype, carrying both changes. This study also showed a notorious eﬀect of the Q730E polymorphism on trimming depending on peptide length, which had been previously noted in vitro using poly-Gly analogs (58). In HLA- B∗27 this was reﬂected in higher abundance of shorter peptides (≤9-mers) with E730, and opposite eﬀects of the E730Q change on P1 residue frequencies among 9-mers compared to longer peptides, depending on their susceptibility to ERAP1 trimming (Figure 2B), with little consequence on the global aﬃnity of the peptidome. The distinct alterations resulting from the length- independent eﬀects of K528R and the length-dependent eﬀects of Q730E introduce a signiﬁcant complexity in the functional features of ERAP1 haplotypes. Thus, at the population level, ERAP1 polymorphism can inﬂuence the HLA-B27 peptidome in very diverse ways, due to the multiplicity of ERAP1 haplotypes and the distinct combinations of functionally diﬀerent mutations within each haplotype. variants or low ERAP1 levels. Like in HLA-B∗27 these changes aﬀected only a fraction of the peptidome and the amounts of many ligands remained essentially unaltered regardless of the ERAP1 context. Given the association of the low activity Hap10 haplotype with BSCR (29) this disease might be favored by an A∗29:02 peptidome with at least a fraction of peptides showing suboptimal length, lower hydrophobicity and decreased aﬃnity. ERAP1 and the HLA-B∗51 Peptidome The following eﬀects were observed in the less active ERAP1 context of B∗51:08, relative to B∗51:01 (Figure 4): (1) the percentage of octamers was decreased, and that of nonamers increased, reﬂecting lower trimming, (2) The Ala2 subpeptidome was increased, at the expense of the Pro2 subpeptidome; the diﬀerential distribution of P1 residues in both subpeptidomes was similar in B∗51:08 and B∗51:01, but the Ala2 subpeptidome showed a higher percentage of peptides with ERAP1-susceptible residues in B∗51:08, according to the lower activity of its ERAP1 background, (3) since the Pro2 subpeptidome has higher aﬃnity, this resulted in lower global aﬃnity of the HLA-B∗51:08 peptidome expressed in the Hap10 context. ERAP1 and the HLA-C∗06:02 Peptidome To my knowledge there are no studies directly addressing the eﬀects of ERAP1 on the C∗06:02 peptidome. However, based on the similar frequency of ERAP1-susceptible P1 residues (A,C, L, M, Y) in C∗06:02 compared to HLA-B∗27 (about 20 and 27%, respectively), it is plausible to predict that the eﬀects of this enzyme on C∗06:02 will also be signiﬁcant. ERAP2 AND THE PEPTIDOMES OF DISEASE-ASSOCIATED MHC-I MOLECULES Unlike ERAP1, ERAP2 is not expressed in at least 25% of individuals due to frequent polymorphisms that impair protein expression (2, 65). Most ERAP2-positive individuals seem to express a single variant, characterized by the presence of Lys at position 392 (65). Thus, it is the presence or absence of ERAP2 and its expression level, rather than polymorphism October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org Frontiers in Immunology \| www.frontiersin.org 7 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro Inﬂuence of ERAP1 polymorphism on the HLA-B27 peptidome. (A) Effects of K528R. Comparison of the B27:05 ligands over-represented (>1-fold) in a 2, N: 2215 peptides) expressing the Hap3 variant of ERAP1 (red), with those over-represented in a cell line (C1R05, N: 2801) expressing the Hap8 variant p3 and Hap8 differ by R127P, I276M, and K528R, but the two former changes have much less inﬂuence on peptide trimming. The comparisons involve gth (upper panel), joint frequencies of P1 residues with low or intermediate+high susceptibility to ERAP1 among 9-mers (second panel) or longer peptides , and theoretical afﬁnity of the total ligands, with the medians indicated by bars. (B) Effects of E730Q. Comparison of the B27:05 ligands over-represented the LG2 cell line (Hap3: red), with those over-represented in LCL 6370 (blue) expressing the Hap2 variant (N: 2,444 and 2,869 peptides, respectively). Hap2 differ only by the E730Q change. Conventions are as in (A). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by e upper panels) or Mann–Whitney tests (lower panels). These data were originally published in reference (102). FIGURE 2 \| Inﬂuence of ERAP1 polymorphism on the HLA-B27 peptidome. (A) Effects of K528R. Comparison of the B27:05 ligands over-represented (>1-fold) in a cell line (LG2, N: 2215 peptides) expressing the Hap3 variant of ERAP1 (red), with those over-represented in a cell line (C1R05, N: 2801) expressing the Hap8 variant (green). Hap3 and Hap8 differ by R127P, I276M, and K528R, but the two former changes have much less inﬂuence on peptide trimming. The comparisons involve peptide length (upper panel), joint frequencies of P1 residues with low or intermediate+high susceptibility to ERAP1 among 9-mers (second panel) or longer peptides (third panel), and theoretical afﬁnity of the total ligands, with the medians indicated by bars. (B) Effects of E730Q. ERAP2 AND THE PEPTIDOMES OF DISEASE-ASSOCIATED MHC-I MOLECULES (B) Comparison of the A29:02 ligands over-represented (>3-fold) in PF97387 (N: 446 peptides), with those equally over-represented in the cell line SWEIG (N: 390 peptides) expressing very low levels of ERAP1 (green). All conventions are as in (A). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (upper panels) or Mann–Whitney tests (lower panels). These data were originally published in reference (66). FIGURE 3 \| Inﬂuence of ERAP1 on the A29:02 peptidome. (A) Comparison of the A29:02 ligands over-represented (>3-fold) in a cell line (PF97387, N: 292 peptides) expressing the Hap2/Hap3 variants of ERAP1 (blue), with those equally over-represented in a cell line (MOU, N: 383) expressing the Hap6/Hap8 variants (red). Both haplotype combinations differ by the R127P and K528R changes. The comparisons involve peptide length (upper panel), mean side chain volume at each peptide position among the 9-mers in these peptide sets (second panel), theoretical afﬁnity of these 9-mers, with the medians indicated by bars (third panel), and hydropathy of the same peptides, estimated by the Grand Average of Hydropathy (GRAVY) index (lower panel). (B) Comparison of the A29:02 ligands over-represented (>3-fold) in PF97387 (N: 446 peptides), with those equally over-represented in the cell line SWEIG (N: 390 peptides) expressing very low levels of ERAP1 (green). All conventions are as in (A). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (upper panels) or Mann–Whitney tests (lower panels). These data were originally published in reference (66). ERAP2 AND THE PEPTIDOMES OF DISEASE-ASSOCIATED MHC-I MOLECULES Comparison of the B27:05 ligands over-represented (>1-fold) in the LG2 cell line (Hap3: red), with those over-represented in LCL 6370 (blue) expressing the Hap2 variant (N: 2,444 and 2,869 peptides, respectively). Hap3 and Hap2 differ only by the E730Q change. Conventions are as in (A). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (three upper panels) or Mann–Whitney tests (lower panels). These data were originally published in reference (102). October 2018 \| Volume 9 \| Article 2463 8 Frontiers in Immunology \| www.frontiersin.org Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro nﬂuence of ERAP1 on the A29:02 peptidome. (A) Comparison of the A29:02 ligands over-represented (>3-fold) in a cell line (PF97387, N: 292 pressing the Hap2/Hap3 variants of ERAP1 (blue), with those equally over-represented in a cell line (MOU, N: 383) expressing the Hap6/Hap8 variants aplotype combinations differ by the R127P and K528R changes. The comparisons involve peptide length (upper panel), mean side chain volume at each ion among the 9-mers in these peptide sets (second panel), theoretical afﬁnity of these 9-mers, with the medians indicated by bars (third panel), and f the same peptides, estimated by the Grand Average of Hydropathy (GRAVY) index (lower panel). (B) Comparison of the A29:02 ligands nted (>3-fold) in PF97387 (N: 446 peptides), with those equally over-represented in the cell line SWEIG (N: 390 peptides) expressing very low levels of n). All conventions are as in (A). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (upper panels) or ey tests (lower panels). These data were originally published in reference (66). FIGURE 3 \| Inﬂuence of ERAP1 on the A29:02 peptidome. (A) Comparison of the A29:02 ligands over-represented (>3-fold) in a cell line (PF97387, N: 292 peptides) expressing the Hap2/Hap3 variants of ERAP1 (blue), with those equally over-represented in a cell line (MOU, N: 383) expressing the Hap6/Hap8 variants (red). Both haplotype combinations differ by the R127P and K528R changes. The comparisons involve peptide length (upper panel), mean side chain volume at each peptide position among the 9-mers in these peptide sets (second panel), theoretical afﬁnity of these 9-mers, with the medians indicated by bars (third panel), and hydropathy of the same peptides, estimated by the Grand Average of Hydropathy (GRAVY) index (lower panel). ERAP2 and HLA-B∗27 altering its enzymatic activity, which associates with disease risk or protection. So far the eﬀects of ERAP2 have been examined on the HLA-B∗27 and A∗29:02 peptidomes, since ERAP2 expression is a risk factor for both AS and BSCR. The eﬀects of ERAP2 expression on the HLA-B∗27 peptidome were initially examined on cell lines expressing highly active ERAP1 variants (Hap1 or Hap2), but diﬀering in the expression October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 9 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro FIGURE 4 \| Inﬂuence of ERAP1 polymorphism on the HLA-B51 peptidome. Comparison of the HLA-B51:01 ligands from the transfectant cell line 721.221-B51 (N: 1,271 peptides) expressing the Hap1/Hap8 variants of ERAP1 (blue), with the B51:08 ligands from LCL BCH-30 (N: 624 peptides) expressing the Hap10 variant (red). The comparisons involve (A) peptide length distribution, (B) percent of peptides with Pro2 or Ala2, (C) Percent residue frequencies at P1 among peptides with Pro2 (upper panel) or Ala2 (lower panel), (D) percent of peptides with P1 residues showing low or intermediate+high susceptibility to ERAP1 trimming among peptides with Pro2 (upper panel) or Ala2 (lower panel), (E) theoretical afﬁnity of B51:01 and B51:08 ligands for their respective B51 subtypes, with the medians indicated by bars. The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (A–D) or Mann–Whitney tests (E). These data were originally published in reference (103). t f ERAP2 (106) Th f thi ﬀ t d th hi h f HLA B∗27 li d d th i tibilit FIGURE 4 \| Inﬂuence of ERAP1 polymorphism on the HLA-B51 peptidome. Comparison of the HLA-B51:01 ligands from the transfectant cell line 721.221-B51 (N: 1,271 peptides) expressing the Hap1/Hap8 variants of ERAP1 (blue), with the B51:08 ligands from LCL BCH-30 (N: 624 peptides) expressing the Hap10 variant (red). The comparisons involve (A) peptide length distribution, (B) percent of peptides with Pro2 or Ala2, (C) Percent residue frequencies at P1 among peptides with Pro2 (upper panel) or Ala2 (lower panel), (D) percent of peptides with P1 residues showing low or intermediate+high susceptibility to ERAP1 trimming among peptides with Pro2 (upper panel) or Ala2 (lower panel), (E) theoretical afﬁnity of B51:01 and B51:08 ligands for their respective B51 subtypes, with the medians indicated by bars. ERAP2 and HLA-B∗27 The statistical signiﬁcance of the changes is indicated by the p-values (left and right panels) or asterisks (middle panel), as estimated by the χ2 (two left panels) or Mann–Whitney tests (right panel). The data in the two left panels and those in the right panel were originally published in Martin-Esteban et al. (101, 106), respectively. frequencies of hydrophobic ones (Figure 6). These changes did not alter the overall aﬃnity of the peptidome. the allosteric activation of ERAP1, as reported for ERAP1/2 heterodimers in vitro (107, 108), since this would presumably induce a more widespread alteration of P1 frequencies. Yet, considering the relatively low eﬃciency of ERAP2 with long substrates (52), the eﬀects on peptide length are compatible with some improvement of ERAP1 trimming in the presence of ERAP2. Thus, our study did not exclude a functional interaction between both enzymes in the processing of the HLA-B∗27 peptidome, but suggested that they largely act as separate entities in live cells. The length eﬀects of ERAP2 were opposite to those observed in HLA-B∗27 and suggested some protection by ERAP2 from ERAP1 trimming, which might be mediated by unproductive binding of long peptides to the former enzyme. That this eﬀect was not observed in HLA-B∗27 can be explained by the distinct nature of HLA-B∗27 ligands. These have about two-fold higher frequency of basic P1 residues, which are trimmed by ERAP2 and are ERAP1-resistant, thus diminishing the chances for unproductive binding of non-cognate ligands to ERAP2. In contrast, aliphatic/aromatic P1 residues, which are generally ERAP1-susceptible but ERAP2-resistant are favored and more frequent in A∗29:02. Moreover, A∗29:02 ligands are more hydrophobic and might bind better ERAP2 than the more polar HLA-B∗27 ligands, due to the relatively low electrostatic potential of the substrate binding site of this enzyme (52). We next addressed the relevance of ERAP2 in distinct ERAP1 contexts (101). When the HLA-B∗27 peptidomes expressed in a low activity ERAP1 background and presence of ERAP2 (Hap10/ERAP2+) and in a high activity ERAP1 context and absence of ERAP2 (Hap1 or Hap2/ERAP2-) were compared, the changes in peptide length and P1 residue usage reproduced those observed in ERAP2-positive cells diﬀering in the activity of their ERAP1 variants (Hap2 vs. Hap10), but, in addition, the amounts of peptides with basic P1 residues were signiﬁcantly lower in the presence of ERAP2. ERAP2 and HLA-B∗27 The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 (A–D) or Mann–Whitney tests (E). These data were originally published in reference (103). or not of ERAP2 (106). The presence of this enzyme aﬀected the peptidome at two levels. The ﬁrst one was an increase in the amounts of 9-mers relative to longer peptides. The second one was a substantial decrease in the amounts of peptides with basic P1 residues (Figure 5), which can be explained by their relatively high frequency among HLA-B∗27 ligands and their susceptibility to ERAP2 trimming, illustrating the role of this enzyme in epitope destruction. The predominant eﬀect of ERAP2 on HLA- B∗27 ligands with basic P1 residues strongly argues in favor of a direct eﬀect of the enzyme on trimming, rather than through October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 10 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro FIGURE 5 \| Inﬂuence of ERAP2 on the HLA-B27 peptidome. Comparison of the B27:05 ligands over-represented (>1.5-fold) in the ERAP2-negative LCL 10151 (N: 764 peptides) expressing the Hap2 variant of ERAP1 (green), with those over-represented in the ERAP2 positive LCL 6370 (N: 817) expressing the same ERAP1 variant (red). The comparisons involve peptide length distribution (left panel), percent frequencies of P1 residues (middle panel), and theoretical afﬁnity of HLA-B27 ligands (N: 4,945), with N-terminal basic or other residues, with the medians indicated by bars. The statistical signiﬁcance of the changes is indicated by the p-values (left and right panels) or asterisks (middle panel), as estimated by the χ2 (two left panels) or Mann–Whitney tests (right panel). The data in the two left panels and those in the right panel were originally published in Martin-Esteban et al. (101, 106), respectively. FIGURE 5 \| Inﬂuence of ERAP2 on the HLA-B27 peptidome. Comparison of the B27:05 ligands over-represented (>1.5-fold) in the ERAP2-negative LCL 10151 (N: 764 peptides) expressing the Hap2 variant of ERAP1 (green), with those over-represented in the ERAP2 positive LCL 6370 (N: 817) expressing the same ERAP1 variant (red). The comparisons involve peptide length distribution (left panel), percent frequencies of P1 residues (middle panel), and theoretical afﬁnity of HLA-B27 ligands (N: 4,945), with N-terminal basic or other residues, with the medians indicated by bars. ERAP2 and HLA-B∗27 This study conﬁrmed the high degree of functional independence of both enzymes in vivo, and also revealed an eﬀect of ERAP2 on lowering the aﬃnity of the HLA-B∗27 peptidome (Figure 5), which was accounted for by the positive contribution of basic P1 residues to HLA-B∗27 binding (109, 110). It is interesting to contextualize these eﬀects, along with those of ERAP1 (Figure 3), with the recent genetic ﬁndings concerning the combined association of the low activity and decreased expression of the Hap10 variant of ERAP1 and the presence and higher expression level of ERAP2 with BSCR (29). Both low ERAP1 activity and the presence of ERAP2 lead to increased peptide length among A∗29:02 ligands. Thus, the eﬀects of both risk factors add to each other in this feature. In contrast, the eﬀects on P1 residue frequencies were diﬀerent: whereas low ERAP1 activity leads to decreased frequency of hydrophobic P1 residues, ERAP2 expression had the opposite eﬀect. These observations might suggest that it is the eﬀect of trimming on peptide length, rather than other features, what relates to BSCR, which in turn might point out to the relevance of particular uveitogenic peptide(s), rather than alterations of general features of A∗29:02, in the pathogenesis of this disease. Frontiers in Immunology \| www.frontiersin.org ERAP2 and HLA-A∗29:02 The eﬀects of ERAP2 on the A∗29:02 peptidome have a particular interest due to the association of this enzyme with BSCR (2), a disease exclusively aﬀecting A∗29:02-positive individuals (3). In a recent study we addressed this issue by comparing the A∗29:02 peptidomes from two pairs of ERAP1-concordant cell lines diﬀering in the expression or not of ERAP2 and expressing, in all cases, high activity ERAP1 variants (93). Two major eﬀects were observed: (1) the presence of ERAP2 resulted in decreased amounts of 9-mers, relative to longer peptides, and (2) the predominant peptides in the presence of ERAP2 showed lower frequencies of ERAP2-susceptible P1 residues and higher Like in HLA-B∗27, the lower frequency of ERAP2-susceptible, but not of ERAP1-susceptible residues in the presence of ERAP2 argues toward a direct and largely independent eﬀect of this enzyme, relative to ERAP1, on peptide trimming. October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 11 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro FIGURE 6 \| Inﬂuence of ERAP2 on the A29:02 peptidome. Comparison of the A29:02 ligands over-represented (>1.5-fold) in the mock-transfected ERAP2-negative LCL PF97387 (PG-GFP, N: 839 peptides) expressing the Hap2/Hap3 variants of ERAP1 (green), with those (red) over-represented in the same ERAP2-transfected LCL (PF-ERAP2, N: 990 peptides). The comparisons involve peptide length distribution (left panel), joint frequencies of P1 residues showing low or intermediate+high susceptibility to ERAP2 trimming (middle panel) or showing low+intermediate or high hydropathy (right panel). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 test. These data were originally published in reference (93). FIGURE 6 \| Inﬂuence of ERAP2 on the A29:02 peptidome. Comparison of the A29:02 ligands over-represented (>1.5-fold) in the mock-transfected ERAP2-negative LCL PF97387 (PG-GFP, N: 839 peptides) expressing the Hap2/Hap3 variants of ERAP1 (green), with those (red) over-represented in the same ERAP2-transfected LCL (PF-ERAP2, N: 990 peptides). The comparisons involve peptide length distribution (left panel), joint frequencies of P1 residues showing low or intermediate+high susceptibility to ERAP2 trimming (middle panel) or showing low+intermediate or high hydropathy (right panel). The statistical signiﬁcance of the changes is indicated by the p-values, as estimated by the χ2 test. These data were originally published in reference (93). The distinct eﬀects of ERAP2 on HLA-B∗27 and A∗29:02 indicate that the functional interaction of the enzyme with MHC- I depends on the speciﬁc binding preferences of each MHC-I allotype. ERAP2 and Other Disease-Associated MHC-I Molecules The eﬀects of ERAP2 on the C∗06:02 and HLA-B∗51 peptidomes have not yet been, to my knowledge, experimentally addressed. Based on the P1 residue frequencies of the C∗06:02 peptidome (Figure 1) it can be inferred that the eﬀects will be smaller than those in HLA-B∗27 and perhaps closer to those in A∗29:02, but this issue clearly warrants further research. y Studying the eﬀects of ERAP2 on the HLA-B∗51 peptidome may be particularly interesting. Unlike other MHC-I-associated diseases, ERAP2 has not been reported to be a risk factor for BD. In addition, the HLA-B∗51 peptidome (Figure 1) includes very few ligands whose P1 residues are susceptible to ERAP2- trimming, such as basic ones, Gln or Ala (12). Therefore, direct eﬀects of ERAP2 on the HLA-B∗51 peptidome through over- trimming of B∗51 ligands are probably very limited. For this reason, this system may be a very suitable one to explore the role of ERAP2 on modulating ERAP1 trimming in live cells, as well as in the generation of MHC-I ligands. The studies are currently ongoing in our laboratory. ERAP2 and HLA-A∗29:02 individually or combined, on peptide trimming and determined how they modulate the enzymatic activity. These issues have been extensively reviewed elsewhere (38, 53, 111–114) Yet, the actual inﬂuence of ERAP1 polymorphism in shaping the constitutive peptidomes of MHC-I molecules remained much less explored. This aspect is quite relevant for two reasons. The ﬁrst one is that ERAP1 is expressed in all individuals but, due to its high polymorphism, the frequency of distinct ERAP1 variants, both within and among populations, is very variable (56). This has likely consequences for the diversity of the antigen handling capabilities among individuals and population groups, a diversity that is ampliﬁed by the distinct inﬂuence of ERAP1 variants on unrelated MHC-I-bound peptidomes, as discussed above. The second reason is that the discovery that ERAP1 polymorphism is a risk factor for MHC-I associated diseases and is usually in epistasis with the MHC susceptibility allele of each disease, sharpens the focus on the functional interaction between ERAP1 polymorphism and particular disease-associated MHC- I molecules, and therefore on the eﬀects of such polymorphism on MHC-I bound peptidomes. The studies summarized in this review allow us to deﬁne the nature of these eﬀects, how they aﬀect the relevant peptidomes, and how they inﬂuence the aﬃnity of each peptidome for its cognate MHC molecule. They also allow us to delineate the range of alterations induced by diﬀerent natural variants of ERAP1, to rank the eﬀects of the individual mutations, which are in turn ampliﬁed by linked polymorphisms aﬀecting ERAP1 levels, and to provide a molecular basis for the inﬂuence of ERAP1 on distinct MHC-I associated diseases. WHAT IMMUNOPEPTIDOMICS STUDIES TEACH US ABOUT ERAP1 AND ERAP2 FUNCTION Yet, most of the enzyme pool in the ER consists of monomeric molecules and it is unclear to what extent both enzymes physically interact in vivo. In addition, the nature of the eﬀects detected on the MHC-I bound peptidomes in live cells can be for the most part explained by separate eﬀects of both enzymes on the peptide pool, and do not seem consistent with a prominent eﬀect mediated by direct activation of ERAP1. The results in live cells do not exclude, and actually support, a functional interdependency of both enzymes in peptide processing. One possible mechanism, suggested by our recent studies on A∗29:02 (93) and by in vitro studies (20) might be based on the unproductive binding of long substrates to ERAP2, impairing or slowing their digestion by ERAP1 on one side, and the capacity of octamers to inhibit ERAP1 trimming of cognate substrates. Digestion of some of these octamers by ERAP2 could counteract their inhibition of ERAP1. The relevance of these eﬀects must be substantiated by direct experimental evidence. However, they may provide a mechanism of functional interaction between both enzymes that obviates the need for a direct activation of ERAP1 by ERAP2, is based on the known enzymatic and peptide handling features of both enzymes, and may be more consistent with the observed alterations of MHC-I-bound peptidomes in live cells. Yet, the association of ERAP1 and ERAP2 with distinct MHC- I-associated diseases means that two enzymes that are closely related to MHC-I function, through their processing of MHC-I ligands, play a role in these diseases. This fact strongly suggests that the pathogenetic role of MHC-I molecules is related to their peptide binding features. Although ERAP1 and ERAP2 are prominent susceptibility factors, the major association is always with the MHC-I molecule. Therefore, these enzymes modulate the strength of, but do not determine, the association of MHC-I with disease. The characterization of their eﬀects on MHC-I-bound peptidomes provides a unique opportunity to deﬁne the ways in which the functional interaction between closely related molecules modulates disease susceptibility. WHAT IMMUNOPEPTIDOMICS STUDIES TEACH US ABOUT ERAP1 AND ERAP2 FUNCTION The expression pattern of ERAP2 at the population level is much simpler, but not less drastic, than ERAP1. With few exceptions (64) only one enzymatic variant is expressed, and a signiﬁcant percentage of individuals do not express ERAP2 (65). Due in part to absence of an orthologous enzyme in mice, the role of ERAP2 is less well characterized. Although its structure, substrate speciﬁcity and handling features are known from in vitro studies (16, 51, 52), the eﬀect of ERAP2 on MHC-I-bound The role of ERAP1 in antigen processing has been widely studied both in mouse and humans. Crystallographic and enzymological studies have greatly contributed to unveiling the mechanism of this enzyme. In addition, multiple in vitro studies addressed the role of polymorphic ERAP1 residues, either October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 12 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro peptidomes in live cells was unknown until its eﬀects on the HLA-B∗27 and A∗29:02 peptidomes were analyzed in the studies reviewed above. They revealed that ERAP2 has signiﬁcant and allotype-dependent eﬀects on both MHC-I molecules, providing a rationale for their association with the diseases for which they are risk factors. deﬁnitive answers to these issues, which can only be obtained by an integrative knowledge of the many factors involved. This fact was suitably stated by Robert and colleagues some years ago in the context of the association of ERAP1 with AS: “Determining how ERAP1 inﬂuences the development of spondyloarthritis may be as complicated as deciphering the role of HLA-B27” (115). Systems biology approaches, such as that recently employed in preeclampsia (116), may at the end be required for a proper understanding of pathogenetic pathways in these diseases. An important issue is the nature of the functional inter- dependence of ERAP1 and ERAP2 in peptide processing. The report that ERAP1/ERAP2 heterodimers existed in vivo and account for a fraction of the enzyme pool in the ER (16) suggested that this association could signiﬁcantly improve the eﬃciency of their joint hydrolytic activity. Several in vitro studies tried to engineer molecular constructs in order to reproduce as closely as possible the heterodimers detected in vivo. Such constructs were more eﬃcient than the separate enzymes and their physical interaction apparently induced an activation of ERAP1 (16, 107, 108). WHAT IMMUNOPEPTIDOMICS STUDIES TEACH US ABOUT ERAP1 AND ERAP2 FUNCTION Several features of this interaction have so far emerged from immunopeptidomics studies: (1) Both ERAP1 polymorphism and ERAP2 expression aﬀect a seemingly moderate percentage of peptides in a qualitative way, (2) a larger percentage of peptides are either positively or negatively aﬀected in their expression levels by these enzymes, depending mostly on their length and P1 residues, (3) many peptides seem to be unaﬀected by ERAP1 or ERAP2, either because these enzymes are not involved in their processing or because their generation/destruction balance is the same on distinct enzymatic backgrounds, (4) the alterations in the peptidomes, specially the quantitative ones, signiﬁcantly re-shape the MHC-I-bound peptidomes depending on the ERAP1/ERAP2 context, (5) these alterations can be large enough to alter, in some cases, the global aﬃnity of the peptidome, (6) the eﬀects are MHC allotype-dependent, since the particular binding speciﬁcity of each MHC molecule selects distinct peptide subsets out of those generated in the particular ERAP1/ERAP2 context, and (7) without excluding a functional inter-dependence of ERAP1 and ERAP2, both enzymes seem to act, at least to a large extent, separately in vivo. WHAT IMMUNOPEPTIDOMICS STUDIES TEACH US ABOUT THE PATHOGENETIC ROLE OF MHC-I MOLECULES The alterations in the peptidome may have functional implications at multiple levels: (1) the generation or destruction of speciﬁc epitopes in a given ERAP1/2 context has obvious consequences in the antigen presenting capacity of individuals, depending on their particular enzymatic background (117), (2) the quantitative alterations in the peptidome induced by ERAP1/2 polymorphism may, in principle, alter the tolerogenic/autoimmune features of MHC-I molecules since speciﬁc T cell activation and immunogenicity depend on epitope density (118, 119), (3) the same alterations and the changes that they induce in the global aﬃnity of the peptidome, may aﬀect NK recognition, which is peptide-dependent and inﬂuenced by aﬃnity (120), or otherwise inﬂuence NK function (33, 121), and (4) changes in the aﬃnity of the peptidome for MHC-I-associated diseases are polygenic disorders of unknown etiology in which innate and adaptive immune mechanisms, some of which are incompletely understood, are likely to play a role. These diseases involve a variety of inﬂammatory pathways and aﬀect distinct organs: they are complex pathological conditions in which multiple genetic, immunological, and environmental factors play deﬁnite, but poorly characterized roles. The diﬃculties imposed by this complexity are well illustrated by the fact that, in spite of intensive research for almost 50 years, the pathogenetic role of HLA-B∗27 in AS remains undeﬁned. Therefore, it is probably naïve to assume that any single-sided methodological or conceptual approach may provide October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org Frontiers in Immunology \| www.frontiersin.org 13 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro the MHC-I molecule may potentially aﬀect its folding kinetics and its stability at the cell surface. In HLA-B27, both folding and surface dissociation leading to generation of non-canonical forms, such as heavy chain homodimers, are thought to inﬂuence the pathogenetic potential of this molecule (122–124). Dissecting the role of each of these factors in the diﬀerent diseases must await further research. Finally, the notorious diﬀerence in the genetic association of ERAP1 and ERAP2 with disease, namely the epistasis of ERAP1 and the susceptibility MHC-I allele in AS, psoriasis and BD, the non epistatic association of ERAP2 in the two former diseases, and, as far as it is known, the lack of association of ERAP2 with BD, may be better explained by the separate activities of both enzymes on the MHC-I peptidomes, than by assuming a strong inﬂuence of ERAP2 on ERAP1 activity. REFERENCES disorders: towards a unifying view. Mol Immunol. (2016) 77:193–204. doi: 10.1016/j.molimm.2016.08.005 disorders: towards a unifying view. Mol Immunol. (2016) 77:193–204. doi: 10.1016/j.molimm.2016.08.005 1. Deitiker P, Atassi MZ. MHC genes linked to autoimmune disease. Crit Rev Immunol. (2015) 35:203–51. doi: 10.1615/CritRevImmunol.2015014510 13. McGonagle D, Aydin SZ, Gul A, Mahr A, Direskeneli H. ‘MHC-I-opathy’- uniﬁed concept for spondyloarthritis and Behcet disease. Nat Rev Rheumatol. (2015) 11:731–40. doi: 10.1038/nrrheum.2015.147 2. Kuiper JJ, Van Setten J, Ripke S, Van ’t SR, Mulder F, Missotten T, et al. A genome-wide association study identiﬁes a functional ERAP2 haplotype associated with birdshot chorioretinopathy. Hum Mol Genet. (2014) 23:6081–7. doi: 10.1093/hmg/ddu307 14. Hatemi G, Karatemiz G, Yazici H. Behcet’s disease: an MHC-I-opathy? Clin Exp Rheumatol. (2017) 35(Suppl. 104):5. 15. Saric T, Chang SC, Hattori A, York IA, Markant S, Rock KL, et al. An IFN-gamma-induced aminopeptidase in the ER, ERAP1, trims precursors to MHC class I-presented peptides. Nat Immunol. (2002) 3:1169–76. doi: 10.1038/ni859 3. Herbort CP Jr, Pavesio C, LeHoang P, Bodaghi B, Fardeau C, Kestelyn P, et al. Why birdshot retinochoroiditis should rather be called ‘HLA-A29 uveitis’? Br J Ophthalmol. (2017) 101:851–5. doi: 10.1136/bjophthalmol-2016-309764 4. Brewerton DA, Hart FD, Nicholls A, Caﬀrey M, James DC, Sturrock RD. Ankylosing spondylitis and HL-A 27. Lancet (1973) 1:904–7. doi: 10.1016/S0140-6736(73)91360-3 16. Saveanu L, Carroll O, Lindo V, Del Val M, Lopez D, Lepelletier Y, et al. Concerted peptide trimming by human ERAP1 and ERAP2 aminopeptidase complexes in the endoplasmic reticulum. Nat Immunol. (2005) 6:689–97. doi: 10.1038/ni1208 5. Schlosstein L, Terasaki PI, Bluestone R, Pearson CM. High association of an HL-A antigen, W27, with ankylosing spondylitis. N Engl J Med. (1973) 288:704–6. doi: 10.1056/NEJM197304052881403 17. York IA, Chang SC, Saric T, Keys JA, Favreau JM, Goldberg AL, et al. The ER aminopeptidase ERAP1 enhances or limits antigen presentation by trimming epitopes to 8-9 residues. Nat Immunol. (2002) 3:1177–84. doi: 10.1038/ni860 6. Ohno S, Ohguchi M, Hirose S, Matsuda H, Wakisaka A, Aizawa M. Close association of HLA-Bw51 with Behcet’s disease. Arch Ophthalmol. (1982) 100:1455–8. doi: 10.1001/archopht.1982.01030040433013 18. Hammer GE, Gonzalez F, James E, Nolla H, Shastri N. In the absence of aminopeptidase ERAAP, MHC class I molecules present many unstable and highly immunogenic peptides. Nat Immunol. (2007) 8:101–8. doi: 10.1038/ni1409 7. Gul A, Ohno S. HLA-B∗51 and Behcet disease. Ocul Immunol Inﬂamm. (2012) 20:37–43. doi: 10.3109/09273948.2011.634978 19. Reeves E, Edwards CJ, Elliott T, James E. WHAT IMMUNOPEPTIDOMICS STUDIES TEACH US ABOUT THE PATHOGENETIC ROLE OF MHC-I MOLECULES The association of ERAP1/2 with distinct MHC-I- dependent inﬂammatory diseases initially suggested that similar mechanism(s) might underlie the joint association of ERAP1/2 and MHC-I to diﬀerent diseases (12). However, the distinct eﬀects of ERAP1/ERAP2 on diﬀerent MHC-I molecules revealed by recent immunopeptidomics studies may be more compatible with the alternative possibility that the MHC allotype-speciﬁc eﬀects of these enzymes actually lead to distinct pathogenetic mechanisms of ERAP/MHC interaction in diﬀerent diseases. This is also supported by the distinct association patterns of ERAP1 (Table 1). Similar ideas have been proposed by McGonagle and colleagues from a consideration of the pathology/immunopathology of some of these diseases (13). FUNDING The work of the author is funded by grant SAF2017-86578-R from the Ministry of Science, Innovation, and Universities of the Government of Spain and by an institutional grant of the Fundación Ramón Areces to the CBMSO. AUTHOR CONTRIBUTIONS The author conﬁrms being the sole contributor of this work and has approved it for publication. REFERENCES Naturally occurring ERAP1 haplotypes encode functionally distinct alleles with ﬁne substrate speciﬁcity. J Immunol. (2013). 191:35–43. doi: 10.4049/jimmunol.1300598 8. de Menthon M, LaValley MP, Maldini C, Guillevin L, Mahr A. HLA-B51/B5 and the risk of Behcet’s disease: a systematic review and meta-analysis of case-control genetic association studies. Arthritis Rheum. (2009) 61:1287–96. doi: 10.1002/art.24642 20. Martin-Esteban A, Gomez-Molina P, Sanz-Bravo A, Lopez de Castro JA. Combined eﬀects of ankylosing spondylitis-associated ERAP1 polymorphisms outside the catalytic and peptide-binding sites on the processing of natural HLA-B27 ligands. J Biol Chem. (2014) 289:3978–90. doi: 10.1074/jbc.M113.529610 9. Henseler T, Christophers E. Psoriasis of early and late onset: characterization of two types of psoriasis vulgaris. J Am Acad Dermatol. (1985) 13:450–6. doi: 10.1016/S0190-9622(85)70188-0 10. Gudjonsson JE, Karason A, Runarsdottir EH, Antonsdottir AA, Hauksson VB, Jonsson HH, et al. Distinct clinical diﬀerences between HLA- Cw∗0602 positive and negative psoriasis patients–an analysis of 1019 HLA-C- and HLA-B-typed patients. J Invest Dermatol. (2006) 126:740–5. doi: 10.1038/sj.jid.5700118 21. The TASK and WTCCC2 Consortia. Interaction between ERAP1 and HLA-B27 in ankylosing spondylitis implicates peptide handling in the mechanism for HLA-B27 in disease susceptibility. Nat Genet. (2011) 43:761–7. doi: 10.1038/ng.873 22. Cortes A, Pulit SL, Leo PJ, Pointon JJ, Robinson PC, Weisman MH, et al. Major histocompatibility complex associations of ankylosing spondylitis are complex and involve further epistasis with ERAP1. Nat Commun. (2015) 6:7146. doi: 10.1038/ncomms8146 11. Kuiper J, Rothova A, de Boer J, Radstake T. The immunopathogenesis of birdshot chorioretinopathy; a bird of many feathers. Prog Retin Eye Res. (2015) 44:99–110. doi: 10.1016/j.preteyeres.2014.11.003 12. Lopez de Castro JA, Alvarez-Navarro C, Brito A, Guasp P, Martin-Esteban A Sanz-Bravo A. Molecular and pathogenic eﬀects of endoplasmic reticulum aminopeptidases ERAP1 and ERAP2 in MHC-I-associated inﬂammatory 23. Strange A, Capon F, Spencer CC, Knight J, Weale ME, Allen MH, et al. A genome-wide association study identiﬁes new psoriasis susceptibility October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 14 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro hypertension in the Northeastern Han Chinese. Gene (2015) 560:211–6. doi: 10.1016/j.gene.2015.02.005 loci and an interaction between HLA-C and ERAP1. Nat Genet. (2010) 42:985–90. doi: 10.1038/ng.694 hypertension in the Northeastern Han Chinese. Gene (2015) 560:211–6. doi: 10.1016/j.gene.2015.02.005 loci and an interaction between HLA-C and ERAP1. Nat Genet. (2010) 42:985–90. doi: 10.1038/ng.694 24. Kirino Y, Bertsias G, Ishigatsubo Y, Mizuki N, Tugal-Tutkun I, Seyahi E, et al. REFERENCES Structure of the complex between human T-cell receptor, viral peptide and HLA-A2. Nature (1996) 384:134–41. doi: 10.1038/384134a0 49. Nguyen TT, Chang SC, Evnouchidou I, York IA, Zikos C, Rock KL, et al. Structural basis for antigenic peptide precursor processing by the endoplasmic reticulum aminopeptidase ERAP1. Nat Struct Mol Biol. (2011) 18:604–13. doi: 10.1038/nsmb.2021 31. Jardetzky T. Not just another Fab: the crystal structure of a TcR-MHC-peptide complex. Structure (1997) 5:159–63. doi: 10.1016/S0969-2126(97)00175-5 50. Kochan G, Krojer T, Harvey D, Fischer R, Chen L, Vollmar M, et al. Crystal structures of the endoplasmic reticulum aminopeptidase-1 (ERAP1) reveal the molecular basis for N-terminal peptide trimming. Proc Natl Acad Sci USA. (2011) 108:7745–50. doi: 10.1073/pnas.1101262108 32. Brooks AG, Boyington JC, Sun PD. Natural killer cell recognition of HLA class I molecules. Rev Immunogenet. (2000) 2:433–48. 51. Birtley JR, Saridakis E, Stratikos E, Mavridis IM. The crystal structure of human endoplasmic reticulum aminopeptidase 2 reveals the atomic basis for distinct roles in antigen processing. Biochemistry (2012) 51:286–95. doi: 10.1021/bi201230p 33. Aldhamen YA, Seregin SS, Rastall DP, Aylsworth CF, Pepelyayeva Y, Busuito CJ, et al. Endoplasmic reticulum aminopeptidase-1 functions regulate key aspects of the innate immune response. PLoS ONE (2013) 8:e69539. doi: 10.1371/journal.pone.0069539 34. Aldhamen YA, Pepelyayeva Y, Rastall DP, Seregin SS, Zervoudi E, Koumantou D, et al. Autoimmune disease-associated variants of extracellular endoplasmic reticulum aminopeptidase 1 induce altered innate immune responses by human immune cells. J Innate Immun. (2015) 7:275–89. doi: 10.1159/000368899 52. Mpakali A, Giastas P, Mathioudakis N, Mavridis IM, Saridakis E, Stratikos E. Structural basis for antigenic peptide recognition and processing by ER aminopeptidase 2. J Biol Chem. (2015) 290:26021–32. doi: 10.1074/jbc.M115.685909 53. Stratikos E, Stern LJ. Antigenic peptide trimming by ER aminopeptidases– insights from structural studies. Mol Immunol. (2013) 55:212–9. doi: 10.1016/j.molimm.2013.03.002 35. Firat E, Saveanu L, Aichele P, Staeheli P, Huai J, Gaedicke S, et al. The role of endoplasmic reticulum-associated aminopeptidase 1 in immunity to infection and in cross-presentation. J Immunol. (2007) 178:2241–8. doi: 10.4049/jimmunol.178.4.2241 54. Hearn A, York IA, Rock KL. The speciﬁcity of trimming of MHC class I-presented peptides in the endoplasmic reticulum. J Immunol. (2009) 183:5526–36. doi: 10.4049/jimmunol.0803663 36. Stamogiannos A, Papakyriakou A, Mauvais FX, van Endert P, Stratikos E. Screening identiﬁes thimerosal as a selective inhibitor of endoplasmic reticulum aminopeptidase 1. ACS Med Chem Lett. (2016) 7:681–5. doi: 10.1021/acsmedchemlett.6b00084 55. Chang SC, Momburg F, Bhutani N, Goldberg AL. REFERENCES Genome-wide association analysis identiﬁes new susceptibility loci for Behcet’s disease and epistasis between HLA-B∗51 and ERAP1. Nat Genet. (2013) 45:202–7. doi: 10.1038/ng.2520 43. Zee RYL, Rivera A, Inostroza Y, Ridker PM, Chasman DI, Romero JR. Gene variation of endoplasmic reticulum aminopeptidases 1 and 2, and risk of blood pressure progression and incident hypertension among 17,255 initially healthy women. Int J Genomics (2018) 2018:2308585. doi: 10.1155/2018/2308585 25. Cortes A, Hadler J, Pointon JP, Robinson PC, Karaderi T, Leo P, et al. Identiﬁcation of multiple risk variants for ankylosing spondylitis through high-density genotyping of immune-related loci. Nat Genet. (2013) 45:730–8. doi: 10.1038/ng.2667 44. Hisatsune C, Ebisui E, Usui M, Ogawa N, Suzuki A, Mataga N, et al. ERp44 exerts redox-dependent control of blood pressure at the ER. Mol Cell (2015) 58:1015–27. doi: 10.1016/j.molcel.2015.04.008 26. Robinson PC, Costello ME, Leo P, Bradbury LA, Hollis K, Cortes A, et al. ERAP2 is associated with ankylosing spondylitis in HLA-B27- positive and HLA-B27-negative patients. Ann Rheum Dis. (2015) 74:1627–9. doi: 10.1136/annrheumdis-2015-207416 45. Tanioka T, Hattori A, Masuda S, Nomura Y, Nakayama H, Mizutani S, et al. Human leukocyte-derived arginine aminopeptidase. The third member of the oxytocinase subfamily of aminopeptidases. J Biol Chem. (2003) 278:32275–83. doi: 10.1074/jbc.M305076200 46. Goto Y, Hattori A, Ishii Y, Tsujimoto M. Reduced activity of the hypertension-associated Lys528Arg mutant of human adipocyte-derived leucine aminopeptidase (A-LAP)/ER-aminopeptidase-1. FEBS Lett. (2006) 580:1833–8. doi: 10.1016/j.febslet.2006.02.041 27. Tsoi LC, Spain SL, Knight J, Ellinghaus E, Stuart PE, Capon F, et al. Identiﬁcation of 15 new psoriasis susceptibility loci highlights the role of innate immunity. Nat Genet. (2012) 44:1341–8. doi: 10.1038/ng.2467 28. Yin X, Low HQ, Wang L, Li Y, Ellinghaus E, Han J, et al. Genome- wide meta-analysis identiﬁes multiple novel associations and ethnic heterogeneity of psoriasis susceptibility. Nat Commun. (2015) 6:6916. doi: 10.1038/ncomms7916 47. Goto Y, Ogawa K, Nakamura TJ, Hattori A, Tsujimoto M. Substrate- dependent nitric oxide synthesis by secreted endoplasmic reticulum aminopeptidase 1 in macrophages. J Biochem. (2015) 157:439–49. doi: 10.1093/jb/mvv001 29. Kuiper JJW, Setten J, Devall M, Cretu-Stanku M, Hiddingh S, OphoﬀRA, et al. Functionally distinct ERAP1 and ERAP2 are a hallmark of HLA-A29- (Birdshot) Uveitis. Hum Mol Genet. (2018). doi: 10.1093/hmg/ddy319 48. Forstermann U, Sessa WC. Nitric oxide synthases: regulation and function. Eur Heart J. (2012) 33:829–37. doi: 10.1093/eurheartj/ehr304 30. Garboczi DN, Ghosh P, Utz U, Fan QR, Biddison WE, Wiley DC. REFERENCES Schittenhelm RB, Sian TC, Wilmann PG, Dudek NL, Purcell AW. Revisiting the arthritogenic peptide theory: quantitative not qualitative changes in the peptide repertoire of HLA-B27 allotypes. Arthritis Rheumatol. (2015) 67:702–13. doi: 10.1002/art.38963 67. Paladini F, Fiorillo MT, Vitulano C, Tedeschi V, Piga M, Cauli A, et al. An allelic variant in the intergenic region between ERAP1 and ERAP2 correlates with an inverse expression of the two genes. Sci Rep. (2018) 8:10398. doi: 10.1038/s41598-018-28799-8 86. Yair-Sabag S, Tedeschi V, Vitulano C, Barnea E, Glaser F, Melamed KD, et al. The peptide repertoire of HLA-B27 may include ligands with lysine at P2 anchor position. Proteomics (2018) 18:e1700249. doi: 10.1002/pmic.201700249 68. WTCC Consortium. Association scan of 14,500 nonsynonymous SNPs in four diseases identiﬁes autoimmunity variants. Nat Genet. (2007) 39:1329– 37. doi: 10.1038/ng.2007.17 87. Herberts CA, Neijssen JJ, de Haan J, Janssen L, Drijfhout JW, Reits EA, et al. Cutting edge: HLA-B27 acquires many N-terminal dibasic peptides: coupling cytosolic peptide stability to antigen presentation. J Immunol. (2006) 176:2697–701. doi: 10.4049/jimmunol.176.5.2697 69. van de Bunt M, Cortes A, IGAS Consortium, Brown MA, Morris AP, McCarthy MI. Evaluating the performance of ﬁne-mapping strategies at common variant gwas loci. PLoS Genet. (2015) 11:e1005535. doi: 10.1371/journal.pgen.1005535 88. Momburg F, Roelse J, Howard JC, Butcher GW, Hammerling GJ, Neefjes JJ. Selectivity of MHC-encoded peptide transporters from human, mouse and rat. Nature (1994) 367:648–51. doi: 10.1038/367648a0 70. Takeuchi M, Ombrello MJ, Kirino Y, Erer B, Tugal-Tutkun I, Seyahi E, et al. A single endoplasmic reticulum aminopeptidase-1 protein allotype is a strong risk factor for Behcet’s disease in HLA-B∗51 carriers. Ann Rheum Dis. (2017) 75:2208–11. doi: 10.1136/annrheumdis-2015-209059 89. Van Endert PM, Riganelli D, Greco G, Fleischhauer K, Sidney J, Sette A, et al. The peptide-binding motif for the human transporter associated with antigen processing. J Exp Med. (1995) 182:1883–95. doi: 10.1084/jem.182.6.1883 71. Lysell J, Padyukov L, Kockum I, Nikamo P, Stahle M. Genetic association with ERAP1 in psoriasis is conﬁned to disease onset after puberty and not dependent on HLA-C∗06. J Invest Dermatol. (2013) 133:411–7. doi: 10.1038/jid.2012.280 90. Petrovsky N, Brusic V. Virtual models of the HLA class I antigen processing pathway. Methods (2004) 34:429–35. doi: 10.1016/j.ymeth.2004.06.005 91. Granados DP, Sriranganadane D, Daouda T, Zieger A, Laumont CM, Caron- Lizotte O, et al. Impact of genomic polymorphisms on the repertoire of human MHC class I-associated peptides. Nat Commun. (2014) 5:3600. doi: 10.1038/ncomms4600 72. REFERENCES Genetic variants in ERAP1 and ERAP2 associated with immune- mediated diseases inﬂuence protein expression and isoform proﬁle. Arthritis Rheumatol. (2018) 70:255–65. doi: 10.1002/art.40369 79. Garcia-Medel N, Sanz-Bravo A, Alvarez-Navarro C, Gomez-Molina P, Barnea E, Marcilla M, et al. Peptide handling by HLA-B27 subtypes inﬂuences their biological behavior, association with ankylosing spondylitis and susceptibility to ERAP1. Mol Cell Proteomics (2014) 13:3367–80. doi: 10.1074/mcp.M114.039214 62. Costantino F, Talpin A, Evnouchidou I, Kadi A, Leboime A, Said- Nahal R, et al. ERAP1 gene expression is inﬂuenced by non-synonymous polymorphisms associated with predisposition to spondyloarthritis. Arthritis Rheumatol. (2015) 67:1525–34. doi: 10.1002/art.39072 80. Guasp P, Alvarez-Navarro C, Gomez-Molina P, Martin-Esteban A, Marcilla M, Barnea E, et al. The peptidome of the Behcet’s disease-associated HLA- B∗51:01 includes two sub-peptidomes diﬀerentially shaped by ERAP1. Arthritis Rheumatol. (2016) 68:505–15. doi: 10.1002/art.39430 63. Evnouchidou I, Birtley J, Seregin S, Papakyriakou A, Zervoudi E, Samiotaki M, et al. A common SNP in ER aminopeptidase 2 induces a speciﬁcity switch that leads to altered antigen processing. J Immunol. (2012) 189:2383–92. doi: 10.4049/jimmunol.1200918 81. Mobbs JI, Illing PT, Dudek NL, Brooks AG, Baker DG, Purcell AW, et al. The molecular basis for peptide repertoire selection in the human leucocyte antigen (HLA) C∗06:02 molecule. J Biol Chem. (2017) 292:17203– 15. doi: 10.1074/jbc.M117.806976 64. Vanhille DL, Hill LD, Hilliard DD, Lee ED, Teves ME, Srinivas S, et al. A novel haplotype structure in a chilean population: implications for ERAP2 protein expression and preeclampsia risk. Mol Genet Genomic Med. (2013) 1:98–107. doi: 10.1002/mgg3.13 82. Jardetzky TS, Lane WS, Robinson RA, Madden DR, Wiley DC. Identiﬁcation of self peptides bound to puriﬁed HLA-B27. Nature (1991) 353:326–29. doi: 10.1038/353326a0 65. Andres AM, Dennis MY, Kretzschmar WW, Cannons JL, Lee-Lin SQ, Hurle B, et al. Balancing selection maintains a form of ERAP2 that undergoes nonsense-mediated decay and aﬀects antigen presentation. PLoS Genet. (2010) 6:e1001157. doi: 10.1371/journal.pgen.1001157 83. Lopez de Castro JA, Alvarez I, Marcilla M, Paradela A, Ramos M, Sesma L, et al. HLA-B27: a registry of constitutive peptide ligands. Tissue Antigens (2004). 63:424–45. doi: 10.1111/j.0001-2815.2004.00220.x 84. Ben Dror L, Barnea E, Beer I, Mann M, Admon A. The HLA- B∗2705 peptidome. Arthritis Rheum. (2010) 62:420–9. doi: 10.1002/art. 27257 66. Alvarez-Navarro C, Martin-Esteban A, Barnea E, Admon A, Lopez de Castro JA. ERAP1 polymorphism relevant to inﬂammatory disease shapes the peptidome of the birdshot chorioretinopathy-associated HLA-A∗29:02 antigen. Mol Cell Proteomics (2015) 14:1770–80. doi: 10.1074/mcp.M115.048959 85. REFERENCES The ER aminopeptidase, ERAP1, trims precursors to lengths of MHC class I peptides by a “molecular ruler” mechanism. Proc Natl Acad Sci USA. (2005) 102:17107–12. doi: 10.1073/pnas.0500721102 37. Schmidt K, Keller C, Kuhl AA, Textor A, Seifert U, Blankenstein T, et al. ERAP1-dependent antigen cross-presentation determines eﬃcacy of adoptive T cell therapy in mice. Cancer Res. (2018) 78:3243–54. doi: 10.1158/0008-5472.CAN-17-1946 56. Ombrello MJ, Kastner DL, Remmers EF. Endoplasmic reticulum-associated amino-peptidase 1 and rheumatic disease: genetics. Curr Opin Rheumatol. (2015) 27:349–56. doi: 10.1097/BOR.0000000000000189 38. Alvarez-Navarro C, Lopez de Castro JA. ERAP1 structure, function and pathogenetic role in ankylosing spondylitis and other MHC-associated diseases. Mol Immunol. (2014) 57:12–21. doi: 10.1016/j.molimm.2013.06.012 57. Evnouchidou I, Kamal RP, Seregin SS, Goto Y, Tsujimoto M, Hattori A, et al. Cutting Edge: coding single nucleotide polymorphisms of endoplasmic reticulum aminopeptidase 1 can aﬀect antigenic peptide generation in vitro by inﬂuencing basic enzymatic properties of the enzyme. J Immunol. (2011) 186:1909–13. doi: 10.4049/jimmunol.1003337 39. Reeves E, James E. The role of polymorphic ERAP1 in autoinﬂammatory disease. Biosci Rep. (2018) 38:BSR20171503. doi: 10.1042/BSR20 171503 58. Stamogiannos A, Koumantou D, Papakyriakou A, Stratikos E. Eﬀects of polymorphic variation on the mechanism of Endoplasmic Reticulum Aminopeptidase 1. Mol Immunol. (2015) 67:426–35. doi: 10.1016/j.molimm.2015.07.010 40. Goto Y, Ogawa K, Nakamura TJ, Hattori A, Tsujimoto M. TLR-mediated secretion of endoplasmic reticulum aminopeptidase 1 from macrophages. J Immunol. (2014) 192:4443–52. doi: 10.4049/jimmunol.1300935 j 59. Dixon AL, Liang L, Moﬀatt MF, Chen W, Heath S, Wong KC, et al. A genome-wide association study of global gene expression. Nat Genet. (2007) 39:1202–7. doi: 10.1038/ng2109 41. Goto Y, Ogawa K, Hattori A, Tsujimoto M. Secretion of endoplasmic reticulum aminopeptidase 1 is involved in the activation of macrophages induced by lipopolysaccharide and interferon-gamma. J Biol Chem. (2011) 286:21906–14. doi: 10.1074/jbc.M111.239111 60. Harvey D, Pointon JJ, Evans DM, Karaderi T, Farrar C, Appleton LH, et al. Investigating the genetic association between ERAP1 and ankylosing spondylitis. Hum Mol Genet. (2009) 18:4204–12. doi: 10.1093/hmg/ddp371 42. Yang S, Liu X, Gao Y, Ding M, Li B, Sun H, et al. Association of single nucleotide polymorphisms in the 3’UTR of ERAP1 gene with essential October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 15 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro 61. Hanson AL, Cuddihy T, Haynes K, Loo D, Morton CJ, Oppermann U, et al. REFERENCES Wisniewski A, Matusiak L, Szczerkowska-Dobosz A, Nowak I, Luszczek W, Kusnierczyk P. The association of ERAP1 and ERAP2 single nucleotide polymorphisms and their haplotypes with psoriasis vulgaris is dependent on the presence or absence of the HLA-C∗06:02 allele and age at disease onset. Hum Immunol. (2018) 79:109–16. doi: 10.1016/j.humimm.2017.11.010 92. Abelin JG, Keskin DB, Sarkizova S, Hartigan CR, Zhang W, Sidney J, et al. Mass spectrometry proﬁling of HLA-associated peptidomes in mono-allelic cells enables more accurate epitope prediction. Immunity (2017) 46:315–26. doi: 10.1016/j.immuni.2017.02.007 73. Bjorkman PJ, Saper MA, Samraoui B, Bennett WS, Strominger JL, Wiley DC. The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature (1987) 329:512–8. doi: 10.1038/329512a0 93. Sanz-Bravo A, Martin-Esteban A, Kuiper JJW, Garcia-Peydro M, Barnea E, Admon A, et al. Allele-speciﬁc alterations in the peptidome underlie the joint association of HLA-A∗29:02 and Endoplasmic Reticulum Aminopeptidase 2 (ERAP2) with Birdshot Chorioretinopathy. Mol Cell Proteomics (2018) 17:1564–77. doi: 10.1074/mcp.RA118.000778 74. Saper MA, Bjorkman PJ, Wiley DC. Reﬁned structure of the human histocompatibility antigen HLA-A2 at 2.6 A resolution. J Mol Biol. (1991) 219:277–319. doi: 10.1016/0022-2836(91)90567-P 75. Garrett TP, Saper MA, Bjorkman PJ, Strominger JL, Wiley DC. Speciﬁcity pockets for the side chains of peptide antigens in HLA-Aw68. Nature (1989) 342:692–6. doi: 10.1038/342692a0 94. Sakaguchi T, Ibe M, Miwa K, Yokota S, Tanaka K, Schonbach C, et al. Predominant role of N-terminal residue of nonamer peptides in their binding to HLA-B∗5101 molecules. Immunogenetics (1997) 46:245–8. doi: 10.1007/s002510050269 76. Madden DR, Gorga JC, Strominger JL, Wiley DC. The structure of HLA-B27 reveals nonamer self-peptides bound in an extended conformation. Nature (1991) 353:321–5. doi: 10.1038/353321a0 95. Sakaguchi T, Ibe M, Miwa K, Kaneko Y, Yokota S, Tanaka K, et al. Binding of 8-mer to 11-mer peptides carrying the anchor residues to slow assembling HLA class I molecules (HLA-B∗5101). Immunogenetics (1997) 45:259–65. doi: 10.1007/s002510050201 77. Madden DR, Gorga JC, Strominger JL, Wiley DC. The three- dimensional structure of HLA-B27 at 2.1 A resolution suggests a general mechanism for tight peptide binding to MHC. Cell (1992) 70:1035–48. doi: 10.1016/0092-8674(92)90252-8 96. Fruci D, Giacomini P, Nicotra MR, Forloni M, Fraioli R, Saveanu L, et al. Altered expression of endoplasmic reticulum aminopeptidases ERAP1 and ERAP2 in transformed non-lymphoid human tissues. J Cell Physiol. (2008) 216:742–9. doi: 10.1002/jcp.21454 78. Ruppert J, Sidney J, Celis E, Kubo RT, Grey HM, Sette A. REFERENCES Prominent role of secondary anchor residues in peptide binding to HLA-A2.1 molecules. Cell (1993) 74:929–37. doi: 10.1016/0092-8674(93)90472-3 October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 16 Disease-Associated MHC-I Peptidomes and ERAP Proteins López de Castro 97. Chen L, Fischer R, Peng Y, Reeves E, McHugh K, Ternette N, et al. Critical role of endoplasmic reticulum aminopeptidase 1 in determining the length and sequence of peptides bound and presented by HLA-B27. Arthritis Rheumatol. (2014) 66:284–94. doi: 10.1002/art.38249 112. Weimershaus M, Evnouchidou I, Saveanu L, van Endert P. (2013). Peptidases trimming MHC class I ligands. Curr Opin Immunol. 25:90–6. doi: 10.1016/j.coi.2012.10.001 113. Stratikos E, Stamogiannos A, Zervoudi E, Fruci D. A role for naturally occurring alleles of endoplasmic reticulum aminopeptidases in tumor immunity and cancer pre-disposition. Front Oncol. (2014) 4:363. doi: 10.3389/fonc.2014.00363 98. Barnea E, Melamed KD, Haimovich Y, Satumtira N, Dorris ML, Nguyen MT, et al. The HLA-B27 peptidome in vivo in spondyloarthritis-susceptible HLA- B27 transgenic rats and the eﬀect of ERAP1 deletion. Mol Cell Proteomics (2017) 16:642–62. doi: 10.1074/mcp.M116.066241 114. Mpakali A, Maben Z, Stern LJ, Stratikos E. Molecular pathways for antigenic peptide generation by ER aminopeptidase 1. Mol Immunol. (2018). doi: 10.1016/j.molimm.2018.03.026 99. Garcia-Medel N, Sanz-Bravo A, Van Nguyen D, Galocha B, Gomez- Molina P, Martin-Esteban A, et al. Functional interaction of the ankylosing spondylitis-associated endoplasmic reticulum aminopeptidase 1 polymorphism and HLA-B27 in vivo. Mol Cell Proteomics (2012) 11:1416–29. doi: 10.1074/mcp.M112.019588 115. Colbert RA, DeLay ML, Klenk EI, Layh-Schmitt G. From HLA-B27 to spondyloarthritis: a journey through the ER. Immunol Rev. (2010). 233:181– 202. doi: 10.1111/j.0105-2896.2009.00865.x 100. Sanz-Bravo A, Campos J, Mazariegos MS, Lopez de Castro JA. Dominant role of the ERAP1 polymorphism R528K in shaping the HLA-B27 peptidome through diﬀerential processing determined by multiple peptide residues. Arthritis Rheumatol. (2015) 67:692–701. doi: 10.1002/art.38980 116. Than NG, Romero R, Tarca AL, Kekesi KA, Xu Y, Xu Z, et al. Integrated systems biology approach identiﬁes novel maternal and placental pathways of preeclampsia. Front Immunol. (2018) 9:1661. doi: 10.3389/ﬁmmu.2018.01661 117. Seregin SS, Rastall DP, Evnouchidou I, Aylsworth CF, Quiroga D, Kamal RP, et al. Endoplasmic reticulum aminopeptidase-1 alleles associated with increased risk of ankylosing spondylitis reduce HLA-B27 mediated presentation of multiple antigens. Autoimmunity (2013) 46:497–508. doi: 10.3109/08916934.2013.819855 101. Martin-Esteban A, Sanz-Bravo A, Guasp P, Barnea E, Admon A, Lopez de Castro JA. REFERENCES Separate eﬀects of the ankylosing spondylitis associated ERAP1 and ERAP2 aminopeptidases determine the inﬂuence of their combined phenotype on the HLA-B∗27 peptidome. J Autoimmun. (2017) 79:28–38. doi: 10.1016/j.jaut.2016.12.008 118. Liu W, Chen YH. High epitope density in a single protein molecule signiﬁcantly enhances antigenicity as well as immunogenicity: a novel strategy for modern vaccine development and a preliminary investigation about B cell discrimination of monomeric proteins. Eur J Immunol. (2005) 35:505–14. doi: 10.1002/eji.200425749 102. Sanz-Bravo A, Alvarez-Navarro C, Martin-Esteban A, Barnea E, Admon A, Lopez de Castro JA. Ranking the contribution of ankylosing spondylitis- associated ERAP1 polymorphisms to shaping the HLA-B∗27 peptidome. Mol Cell Proteomics (2018) 17:1308–23. doi: 10.1074/mcp.RA117.000565 103. Guasp P, Barnea E, Gonzalez-Escribano MF, Jimenez-Reinoso A, Regueiro JR, Admon A, et al. The Behcet’s disease-associated variant of the aminopeptidase ERAP1 shapes a low aﬃnity HLA-B∗51 peptidome by diﬀerential subpeptidome processing. J Biol Chem. (2017) 292:9680–9. doi: 10.1074/jbc.M117.789180 119. Gonzalez PA, Carreno LJ, Coombs D, Mora JE, Palmieri E, Goldstein B, et al. T cell receptor binding kinetics required for T cell activation depend on the density of cognate ligand on the antigen-presenting cell. Proc Natl Acad Sci USA. (2005) 102:4824–9. doi: 10.1073/pnas.0500922102 104. Guasp P, Alvarez-Navarro C, Gomez-Molina P, Martin-Esteban A, Marcilla M, Barnea E, et al. Heterozygosity of the 721.221-B∗51:01 cell line used in the study by Guasp et (Arthritis Rheumatol, February 2016). Arthritis Rheumatol. (2017) 69:686. doi: 10.1002/art.40073 120. Cifaldi L, Romania P, Falco M, Lorenzi S, Meazza R, Petrini S, et al. ERAP1 regulates natural killer cell function by controlling the engagement of inhibitory receptors. Cancer Res. (2015) 75:824–34. doi: 10.1158/0008-5472.CAN-14-1643 121. Rastall DPW, Alyaquob FS, O’Connell P, Pepelyayeva Y, Peters D, Godbehere-Roosa S, et al. Mice expressing human ERAP1 variants associated with ankylosing spondylitis have altered T-cell repertoires and NK cell functions, as well as increased in utero and perinatal mortality. Int Immunol. (2017) 29:277–89. doi: 10.1093/intimm/dxx035 105. Vilches C, Bunce M, de Pablo R, Murray AK, McIntyre CA, Kreisler M. Complete coding regions of two novel HLA-B alleles detected by phototyping (PCR-SSP) in the British caucasoid population: B∗5108 and B∗5002. Tissue Antigens (1997) 50:38–41. doi: 10.1111/j.1399-0039.1997.tb02831.x 106. Martin-Esteban A, Guasp P, Barnea E, Admon A, Lopez de Castro JA. Functional interaction of the ankylosing spondylitis associated endoplasmic reticulum aminopeptidase 2 with the HLA-B∗27 peptidome in human cells. Arthritis Rheumatol. (2016) 68:2466–75. doi: 10.1002/art.39734 122. Colbert RA, Tran TM, Layh-Schmitt G. REFERENCES HLA-B27 misfolding and ankylosing spondylitis. Mol Immunol. (2014) 57:44–51. doi: 10.1016/j.molimm.2013.07.013 123. Shaw J, Hatano H, Kollnberger S. The biochemistry and immunology of non-canonical forms of HLA-B27. Mol Immunol. (2014) 57:52–8. doi: 10.1016/j.molimm.2013.05.243 107. Evnouchidou I, Weimershaus M, Saveanu L, van Endert P. ERAP1-ERAP2 dimerization increases peptide-trimming eﬃciency. J Immunol. (2014) 193:901–8. doi: 10.4049/jimmunol.1302855 108. Chen H, Li L, Weimershaus M, Evnouchidou I, van Endert P, Bouvier M. ERAP1-ERAP2 dimers trim MHC I-bound precursor peptides; implications for understanding peptide editing. Sci Rep. (2016) 6:28902. doi: 10.1038/srep28902 124. Ridley A, Hatano H, Wong-Baeza I, Shaw J, Matthews KK, Al Mossawi H, et al. Activation-induced killer cell immunoglobulin- like receptor 3DL2 binding to HLA-B27 licenses pathogenic T cell diﬀerentiation in spondyloarthritis. Arthritis Rheumatol. (2016) 68:901–14. doi: 10.1002/art.39515 109. Lamas JR, Paradela A, Roncal F, Lopez de Castro JA. Modulation at multiple anchor positions of the peptide speciﬁcity of HLA-B27 subtypes diﬀerentially associated with ankylosing spondylitis. Arthritis Rheum. (1999) 42:1975–85. doi: 10.1002/1529-0131(199909)42:9<1975::AID-ANR25>3.0.CO;2-V Conﬂict of Interest Statement: The author declares that the research was conducted in the absence of any commercial or ﬁnancial relationships that could be construed as a potential conﬂict of interest. 110. Hillig RC, Huelsmeyer M, Saenger W, Welﬂe K, Misselwitz R, Welﬂe H, et al. Thermodynamic and structural analysis of peptide-and allele-dependent properties of two HLA-B27 subtypes exhibiting diﬀerential disease association. J Biol Chem. (2004) 279:652–63. doi: 10.1074/jbc.M307457200 Copyright © 2018 López de Castro. 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. 111. Shastri N, Cardinaud S, Schwab SR, Serwold T, Kunisawa J. All the peptides that ﬁt: the beginning, the middle, and the end of the MHC class I antigen-processing pathway. Immunol Rev. (2005).207:31–41. doi: 10.1111/j.0105-2896.2005.00321.x October 2018 \| Volume 9 \| Article 2463 Frontiers in Immunology \| www.frontiersin.org 17
https://openalex.org/W4317254401	https://hal.science/hal-04122780/document	English	null	X-SHYNE: X-shooter spectra of young exoplanet analogs	Astronomy & astrophysics	2,023	cc-by	10,343	To cite this version: Simon Petrus, Gaël Chauvin, M. Bonnefoy, Pascal Tremblin, B. Charnay, et al.. X-SHYNE: X-shooter spectra of young exoplanet analogs: I. A medium-resolution 0.65-2.5 µ m one-shot spectrum of VHS 1256-1257 b. Astronomy and Astrophysics - A&A, 2023, 670, 10.1051/0004-6361/202244494. hal- 04122780 X-SHYNE: X-shooter spectra of young exoplanet analogs: I. A medium-resolution 0.65-2.5 µ m one-shot spectrum of VHS 1256-1257 b Simon Petrus, Gaël Chauvin, M. Bonnefoy, Pascal Tremblin, B. Charnay, P. Delorme, Gabriel Dominique Marleau, Amelia Bayo, Elena Manjavacas, Anne Marie Lagrange, et al. To cite this version: Simon Petrus, Gaël Chauvin, M. Bonnefoy, Pascal Tremblin, B. Charnay, et al.. X-SHYNE: X-shooter spectra of young exoplanet analogs: I. A medium-resolution 0.65-2.5 µ m one-shot spectrum of VHS 1256-1257 b. Astronomy and Astrophysics - A&A, 2023, 670, 10.1051/0004-6361/202244494. hal- 04122780 X-SHYNE: X-shooter spectra of young exoplanet analogs: I. A medium-resolution 0.65-2.5 µ m one-shot spectrum of VHS 1256-1257 b Simon Petrus, Gaël Chauvin, M. Bonnefoy, Pascal Tremblin, B. Charnay, P. Delorme, Gabriel Dominique Marleau, Amelia Bayo, Elena Manjavacas, Anne Marie Lagrange, et al. To cite this version: Simon Petrus, Gaël Chauvin, M. Bonnefoy, Pascal Tremblin, B. Charnay, et al.. X-SHYNE: X-shooter spectra of young exoplanet analogs: I. A medium-resolution 0.65-2.5 µ m one-shot spectrum of VHS 1256-1257 b. Astronomy and Astrophysics - A&A, 2023, 670, 10.1051/0004-6361/202244494. hal- 04122780 Distributed under a Creative Commons Attribution 4.0 International License 1. Introduction The so-called L–T transition is evidenced by a sharp bluer- ing of near-infrared colors thought to appear at nearly-constant eﬀective temperatures in mature ﬁeld dwarfs (Filippazzo et al. 2015). The transition appears to be even more extreme for young objects, which are redder and underluminous by up to a few mag- nitudes compared to mature ﬁeld dwarf counterparts. Sophis- ticated models of clouds of diﬀerent compositions (silicates, sulﬁtes, etc.) have been considered for more than a decade to establish the basis of our understanding of atmospheric physics and reproduce the collected spectra. Although they manage to reproduce the spectra of massive late M- to early L-type young brown dwarfs and exoplanets well, they fail to ﬁt the luminosity and the spectra of cooler L–T transition planets (Bonnefoy et al. 2016). The problem likely arises from uncertainties in the cloud model itself in terms of composition, particle size, and density, and also a possible cloud deck inhomogeneity. The modiﬁca- tion of the vertical cloud distribution and cloud particle distri- bution is corroborated by recent variability studies that show higher-amplitude variability in young L dwarfs than old ones (Metchev et al. 2015; Manjavacas et al. 2018). Additional pro- cesses such as desequilibrium chemistry (Skemer et al. 2014) and related thermo-chemical instability (Tremblin et al. 2015) are now also included in some models. The resulting warm- ing up of the deep atmosphere along the L–T transition is pro- posed to explain the properties of these objects, including their rotational spectral modulations (Tremblin et al. 2020). However, today’s low-resolution spectra of imaged exoplanets covering a small portion of the planet spectral energy distribution (SED) Direct imaging and interferometry of young (5–200 Myr) planets can yield high-quality spectra made of tens to thousands of data points in a few hours of telescope time (Biller et al. 2018). About a dozen exoplanets have been directly characterized. The major- ity of them seem to deviate from the well-deﬁned sequence of M, L, and T brown dwarfs in the mature ﬁeld, indicative of diﬀerent physical and atmospheric properties. Since 2001, a few tens of free-ﬂoating objects sharing the same age and mass range as imaged exoplanets have also been identiﬁed in young star-forming regions (Lucas et al. 2001) and more recently in young nearby associations where most imaged exoplanets are identiﬁed (e.g., Faherty et al. 2016, and refer- ences therein). ABSTRACT We present simultaneous 0.65–2.5 µm medium resolution (3300 ≤Rλ ≤8100) VLT/X-shooter spectra of the relatively young (150–300 Myr) low-mass (19 ± 5MJup) L–T transition object VHS 1256−1257 b, a known spectroscopic analog of HR8799d. The companion is a prime target for the JWST Early Release Science (ERS) and one of the highest-amplitude variable brown dwarfs known to date. We compare the spectrum to the custom grids of cloudless ATMO models, exploring the atmospheric composition with the Bayesian inference tool ForMoSA. We also reanalyze low-resolution HST/WFC3 1.10–1.67 µm spectra at minimum and maximum variability to contextualize the X-shooter data interpretation. The models reproduce the slope and most molecular absorption from 1.10 to 2.48 µm self-consistently, but they fail to provide a radius and a surface gravity consistent with evolutionary model predictions. They do not reproduce the optical spectrum and the depth of the K I doublets in the J band consistently. We derived Teﬀ= 1380±54 K, log(g) = 3.97±0.48 dex, [M/H] = 0.21±0.29, and C/O > 0.63. Our inversion of the HST/WFC3 spectra suggests a relative change of 27+6 −5 K of the disk-integrated Teﬀcorrelated with the near-infrared brightness. Our data anchor the characterization of that object in the near-infrared and could be used jointly to the ERS mid-infrared data to provide the most detailed characterization of an ultracool dwarf to date. Key words. techniques: spectroscopic – planets and satellites: atmospheres – planets and satellites: formation – infrared: planetary systems Key words. techniques: spectroscopic – planets and satellites: atmospheres – planets and satellites: formation – infrared: planetary systems Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication. ⋆Based on observations collected at the European Organization for Astronomical Research in the Southern Hemisphere under ESO pro- gramme 0101.C-0290. HAL Id: hal-04122780 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 A&A 670, L9 (2023) https://doi.org/10.1051/0004-6361/202244494 c⃝The Authors 2023 Astronomy & Astrophysics A&A 670, L9 (2023) https://doi.org/10.1051/0004-6361/202244494 c⃝The Authors 2023 LETTER TO THE EDITOR Open Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication. L9, page 1 of 8 3. Results The X-shooter extracted spectrum of VHS 1256−1257 b is reported in Fig. 1. We decided to exclude data shorter than 0.65 µm due to the low S/N at these wavelengths. We performed an empirical analysis that conﬁrmed the young age of this target (see Appendix B). To explore the spectral diversity of the photometric and spectroscopic observations, we used ForMoSA, a tool based on a forward-modeling approach that compares observations with grids of precomputed synthetic atmospheric models using Bayesian inference methods. This code is presented in Petrus et al. (2020) and Petrus et al. (2021), but it has been updated for this study1. The grids of synthetic spectra used as input are now provided in the standardized NetCDF4 format through the use of the xarray2 module, created and maintained by the community of meteorologists and geophysicists, which allows N-dimensional grids to be labeled, as well as parallel computations on these grids (interpolations, extrapolations, and arithmetic). p g p y p In this Letter, we focus on the young L–T transition object VHS J125601.92−125723.9 b (hereafter VHS 1256−1257 b; see Table 1). This companion was detected at 8.06 ± 0.03 as (pro- jected physical separation of 179±9 au) from the tight 0.1 as a binary composed of two equal-magnitude, M7.5 components VHS J125601.92−125723.9 AB (Gauza et al. 2015; Stone et al. 2016). VHS 1256−1257 b is an excellent spectral analog of the emblematic exoplanet HR8799d (Bonnefoy et al. 2016) and one of the ﬁrst targets observed with the James Webb Space Telescope (JWST) which provides 3–28 µm medium-resolution spectra complementary to the X-SHYNE data (Miles et al. 2022). In this study, we present the X-SHYNE observations and the data reduction of VHS 1256−1257 b in Sect. 2. In Sect. 3, we describe our empirical study and the forward modeling analysis of the companion spectrum. In Sect. 4, we discuss atmospheric properties and prospects for future synergy with the JWST data. We used the grids of synthetic spectra produced by the last generation of the ATMO (Tremblin et al. 2015) models. Its speci- ﬁcities and the parameter space that it explores are described in Appendix C. To limit the impact of a possible bias on our estimates of the atmospheric parameters, we applied the strat- egy described by Petrus et al. (2020), which consists in deﬁn- ing diﬀerent optimal wavelength ranges for diﬀerent adjustments according to the speciﬁc parameter studied. 1. Introduction They are seconded by a population of ultra- wide orbit (≥100 au) companions to low-mass stars sharing the same properties (e.g., Béjar et al. 2008; Naud et al. 2014; Artigau et al. 2015). Contrary to imaged planets blurred into the stellar halo, these objects can be characterized at high signal- to-noise ratio (S/N) with seeing-limited spectrographs over a broad spectral range. They have been found to be precious empirical templates to imaged exoplanets with similar place- ments in color–magnitude diagrams and spectral features at low-resolving powers (e.g., Bonnefoy et al. 2010; Chilcote et al. 2017; Chauvin et al. 2017). ⋆Based on observations collected at the European Organization for Astronomical Research in the Southern Hemisphere under ESO pro- gramme 0101.C-0290. L9, page 1 of 8 A&A 670, L9 (2023) Table 1. Physical properties of VHS J125601.92−125723.9 AB and b. ing to resolving powers Rλ = λ/∆λ = 3300, 6700, and 8100, respectively. The slits were oriented perpendicular to the position angles of the companion in order to mitigate the ﬂux contami- nation of the host stars. The target was observed following an ABBA strategy to evaluate and remove the sky emission during the data processing step. The observing conditions and details of the observing log are reported in Appendix A. We used the ESO reflex framework (Freudling et al. 2013) to run the X-shooter pipeline version 2.9.3 on the raw data (Modigliani et al. 2010). The pipeline produces two-dimensional, curvature-corrected, and ﬂux-calibrated spectra for each target and epoch of obser- vation (trace). The spectra were extracted from the traces using a custom python script. The ﬂux in each wavelength channel at the position of the source was averaged within 720 mas aperture in the UVB and VIS arms, and a 1120 mas aperture in the NIR arm. The script computed the noise at the position of the source into each spectral channel following the procedure described in Petrus et al. (2020). The residual nonlinear pixels in the spectra were removed using the kappa-sigma clipping method. The telluric corrections were evaluated and removed using the molecfit package (Smette et al. 2015; Kausch et al. 2015). The spectra at each epoch were corrected from the barycentric veloc- ity and ﬂux-calibrated using a spectro-photometric standard. Finally, we merged our two spectra into one spectrum. Table 1. Physical properties of VHS J125601.92−125723.9 AB and b. Properties Binary Companion Ref. 3. Results Thus, to estimate the eﬀective temperature Teﬀand the radius, we performed a ﬁt between 1.10 and 2.48 µm, masking out the optical part of the spectrum that is not well reproduced by the models (panel E of Fig. 2). The lack of molecules such as CrH in the model participate in this nonreproducibility, which needs to be inves- tigated more. The radius was estimated using the dilution fac- tor Ck = (R/d)2, with R being the target’s radius and d the dis- tance. To estimate the surface gravity, the metallicity, the radial 1 https://pypi.org/project/ForMoSA/ 2 xarray.pydata.org/en/stable/ 1. Introduction Medium spectral resolution enables a better deblending of the atomic and the molecular lines and it oﬀer the rich perspective to untangle temperature, age, atmospheric composition, cloud properties, and disequilibrium chemistry in the spectra. The detailed study of medium resolution spectra will considerably improve the current generation of planetary atmosphere models. In this Letter, we focus on the young L–T transition object VHS J125601.92−125723.9 b (hereafter VHS 1256−1257 b; see Table 1). This companion was detected at 8.06 ± 0.03 as (pro- jected physical separation of 179±9 au) from the tight 0.1 as a binary composed of two equal-magnitude, M7.5 components VHS J125601.92−125723.9 AB (Gauza et al. 2015; Stone et al. 2016). VHS 1256−1257 b is an excellent spectral analog of the emblematic exoplanet HR8799d (Bonnefoy et al. 2016) and one of the ﬁrst targets observed with the James Webb Space Telescope (JWST) which provides 3–28 µm medium-resolution spectra complementary to the X-SHYNE data (Miles et al. 2022). In this study, we present the X-SHYNE observations and the data reduction of VHS 1256−1257 b in Sect. 2. In Sect. 3, we describe our empirical study and the forward modeling analysis of the companion spectrum. In Sect. 4, we discuss atmospheric properties and prospects for future synergy with the JWST data. L9, page 2 of 8 1. Introduction Spectral type M7.5 ± 0.5 L7±1.5 1 Luminosity (log L⊙) −2.95 ± 0.07 −4.54 ± 0.07 2 Teﬀ(K) 290288 −53 1240 ± 50 2, 3 log(g) (dex) 5.22+0.01 −0.02 4.55+0.15 −0.11 2, 3 Mass (MJup) 114+7 −5 19 ± 5 2, 3 Radius (RJup) 1.33+0.07 −0.05 1.17 ± 0.04 2, 3 2MASSJ 11.02 ± 0.02 16.66 ± 0.29 4 2MASSH 10.47 ± 0.02 15.60 ± 0.21 4 2MASSK 10.04 ± 0.02 14.57 ± 0.12 4 Distance (pc) 22.2+1.1 −1.2 2 Age (Myr) 150–300 1, 2 Projected sep. (as) 0.109±0.002 8.06±0.03 3 Physical sep. (au) 2.4±0.1 179±9 2 Notes. The companion Teﬀ, mass, radius, and surface gravity are updated from evolutionary model predictions (Baraﬀe et al. 2003). References. 1 – Gauza et al. (2015), 2 – Dupuy et al. (2020), 3 – this work, 4 – Cutri et al. (2003). Notes. The companion Teﬀ, mass, radius, and surface gravity are updated from evolutionary model predictions (Baraﬀe et al. 2003). References. 1 – Gauza et al. (2015), 2 – Dupuy et al. (2020), 3 – this work, 4 – Cutri et al. (2003). Notes. The companion Teﬀ, mass, radius, and surface gravity are updated from evolutionary model predictions (Baraﬀe et al. 2003). References. 1 – Gauza et al. (2015), 2 – Dupuy et al. (2020), 3 – this work, 4 – Cutri et al. (2003). do not allow one to explore or remove the atmospheric models’ degeneracies. In 2018, we initiated the X-shooter spectra of YouNg Exo- planet analogs program (X-SHYNE) to acquire simultaneous medium-resolution 0.3–2.5 µm spectra of forty young L- and T- type exoplanet analogs that overlap the population of imaged planetary companions in terms of mass, luminosity, and age. Medium spectral resolution enables a better deblending of the atomic and the molecular lines and it oﬀer the rich perspective to untangle temperature, age, atmospheric composition, cloud properties, and disequilibrium chemistry in the spectra. The detailed study of medium resolution spectra will considerably improve the current generation of planetary atmosphere models. In 2018, we initiated the X-shooter spectra of YouNg Exo- planet analogs program (X-SHYNE) to acquire simultaneous medium-resolution 0.3–2.5 µm spectra of forty young L- and T- type exoplanet analogs that overlap the population of imaged planetary companions in terms of mass, luminosity, and age. 2. Observation and data reduction The adopted param- eters from each run is given in the Table D.1. 2. Observation and data reduction perature, surface gravity, and mass: L = −5.05 ± 0.22 dex, Teﬀ= 880+140 −110 K, log(g) = 4.25+0.35 −0.10, and M = 11.2+1.8 −9.7MJup, respectively. Their solutions led to a rather cool atmosphere and unusually under-luminous object given its spectral type. The par- allax measurement was recently revised by Dupuy et al. (2020) based on CFHT and Pan-STARRS observations placing this sys- tem at larger distances: 51.6 ± 3.0 mas (22.2+1.1 −1.2 pc). This new estimate implies higher values for the mass (19 ± 5 MJup), sur- face gravity (log(g) = 4.55+0.15 −0.11), radius (1.17+0.04 −0.04 RJup), and a warmer temperature (1240 ± 50 K) from the Saumon & Marley (2008) hybrid evolutionary models. It brings the absolute mag- nitudes of all three components in better agreement with known young objects. From a ﬁrst atmospheric model analysis using near-infared spectra (NTT/SofI, 0.95–2.52 µm, Rλ = 600) and L spectra (Keck/NIRSPEC at LP band, Rλ = 1300), Miles et al. (2018) had already derived a warmer temperature of 1240 K with photospheric clouds for VHS 1256−1257 b although sur- face gravity and radius estimates were still aﬀected by the early, distance determination. From their LP band spectra, Miles et al. (2018) have also reported the detection of low abun- dances of CH4, suggesting that nonequilibrium chemistry pro- cesses are at play between CO and CH4. The upper atmo- spheres of their best-ﬁt models depart from equilibrium abun- dances of CH4 by factors of 10–100. Lastly, Hoch et al. (2022) perature, surface gravity, and mass: L = −5.05 ± 0.22 dex, Teﬀ= 880+140 −110 K, log(g) = 4.25+0.35 −0.10, and M = 11.2+1.8 −9.7MJup, respectively. Their solutions led to a rather cool atmosphere and unusually under-luminous object given its spectral type. The par- allax measurement was recently revised by Dupuy et al. (2020) based on CFHT and Pan-STARRS observations placing this sys- tem at larger distances: 51.6 ± 3.0 mas (22.2+1.1 −1.2 pc). This new estimate implies higher values for the mass (19 ± 5 MJup), sur- face gravity (log(g) = 4.55+0.15 −0.11), radius (1.17+0.04 −0.04 RJup), and a warmer temperature (1240 ± 50 K) from the Saumon & Marley (2008) hybrid evolutionary models. It brings the absolute mag- nitudes of all three components in better agreement with known young objects. 2. Observation and data reduction For these last two conﬁgurations, the dilution factor Ck was calculated ana- lytically using the relation of Cushing et al. (2008). These dif- ferent ﬁts are shown in Fig. 2. We also performed independent ﬁts on the J, H, and K band, in addition to the LP spectrum of VHS 1256−12257 b from Miles et al. 2018. The adopted param- eters from each run is given in the Table D.1. 4. Discussion In the context of the discovery of VHS 1256−1257 b, Gauza et al. (2015) led a ﬁrst empirical characterization to derive perature, surface gravity, and mass: L = −5.05 ± 0.22 dex, Teﬀ= 880+140 −110 K, log(g) = 4.25+0.35 −0.10, and M = 11.2+1.8 −9.7MJup, respectively. Their solutions led to a rather cool atmosphere and unusually under-luminous object given its spectral type. The par- allax measurement was recently revised by Dupuy et al. (2020) based on CFHT and Pan-STARRS observations placing this sys- tem at larger distances: 51.6 ± 3.0 mas (22.2+1.1 −1.2 pc). This new estimate implies higher values for the mass (19 ± 5 MJup), sur- face gravity (log(g) = 4.55+0.15 −0.11), radius (1.17+0.04 −0.04 RJup), and a warmer temperature (1240 ± 50 K) from the Saumon & Marley (2008) hybrid evolutionary models. It brings the absolute mag- nitudes of all three components in better agreement with known young objects. From a ﬁrst atmospheric model analysis using near-infared spectra (NTT/SofI, 0.95–2.52 µm, Rλ = 600) and L spectra (Keck/NIRSPEC at LP band, Rλ = 1300), Miles et al. (2018) had already derived a warmer temperature of 1240 K with photospheric clouds for VHS 1256−1257 b although sur- face gravity and radius estimates were still aﬀected by the S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b Fig. 1. X-shooter optical and near-infrared spec- tra of VHS 1256-1257 b (black). Zooms are pro- vided in order to show the atomic and molecular absorptions that are detected in our data. 2. Observation and data reduction From a ﬁrst atmospheric model analysis using near-infared spectra (NTT/SofI, 0.95–2.52 µm, Rλ = 600) and L spectra (Keck/NIRSPEC at LP band, Rλ = 1300), Miles et al. (2018) had already derived a warmer temperature of 1240 K with photospheric clouds for VHS 1256−1257 b although sur- face gravity and radius estimates were still aﬀected by the early, distance determination. From their LP band spectra, Miles et al. (2018) have also reported the detection of low abun- dances of CH4, suggesting that nonequilibrium chemistry pro- cesses are at play between CO and CH4. The upper atmo- spheres of their best-ﬁt models depart from equilibrium abun- dances of CH4 by factors of 10–100. Lastly, Hoch et al. (2022) velocity, and the rotational velocity, we chose to take advantage of the absorption doublet K I with the higher S/N which is sensi- tive to these parameters. We therefore ﬁt our data between 1.225 and 1.275 µm, without the continuum, to avoid the contribution of the pseudo-continuum and molecular absorption to the ﬁt. We also ﬁt our data between 2.28 and 2.41 µm to exploit the CO overtones, which are particularly sensitive to C/O. For these last two conﬁgurations, the dilution factor Ck was calculated ana- lytically using the relation of Cushing et al. (2008). These dif- ferent ﬁts are shown in Fig. 2. We also performed independent ﬁts on the J, H, and K band, in addition to the LP spectrum of VHS 1256−12257 b from Miles et al. 2018. The adopted param- eters from each run is given in the Table D.1. velocity, and the rotational velocity, we chose to take advantage of the absorption doublet K I with the higher S/N which is sensi- tive to these parameters. We therefore ﬁt our data between 1.225 and 1.275 µm, without the continuum, to avoid the contribution of the pseudo-continuum and molecular absorption to the ﬁt. We also ﬁt our data between 2.28 and 2.41 µm to exploit the CO overtones, which are particularly sensitive to C/O. For these last two conﬁgurations, the dilution factor Ck was calculated ana- lytically using the relation of Cushing et al. (2008). These dif- ferent ﬁts are shown in Fig. 2. We also performed independent ﬁts on the J, H, and K band, in addition to the LP spectrum of VHS 1256−12257 b from Miles et al. 2018. 2. Observation and data reduction VHS 1256−1257 b was observed on May 28, 2018 with the X-shooter seeing-limited medium-resolution spectrograph mounted at the VLT/UT2 Cassegrain focus (Vernet et al. 2011). The wide wavelength coverage of the instrument (0.300– 2.480 µm) obtained at one shot prevents uncertainties arising from the collage of spectra of this variable object obtained on separate bands on diﬀerent nights. We chose the 1.6′′, 1.2′′ and 0.6′′-wide slits for the UVB, VIS, and NIR arms correspond- L9, page 2 of 8 S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b Fig. 1. X-shooter optical and near-infrared spec- tra of VHS 1256-1257 b (black). Zooms are pro- vided in order to show the atomic and molecular absorptions that are detected in our data. 0.7 0.8 0.9 1.0 1.1 Wavelength ( m) 20 0 20 Residuals ( ) 0.7 1.0 1.3 1.6 Normalized flux (E) Best fit (ATMO) X-Shooter data 3.0 3.2 3.4 3.6 3.8 4.0 10 0 10 Residuals ( ) 1.4 1.6 1.8 Normalized flux (D) 2.290 2.295 2.300 2.305 2.310 2.315 2.320 2.325 2.330 2.5 0.0 2.5 Residuals ( ) 1.7 2.0 2.3 Normalized flux (C) 1.240 1.242 1.244 1.246 1.248 1.250 1.252 1.254 1.256 2.5 0.0 2.5 Residuals ( ) 1.6 1.8 2.0 Normalized flux (B) 1.2 1.4 1.6 1.8 2.0 2.2 2.4 10 010 Residuals ( ) 1.0 1.5 2.0 2.5 Normalized flux (A) Fig. 2. X-shooter spectra of VHS 1256−1257 b (black) compared to the best-ﬁt ATMO model (red). Residuals from the best ﬁt are shown at the bottom of each panel. The ﬁt is pro- vided between (A) 1.10 and 2.48 µm; (B) 1.225 and 1.275 µm, a zoom around the K I doublet is shown here; (C) 2.28 and 2.41 µm, a zoom around the CO overtones is shown here; and (D) 2.90 and 4.14 µm. (E) 0.65–2.48 µm is the ﬁt evidencing the model departure at optical wave- lengths which are only shown here. velocity, and the rotational velocity, we chose to take advantage of the absorption doublet K I with the higher S/N which is sensi- tive to these parameters. We therefore ﬁt our data between 1.225 and 1.275 µm, without the continuum, to avoid the contribution of the pseudo-continuum and molecular absorption to the ﬁt. We also ﬁt our data between 2.28 and 2.41 µm to exploit the CO overtones, which are particularly sensitive to C/O. 2. Observation and data reduction The ﬁt is p vided between (A) 1.10 and 2.48 µm; (B) 1.2 and 1.275 µm, a zoom around the K I doub is shown here; (C) 2.28 and 2.41 µm, a zoo around the CO overtones is shown here; and ( 2.90 and 4.14 µm. (E) 0.65–2.48 µm is the evidencing the model departure at optical wav lengths which are only shown here. Fig. 1. X-shooter optical and near-infrared spec- tra of VHS 1256-1257 b (black). Zooms are pro- vided in order to show the atomic and molecular absorptions that are detected in our data. 0.7 0.8 0.9 1.0 1.1 Wavelength ( m) 20 0 20 Residuals ( ) 0.7 1.0 1.3 1.6 Normalized flux (E) Best fit (ATMO) X-Shooter data 3.0 3.2 3.4 3.6 3.8 4.0 10 0 10 Residuals ( ) 1.4 1.6 1.8 Normalized flux (D) 2.290 2.295 2.300 2.305 2.310 2.315 2.320 2.325 2.330 2.5 0.0 2.5 Residuals ( ) 1.7 2.0 2.3 Normalized flux (C) 1.240 1.242 1.244 1.246 1.248 1.250 1.252 1.254 1.256 2.5 0.0 2.5 Residuals ( ) 1.6 1.8 2.0 Normalized flux (B) 1.2 1.4 1.6 1.8 2.0 2.2 2.4 10 010 Residuals ( ) 1.0 1.5 2.0 2.5 Normalized flux (A) Fig. 2. X-shooter spectra of VHS 1256−1257 b (black) compared to the best-ﬁt ATMO model (red). Residuals from the best ﬁt are shown at the bottom of each panel. The ﬁt is pro- vided between (A) 1.10 and 2.48 µm; (B) 1.225 and 1.275 µm, a zoom around the K I doublet is shown here; (C) 2.28 and 2.41 µm, a zoom around the CO overtones is shown here; and (D) 2.90 and 4.14 µm. (E) 0.65–2.48 µm is the ﬁt evidencing the model departure at optical wave- lengths which are only shown here. Fig. 2. X-shooter spectra of VHS 1256−1257 b (black) compared to the best-ﬁt ATMO model (red). Residuals from the best ﬁt are shown at the bottom of each panel. The ﬁt is pro- vided between (A) 1.10 and 2.48 µm; (B) 1.225 and 1.275 µm, a zoom around the K I doublet is shown here; (C) 2.28 and 2.41 µm, a zoom around the CO overtones is shown here; and (D) 2.90 and 4.14 µm. (E) 0.65–2.48 µm is the ﬁt evidencing the model departure at optical wave- lengths which are only shown here. 2. Observation and data reduction 0.7 0.8 0.9 1.0 1.1 Wavelength ( m) 20 0 20 Residuals ( ) 0.7 1.0 1.3 1.6 Normalized flux (E) Best fit (ATMO) X-Shooter data 3.0 3.2 3.4 3.6 3.8 4.0 10 0 10 Residuals ( ) 1.4 1.6 1.8 Normalized flux (D) 2.290 2.295 2.300 2.305 2.310 2.315 2.320 2.325 2.330 2.5 0.0 2.5 Residuals ( ) 1.7 2.0 2.3 Normalized flux (C) 1.240 1.242 1.244 1.246 1.248 1.250 1.252 1.254 1.256 2.5 0.0 2.5 Residuals ( ) 1.6 1.8 2.0 Normalized flux (B) 1.2 1.4 1.6 1.8 2.0 2.2 2.4 10 010 Residuals ( ) 1.0 1.5 2.0 2.5 Normalized flux (A) Fig. 2. X-shooter spectra of VHS 1256−1257 b (black) compared to the best-ﬁt ATMO model (red). Residuals from the best ﬁt are shown at the bottom of each panel. The ﬁt is pro- vided between (A) 1.10 and 2.48 µm; (B) 1.225 and 1.275 µm, a zoom around the K I doublet is shown here; (C) 2.28 and 2.41 µm, a zoom around the CO overtones is shown here; and (D) 2.90 and 4.14 µm. (E) 0.65–2.48 µm is the ﬁt evidencing the model departure at optical wave- lengths which are only shown here. Fig. 1. X-shooter optical and near-i tra of VHS 1256-1257 b (black). Zo vided in order to show the atomic an absorptions that are detected in our Fig. 1. X-shooter optical and near-infrared spe tra of VHS 1256-1257 b (black). Zooms are p vided in order to show the atomic and molecu absorptions that are detected in our data. 0.7 0.8 0.9 1.0 1.1 Wavelength ( m) 20 0 20 Residuals ( ) 0.7 1.0 1.3 1.6 Normalized flux (E) Best fit (ATMO) X-Shooter data 3.0 3.2 3.4 3.6 3.8 4.0 10 0 10 Residuals ( ) 1.4 1.6 1.8 Normalized flux (D) 2.290 2.295 2.300 2.305 2.310 2.315 2.320 2.325 2.330 2.5 0.0 2.5 Residuals ( ) 1.7 2.0 2.3 Normalized flux (C) 1.240 1.242 1.244 1.246 1.248 1.250 1.252 1.254 1.256 2.5 0.0 2.5 Residuals ( ) 1.6 1.8 2.0 Normalized flux (B) 1.2 1.4 1.6 1.8 2.0 2.2 2.4 10 010 Residuals ( ) 1.0 1.5 2.0 2.5 Normalized flux (A) Fig. 2. X-shooter spectra of VHS 1256−125 (black) compared to the best-ﬁt ATMO mod (red). Residuals from the best ﬁt are show at the bottom of each panel. 4. Discussion In the context of the discovery of VHS 1256−1257 b, Gauza et al. (2015) led a ﬁrst empirical characterization to derive the spectral type, but also the bulk properties of the compan- ion. Comparing the luminosity of VHS 1256−1257 b, derived from the near-infrared photometry, bolometric correction and ﬁrst parallax estimate (π = 78.8 ± 6.4 mas, 12.7 ± 1.0 pc), to the predictions of the BT-SETTL evolutionary models, they derived the following physical values for the luminosity, eﬀective tem- L9, page 3 of 8 A&A 670, L9 (2023) 1.2 1.3 1.4 1.5 1.6 Wavelength ( m) 1 2 3 4 5 6 7 8 Flux (W. m 2. m 1) 1e 16 X-Shooter (this work) Data HST G141 (minimum luminosity) Best fit (minimum luminosity) Data HST G141 (maximum luminosity) Best fit (maximum luminosity) 1300 1310 1320 1330 1340 1350 Teff (K) 1.2 1.3 1.4 1.5 1.6 Wavelength ( m) 1 2 3 4 5 6 7 8 Flux (W. m 2. m 1) 1e 16 X-Shooter (this work) Data HST G141 (minimum luminosity) Best fit (minimum luminosity) Data HST G141 (maximum luminosity) Best fit (maximum luminosity) Fig. 3. Comparison with HST data. Left: ForMoSA ﬁts (dash lines) of HST G141 data (solid lines). We used the maximum (blue) and minimum (red) amplitude data from the sequence of Bowler et al. (2020). We also compared these HST data with our X-shooter data. Right: comparison between Teﬀposteriors from the HST’s ﬁts (red and blue) and the ﬁt of our X-shooter data (gray). estimate Teﬀ ∼ 1200-1300 K, log(g) ∼3.25–3.75 dex, and C/O ∼0.59 with the inversion of their Keck/OSIRIS spec- trum with a custom version of the model of atmosphere PHOENIX. estimate Teﬀ ∼ 1200-1300 K, log(g) ∼3.25–3.75 dex, and C/O ∼0.59 with the inversion of their Keck/OSIRIS spec- trum with a custom version of the model of atmosphere PHOENIX. O h i f d d li l i f h The [M/H] and the C/O that we measured are very similar to those measured for HR8799 b (C/O = 0.61+0.09 −0.03, Barman et al. 2015), HR8799 c (C/O = 0.65+0.10 −0.05, Konopacky et al. 2013), and HR8799 e ([M/H] = 0.48+0.25 −0.29 and C/O = 0.60+0.07 −0.08, Mollière et al. 2020). 4. Discussion This conﬂict between the physical properties’ estimate from models of atmosphere and from evolutionary models are well known and constitutes one of the current main challenges in the characterization of planetary atmospheres. VHS 1256−1257 b has been identiﬁed as one of the most variable substellar objects (19.3% spectroscopic variations between 1.1 and 1.7 µm) by Bowler et al. (2020), thanks to a sequence of 11 observations (8.5 h) performed with HST. The presence of inhomogeneous cloud covers has been proposed and explored for this companion by Zhou et al. (2020), but the existence of temperature ﬂuctuations arising in a convective atmosphere could be an alternative explanation (Tremblin et al. 2020). To explore the impact of this spectroscopic variability, we applied ForMoSA using the ATMO models on the faintest and brightest state of this spectroscopic sequence to estimate Teﬀand we compared them with the Teﬀestimated from our X-shooter data. For these ﬁts, we calculated the dilution factor Ck analyt- ically to avoid any possible diﬀerence in ﬂux calibration, and we calculated the likelihood using the same wavelength range. Figure 3 illustrates the X-shooter data, over-plotted with the ﬁts of the HST data (left panel) and the corresponding Teﬀposte- riors (right panel). The diﬀerence in temperatures between the faintest and brightest states could be indicative of a temperature ﬂuctuation in the atmosphere of VHS 1256−1257 b. Modulations of the temperature gradient in the region around one bar could also explain the possible CH4 variability detection observed by Bowler et al. (2020). Finally, we also note that the Teﬀestimate from our X-Shooter data, using the same wavelength range as the HST data (gray posterior), is coherent with the Teﬀestimate with the brightest state of the HST spectra. Finally, accessing spectral resolution up to 8100 for the K band, for which water and carbon monoxide are detected, enabled us to constrain the C/O ratio between solar and super-solar val- ues with C/O ≥0.63. The metallicity appears to be solar to super-solar for all of the ﬁts (except for the K band and H band). 4. Discussion Thesolartosuper-solarcompositionofVHS 1256−1257 b could indicate a speciﬁc atmospheric metal enrichment of solids and gas during its phases of formation and evolution if it formed within a disk by core accretion or even gravitational instability (Mollière et al. 2022). However, the existence of such a massive planet-forming disk, as well as the presence of giant planets, is not expected given the very low mass of the central binary (Burn et al. 2021). In addition, the hierarchical architecture of the whole sys- tem, a tight equal-mass binary with a planetary-mass companion orbitingatwideorbit,stronglysupportsastellar-likeformationby gravo-turbulent fragmentation (Padoan & Nordlund 2004). Fur- ther spectroscopic observations to directly measure of the atmo- spheric composition of the central binary will be important. More extendedatmospheremodelingwillbecrucialtoconﬁrmthecom- position of VHS 1256−1257 b, but also its photometric temporal variability. Our atmospheric forward-modeling analysis of the X-shooter spectra conﬁrms the temperature of 1326 ± 2 K with ATMO models for a full ﬁtting applied between 1.10 and 2.48 µm. Nevertheless, this estimate varies signiﬁcantly with the choice of the wavelength ranges used for the ﬁt from 1145 ± 13 K for the LP band to 1417 ± 17 K for the K band. For the surface gravity determination, we focused on the ﬁtting of the gravity-sensitive K I lines between 1.225 and 1.275 µm. The resulting solution indicates a surface gravity of log(g) = 4.25 ± 0.20 dex, but this estimate tends to be lower (<4.0 dex) when the ﬁt is performed on a larger wavelength range (J, H, and K bands). Although the eﬀective temperature and surface gravity are relatively consistent at 2σ with the evolutionary model, the radius estimate (R between 0.71 and 0.91 RJup) given by the ATMO models is not, and deﬁnitively needs to be further investigated for this family of models. At the moment, the pressure levels that are impacted by the reduced temperature gradient is assumed to be a function of the surface gravity (see Appendix C for details); an adjustment of the size of the layer impacted by the reduced temperature gradient might be needed to get a radius in better agreement with evolutionary models. We also notice that the radius estimated from the LP band is consistent with the evolutionary model, but the surface gravity is inconsistent with that obtained on the other band and too low with respect to evolutionary model predictions. S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b Paris-Sud, UVSQ, i i é i S l 91191 Gif 5 Maison de la Simulation, CEA, CNRS, Univ. Paris-Su S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b 2004, ApJ, 617, 559 Petrus, S., Bonnefoy, M., Chauvin, G., et al. 2020, A&A, 633, A124 Petrus, S., Bonnefoy, M., Chauvin, G., et al. 2021, A&A, 648, A59 Saumon, D., & Marley, M. S. 2008, ApJ, 689, 1327 Skemer, A. J., Marley, M. S., Hinz, P. M., et al. 2014, ApJ, 792, 17 Smette, A., Sana, H., Noll, S., et al. 2015, A&A, 576, A77 Stone, J. M., Skemer, A. J., Kratter, K. M., et al. 2016, ApJ, 818, L12 Suárez, G., & Metchev, S. 2022, MNRAS, 513, 5701 Tremblin, P., Amundsen, D. S., Mourier, P., et al. 2015, ApJ, 804, L17 Tremblin, P., Amundsen, D. S., Chabrier, G., et al. 2016, ApJ, 817, L19 Tremblin, P., Padioleau, T., Phillips, M. W., et al. 2019, ApJ, 876, 144 Tremblin, P., Phillips, M. W., Emery, A., et al. 2020, A&A, 643, A23 Venot, O., Hébrard, E., Agúndez, M., et al. 2012, A&A, 546, A43 Vernet, J., Dekker, H., D’Odorico, S., et al. 2011, A&A, 536, A105 Zhou, Y., Bowler, B. P., Morley, C. V., et al. 2020, AJ, 160, 77 Strategies, Processes, and Systems III, eds. D. R. Silva, A. B. Peck, & B Soifer, SPIE Conf. Ser., 7737, 773728 Mollière, P., Stolker, T., Lacour, S., et al. 2020, A&A, 640, A131 Mollière, P., Molyarova, T., Bitsch, B., et al. 2022, ApJ, 934, 74 Naud, M.-E., Artigau, É., Malo, L., et al. 2014, ApJ, 787, 5 Padoan, P., & Nordlund, Å. 2004, ApJ, 617, 559 Petrus, S., Bonnefoy, M., Chauvin, G., et al. 2020, A&A, 633, A124 Petrus, S., Bonnefoy, M., Chauvin, G., et al. 2021, A&A, 648, A59 Saumon, D., & Marley, M. S. 2008, ApJ, 689, 1327 Skemer, A. J., Marley, M. S., Hinz, P. M., et al. 2014, ApJ, 792, 17 Smette, A., Sana, H., Noll, S., et al. 2015, A&A, 576, A77 Stone, J. M., Skemer, A. J., Kratter, K. M., et al. 2016, ApJ, 818, L12 Suárez, G., & Metchev, S. 2022, MNRAS, 513, 5701 Tremblin, P., Amundsen, D. S., Mourier, P., et al. 2015, ApJ, 804, L17 Tremblin, P., Amundsen, D. S., Chabrier, G., et al. 2016, ApJ, 817, L19 Tremblin, P., Padioleau, T., Phillips, M. W., et al. 2019, ApJ, 876, 144 Tremblin, P., Phillips, M. W., Emery, A., et al. 2020, A&A, 643, A23 Venot, O., Hébrard, E., Agúndez, M., et al. S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b MIRI at medium spectral resolution over a broad wavelength range from 1 to 28 µm (Rλ ∼1500–3000), respectively. The characterization of these spectra using the forward modeling approach will be treated in a future paper. This will be par- ticularly interesting to accurately constrain the eﬀective tem- perature of VHS1256 b, the abundances of important secondary gases (CO, H2O, and CH4), the importance of thermo-chemical instability, patchy cloud decks, and nonequilibrium chemistry processes at play. As proposed by Tremblin et al. (2020), the detection of direct cloud spectral signatures such as the silicate absorption feature at 10 µm, as detected for ultracool dwarfs observed with Spitzer (Suárez & Metchev 2022), will help to conﬁrm the presence of clouds and their contribution to the spec- tral energy distribution and modulation of VHS 1256−1257 b. Such a signature has been detected in the spectrum provided by JWST and will be studied thoroughly in the future (Miles et al. 2022). The ERS program will serve as a benchmark for future characterization studies of young brown dwarfs and exoplane- tary atmospheres. Lucas, P. W., Roche, P. F., Allard, F., & Hauschildt, P. H. 2001, MNRAS, 326, 695 Manjavacas, E., Apai, D., Zhou, Y., et al. 2018, AJ, 155, 11 Manjavacas, E., Lodieu, N., Béjar, V. J. S., et al. 2020, MNRAS, j j Martin, E. C., Mace, G. N., McLean, I. S., et al. 2017, ApJ, 838, 73 Metchev, S. A., Heinze, A., Apai, D., et al. 2015, ApJ, 799, 154 Miles, B. E., Skemer, A. J., Barman, T. S., Allers, K. N., & Stone, J. M. 2018, ApJ, 869, 18 Miles, B. E., Biller, B. A., Patapis, P., et al. 2022, AAS J., submitted [arXiv:2209.00620] Modigliani, A., Goldoni, P., Royer, F., et al. 2010, in Observatory Operations: Strategies, Processes, and Systems III, eds. D. R. Silva, A. B. Peck, & B. T. Modigliani, A., Goldoni, P., Royer, F., et al. 2010, in Observatory Opera Strategies, Processes, and Systems III, eds. D. R. Silva, A. B. Peck, & Soifer, SPIE Conf. Ser., 7737, 773728 Mollière, P., Stolker, T., Lacour, S., et al. 2020, A&A, 640, A131 Mollière, P., Molyarova, T., Bitsch, B., et al. 2022, ApJ, 934, 74 Naud, M.-E., Artigau, É., Malo, L., et al. 2014, ApJ, 787, 5 Padoan, P., & Nordlund, Å. S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b 2012, A&A, 546, A43 Vernet, J., Dekker, H., D’Odorico, S., et al. 2011, A&A, 536, A105 Zhou, Y., Bowler, B. P., Morley, C. V., et al. 2020, AJ, 160, 77 y p Naud, M.-E., Artigau, É., Malo, L., et al. 2014, ApJ, 787, 5 g Padoan, P., & Nordlund, Å. 2004, ApJ, 617, 559 aumon, D., & Marley, M. S. 2008, ApJ, 689, 1327 kemer, A. J., Marley, M. S., Hinz, P. M., et al. 2014, ApJ, 792, remblin, P., Amundsen, D. S., Mourier, P., et al. 2015, ApJ, 804, Acknowledgements. We would like to thank the staﬀof ESO VLT for their support at the telescope at Paranal and La Silla, and the preparation of the observation at Garching. This publication made use of the SIMBAD and VizieR database operated at the CDS, Strasbourg, France. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www. cosmos.esa.int/gaia), processed by the Gaia Data Processing and Anal- ysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/ consor576utium). Funding for the DPAC has been provided by national insti- tutions, in particular the institutions participating in the Gaia Multilateral Agree- ment. We acknowledge support in France from the French National Research Agency (ANR) through project grants ANR-14-CE33-0018 and ANR-20-CE31- 0012. S.-P. acknowledges the support of ANID, – Millennium Science Initiative Program – NCN19_171. G-DM acknowledges the support of the DFG priority program SPP 1992 “Exploring the Diversity of Extrasolar Planets” (MA 9185/1) and from the Swiss National Science Foundation under grant 200021_204847 “PlanetsInTime”. Parts of this work have been carried out within the framework of the NCCR PlanetS supported by the Swiss National Science Foundation. 1 Instituto de Física y Astronomía, Facultad de Ciencias, Universidad de Valparaíso, Av. Gran Bretaña 1111, Valparaíso, Chile e-mail: simon.petrus@npf.cl 2 2 Núcleo Milenio Formacíon Planetaria – NPF, Universidad de Val- paraíso, Av. Gran Bretaña 1111, Valparaíso, Chile paraíso, Av. Gran Bretaña 1111, Valparaíso, Chile 3 Laboratoire Lagrange, Université Côte d’Azur, CNRS, Observ de la Côte d’Azur, 06304 Nice, France 4 p , , p 3 Laboratoire Lagrange, Université Côte d’ de la Côte d’Azur, 06304 Nice, France 4 de la Côte d’Azur, 06304 Nice, France 4 p 5 Maison de la Simulation, CEA, CNRS, Univ. Paris-Sud Université Paris-Saclay, 91191 Gif-sur-Yvette, France 6 p 5 Maison de la Simulation, CEA, CNRS, Univ. 1 Instituto de Física y Astronomía, Facultad de Ciencias, Universidad de Valparaíso, Av. Gran Bretaña 1111, Valparaíso, Chile e-mail: simon.petrus@npf.cl 4. Discussion In the context of the cloudless ATMO models, the adiabatic index value of γ = 1.02 ± 0.01 is consistent with a decrease in the adiabatic index in the lower part of the atmospheres, as well as processes of nonequilibrium chemistry and quench CO/CH4 (i.e., they prevent the formation of CH4), as is expected for young, late L dwarfs subject to a CO/CH4 thermo-chemical instability in this scenario (Tremblin et al. 2016). Finally, we also derived the radial velocity (RV) between −4.58 and 8.77 km s−1 and the projected rotational velocities (v sin(i)) < 37 km s−1 (limited by the spec- tral resolution of X-shooter), compatible with the NIRSPEC data (RV = 2.1+1.6 −1.7 km s−1 and v sin(i) = 13.5+3.6 −4.1 km s−1, Bryan et al. 2018. VHS 1256−1257 AB and b was a prime target of the JWST Early Release Science (ERS) program (Hinkley et al. 2022, PI: S. Hinkey) that acquired high-ﬁdelity spectra with NIRSpec and L9, page 4 of 8 References , , , Université Paris-Saclay, 91191 Gif-sur-Yvette, France 6 y 6 LESIA, Observatoire de Paris, Université PSL, CNRS, Sorbonne Université, Univ. Paris Diderot, Sorbonne Paris Cité, 5 Place Jules Janssen, 92195 Meudon, France 7 Allers, K. N., & Liu, M. C. 2013, ApJ, 772, 79 , , , , p , , Artigau, É., Gagné, J., Faherty, J., et al. 2015, ApJ, 806, 254 Artigau, É., Gagné, J., Faherty, J., et al. 2015, ApJ, 806, 254 Baraﬀe, I., Chabrier, G., Barman, T. S., Allard, F., & Hauschildt, P. H. 2003, A&A, 402, 701 7 Fakultät für Physik, Universität Duisburg-Essen, Lotharstraße 1, 47057 Duisburg, Germany Barman, T. S., Konopacky, Q. M., Macintosh, B., & Marois, C. 2015, ApJ, 804, 61 8 Institut für Astronomie und Astrophysik, Universität Tübingen, Auf der Morgenstelle 10, 72076 Tübingen, Germany Béjar, V. J. S., Zapatero Osorio, M. R., Pérez-Garrido, A., et al. 2008, ApJ, 673, Béjar, V. J. S., Zapatero Osorio, M. R., Pérez-Garrido, A., et al. 2008, ApJ, 673, L185 9 Physikalisches Institut, Universität Bern, Gesellschaftsstr. 6, 3012 Bern, Switzerland Biller, B. A., & Bonnefoy, M. 2018, in Handbook of Exoplanets (Springer International Publishing AG), 101 10 Max-Planck-Institut für Astronomie, Königstuhl 17, 69117 Heidel- berg, Germany Bonnefoy, M., Chauvin, G., Rojo, P., et al. 2010, A&A, 512, A52 Bonnefoy, M., Chauvin, G., Rojo, P., et al. 2010, A&A, 512, A52 Bonnefoy, M., Chauvin, G., Lagrange, A. M., et al. 2014, A&A, 562, A127 Bonnefoy, M., Chauvin, G., Lagrange, A. M., et al. 2014, A&A, 562, A127 g y 11 European Southern Observatory, Karl Schwarzschild-Straße 2, 85748 Garching bei München, Germany efoy, M., Zurlo, A., Baudino, J. L., et al. 2016, A&A, 587, A58 Bonnefoy, M., Zurlo, A., Baudino, J. L., et al. 2016, A&A, 587, A58 Bowler, B. P., Zhou, Y., Morley, C. V., et al. 2020, ApJ, 893, L30 Bowler, B. P., Zhou, Y., Morley, C. V., et al. 2020, ApJ, 893, L30 12 AURA for the European Space Agency (ESA), ESA Oﬃce, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA 13 Bryan, M. L., Benneke, B., Knutson, H. A., Batygin, K., & Bowler, B. P. 2018, N t A t 2 138 Bryan, M. L., Benneke, B., Knutson, H. A., Batygin, K., & Bowler, B. P. 2018, Nat. Astron., 2, 138 Nat. Astron., 2, 138 Burn, R., Schlecker, M., Mordasini, C., et al. 2021, A&A, 656, A72 Chauvin, G., Desidera, S., Lagrange, A. M., et al. 2017, A&A, 605, L9 13 LESIA, Observatoire de Paris, Université PSL, CNRS, Sorbonne Université, Université de Paris, 5 Place Jules Janssen, 92195 Meudon, France hilcote, J., Pueyo, L., De Rosa, R. J., et al. 2017, AJ, 153, 182 Cushing, M. C., Rayner, J. T., & Vacca, W. D. 2005, ApJ, 623, 1115 Cushing, M. C., Marley, M. S., Saumon, D., et al. 2008, ApJ, 678, 1372 14 Institute for Astronomy The University of Edinburgh Royal Obser- vatory Blackford Hill, Edinburgh EH9 3HJ, UK Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, VizieR Online Data Catalog: II/246 A&A 670, L9 (2023) Appendix A: X-Shooter observing log Catalog: II/246 y g 15 Núcleo de Astronomía, Facultad de Ingeniería y Ciencias, Universi- dad Diego Portales, Av. Ejército 441, Santiago, Chile Dupuy, T. J., Liu, M. C., Magnier, E. A., et al. 2020, Res. Notes Am. Astron. Soc., 4, 54 16 Centro de Astrofísica y Tecnologías Aﬁnes (CATA), Casilla 36-D, Santiago, Chile Faherty, J. K., Riedel, A. R., Cruz, K. L., et al. 2016, ApJS, 225, 10 y p Filippazzo, J. C., Rice, E. L., Faherty, J., et al. 2015, ApJ, 810, 158 Freudling, W., Romaniello, M., Bramich, D. M., et al. 2013, A&A, 559, A96 g 17 Department of Astronomy and Carl Sagan Institute, Cornell Univer- sity, 122 Sciences Drive, Ithaca, NY 14853, USA g Gauza, B., Béjar, V. J. S., Pérez-Garrido, A., et al. 2015, ApJ, 804, 96 j p Hinkley, S., Carter, A. L., Ray, S., et al. 2022, PASP, 134, 095003 18 Astrophysics Group, School of Physics and Astronomy, University of Exeter, Exeter EX4 4QL, UK Hoch, K. K. W., Konopacky, Q. M., Barman, T. S., et al. 2022, AJ, 164, 155 Kausch, W., Noll, S., Smette, A., et al. 2015, A&A, 576, A78 19 Center for Astrophysics and Space Sciences, University of Califor- nia, San Diego, La Jolla, CA 92093, USA Konopacky, Q. M., Barman, T. S., Macintosh, B. A., & Marois, C. 2013, Science, 339, 1398 L9, page 5 of 8 L9, page 5 of 8 Appendix A: X-Shooter observing log Table A.1 reports the observing conditions during our data acquirement. Table A.1. Observing log Seq Date UT Start-Time DIT NDIT NEXP <Seeing> Airmass S/N (yyyy-mm-dd) (hh:mm) (s) (") 1 2018-05-28 01:36 670/670/234 1/1/3 12/12/12 0.68 1.05 <1/2/17 2 2018-05-28 02:36 670/670/234 1/1/3 12/12/12 0.74 1.17 <1/2/26 Notes. The seeing is measured at 0.5 µm and given for the visible arm. The Detector Integration Time (DIT) values refer to the individual exposure time per frame in the UVB, VIS, and NIR arms respectively. The number of individual frames per exposure (NDIT) and the number of exposures (NEXP) are also given for each arms. Table A.1. Observing log Table A.1. Observing log Notes. The seeing is measured at 0.5 µm and given for the visible arm. The Detector Integration Time (DIT) values refer to the individual exposure time per frame in the UVB, VIS, and NIR arms respectively. The number of individual frames per exposure (NDIT) and the number of exposures (NEXP) are also given for each arms. Notes. The seeing is measured at 0.5 µm and given for the visible arm. The Detector Integration Time (DIT) values refer to the individual exposure time per frame in the UVB, VIS, and NIR arms respectively. The number of individual frames per exposure (NDIT) and the number of exposures (NEXP) are also given for each arms. L9, page 6 of 8 S. Petrus et al.: X-SHYNE. I. A medium-resolution 0.65–2.5 µm spectrum of VHS 1256-1257 b Appendix B: Empirical analysis Fig. B.1. Equivalent widths of the detected K I lines of VHS 1256 b compared to the values of the sample of old (FLD-G) and young, low gravity (VL-G) and intermediate gravity (INT-G), brown dwarfs of Martin et al. (2017). Appendix B: Empirical analysis Appendix B: Empirical analysis Table C.1. Parameters’ space of the grid of synthetic spectra ATMO. We note that Teﬀis the eﬀective temperature, log(g) is the surface gravity, [M/H] is the metallicity, C/O is the carbon-oxygen ratio, and γ is the reduced adiabatic index. Table C.1. Parameters’ space of the grid of synthetic spectra ATMO. We note that Teﬀis the eﬀective temperature, log(g) is the surface gravity, [M/H] is the metallicity, C/O is the carbon-oxygen ratio, and γ is the reduced adiabatic index. ATMO Teﬀ(K) Range [800, 3000] Step 50 log(g) (dex) Range [2.5, 5.5] Step 0.5 [M/H] Range [-0.6, 0.6] Step 0.3 C/O Range [0.3, 0.7] Step 0.25 γ Range [1.01, 1.05] Step 0.02 for their data Rλ ∼2000). For each K I line, the EW calculated for VHS1256 b is substantially weaker than the one of ﬁeld and intermediate-gravity (INT-G), which conﬁrms the VL-G classi- ﬁcation and young age determined by Gauza et al. (2015) based on the H2OD and H-continuum spectral indices. Fig. B.1. Equivalent widths of the detected K I lines of VHS 1256 b compared to the values of the sample of old (FLD-G) and young, low gravity (VL-G) and intermediate gravity (INT-G), brown dwarfs of Martin et al. (2017). Appendix C: Grid description In this appendix we brieﬂy describe the main properties of the ATMO (Tremblin et al. 2015) models and the grids that they gen- erated and that have been used in our study (see the Table C.1). These grids are publicly available at https://opendata. erc-atmo.eu. The ATMO models assume that clouds are not needed to reproduce the shape of the SED of brown dwarfs (apart from the 10-µm silicate feature). Tremblin et al. (2016) have proposed that diabatic convective processes (Tremblin et al. 2019) induced by out-of-equilibrium chemistry of CO/CH4 and N2/NH3 can reduce the temperature gradient in the atmosphere and reproduce the reddening previously thought to occur by clouds. The grids assume a modiﬁcation of the temperature gra- dient with an eﬀective adiabatic index. The levels modiﬁed are in between two and 200 bar at log(g) = 5.0 and are scaled by ×10log(g)−5 at other surface gravities. Out-of-equilirium chem- istry is used with kzz = 105 cm2/s at log(g) = 5.0 and are scaled by ×102(5−log(g)) at other surface gravities. The mixing length is assumed to be 2 scale height at 20 bar and higher pressures at log(g) = 5.0 and is scaled down by the ratio between the local pressure and the pressure at 20 bars for lower pressures. The 20 bars limit is scaled by ×10log(g)−5 at other surface gravities. The chemistry includes 277 species and out-of-equilibrium chem- istry has been performed using the model of Venot et al. (2012), and rainout is assumed to occur for species that are not included in the out-of-equilibrium model. Opacity sources include H2-H2, H2-He, H2O, CO2, CO, CH4, NH3, Na, K, Li, Rb, Cs, TiO, VO, FeH, PH3, H2S, HCN, C2H2, SO2, Fe, and H−, and the Rayleigh scattering opacities for H2, He, CO, N2, CH4, NH3, H2O, CO2, H2S, and SO2. The grids explore the following parameters: eﬀec- tive temperatures between 500 and 3000K (step 100K); log(g) between 2.5 and 5.5 (step 0.5); an eﬀective adiabatic index with three values of 1.05, 1.03, and 1.01; metallicity with ﬁve values of -0.6, -0.3, 0, +0.3, and +0.6; a C/O ratio with three values of 0.3, solar, and 0.7. The last version has been used to generate a grid of synthetic spectra at medium resolution (Rλ ∼10000) from 0.83 to 2.5 µm as well as lower resolution synthetic spec- tra (Rλ ∼1000) between 0.8 and 20 µm. Table A.1. Observing log The medium spectral resolution of our data allowed us to identify several strong absorption lines between 0.8 and 1.3 µm, such as neutral cesium (Cs, 0.852 µm), as well as neutral potas- sium (K I) doublets (1.168, 1.177 µm and 1.243, 1.254 µm). The sodium (Na I) doublet (1.138 and 1.141 µm) is also detected. These lines were not resolved in the early study of Gauza et al. (2015) given the lower spectral resolution of the GTC/OSIRIS optical and NTT/SofI nIR spectra. We therefore derived the equivalent width (EW) of the KI lines and compared them with the ones obtained by Martin et al. (2017) for a sample of old and young (low gravity and intermediate gravity) brown dwarfs as shown in Fig. B.1. Appendix C: Grid description The EW of alkaline lines in the NIR has been shown to cor- relate with surface gravity (Cushing et al. 2005; Allers & Liu 2013; Bonnefoy et al. 2014; Manjavacas et al. 2020). The strength of these lines increases with the surface gravity due to a higher pressure in the photosphere. Therefore, they are great tracers of age. To calculate these EWs, we have exploited the resolution of X-Shooter by applying the following procedure. We constructed 100 spectra whose ﬂux at each wavelength was randomly generated with a Gaussian law with a mean as the ini- tial data value and a standard deviation as the data error. For each spectrum, we subtracted a Lorentzian law ﬁtted on the line to the spectrum, which was then degraded to a resolution Rλ = 250 to accurately estimate the pseudo-continuum at the position of the line. The EW was calculated with the following: EW = n X i=1 1 −f(λi) fc(λi) ∆λi, (B.1) (B.1) (B.1) with f(λi) and fc(λi) being the ﬂux of the line and the pseudo-continuum, respectively, for each wavelength λi. Fur- thermore, ∆λi is the wavelength step. We estimated the ﬁnal EW from the mean of the series of these 100 EWs and the ﬁnal error from its standard deviation. The Figure B.1 compares the EW of each K I line calculated for VHS1256 b and the EWs calculated by Martin et al. (2017) who used the method from Allers & Liu (2013), who estimated the pseudo-continuum with a linear ﬁt to the ﬂux in two contin- uum windows deﬁned at each side of the line (more appropriate L9, page 7 of 8 L9, page 7 of 8 A&A 670, L9 (2023) a Low resolution version of the ATMO model. b With continuum subtracted Appendix D: Estimates of atmospheric and dynamic parameters chosen to leave it unconstrained by deﬁning its ﬁnal value based on the extreme estimates from all of the ﬁts. – For most of the ﬁts, the C/O converges to the upper edge of the grid. We deﬁned its adopted values as the most conserva- tive lower limit. As illustrated in the Table D.1, the estimate of each parameter is strongly dependent on the wavelength range used for the ﬁt. To take these variations into account and propose a conserva- tive estimate of the atmospheric parameters of VHS1256 b, we followed the method described in Petrus et al. (2020): As illustrated in the Table D.1, the estimate of each parameter is strongly dependent on the wavelength range used for the ﬁt. – For most of the ﬁts, the γ converges to the lower edge of the grid. We deﬁned its adopted values as the most conservative upper limit. To take these variations into account and propose a conserva- tive estimate of the atmospheric parameters of VHS1256 b, we followed the method described in Petrus et al. (2020): – Because the ﬁt between 0.670 and 1.100 µm does not seem to reproduce the data, it has been excluded to estimate the radius. Its adopted value is based on the extreme estimates from all of the other ﬁts. – The ﬁnal Teﬀis mainly constrained by the low-resolution spectral information. It is deﬁned by the extreme estimates from the ﬁts that used the JHK, J, H, and K bands. – The ﬁnal log(g) is constrained by the low-resolution spectral information, but also by the molecular and atomic absorp- tion. It is deﬁned by the extreme estimates from the ﬁts that used the [1.225-1.275] µm wavelength range, as well as the JHK, J, H, and K bands. – The adopted value of the radial velocity is deﬁned as the extreme estimates from all of the ﬁts. – For the same reason as for the radius, we excluded the ﬁt on the RI band to estimate the bolometric luminosity. Its adopted value is deﬁned as the extreme estimates from all of the other ﬁts. – Two modes seem to be observed for the metallicity, one solar (∼0.0) and one super-solar (∼0.5) mode. That is why we have – Two modes seem to be observed for the metallicity, one solar (∼0.0) and one super-solar (∼0.5) mode. That is why we have Table D.1. Appendix D: Estimates of atmospheric and dynamic parameters Atmospheric parameters of VHS 1256-1257 b inferred from diﬀerent wavelength ranges, using ForMoSA and the ATMO model grid. We use NC for "no constraint". The error bars given here are purely statistical and derived from the propagation of the small error bars of our data through the Bayesian inversion. The method used to estimate the ﬁnal values is described in Appendix D. Table D.1. Atmospheric parameters of VHS 1256-1257 b inferred from diﬀerent wavelength ranges, using ForMoSA and the ATMO model grid. We use NC for "no constraint". The error bars given here are purely statistical and derived from the propagation of the small error bars of our data through the Bayesian inversion. The method used to estimate the ﬁnal values is described in Appendix D. λmin −λmax Teﬀ log(g) [M/H] C/O γ R RV v sin(i) log(L/L⊙) (µm) (K) (dex) (RJup) (km s−1) (km s−1) (dex) 1.100–2.480 (JHK) 1326+2 −2 3.75+0.02 −0.02 0.45+0.03 −0.02 0.63+0.01 −0.01 1.013+0.001 −0.001 0.91+0.01 −0.01 4.47+0.39 −0.38 — −4.63+0.01 −0.01 0.670–1.100 (RI) 1361+4 −3 5.49+0.01 −0.01 0.42+0.01 −0.02 > 0.69 < 1.02 0.50+0.01 −0.01 6.04+1.36 −0.59 — −5.10+0.01 −0.01 1.100–1.356 (J) 1414+6 −21 3.90+0.09 −0.08 0.44+0.07 −0.07 > 0.68 1.02+0.01 −0.01 0.72+0.03 −0.01 3.32+0.75 −0.82 — −4.73+0.01 −0.01 1.410–1.810 (H) 1390+6 −7 3.50+0.02 −0.01 -0.05+0.04 −0.03 > 0.66 1.04+0.01 −0.01 0.82+0.01 −0.01 3.80+0.43 −0.61 — −4.64+0.01 −0.01 1.952–2.478 (K) 1417+17 −17 4.02+0.05 −0.05 0.08+0.05 −0.05 > 0.68 1.02+0.01 −0.01 0.81+0.01 −0.01 1.60+1.13 −0.92 — −4.62+0.01 −0.01 2.90–4.14 (LP)a 1145+13 −10 < 2.69 0.47+0.08 −0.09 0.39+0.06 −0.05 1.03+0.01 −0.01 1.25+0.02 −0.02 5.62+3.15 −3.25 — −4.61+0.01 −0.01 1.12–1.65 (HST min) 1309+3 −3 3.34+0.10 −0.11 NC NC < 1.02 — — — — 1.12–1.65 (HST max) 1336+5 −3 3.40+0.08 −0.09 NC NC < 1.02 — — — — 1.12–1.65 (X-Shooter) 1331+2 −2 3.70+0.03 −0.03 NC NC < 1.02 — — — — 2.280–2.410 (12CO) 1479+72 −63 4.21+0.18 −0.18 0.02+0.11 −0.10 > 0.63 < 1.02 — 4.04+1.27 −1.23 — — 1.225–1.275 (K I)b 1573+36 −46 4.25+0.20 −0.21 > 0.32 0.43+0.11 −0.09 < 1.015 — −1.32+4.33 −3.26 < 37 — Adopted values 1380±54 3.97±0.48 0.21±0.29 > 0.63 < 1.02 0.99±0.28 1.41±5.99 < 37 −4.67±0.07 a Low resolution version of the ATMO model. b With continuum subtracted L9, page 8 of 8
https://openalex.org/W2945177466	https://munin.uit.no/bitstream/10037/17415/3/article.pdf	English	null	Co-occurring change in children's conduct problems and maternal depression: Latent class individual participant data meta-analysis of the Incredible Years parenting program	Development and psychopathology	2,019	cc-by	12,028	Abstract Children vary in the extent to which they benefit from parenting programs for conduct problems. How does parental mental health change if children benefit less or more? We assessed whether changes in conduct problems and maternal depressive symptoms co-occur following participation in the Incredible Years parenting program. We integrated individual participant data from 10 randomized trials (N = 1280; children aged 2–10 years) and distinguished latent classes based on families’ baseline and post-test conduct problems and maternal depres- sive symptoms, using repeated measures latent class analysis (RMLCA) and latent transition analysis (LTA). Classes differed mainly in severity of conduct problems and depression (RMLCA; 4 classes). Conduct problems reduced in all classes. Depressive symptoms did not change in most classes, except in a class of families where conduct problems and depression were particularly severe. Incredible Years led to a greater likelihood of families with particularly severe conduct problems and depression moving to a class with mild problems (LTA; 3 classes). Our findings suggest that for the majority of families, children’s conduct problems reduce, but maternal depressive symp- toms do not, suggesting relative independence, with the exception of families with severe depression and severe conduct problems where changes for the better do co-occur. ords: conduct problems, individual participant data meta-analysis, maternal depression, parenting program (Received 8 March 2019; revised 9 May 2019; accepted 11 May 2019) (Received 8 March 2019; revised 9 May 2019; accepted 11 May 2019) (Received 8 March 2019; revised 9 May 2019; accepted 11 May 2019) Our specific objectives were to (a) ascertain any co-occurrence of change in conduct problems and maternal depression and (b) assess the influence of participation in a parenting program on this co-occurrence. Specifically, we sought to identify the exis- tence of any subgroup of families where participation in a parent- ing program results in co-occurring improvement in both children’s conduct problems and maternal depressive symptoms and any subgroup of families where a lack of improvement in children’s conduct problems co-occurs with increased maternal depressive symptoms. We use the term maternal depression because most parents in the pooled sample were mothers (98%). Children’s conduct problems and parental depressive symptoms often coexist whilst also augmenting each other (Goodman et al., 2011; Silberg, Maes, & Eaves, 2010). When parents partic- ipate in a parenting program to reduce their child’s conduct prob- lems, changes in conduct problems and in parental depressive symptoms may be related. Author for Correspondence: Patty Leijten, Research Institute for Child Development and Education & Research Priority Area Yield, University of Amsterdam, Nieuwe Achtergracht 127, 1018 WS Amsterdam, The Netherlands. Email: p.leijten@uva.nl. Cite this article: Leijten P et al (2019). Co-occurring change in children’s conduct problems and maternal depression: Latent class individual participant data meta-analysis of the Incredible Years parenting program. Development and Psychopathology 31, 1851–1862. https://doi.org/10.1017/S0954579419001068 Development and Psychopathology (2019), 31, 1851–1862 Development and Psychopathology (2019), 31, 1851–1862 Development and Psychopathology (2019), 31, 1851–1862 doi:10.1017/S0954579419001068 Co-occurring change in children’s conduct problems and maternal depression: Latent class individual participant data meta-analysis of the Incredible Years parenting program Co-occurring change in children’s conduct problems and maternal depression: Latent class individual participant data meta-analysis of the Incredible Years parenting program Patty Leijten1, Frances Gardner2, G.J. Melendez-Torres3, Joyce Weeland1, Judy Hutchings4, Sabine Landau5, Sinéad McGilloway6, Geertjan Overbeek1, Jolien van Aar1, Ankie Menting7, Bram Orobio de Castro7, Vashti Berry8, Maria Filomena Gaspar9, Ulf Axberg10, Willy-Tore Mørch11 and Stephen Scott5 Patty Leijten1, Frances Gardner2, G.J. Melendez-Torres3, Joyce Weeland1, Judy Hutchings4, Sabine Landau5, Sinéad McGilloway6, Geertjan Overbeek1, Jolien van Aar1, Ankie Menting7, Bram Orobio de Castro7, Vashti Berry8, Maria Filomena Gaspar9, Ulf Axberg10, Willy-Tore Mørch11 and Stephen Scott5 1Research Institute for Child Development and Education, University of Amsterdam, Amsterdam, Netherlands; 2Centre for Evidence-Based Intervention, Department of Social Policy and Intervention, University of Oxford, Oxford, United Kingdom; 3DECIPHer Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement, Cardiff University, Cardiff, United Kingdom; 4School of Psychology, Bangor University, Bangor, United Kingdom; 5Institute of Psychiatry, Psychology & Neuroscience, King’s College London, London, United Kingdom; 6Centre for Mental Health and Community Research, Maynooth University, Maynooth, Ireland; 7Department of Developmental Psychology, Utrecht University, Utrecht, Netherlands; 8College of Medicine and Health, University of Exeter, Exeter, United Kingdom; 9Centre for Social Studies, University of Coimbra, Coimbra, Portugal; 10Department of Psychology, University of Gothenburg, Gothenburg, Sweden and 11Regional Center for Children and Adolescents, University of Tromsø, Tromsø, Norway is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/ distribution, and reproduction in any medium, provided the original work is properly cited. Parenting Programs for Children’s Conduct Problems h d d Parenting programs are the primary strategy used to reduce child- ren’s conduct problems. Most programs are based on relational and social learning perspectives. Relational perspectives posit that conduct problems originate from a lack of warmth and nur- turing from parents, distorting the affectionate bond between par- ents and children’s cognitive models on social relationships (e.g., Cassidy & Shaver, 2008; Maccoby & Martin, 1983). Social learn- ing perspectives posit that conduct problems are shaped by coer- cive parent–child interactions in which parents and children unwittingly reinforce aversive behavior in each other, leading to interaction patterns that become increasingly difficult to manage (e.g., Patterson, Reid, & Dishion,1992). Parenting programs based on these perspectives support parents to improve their relationship with their child, for example through positive involvement and child-led play, and to reinforce positive child behavior and avoid reinforcement of conduct problems, for example through proactive and nonharsh discipline techniques (Kaehler, Jacobs, & Jones, 2016). There is robust evidence that these programs reduce child- ren’s conduct problems at various stages of its development (Leijten et al., 2019; Weisz & Kazdin, 2017). Conduct problems, in turn, can contribute to maternal depres- sive symptoms, by placing a considerable burden on parents and challenging parental self-efficacy (Meltzer, Ford, Goodman, & Vostanis, 2011). More specifically, mothers may feel like they have little control over their child’s behavior (Harrison & Sofronoff, 2002) or that they are to blame for their child’s disrup- tive behavior (Moses, 2010). Mothers may develop feelings of helplessness, thereby losing the confidence, and potentially the motivation, to redirect or correct the child’s behavior—experi- ences that may contribute to depressive symptoms (Jones & Prinz, 2005; Weaver, Shaw, Dishion, & Wilson, 2008). Little is known about possible co-occurrences of intervention benefits, or a lack thereof. For example, do positive changes in parental mental health depend on how much the program improves children’s conduct problems, and vice versa? In the pre- sent study, we sought to examine whether maternal depressive symptoms change together with change in children’s conduct problems. While the main effects of parenting programs on child- ren’s conduct problems and parental depressive symptoms are well studied (e.g., Bennett, Barlow, Huband, Smailagic, & Roloff, 2013; Leijten et al., 2018), the extent to which these effects depend on each other is largely unknown. Effective parenting programs may therefore benefit both child- ren’s conduct problems and maternal depressive symptoms. Abstract factors) may help account for heterogeneity in intervention out- comes (e.g., Connell, Stormshak, Dishion, Fosco, & Van Ryzin, 2018; Fosco, Van Ryzin, Stormshak, & Dishion, 2014; Gardner et al., 2009; Hyde, Waller, & Burt, 2014; Lochman, Dishion, Boxmeyer, Powell, & Qu, 2017; Pelham, Dishion, Tein, Shaw, & Wilson 2017; Shaw et al., 2016; Smith et al., 2018). In addition, Tom’s work on peer deviancy training was ground-breaking both in showing how plausible, theory-based interventions carry the potential for harm and in interrogating the mechanisms and conditions under which youth did not benefit (e.g., Dishion, McCord, & Poulin, 1999). While contributing much to a lively literature on heterogeneity in intervention benefits, he also recognized that the field has not been very good at identifying the families that benefit less or more. He advocated for methodolog- ically innovative ways to explore differential intervention benefits and identify subgroups of families in which interventions may have harmful (collateral) effects. In this paper, we use such novel approaches, repeated measures latent class analysis and latent tran- sition analysis, to explore co-occurring change in children’s and mothers’ mental health in response to participation in a parenting program designed to address children’s conduct problems. mothers more often have a genetic vulnerability to developmental mental health problems themselves. Second, ‘third variable’ family characteristics such as socioeconomic disadvantage and its accompanying factors, such as living in a more deprived and vio- lent neighborhood, and increased risks for physical illnesses, con- tribute to both maternal depressive symptoms and children’s conduct problems, explaining why the two often co-occur (Galea et al., 2007; Shaw & Shelleby, 2014). Third, maternal depressive symptoms may be a barrier for parents to engage in, or successfully complete, programs to support optimal parenting practices, thereby limiting the opportunity for children to benefit from these programs (Rostad, Moreland, Valle, & Chaffin, 2018). p g More importantly for purposes of this study, maternal depres- sive symptoms and children’s conduct problems also influence each other more directly. The debilitating nature of depression makes the, already difficult, job of parenting more challenging, negatively affecting the quality of the parenting that children receive (Lovejoy, Graczyk, O’Hare, & Neuman, 2000). Parenting Programs for Children’s Conduct Problems h d d For example, familiarization with techniques that help to redirect the child’s behavior may increase mothers’ perceived control over that behavior and reduce feelings of helplessness and inade- quacy (Bennett et al., 2013). In turn, less disruptive and more sociable child behavior provides mothers with experiences of con- nectedness and warmth (Combs-Ronto, Olson, Lunkenheimer, & Sameroff, 2009; Wiggins, Sofronoff, & Sanders, 2009), which could diminish depressive symptoms. Lastly, the increased activ- ity, often rewarding activity, that comes with participating in a parenting program (e.g., attending weekly sessions and working successfully on practice assignments) may for some mothers Abstract Already in infancy, mothers who suffer from depression more often engage with their children in ways that are less sensitively attuned (e.g., Murray, Fiori-Cowley, Hooper, & Cooper, 1996), compro- mising mother–child bonding with long lasting negative effects on children’s mental health (Moehler, Brunner, Wiebel, Reck, & Resch, 2006). In addition, depressive symptoms may distort moth- ers’ perceptions of their children’s behavior, making them vulner- able to negative thoughts, such as thinking that their children do not appreciate their efforts or do not wish to be close to them (Callender, Olson, Choe, & Sameroff, 2012). Furthermore, depres- sive symptoms make it more challenging for mothers to be sensi- tive to their children’s needs and to provide warmth and consistent, nonharsh discipline (Dette-Hagenmeyer & Reichle, 2014; Lovejoy et al., 2000), something children need in order to prevent their conduct problems from intensifying (Patterson et al., 1992). Evidence that treatment for maternal depression has positive effects on children’s conduct problems (e.g., Cuijpers, Weitz, Karyotaki, Garber, & Andersson, 2015), further supports the causal link between maternal depressive symptoms and children’s conduct problems. Abstract In other words, parental depressive symptoms may increase or decrease depending on improvements, or lack thereof, in children’s conduct problems. The purpose of this study, which is based on a large pooled sample of 10 random- ized trials, was to identify whether and how changes in children’s conduct problems and maternal depressive symptoms co-occur. We gratefully conducted this study to honor the legacy of Dr. Thomas Dishion. Tom never took it for granted that families and children would benefit from well-intended intervention programs. He was highly productive in testing the sources of differential responses of families to parenting programs, and he encouraged colleagues and mentees to explore how characteristics at multiple levels (e.g., neighborhood, cultural, family, child, and genetic © Cambridge University Press 2019. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licenc 4.0/), which permits unrestricted re-use, distribution, and reproduction in any medium, provided the original work is properly cited. 1852 Patty Leijten et al. factors) may help account for heterogeneity in intervention out- comes (e.g., Connell, Stormshak, Dishion, Fosco, & Van Ryzin, 2018; Fosco, Van Ryzin, Stormshak, & Dishion, 2014; Gardner et al., 2009; Hyde, Waller, & Burt, 2014; Lochman, Dishion, Boxmeyer, Powell, & Qu, 2017; Pelham, Dishion, Tein, Shaw, & Wilson 2017; Shaw et al., 2016; Smith et al., 2018). In addition, Tom’s work on peer deviancy training was ground-breaking both in showing how plausible, theory-based interventions carry the potential for harm and in interrogating the mechanisms and conditions under which youth did not benefit (e.g., Dishion, McCord, & Poulin, 1999). While contributing much to a lively literature on heterogeneity in intervention benefits, he also recognized that the field has not been very good at identifying the families that benefit less or more. He advocated for methodolog- ically innovative ways to explore differential intervention benefits and identify subgroups of families in which interventions may have harmful (collateral) effects. In this paper, we use such novel approaches, repeated measures latent class analysis and latent tran- sition analysis, to explore co-occurring change in children’s and mothers’ mental health in response to participation in a parenting program designed to address children’s conduct problems. Symptoms This increases the likelihood that our findings can be generalized to other established parenting programs (e.g., Parent Management Training—Oregon, Forgatch & Patterson, 2010), provided they use a similar delivery approach (e.g., Parents Plus Program; Sharry, Hampson, & Fanning, 2003). Findings may in part be different for programs that adopt another approach, such as those delivered individually, in some cases at home and/ or using video-feedback (e.g., Parent–child Interaction Therapy; Zisser & Eyberg, 2010; Family Check-Up; Dishion & Kavanagh, 2003). In addition, the Incredible Years program has shown robust effects in independent evaluations across a number of countries (Leijten, Melendez-Torres, Knerr, & Gardner, 2016), and it has been widely disseminated in several European countries. This fur- ther increases the relevance of our findings. Importantly, all of the European trials of the Incredible Years program were conducted independently of the US-based program developer (Webster- Stratton), and those European countries in which Incredible Years programs have been delivered tend to have relatively similar health and social care systems when compared with, for example, the USA. This allows for sufficient homogeneity in the usual services that children received across trials, and thus supports the comparability of the synthesized data. Lastly, the existence of active research net- works for Incredible Years in Europe was helpful in obtaining data from a near-complete set of trials. Co-occurring Change. We chose to study co-occurring change because we hypothesized that changes in children’s conduct prob- lems and maternal depressive symptoms in part depend on each other. The effects of parenting programs on children’s conduct problems and maternal depressive symptoms are typically ana- lyzed either in separate models or in a mediation model, with reduced maternal depressive symptoms as a putative mediator of program effects on children’s conduct problems (e.g., Hutchings et al., 2012; Shaw et al., 2009). Such a mediation model assumes that changes in maternal depression precede changes in children’s conduct problems. Yet, arguably new skills are learned and practiced as early as after the very first session of a parenting program, and hence, changes in mothers’ attitudes and behaviors may affect children’s behavior and maternal mental health from very early on, and almost simultaneously. This is con- sistent with clinical impressions from parenting groups, where parents often report feeling rewarded by success in early home practice assignments, for example seeing their child’s pleasure during responsive play. Symptoms Several mechanisms explain the co-occurrence of children’s con- duct problems and maternal depressive symptoms. First, maternal depression puts children at risk for conduct problems through genetic transmission (Singh et al., 2011); children of depressed 1853 Development and Psychopathology function as some form of behavioral activation, a key component of effective treatment for depression (Cuijpers, Van Straten, & Warmerdam, 2007). Indeed, there is some evidence, albeit less consistent than for children’s conduct problems, that parenting programs reduce maternal depressive symptoms (Bennett et al., 2013; but see Leijten et al., 2018, for conflicting evidence). Importantly, rather than being a secondary outcome of parenting programs, reduced levels of maternal depression may be one pos- sible mechanism accounting for the positive effects of parenting programs on children’s conduct problems (Hutchings, Bywater, Williams, Lane, & Whitaker, 2012; Shaw, Connell, Dishion, Wilson, & Gardner, 2009). function as some form of behavioral activation, a key component of effective treatment for depression (Cuijpers, Van Straten, & Warmerdam, 2007). Indeed, there is some evidence, albeit less consistent than for children’s conduct problems, that parenting programs reduce maternal depressive symptoms (Bennett et al., 2013; but see Leijten et al., 2018, for conflicting evidence). Importantly, rather than being a secondary outcome of parenting programs, reduced levels of maternal depression may be one pos- sible mechanism accounting for the positive effects of parenting programs on children’s conduct problems (Hutchings, Bywater, Williams, Lane, & Whitaker, 2012; Shaw, Connell, Dishion, Wilson, & Gardner, 2009). In addition to increasing statistical power, IPD meta-analysis maximizes transparency and minimizes bias, particularly selective outcome reporting and publication bias, by including all available data. Because these bias-reducing benefits of IPD only apply if data from a large proportion of eligible trials can be obtained and included, we carefully predefined the trials we sought to include; we focused only on trials of the Incredible Years parent- ing program in Europe. A key advantage of including trials of Incredible Years, for the current study, is the similarity of the Incredible Years program con- tent to that of many other established parenting programs (see Kaehler et al., 2016, for an overview), even though its strong empha- sis on a collaborative group approach is somewhat different from some other programs. Symptoms These change processes cannot be cap- tured by traditional mediation models that assume a period of change in parents first, before change in children starts (Weeland et al., 2018a, 2018b). Instead, we hypothesized the exis- tence of subgroups of families with different patterns of co-occurring change, or lack thereof, in children’s conduct prob- lems and maternal depressive symptoms. More specifically, we hypothesized that participation in a parenting program for child- ren’s conduct problems predicts these patterns such that the pro- gram leads to co-occurring improvement in maternal depressive symptoms and children’s conduct problems for some families and to increased maternal depressive symptoms in families where children’s conduct problems fail to improve. Design and procedure We obtained individual participant data from an almost complete set (88%) of 15 trials on the effects of the Incredible Years parenting program for children ages 2–10 years across six countries in Europe. The study protocol for integration of the first 14 trials is published on www.spi.ox.ac.uk/parentingIPD, and the details of how the data were harmonized are published in Leijten and colleagues (2018). Data from the 15th trial (Weeland, Chhangur, et al., 2017) were added later, following the same data harmonization procedures because the study finished after the original data integration project. Procedures were approved by the Departmental Research Ethics Committee of the University of Oxford. Individual Participant Data (IPD) Meta-Analysis The study of individual differences in program benefits, especially in multivariate models that test co-occurring change in two out- comes, requires a sufficiently large sample size to identify co-occurring change patterns. We achieved this by using IPD meta-analysis. This approach synthesizes individual family level data from multiple trials, allowing us to use both variance between and within trials to estimate differential program benefits, increas- ing statistical power and generalizability (Curran & Hussong, 2009). In our case, this meant we were able to use the individual family level data from a large pooled sample to identify clusters of families based on co-occurring change in conduct problems and maternal depression. In addition, because all trials had randomly allocated families to conditions, any differences in clusters between conditions can be interpreted as causal effects of the par- enting program. y Ten of the 15 trials were eligible for inclusion in the present study. Of the five trials that were not eligible, two trials (Morpeth et al., 2017; Weeland, Chhangur, et al., 2017) contained no data on maternal depression; two trials (Leijten et al., 2017; Menting et al., 2014) contained data on maternal depressive symptoms at baseline only; and one trial contained no data on children’s conduct problems (because of the children’s young age—toddlerhood; Hutchings, Griffith, Bywater, & Williams, 2017). Details of the included trials are in Table 1. 1854 Patty Leijten et al. Table 1. Characteristics of the included trials in the individual participant data meta-analysis Table 1. Maternal depression The most frequently used measure for maternal depressive symp- toms was the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961); one trial used the Brief Symptom Inventory—depression subscale (BSI; Derogatis & Spencer, 1982); and one trial used the General Health Questionnaire (GHQ; Williams & Goldberg, 1988). BSI and GHQ scores were converted to BDI scores using norm deviation scores. Internal consistency ranged α = .87 to α = .93. Individual Participant Data (IPD) Meta-Analysis Characteristics of the included trials in the individual participant data meta-analysis Lead author (year) Country n Child age (M) % low income % ethnic minority Baseline conduct problemsa (M) Baseline maternal depressive symptomsb (M) Axberg (2012) Sweden 62 3–8 (5.97) 41 0 112–201 (155.01) 2–37 (14.58) Gardner (2006) England 76 2–9 (5.93) 64 2 55–241 (161.47) 0–47 (17.14) Hutchings (2007) Wales 153 3–4 (3.84) 79 1 75–218 (145.06) 0–50 (16.44) Larsson (2009) Norway 75 3–8 (6.58) 25 1 111–213 (158.04) 0–34 (7.82) McGilloway (2012) Ireland 149 2–7 (4.84) 47 6 82–235 (158.54) 0–57 (16.03) Scott (2001) England 141 2–0 (5.67) 58 15 81–243 (162.52) 0–39 (14.70) Scott (2010a) England 174 4–6 (5.50) 44 75 42–192 (104.78) 0–28 (4.75) Scott (2010b) England 112 4–6 (5.21) 44 40 51–252 (128.42) 0–38 (8.23) Scott (2013) England 214 3–7 (6.07) 80 19 40–212 (136.13) 0–20 (4.32) Seabra-Santos (2016) Portugal 124 3–6 (4.66) 0 0 64–221 (127.61) 0–32 (10.08) aPossible range of conduct problems (Eyberg Child Behavior Inventory) 36–252. bPossible range of depressive symptoms (Beck Depression Inventory) 0–36. aPossible range of conduct problems (Eyberg Child Behavior Inventory) 36–252. bPossible range of depressive symptoms (Beck Depression Inventory) 0–36. trials used the Parental Account of Children’s Symptoms (PACS; Taylor, Schachar, Thorley, & Wieselberg, 1986). We con- verted PACS scores into ECBI scores using norm deviation scores (see Leijten et al., 2018, for this procedure). ECBI and PACS scores correlated r = .71. Internal consistency ranged α = .79 to α = .95. Participants The total integrated sample from 10 trials included 1280 families (753 intervention; 527 control condition—some trials used a 2:1 allocation ratio). Children ranged in age from 2 to 10 years (M = 5.33; SD = 1.30; 36% girls). The socioeconomic status of par- ticipant families was diverse (52% low income; 61% low educa- tional level; 36% single parent; 29% no employed parent in the household), and 20% reported to be from an ethnic minority. We used data from one parent (98% mothers) because most trials included data from only one parent. In terms of categorical defi- nitions, 43% of the children showed severe conduct problems (see Burns & Patterson, 2001, for norm scores), and 19% of the moth- ers showed moderate to severe depression (see Veerman, Dowrick, Ayuso-Mateos, Dunn, & Barendregt, 2009, for norm scores). Analytic Strategy Families in the intervention condition were offered the Incredible Years parenting program (Webster-Stratton & Reid, 2010). The number of sessions ranged 12 to 19 across trials, depending on the timing and context of the trials; older versions and prevention (versus treatment) versions of Incredible Years include fewer ses- sions. Across trials, parents attended on average 68% of the ses- sions (range Mpercent sessions attended = 55–91%). With regard to the professionals delivering the sessions, 63% were trained as clin- ical psychologists and 31% were already certified Incredible Years trainers before the start of the trial. To monitor program fidelity, 100% of the staff completed checklists of completed and omitted session activities after each session, and 80% of the staff received weekly or fortnightly supervision. The control condition was either a waitlist (7 trials), minimal intervention (e.g., a phone helpline; 2 trials), or care as usual (1 trial). There are several approaches to identify patterns of co-occurring change in two outcomes (here, conduct problems and maternal depression). One approach is to model classes with similar scores over time, using repeated measures latent class analysis (RMLCA); another is to model families’ transitions between classes that exist at either time point (i.e., baseline and posttest), using latent tran- sition analysis (LTA). We chose to conduct both approaches, first, to make sure our findings do not hinge on the approach chosen, and second, because each approach explores slightly different aspects of the research question. Specifically, RMLCA answers the question, first, how changes in children’s conduct problems and maternal depression co-occur over time, and second, whether the Incredible Years program increases the likelihood of certain patterns of co-occurring change. LTA answers the question whether the Incredible Years program, relative to control, causes parents to move classes between baseline and post-test times, for example to classes defined by less severe problems, or to clas- ses defined by equally severe conduct problems and more severe maternal depression (if such a class exists). Conduct problems The most frequently used measure for parent-reported children’s conduct problems in children was the Eyberg Child Behavior Inventory—Intensity Scale (ECBI; Eyberg & Ross, 1978). Two RMLCA treats all manifest indicators, some of which relate to repeated measurement occasions of the same variable, as describing Development and Psychopathology 1855 and posttest, with the same within-class mean level of conduct problems and depression at both time points. However, examina- tion of two-class and three-class solutions estimated at baseline and posttest indicated that while maternal depression appeared to be commensurate between classes, conduct problems decreased across all classes. We therefore did not require conduct problem means to be equal between baseline and posttest between analogous classes. This means that the classes do not have the same meaning at both time points. In our case, we were primarily interested in between-condition differences in to what classes family moved. We examined how intervention status predicted movement between classes by testing a one-step interaction model between class membership at baseline and intervention status in predicting class membership at posttest. We directly estimated probabilities of movement between classes for intervention and control families using a probability parametrization. underlying response patterns in the data. That is to say, respon- dents can be categorized as belonging to a class characterized by both baseline and post-test responses to a set of indicators (here, conduct problems and maternal depression). It is then possible to test whether a covariate (here, Incredible Years vs. control) pre- dicts ‘belonging’ to a class. In our case, this allowed us to distin- guish different classes based on co-occurring change in conduct problems and maternal depressive symptoms, thereby potentially identifying, first, how changes in maternal depressive symptoms depend on changes in children’s conduct problems, and second, whether participation in the Incredible Years program makes some patterns of co-occurring change more likely than others. p g g y In contrast, LTA describes classes separately for each measure- ment wave (e.g., a class of families with clinical levels of both con- duct problems and maternal depression and a class of families with mild levels of both), and it seeks to understand how and if families move between latent classes (e.g., from a class of clinical problems to a class of mild problems). In an LTA model, classes are estimated based on baseline conduct problems and depression scores, and again based on post-test conduct problems and depression scores. Descriptive Results p We chose to combine RMLCA and LTA because they are com- plementary in their approaches to examining patterns of change through belonging to a certain class (RMLCA) and patterns of change as indicated by moving between classes (LTA). For both RMLCA and LTA, the number of classes in each model is deter- mined by a combination of significance testing, fit indices (e.g., the Bayesian information criterion [BIC] and scaled relative entropy), and theoretically informed judgment. We predomi- nantly focused on the degree to which improvements in model fit slowed down when we tested models with more classes. Across trials, baseline scores for children’s conduct problems and maternal depressive symptoms encompassed almost the full pos- sible range, from no or minimal symptoms to severe clinical levels of conduct problems and depression (Table 1). This reflects the fact that the 10 trials targeted a wide range of families, from pre- vention samples at schools in inner-city London and Sure Start samples in Wales, to treatment samples in outpatient psychiatric clinics in Norway and Sweden. This variation in conduct prob- lems and maternal depressive symptoms was both a key benefit and central to our analyses—the ability to identify classes based on conduct problems and depressive symptoms depends on var- iation in conduct problems and depression. For the RMLCA, we allowed the residual variances for each indicator to vary across classes, and tested models with two to four classes. We chose a four-class model based on model fit and interpretability, and we related intervention status to class member- ship in a one-step model, converting multinomial regression coef- ficients into probabilities of membership by intervention status. Baseline levels of children’s conduct problems and maternal depression correlated such that, as expected, mothers of children with more severe conduct problems experienced more depressive symptoms (r = .32, p <.001). An overview of the main effects of Incredible Years in the original full pooled sample, including a main effect on reduced conduct problems (β = −.35, p <.001), and no main effect on maternal depressive symptoms (β = −.08, p = .095), is presented in Leijten and colleagues (2018). p p y For the LTA, we tested between two and four classes at baseline, and then compared baseline classes with classes estimated on post- test data. Conduct problems The role of the covariate (here, Incredible Years vs. control) is to test if patterns of movement between latent classes differ over the levels of the covariate; that is, are families in the intervention condition more likely to move from one class to another as compared with families in the control condition? In our case, we were specifically interested to ascertain: (a) if families in the intervention condition were more likely to move to a class with less severe conduct problems and maternal depression, indi- cating co-occurring improvement (if such a class exists); (b) if families in the intervention condition who moved to a class with similar, or more severe conduct problems, did so more often if this class was characterized by more severe maternal depression, indicating that if Incredible Years fails to improve children’s conduct problems, this co-occurs with a worsening of maternal depression (and vice versa). g p y p We accounted for the fact that data came from 10 different tri- als by using clustered standard errors, as the number of parame- ters to be estimated in each model was greater than the number of trials (or level 2 clusters), a situation to be avoided in multilevel structural equation modeling. Use of fixed effects (i.e., dummy variables by trial) would have ‘flattened out’ substantively infor- mative heterogeneity in risk distributions between classes. All analyses assumed, as is standard in most latent variable models estimated using full information maximum likelihood, that data were missing at random; that is, that missingness was a function of other variables in the model. This is a weaker assumption than ‘missing completely at random’ but a stronger assumption that ‘not missing at random,’ which would require a number of addi- tional modeling assumptions to implement. All analyses were conducted in Mplus version 8.2 (Muthén & Muthén, 2018). Descriptive Results We fixed indicator variances to be equal across classes and across waves because of the classes estimated and to preserve inter- pretability between classes and between waves. A four-class solu- tion did not converge satisfactorily, so we retained two-class and three-class solutions. A common assumption in LTA is that the measurement of classes is identical at baseline and posttest, even if a family’s class membership changes at baseline and posttest. In other words, the assumption is that, for example, a class of severe conduct problems and clinical depression exists at both baseline Identifying Class Membership Using Repeated Measures Latent Class Analysis (RMLCA) Because significance testing did not suggest meaningful differ- ences between classes in the RMLCA, we identified a four class 1856 Patty Leijten et al. Patty Leijten et al. Table 2. Co-Occurrence of children’s conduct problems and parental depression in families with varying levels of baseline problem severity, based on RMLCA Class 1 Class 2 Class 3 Class 4 Conduct problems Baseline M (S2) 116.77 (1328.08) 128.69 (908.33) 151.70 (1042.4) 162.00 (1272.29) Post-test M (S2) 99.5 (1052.6) 109.03 (649.29) 131.52 (1091.84) 142.67 (1670.00) Maternal depression Baseline M (S2) 0.77 (0.66) 5.42 (8.47) 12.64 (35.09) 26.60 (89.78) Post-test M (S2) 0.78 (0.64) 4.35 (7.21) 10.01 (32.23) 21.90 (137.21) Probability Incredible Years 14% 31% 36% 20% Control 19% 34% 31% 16% Note: Incredible Years significantly predicted class membership. Families who received Incredible Years were more likely to be in Class 3 or 4. RMLCA = repeated measures latent class analysis. al depression in families with varying levels of baseline problem severity, based on RMLCA Table 3. Model fit of the repeated measures latent class analysis (RMLCA) and the latent transition analysis (LTA) RMLCA LTA: Baseline LTA: Post-test # Classes BIC Entropy BIC Entropy BIC Entropy 2 37968 0.74 20741 0.80 18094 0.86 3 37472 0.76 20604 0.82 17948 0.87 4 37357 0.70 20567 0.81 17868 0.86 Table 3. Model fit of the repeated measures latent class analysis (RMLCA) and the latent transition analysis (LTA) model as most theoretically and substantively salient (Table 2). Classes differed mainly in terms of severity of conduct problems and maternal depressive symptoms. Conduct problems and maternal depressive symptoms were least severe in Class 1 and most severe in Class 4. The association between (baseline) con- duct problems and maternal depressive symptoms seemed rela- tively linear between Classes 1, 2, and 3, with mothers of children with more conduct problems experiencing more depres- sive symptoms, but families in Class 4 somewhat diverged from this pattern. While children’s conduct problems in Class 4 were only somewhat more severe than in Class 3 (MClass 4 = 162; MClass 3 = 153; a 0.25 SD difference), maternal depressive symp- toms were much more severe than in in Class 3 (MClass 3 = 13; MClass 4 = 27; a 1.35 SD difference). Identifying Change in Class Membership Using Latent Transition Analysis (LTA) LTA, where classes are identified separately at baseline and at post- test (as opposed to the RMLCA, where classes are based on baseline and post-test scores combined), sheds light on the likelihood that families change from one class to another, and whether these changes are predicted by intervention status. LTA identified three classes at baseline (Table 4) that were similar to Classes 2, 3, and 4 identified by the RMLCA. The largest group at baseline (66%) included families with moderate levels of conduct problems (M = 132; around 80th percentile; Burns & Patterson, 2001) and minimal depression (M = 5; 0–13 is considered minimal). A smaller group included families with severe levels of conduct problems and mild to moderate depression (M = 19; 14–19 is considered mild). Lastly, a small but meaningfully separate group (7%; N = 90) showed similar levels of conduct problems as families in Class 3, but they demonstrated clinical levels of depression (M = 35; > 29 is considered severe). Classes were largely similar at posttest, with slightly lower levels of conduct problems in all classes. Maternal depression scores were held constant for classes at baseline and classes at posttest due to a lack of change on this measure. Intervention status predicted class membership such that fam- ilies randomized to the Incredible Years parenting program, rela- tive to families in the control condition, were more likely to be in Class 3 or 4, than in Class 1 or 2. This suggests that a pattern of severe baseline problems and a reduction in conduct problems (Class 3), or severe baseline problems and a reduction in both conduct problems and maternal depression (Class 4) was more typical in families who had participated in Incredible Years. Identifying Class Membership Using Repeated Measures Latent Class Analysis (RMLCA) This suggests that among the children with severe conduct problems, two family types can be distinguished: those in which maternal depressive symp- toms are elevated, but mild, and those in which maternal depres- sive symptoms are moderate to clinical. Details on model fit and class selection are presented in Table 3. problems reduced to a similar extent in all classes, and depressive symptoms typically did not change, except for a reduction in Class 4 (i.e., in families with the most severe baseline problems). There was no evidence for a subgroup of families where conduct problems did not improve, or where they worsened. In line with this, there was no subgroup of families where maternal depressive symptoms intensified (in co-occurrence with a lack of improved conduct problems). Co-Occurring Change in Conduct Problems and Maternal Depression (RMLCA) There was little evidence of co-occurring change in conduct prob- lems and maternal depressive symptoms (Figure 1). Conduct p g Intervention status predicted the likelihood that families moved to another class. This was especially the case for families Development and Psychopathology 1857 Figure 1. Co-occurring change in four classes of families in conduct problems (left) and maternal depression (right). Intervention status (Incredible Years [IY] versus control) predicts class membership: Class 4 IY 20% Control 16%; Class 3 IY 36% Control 31%; Class 2 IY 31% Control 34%; Class 1 IY 14% Control 19%. Figure 1. Co-occurring change in four classes of families in conduct problems (left) and maternal depression (right). Intervention status (Incredible Years [IY] versus control) predicts class membership: Class 4 IY 20% Control 16%; Class 3 IY 36% Control 31%; Class 2 IY 31% Control 34%; Class 1 IY 14% Control 19%. Table 4. Baseline probability for each class, based on latent class analysis Class 1 Class 2 Class 3 Baseline probability 66% 26% 7% Conduct problemsa Baseline M 131.84 156.4 162.8 Post-test M 113.79 142.2 151.0 S2 1157.00 1157.00 1157.00 Maternal depressionb Baseline M (S2) 4.92 (20.26) 18.36 (20.26) 35.24 (20.26) aVariances (S2) in conduct problems scores were held constant. bMean maternal depression score was held constant at baseline and post-test, because there was no evidence of ‘secular’ decrease between assessment points. Table 4. Baseline probability for each class, based on latent class analysis Table 4. Baseline probability for each class, based on latent class analysis in the Incredible Years parenting program for their children’s conduct problems. We identified different response patterns to determine whether maternal depressive symptoms improve if children’s conduct problems improve and whether maternal depressive symptoms worsen if children’s conduct problems fail to improve. Our findings provide limited evidence for co-occurring change and showed that families with the most severe co-occurring conduct problems and maternal depression were most likely to benefit. in Class 3—those with the most severe conduct problems and maternal depression. Families in this group who were assigned to Incredible Years had a 32% likelihood to move one class down (to Class 2: severe conduct problems and mild depressive symptoms) and a 29% likelihood to move two classes down (to Class 1: mild conduct problems and minimal depressive symp- toms). Co-Occurring Change in Conduct Problems and Maternal Depression (RMLCA) Families from this group who were assigned to the control condition had a likelihood of 58% and 6% to move to Class 2 and Class 1, respectively. In other words, intervention and control families in Class 3 had a similar likelihood to move to another class (intervention 61%; control 64%), but intervention families had a larger chance to have a more substantial move, indicating greater improvement (i.e., from Class 3 to Class 1). y The latent longitudinal classes we identified were mostly defined by problem severity, not by change. Specifically, among families of children with subclinical to severe conduct problems, we identified two types: those in which maternal depression is elevated, but sub- clinical, and those in which maternal depression is subclinical to clinical. This could be interpreted such that some mothers are more negatively affected by their child’s conduct problems than others and/or that conduct problems in some children are at least partly driven by maternal depression, while conduct problems in other children have different origins. In line with previous find- ings indicating larger effects in treatment settings, compared with prevention settings (e.g., Leijten et al., 2019; Weisz & Kazdin, 2017), families with more severe problems at baseline were most likely to benefit from Incredible Years in terms of improvement in both conduct problems and maternal depressive symptoms. For most other families, conduct problems improved, but maternal depressive symptoms did not change. There was no evidence that maternal depressive symptoms intensified in families of children where conduct problems showed less improvement. This finding provides some evidence for co-occurring change of conduct problems and maternal depressive symptoms such that when Incredible Years reduced children’s conduct problems more strongly, it also reduced maternal depression more strongly. Families from Class 2 and 1, with milder problems than families from Class 3 and 4, were less likely to change to another class, regardless of intervention status (Figure 2). There was thus no evi- dence for co-occurring change in conduct problems and maternal depressive symptoms for families with milder problems, due to an overall lack of change in these families. Discussion We tested whether children’s conduct problems and maternal depressive symptoms change together when parents participate 1858 Patty Leijten et al. Figure 2. Results of the latent transition analysis: intervention and control families in Class 3 had a similar likelihood to move to another class, but intervention families had a larger chance for a move indicating greater improvement (i.e., from Class 3 to Class 1). Change is more similar in Class 2, and it is largely absent in Class 1. 1858 Patty Leijten et al Figure 2. Results of the latent transition analysis: intervention and control families in Class 3 had a similar likelihood to move to another class, but intervention families had a larger chance for a move indicating greater improvement (i.e., from Class 3 to Class 1). Change is more similar in Class 2, and it is largely absent in Class 1. behavior. These perceptions of improvement may prevent mater- nal depressive symptoms from worsening. Incredible Years led to co-occurring improvement in conduct problems and maternal depressive symptoms in a small subgroup of families with the most severe problems, at both the child and parent levels. It is well-known that children with more severe problems at baseline often benefit more in terms of reduced con- duct problems (e.g., Leijten et al., 2019; Menting, Orobio de Castro, & Matthys, 2013), although this effect is not always iden- tified in individual trials (e.g., Seabra-Santos et al., 2016; Weeland, Chhangur et al., 2017). Our findings suggest that these families may improve on multiple fronts at the same time—on both child and maternal mental health. This may be because these fam- ilies have the most room for improvement. In addition, especially in families most vulnerable for parent and child mental health problems, symptoms of children’s conduct problems and parental depression might form a feedback system in which these symp- toms reinforce each other—a feedback system that parenting pro- grams strive to break (Gonzalez & Jones, 2016). An alternative explanation for our findings is that even if maternal depressive symptoms are negatively affected by a lack of substantial improvement in children’s conduct problems, par- enting programs may provide other benefits for maternal mental health that counterbalance this. For example, most parents like the group format of Incredible Years, perceiving it as supportive, nonjudgmental, and encouraging (e.g., Furlong & McGilloway, 2012; Seabra-Santos, Gaspar, Azevedo, Homem, & Pimentel, 2011). Development and Psychopathology By using novel theoretical and methodo- logical approaches to continuously improve our understanding of families’ responses to interventions, we can keep honoring Tom’s commitment to providing our youth with the best possible inter- ventions for their mental health. p g p g We tested our research questions in two ways, using RMLCA and LTA, in a well-powered pooled sample of 1280 families. Neither approach identified families where limited improvement in children’s conduct problems co-occurred with worsened maternal depression. This strengthens our confidence that such a group indeed did not exist in the included samples. There may be individual families that show this pattern, but our analyses provide no evidence that this is a systematic, re-occurring pattern in a significant number of families. Because of the variety in set- tings and countries where Incredible Years was implemented (i.e., from families seeking help at outpatient psychiatric clinics to fam- ilies in socioeconomically deprived neighborhoods) and countries (i.e., six different countries), our findings do not pertain to a lim- ited number of specific, highly controlled settings, and they potentially generalize across settings and countries. Acknowledgments. We gratefully acknowledge the researchers, practition- ers, and families who participated in each of the original trials on the effects of the Incredible Years parenting program. Financial Support. This study was partly funded by the National Institute of Health Research (NIHR) Public Health Research Programme (PI Gardner). In addition, Vashti Berry’s time is supported by the NIHR Collaboration for Leadership in Applied Health Research and Care South West Peninsula (NIHR CLAHRC South West Peninsula). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. This work was also supported by The Centre for the Development and Evaluation of Complex Interventions for Public Health Improvement (DECIPHer), a UKCRC Public Health Research Centre of Excellence. Joint funding (MR/KO232331/1) from the British Heart Foundation, Cancer Research UK, Economic and Social Research Council, Medical Research Council, the Welsh Government and the Wellcome Trust, under the auspices of the UK Clinical Research Collaboration, is gratefully acknowledged. G.J. Melendez-Torres is on the staff of DECIPHer. p y g g At the same time, several limitations of our pooled data merit attention. First, both conduct problems and maternal depressive symptoms were measured using parent report. Development and Psychopathology Development and Psychopathology sustain in the months and years after the program (Van Aar, Leijten, Orobio de Castro, & Overbeek, 2017), longer-term effects of parenting programs on parental mental health are less system- atically studied. Our findings do not necessarily imply that families where par- ents perceive no or little improvement in their child’s behavior are not harmed in other ways than intensified maternal depression. Parents make a substantial investment by participating in 12 to 19 weekly parenting group sessions. When the program does not provide the benefits parents expected or hoped for, this could perhaps increase parents’ dysfunctional attributions that they do not have the ability to influence their child’s behavior or that their child is unwilling to change (e.g., Sawrikar & Dadds, 2018), or it may lead to a reluctance to seek mental health support in the future. Our IPD meta-analysis unfortunately did not allow us to test the likelihood of these outcomes. That said, our finding that families with less severe problems derived little benefit from Incredible Years suggests that screening families on both child and maternal mental health may be helpful to identify families who might be less likely to benefit from the Incredible Years parenting program. Third, as in most parenting program evaluation studies, most data came from mothers. It is of yet unclear whether the relation between parental depressive symptoms and children’s conduct problems is mainly similar or different for mothers and fathers (e.g., Connell & Goodman, 2002; Gross, Shaw, Moilanen, Dishion, & Wilson, 2008). The last decade has seen an increase in recommendations for how researchers can engage fathers more (e.g., Panter-Brick et al., 2014), and we encourage replica- tion of our findings once data from sufficient numbers of trials with larger numbers of fathers can be synthesized. This study is part of an initiative to honor the legacy of Dr. Thomas Dishion, who, through continuous critical appraisal of even our most promising interventions for youth mental health, contributed to a more refined understanding of the processes that underlie effective and ineffective intervention efforts. Our findings suggest that, at least for the Incredible Years parenting program, there is no evidence to suggest that maternal depressive symptoms intensify in families where children fail to benefit from the parenting program. Development and Psychopathology While effects of parenting programs on observed child behavior tend to be of sim- ilar magnitude as effects on parent-reported child behavior (Menting et al., 2013), parent-reported and observed measures of child behavior correlate only modestly (Moens, Weeland, Van der Giessen, Chhangur, & Overbeek, 2018). In our case, using parent-reports of conduct problems seems both a strength and a limitation. On the one hand, using parents as informants is in line with our particular interest in how parents’ perceptions of change in child behavior would relate to changes in maternal depressive symptoms. On the other hand, we cannot rule out the possibility that co-occurring change is explained by parental factors—especially maternal depression may influence mothers’ perceptions of children’s conduct problems (Callender et al., 2012). It would therefore have been good to corroborate children’s level of conduct problems with observational measures. In our pooled dataset, these were available for only a small subset of the trials and would therefore not have given us the required stat- istical power. Moreover, observational data tends to be more dif- ficult to harmonize across trials than questionnaires because of their different scales (e.g., frequency counts versus Likert scales) and lack of norm scores. Conflict of Interest. Judy Hutchings and Maria Filomena Gaspar report per- sonal fees for the delivery of leader training for Incredible Years. The remain- ing authors declare no competing or potential conflicts of interest. Axberg, U., & Broberg, A. G. (2012). Evaluation of “The Incredible Years” in Sweden: The transferability of an American parent-training program to Sweden. Scandinavian Journal of Psychology, 53, 224–232. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. Bennett, C., Barlow, J., Huband, N., Smailagic, N., & Roloff, V. (2013). Group-based parenting programs for improving parenting and psychosocial functioning: A systematic review. Journal of the Society for Social Work and Research, 4, 300–332. Axberg, U., & Broberg, A. G. (2012). Evaluation of “The Incredible Years” in Sweden: The transferability of an American parent-training program to Sweden. Scandinavian Journal of Psychology, 53, 224–232. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. Bennett, C., Barlow, J., Huband, N., Smailagic, N., & Roloff, V. (2013). Group-based parenting programs for improving parenting and psychosocial functioning: A systematic review. Journal of the Society for Social Work and Research, 4, 300–332. Discussion The social support and normalization of family mental health problems that the group format provides may prevent depressive symptoms to worsen in mothers whose children fail to benefit from the program. p g Our findings may not necessarily generalize to individually delivered parenting programs, despite the often large overlap in program content (e.g., Parent–child Interaction Therapy; Zisser & Eyberg, 2010). While group-delivered programs have the ben- efit of offering observational learning and social support, individually-delivered programs have the benefit of being able to more flexibly adapt program content and pace to address intra- personal issues such as maternal depression, in some cases in the privacy of parents’ homes (Niec, Hemme, Yopp & Brestan, 2005). Some studies suggest parents prefer individually delivered pro- grams (Wymbs et al., 2016). Head-to-head comparisons of group- versus individually-delivered programs suggest similar lev- els of effectiveness for both conduct problems and parental men- tal health (Niec, Barnett, Prewett, & Shanley, 2016; Sonuga-Barke et al., 2018), but more, and better powered, comparisons are needed to understand whether some families benefit more from some formats than others. Our finding that conduct problems improved in all classes (i.e., subgroups of families) may at first seem to contradict findings that considerable numbers of children do not benefit from parenting programs. We do not think that this is the case. In this age period, children’s conduct problems typically reduce in both intervention and control conditions (e.g., Van Aar et al., 2019) as part of nor- mal maturational changes (Sentse, Kretschmer, De Haan, & Prinzie, 2017; Shaw, Lacourse, & Nagin, 2005). This has implica- tions for generating hypotheses, and interpreting findings, about potential collateral effects of children’s failure to improve more in a parenting program than they would had they not received the program. Our findings suggest that even if children do not improve more than would be expected due to average maturation effects, their parents on average still perceive improvement in their 1859 References R., & Spencer, P. M. (1982). The Brief Symptom Inventory: Administration, Scoring and Procedures Manual 1. Baltimore, MD: Clinical Psychometrics Research. Kaehler, L. A., Jacobs, M., & Jones, D. J. (2016). Distilling common history and practice elements to inform dissemination: Hanf-Model BPT programs as an example. Clinical Child and Family Psychology Review, 19, 236–258. Dette-Hagenmeyer, D. E., & Reichle, B. (2014). Parents’ depressive symptoms and children’s adjustment over time are mediated by parenting, but differ- entially for fathers and mothers. European Journal of Developmental Psychology, 11, 196–210. Larsson, B., &. Fossum, S., Clifford, G., Drugli, M. B., Handegård, B. H., & Mørch, W. T. (2009). Treatment of oppositional defiant and conduct prob- lems in young Norwegian children: Results of a randomized controlled trial. European Child & Adolescent Psychiatry, 18, 42–52. Dishion, T. J., & Kavanagh, K. (2003). Intervening in adolescent problem behav- ior: A family-centered approach. New York, NY: Guilford. Leijten, P., Gardner, F., Landau, S., Harris, V., Mann, J., Hutchings, J., … & Scott, S. (2018). Harnessing the power of individual participant data in a meta-analysis of the benefits and harms of the Incredible Years parenting program. Journal of Child Psychology and Psychiatry, 59, 99–109. Dishion, T. J., McCord, J., & Poulin, F. (1999). When interventions harm: Peer groups and problem behavior. American Psychologist, 54, 755–764. Eyberg, S. M., & Ross, A. W. (1978). Assessment of child behavior problems: The validation of a new inventory. Journal of Clinical Child Psychology, 7, 113–116. Leijten, P., Gardner, F., Melendez-Torres, G. J., Van Aar, J., Hutchings, J., Schulz, S., Knerr, W. & Overbeek, G. (2019). Meta-analyses: Key parenting program components for disruptive child behavior. Journal of the American Academy of Child and Adolescent Psychiatry, 58, 180–190. Forgatch, M. S., & Patterson, G. R. (2010). Parent management training– Oregon model: An intervention for antisocial behavior in children and ado- lescents. In J. R. Weisz & A. E. Kazdin (Eds.), Evidence-based psychothera- pies for children and adolescents (pp. 159–178). New York, NY: Guilford Press. Leijten, P., Melendez-Torres, G. J., Knerr, W., & Gardner, F. (2016). Transported versus homegrown parenting interventions for reducing dis- ruptive child behavior: A multilevel meta-regression study. Journal of the American Academy of Child and Adolescent Psychiatry, 55, 233–243. Fosco, G. M., Van Ryzin, M., Stormshak, E. A., & Dishion, T. J. (2014). Putting theory to the test: Examining family context, caregiver motivation, and con- flict in the Family Check-Up model. References Handbook of attachment: Theory, research, and clinical applications. New York, NY: Guilford. Combs-Ronto, L. A., Olson, S. L., Lunkenheimer, E. S., & Sameroff, A. J. (2009). Interactions between maternal parenting and children’s early dis- ruptive behavior: Bidirectional associations across the transition from pre- school to school entry. Journal of Abnormal Child Psychology, 37, 1151– 1163. Harrison, C., & Sofronoff, K. (2002). ADHD and parental psychological dis- tress: Role of demographics, child behavioral characteristics, and parental cognitions. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 703–711. Hutchings, J., Bywater, T., Williams, M. E., Lane, E., & Whitaker, C. J. (2012). Improvements in maternal depression as a mediator of child behaviour change. Psychology, 3, 795–801. Connell, A. M., & Goodman, S. H. (2002). The association between psychopathology in fathers versus mothers and children’s internalizing and externalizing behavior problems: A meta-analysis. Psychological Bulletin, 128, 746–73. Hutchings, J., Griffith, N., Bywater, T., & Williams, M. E. (2017). Evaluating the Incredible Years Toddler Parenting Programme with parents of toddlers in disadvantaged (Flying Start) areas of Wales. Child: Care, Health and Development, 43, 104–113. Connell, A. M., Stormshak, E., Dishion, T., Fosco, G., & Van Ryzin, M. (2018). The Family Check Up and adolescent depression: An examination of treat- ment responders and non-responders. Prevention Science, 19, 16–26. Hutchings, J., Bywater, T., Daley, D., Gardner, F., Whitaker, C., Jones, K., … & Edwards, R. T. (2007). Parenting intervention in Sure Start services for children at risk of developing conduct disorder: Pragmatic randomised con- trolled trial. British Medical Journal, 334, 678–682. Cuijpers, P., Van Straten, A., & Warmerdam, L. (2007). Behavioral activation treatments of depression: A meta-analysis. Clinical Psychology Review, 27, 318–326. Cuijpers, P., Weitz, E., Karyotaki, E., Garber, J., & Andersson, G. (2015). The effects of psychological treatment of maternal depression on children and parental functioning: a meta-analysis. European Child & Adolescent Psychiatry, 24, 237–245. Hyde, L. W., Waller, R., & Burt, S. A. (2014). Commentary: Improving treat- ment for youth with callous-unemotional traits through the intersection of basic and applied science–reflections on Dadds et al.(2014). Journal of Child Psychology and Psychiatry, 55, 781–783. Jones, T. L., & Prinz, R. J. (2005). Potential roles of parental self-efficacy in parent and child adjustment: A review. Clinical Psychology Review, 25, 341–363. Curran, P. J., & Hussong, A. M. (2009). Integrative data analysis: The simulta- neous analysis of multiple data sets. Psychological Methods, 14, 81–100. Derogatis, L. References Note: References marked with an asterisk indicate trials from which data were included in the present study. Axberg, U., & Broberg, A. G. (2012). Evaluation of “The Incredible Years” in Sweden: The transferability of an American parent-training program to Sweden. Scandinavian Journal of Psychology, 53, 224–232. Second, most trials included only one post-intervention assess- ment. We were therefore able to test co-occurring change during the parenting program only (i.e., between baseline and posttest). We were unable to test co-occurring change after the program had ended and the order and direction of effects. While the effects of parenting programs on children’s conduct problems on average Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. Bennett, C., Barlow, J., Huband, N., Smailagic, N., & Roloff, V. (2013). Group-based parenting programs for improving parenting and psychosocial functioning: A systematic review. Journal of the Society for Social Work and Research, 4, 300–332. Bennett, C., Barlow, J., Huband, N., Smailagic, N., & Roloff, V. (2013). Group-based parenting programs for improving parenting and psychosocial functioning: A systematic review. Journal of the Society for Social Work and Research, 4, 300–332. Patty Leijten et al. 1860 Gonzalez, M. A., & Jones, D. J. (2016). Cascading effects of BPT for child internalizing problems and caregiver depression. Clinical Psychology Review, 50, 11–21. Burns, G. L., & Patterson, D. R. (2001). Normative data on the Eyberg Child Behavior Inventory and Sutter-Eyberg Student Behavior Inventory: Parent and teacher rating scales of disruptive behavior problems in children and adolescents. Child & Family Behavior Therapy, 23, 15–28. Goodman, S. H., Rouse, M. H., Connell, A. M., Broth, M. R., Hall, C. M., & Heyward, D. (2011). Maternal depression and child psychopathology: A meta-analytic review. Clinical Child and Family Psychology Review, 14, 1–27. Callender, K. A., Olson, S. L., Choe, D. E., & Sameroff, A. J. (2012). The effects of parental depressive symptoms, appraisals, and physical punishment on later child externalizing behavior. Journal of Abnormal Child Psychology, 40, 471–483. Gross, H. E., Shaw, D. S., Moilanen, K. L., Dishion, T. J., & Wilson, M. N. (2008). Reciprocal models of child behavior and depressive symptoms in mothers and fathers in a sample of children at risk for early conduct prob- lems. Journal of Family Psychology, 22, 742–751. Cassidy, J., & Shaver, P. R. (2008). References Scott, S., Sylva, K., Doolan, M., Price, J., Jacobs, B., Crook, C., & Landau, S. (2010b). Randomised controlled trial of parent groups for child antisocial behaviour targeting multiple risk factors: The SPOKES project. Journal of Child Psychology and Psychiatry, 51, 48–57. Meltzer, H., Ford, T., Goodman, R., & Vostanis, P. (2011). The burden of car- ing for children with emotional or conduct disorders. International Journal of Family Medicine, 2011, 1–8. Menting, A. T. A., Orobio de Castro, B., & Matthys, W. (2013). Effectiveness of the Incredible Years parent training to modify disruptive and prosocial child behavior: A meta-analytic review. Clinical Psychology Review, 33, 901–913. Seabra-Santos, M. J., Gaspar, M. F., Azevedo, A., Homem, T. C., & Pimentel, M. (2011). Incredible Years parent training program in ADHD: Parents’ attendance and satisfaction. ADHD Attention Deficit Hyperactivity Disorder, 3, 168–169. Seabra-Santos, M. J., Gaspar, M. F., Azevedo, A. F., Homem, T. C., Guerra, J., Martins, V., … & Moura-Ramos, M. (2016). Incredible Years parent train- ing: What changes, for whom, how, for how long? Journal of Applied Developmental Psychology, 44, 93–104. Menting, A. T. A., Orobio de Castro, B., Wijngaards-de Meij, L. D. N. V., & Matthys, W. (2014). A trial of parent training for mothers being released from incarceration and their children. Journal of Clinical Child & Adolescent Psychology, 43, 381–396. Moehler, E., Brunner, R., Wiebel, A., Reck, C., & Resch, F. (2006). Maternal depressive symptoms in the postnatal period are associated with long-term impairment of mother–child bonding. Archives of Women’s Mental Health, 9, 273–278. Sentse, M., Kretschmer, T., De Haan, A., & Prinzie, P. (2017). Conduct prob- lem trajectories between age 4 and 17 and their association with behavioral adjustment in emerging adulthood. Journal of Youth and Adolescence, 46, 1633–1642. Sharry, J., Hampson, G., & Fanning, M. (2003). Parents Plus—’The Early Years’ Programme: A video-based parenting guide to promoting young children’s development and to preventing and managing behaviour prob- lems. Available: Parents Plus, c/o Mater Child Guidance Clinic, Mater Hospital, North Circular Road, Dublin 7, Ireland. [www.parentsplus.ie]. Moens, M. A., Weeland, J., Van der Giessen, D., Chhangur, R. R., & Overbeek, G. (2018). In the eye of the beholder? Parent-observer discrepan- cies in parenting and child disruptive behavior assessments. Journal of Abnormal Child Psychology, 46, 1147–1159. Morpeth, L., Blower, S., Tobin, K., Taylor, R. S., Bywater, T., Edwards, R. T., … & Berry, V. (2017). References Development and Psychopathology, 26, 305–318. Leijten, P., Raaijmakers, M. A. J., Orobio de Castro, B., Van den Ban, E., & Matthys, W. (2017). Effectiveness of the Incredible Years parenting inter- vention for families with socioeconomically disadvantaged and ethnic minority backgrounds. Journal of Clinical Child and Adolescent Psychology, 46, 59–73. Furlong, M., & McGilloway, S. (2012). The Incredible Years Parenting program in Ireland: A qualitative analysis of the experience of disadvantaged parents. Clinical Child Psychology and Psychiatry, 17, 616–630. Lochman, J. E., Dishion, T. J., Boxmeyer, C. L., Powell, N. P., & Qu, L. (2017). Variation in response to evidence-based group preventive intervention for disruptive behavior problems: A view from 938 coping power sessions. Journal of Abnormal Child Psychology, 45, 1271–1284. Galea, S., Ahern, J., Nandi, A., Tracy, M., Beard, J., & Vlahov, D. (2007). Urban neighborhood poverty and the incidence of depression in a population- based cohort study. Annals of Epidemiology, 17, 171–179. Lovejoy, M. C., Graczyk, P. A., O’Hare, E., & Neuman, G. (2000). Maternal depression and parenting behavior: A meta-analytic review. Clinical Psychology Review, 20, 561–592. Gardner, F., Connell, A., Trentacosta, C. J., Shaw, D. S., Dishion, T. J., & Wilson, M. N. (2009). Moderators of outcome in a brief family-centered intervention for preventing early problem behavior. Journal of Consulting and Clinical Psychology, 77, 543–553. Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family: Parent–child interaction. In P. H. Mussen (Series Ed.) & E. M. Hetherington (Vol. Ed.). Handbook of child psychology: Vol. 4. Socialization, personality, and social development (4th ed., pp. 1–101). New York, NY: Wiley. Gardner, F., Burton, J., & Klimes, I. (2006). Randomised controlled trial of a parenting intervention in the voluntary sector for reducing child conduct problems: Outcomes and mechanisms of change. Journal of Child Psychology and Psychiatry, 47, 1123–1132. Development and Psychopathology 1861 Scott, S., Sylva, K., Beckett, C., Doolan, M., Kallitsoglou, A., … & Ford, T. (2013). Which type of parenting programme best improves child behaviour and reading? The Helping Children Achieve trial. Follow-up of the helping children achieve trial. Nuffield Foundation, London. McGilloway, S., Mhaille, G. N., Bywater, T., Furlong, M., Leckey, Y., Kelly, P., … & Donnelly, M. (2012). A parenting intervention for childhood behav- ioral problems: A randomised controlled trial in disadvantaged community-based settings. Journal of Consulting and Clinical Psychology, 80, 116–127. References The effectiveness of the Incredible Years pre-school par- enting programme in the United Kingdom: A pragmatic randomised con- trolled trial. Child Care in Practice, 23, 141–161. Shaw, D. S., Connell, A., Dishion, T. J., Wilson, M. N., & Gardner, F. (2009). Improvements in maternal depression as a mediator of intervention effects on early childhood problem behavior. Development and Psychopathology, 21, 417–439. Moses, T. (2010). Exploring parents’ self-blame in relation to adolescents’ mental disorders. Family Relations, 59, 103–120. Shaw, D. S., Lacourse, E., & Nagin, D. S. (2005). Developmental trajectories of conduct problems and hyperactivity from ages 2 to 10. Journal of Child Psychology and Psychiatry, 46, 931–942. Murray, L., Fiori-Cowley, A., Hooper, R., & Cooper, P. (1996). The impact of postnatal depression and associated adversity on early mother-infant inter- actions and later infant outcome. Child Development, 67, 2512–2526. Shaw, D. S., & Shelleby, E. C. (2014). Early-starting conduct problems: Intersection of conduct problems and poverty. Annual Review of Clinical Psychology, 10, 503–528. Muthén, L. K., & Muthén, B. O. (2018). Mplus Version 8.0. Los Angeles, CA: Muthén & Muthén. Shaw, D. S., Sitnick, S. L., Brennan, L. M., Choe, D. E., Dishion, T. J., Wilson, M. N., & Gardner, F. (2016). The long-term effectiveness of the Family Check-Up on school-age conduct problems: Moderation by neigh- borhood deprivation. Development and Psychopathology, 28, 1471–1486. Niec, L. N., Barnett, M. L., Prewett, M. S., & Shanley Chatham, J. R. (2016). Group parent–child interaction therapy: A randomized control trial for the treatment of conduct problems in young children. Journal of Consulting and Clinical Psychology, 84, 682–698. Niec, L. N., Hemme, J. M., Yopp, J. M., & Brestan, E. V. (2005). Parent–child interaction therapy: The rewards and challenges of a group format. Cognitive and Behavioral Practice, 12, 113–125. Silberg, J. L., Maes, H., & Eaves, L. J. (2010). Genetic and environmental influ- ences on the transmission of parental depression to children’s depression and conduct disturbance: An extended Children of Twins study. Journal of Child Psychology and Psychiatry, 51, 734–744. Panter-Brick, C., Burgess, A., Eggerman, M., McAllister, F., Pruett, K., & Leckman, J. F. (2014). Practitioner review: Engaging fathers–recommenda- tions for a game change in parenting interventions based on a systematic review of the global evidence. Journal of Child Psychology and Psychiatry, 55, 1187–1212. Singh, A. L., D’onofrio, B. M., Slutske, W. S., Turkheimer, E., Emery, R. E., Harden, K. P., … & Martin, N. G. References (2011). Parental depression and offspring psychopathology: A Children of Twins study. Psychological Medicine, 41, 1385–1395. Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). Antisocial boys (Vol. 4). Eugene, OR: Castalia. Smith, J. D., Berkel, C., Hails, K. A., Dishion, T. J., Shaw, D. S., & Wilson, M. N. (2018). Predictors of participation in the Family Check-Up program: A randomized trial of yearly services from age 2 to 10 years. Prevention Science, 5, 652–662. Pelham, W. E., Dishion, T. J., Tein, J. Y., Shaw, D. S., & Wilson, M. N. (2017). What doesn’t work for whom? Exploring heterogeneity in responsiveness to the family check-up in early childhood using a mixture model approach. Prevention Science, 18, 911–922. Sonuga-Barke, E. J., Barton, J., Daley, D., Hutchings, J., Maishman, T., Raftery, J., … & Little, L. (2018). A comparison of the clinical effectiveness and cost of specialised individually delivered parent training for preschool attention-deficit/hyperactivity disorder and a generic, group-based pro- gramme: A multi-centre, randomised controlled trial of the New Forest Parenting Programme versus Incredible Years. European Child & Adolescent Psychiatry, 27, 797–807. Rostad, W. L., Moreland, A. D., Valle, L. A., & Chaffin, M. J. (2018). Barriers to participation in parenting programs: The relationship between parenting stress, perceived barriers, and program completion. Journal of Child and Family Studies, 27, 1264–1274. Sawrikar, V., Hawes, D. J., Moul, C., & Dadds, M. R. (2018). The role of parental attributions in predicting parenting intervention outcomes in the treatment of child conduct problems. Behaviour Research and Therapy, 111, 64–71. Taylor, E., Schachar, R., Thorley, G., & Wieselberg, M. (1986). Conduct disor- der and hyperactivity: I. Separation of hyperactivity and antisocial conduct in British child psychiatric patients. The British Journal of Psychiatry, 149, 760–767. Scott, S., O’Connor, T. G., Futh, A., Matias, C., Price, J., & Doolan, M. (2010a). Impact of a parenting program in a high-risk, multi-ethnic com- munity: The PALS trial. Journal of Child Psychology and Psychiatry, 51, 1331–1341. Van Aar, J., Leijten, P., Orobio de Castro, B., & Overbeek, G. (2017). Sustained, fade-out or sleeper effects? A systematic review and meta-analysis of parent- ing interventions for disruptive child behavior. Clinical Psychology Review, 51, 153–163. Scott, S., Spender, Q., Doolan, M., Jacobs, B., & Aspland, H. (2001). Multicentre controlled trial of parenting groups for childhood antisocial behaviour in clinical practice. British Medical Journal, 323, 194–198. References Van Aar, J., Leijten, P., Orobio de Castro, B., Weeland, J., Matthys, W., Chhangur, R., & Overbeek, G. (2019). Families who benefit and families 1862 Patty Leijten et al. who do not: Integrating person-and variable-centered analyses of parenting intervention responses. Journal of the American Academy of Child & Adolescent Psychiatry. Advance online publication. https://doi.org/10.1016/ j.jaac.2019.02.004 Weeland, J., Chhangur, R. R., Jaffee, S. R., Van der Giessen, D., Matthys, W., Orobio de Castro, B., & Overbeek, G. (2018b). What changes when? A reply to Beauchaine and Slep. Development and Psychopathology, 30, 1535–1540. Weeland, J., Chhangur, R. R., Van der Giessen, D., Matthys, W., Orobio de Castro, B., & Overbeek, G. (2017). Intervention effectiveness of the Incredible Years: New insights into sociodemographic and intervention- based moderators. Behavior Therapy, 48, 1–18. Veerman, J. L., Dowrick, C., Ayuso-Mateos, J. L., Dunn, G., & Barendregt, J. J. (2009). Population prevalence of depression and mean Beck Depression Inventory Score. The British Journal of Psychiatry, 195, 516–519. Weaver, C. M., Shaw, D. S., Dishion, T. J., & Wilson, M. N. (2008). Parenting self-efficacy and problem behavior in children at high risk for early conduct problems: The mediating role of maternal depression. Infant Behavior and Development, 31, 594–605. Weisz, J. R., & Kazdin, A. E. (Eds.). (2017). Evidence-based psychotherapies for children and adolescents. New York, NY: Guilford Press. Wiggins, T. L., Sofronoff, K., & Sanders, M. R. (2009). Pathways Triple P-Positive Parenting Program: Effects on parent–child relationships and child behavior problems. Family Process, 48, 517–530. Webster-Stratton, C., & Reid, M. J. (2010). The Incredible Years Parents, Teachers, and Children Training Series: A multifaceted treatment approach for children with conduct disorders. In J. R. Weisz & A. E. Kazdin, (Eds.), Evidence-based psychotherapies for children and adolescents (pp. 194–210). New York, NY: Guilford Press. Williams, P., & Goldberg, D. P. (1988). A user’s guide to the General Health Questionnaire. Berkshire: NFER, Nelson. Wymbs, F. A., Cunningham, C. E., Chen, Y., Rimas, H. M., Deal, K., Waschbusch, D. A., & Pelham Jr, W. E. (2016). Examining parents’ prefer- ences for group and individual parent training for children with ADHD symptoms. Journal of Clinical Child & Adolescent Psychology, 45, 614–631. Weeland, J., Chhangur, R. R., Jaffee, S. R., Van der Giessen, D., Matthys, W., Orobio de Castro, B., & Overbeek, G. (2018a). Does the Incredible Years reduce child externalizing problems through improved parenting? References The role of child negative affectivity and serotonin transporter linked polymor- phic region (5-HTTLPR) genotype. Development and Psychopathology, 30, 93–112. Zisser, A., & Eyberg, S. M. (2010). Parent–child interaction therapy and the treatment of disruptive behaviors. In J. R. Weisz & A. E. Kazdin, Evidence-based psychotherapies for children and adolescents (pp. 179– 193). New York, NY: Guilford Press.
https://openalex.org/W3196631164	https://www.frontiersin.org/articles/10.3389/fpls.2021.721980/pdf	English	null	Unraveling the Genetic Basis of Fertility Restoration for Cytoplasmic Male Sterile Line WNJ01A Originated From Brassica juncea in Brassica napus	Frontiers in plant science	2,021	cc-by	15,920	Edited by: Kun Lu, Edited by: Kun Lu, Southwest University, China Crosses that lead to heterosis have been widely used in the rapeseed (Brassica napus L.) industry. Cytoplasmic male sterility (CMS)/restorer-of-fertility (Rf) systems represent one of the most useful tools for rapeseed production. Several CMS types and their restorer lines have been identiﬁed in rapeseed, but there are few studies on the mechanisms underlying fertility restoration. Here, we performed morphological observation, map-based cloning, and transcriptomic analysis of the F2 population developed by crossing the CMS line WNJ01A with its restorer line Hui01. Parafﬁn- embedded sections showed that the sporogenous cell stage was the critical pollen degeneration period, with major sporogenous cells displaying loose and irregular arrangement in sterile anthers. Most mitochondrial electron transport chain (mtETC) complex genes were upregulated in fertile compared to sterile buds. Using bulked segregant analysis (BSA)-seq to analyze mixed DNA pools from sterile and fertile F2 buds, respectively, we identiﬁed a 6.25 Mb candidate interval where Rfw is located. Using map-based cloning experiments combined with bacterial artiﬁcial chromosome (BAC) clone sequencing, the candidate interval was reduced to 99.75 kb and two pentatricopeptide repeat (PPR) genes were found among 28 predicted genes in this interval. Transcriptome sequencing showed that there were 1679 DEGs (1023 upregulated and 656 downregulated) in fertile compared to sterile F2 buds. The upregulated differentially expressed genes (DEGs) were enriched in the Kyoto Encyclopedia of Genes and Genomes (KEGG) lysine degradation pathway and phenylalanine metabolism, and the downregulated DEGs were enriched in cutin, suberine, and wax biosynthesis. Furthermore, 44 DEGs were involved in pollen and anther development, such as tapetum, microspores, and pollen wall development. All of them were upregulated except a few such as POE1 genes (which encode Pollen Ole e I allergen and extensin family proteins). There were 261 speciﬁcally expressed DEGs Reviewed by: Zhi Liu, Southwest University, China Jinjin Jiang, Yangzhou University, China Shengwu Hu, Northwest A&F University, China Xiyue Song, Northwest A&F University, China Reviewed by: Zhi Liu, Southwest University, China Jinjin Jiang, Yangzhou University, China Shengwu Hu, Northwest A&F University, China Xiyue Song, Northwest A&F University, China Correspondence: Chao Liu liuchao@mail.hzau.edu.cn Correspondence: Chao Liu liuchao@mail.hzau.edu.cn Specialty section: This article was submitted to Plant Breeding, a section of the journal Frontiers in Plant Science Received: 08 June 2021 Accepted: 05 August 2021 Published: 31 August 2021 Unraveling the Genetic Basis of Fertility Restoration for Cytoplasmic Male Sterile Line WNJ01A Originated From Brassica juncea in Brassica napus Qian Yang, Xiaoyi Nong, Jize Xu, Fan Huang, Fang Wang, Jiangsheng Wu, Chunyu Zhang and Chao Liu* Qian Yang, Xiaoyi Nong, Jize Xu, Fan Huang, Fang Wang, Jiangsheng Wu, Chunyu Zhang and Chao Liu* ORIGINAL RESEARCH published: 31 August 2021 doi: 10.3389/fpls.2021.721980 ORIGINAL RESEARCH published: 31 August 2021 doi: 10.3389/fpls.2021.721980 Yang Q, Nong X, Xu J, Huang F, Wang F, Wu J, Zhang C and Liu C (2021) Unraveling the Genetic Basis of Fertility Restoration for Cytoplasmic Male Sterile Line WNJ01A Originated From Brassica juncea in Brassica napus. Front. Plant Sci. 12:721980. doi: 10.3389/fpls.2021.721980 College of Plant Science and Technology, Huazhong Agricultural University, Wuhan, China INTRODUCTION R. sativus encodes a PPR protein that downregulates orf138 protein at the translation level (Uyttewaal et al., 2008). Another B. napus CMS system, nap, is rarely used to produce hybrid seeds because the sterility is easily aﬀected by temperature change in most cases. In this system, orf222 is the key sterility gene, and it is co-transcribed with both orf139 and an exon of a trans-spliced gene, nad5c (L’Homme et al., 1997). Its restorer gene, Rfn, on chromosome A09 of B. napus, encodes a PPR protein. However, the accumulation of orf222 transcripts was not obviously changed in the restorer line or in transgenic fertility-restored lines compared to the nap CMS line (Liu et al., 2017). Rapeseed (Brassica napus L.) is one of the most important oil crops worldwide. Crosses that lead to heterosis, involving signiﬁcantly increased seed yield, have been widely used for rapeseed production in the past decades (Morrison et al., 2016; Shen et al., 2017). Cytoplasmic male sterility (CMS) combined with its maintainer line and restorer line is a highly valuable resource to produce hybrid seeds of rapeseed (Li et al., 2015). Plant CMS is generally caused by new chimeric open reading frames (ORFs) in the mitochondrial genome, and they are usually generated through mitochondrial genome rearrangement (Tang H. et al., 2017). These novel ORFs are usually homologous or co-transcribed with the genes encoding proteins of the mitochondrial electron transport chain (mtETC) complexes or ATP synthase, so these systems do not function normally (Bohra et al., 2016). Restorer-of-fertility (Rf ) genes in the nuclear genomes of the restorer lines can downregulate the CMS genes to reverse male sterility. Several Rf genes, which encode diverse functional proteins, have been identiﬁed. For instance, maize Rf2 and Rf4 encode an aldehyde dehydrogenase and a bHLH transcription factor (TF) for T-CMS and C-CMS, respectively (Liu and Schnable, 2002); rice RF17 and RF2 encode an acetyl-carrier protein and a glycine-rich protein for CW- CMS and LD-CMS, respectively (Fujii and Toriyama, 2009); and sugar beet Rf1 encodes a putative M48 family peptidase for Owen-CMS (Kitazaki et al., 2015). However, most cloned plant Rf genes encode pentatricopeptide repeat (PPR)-containing proteins (Kubo et al., 2020). These proteins are RNA-binding factors that participate in mRNA processing after transport to a mitochondrion (Schmitz-Linneweber and Small, 2008). During the fertility restoration of CMS, they downregulate sterility genes by cleaving their transcripts or inhibiting their translation (Gaborieau et al., 2016). INTRODUCTION In recent years, new CMS systems from Brassica or related species have been discovered and introduced into B. napus. The hau CMS system originated from Brassica juncea and its sterility gene, orf288, encodes a cytotoxic protein that causes aborted pollen development (Wan et al., 2008; Jing et al., 2012). orf288 transcription was upregulated in the male-sterile line compared to the restorer line, but there was no signiﬁcant diﬀerence at the protein level (Heng et al., 2018). A recent research reported that the restorer gene Rfh of hau CMS has been cloned (Wang H. et al., 2021). Nsa CMS is a novel B. napus CMS system derived from the cytoplasm of Sinapis arvensis, and its restorer line NR1 was developed from a disomic alien addition line of B. napus and S. arvensis (Wei et al., 2010). Three ORFs (orf224, orf309, and orf346), which possess chimeric and transmembrane structures, are the candidate sterility genes of Nsa CMS (Sang et al., 2019). It was further conﬁrmed that ORF346 is the key gene for pollen abortion in Nsa CMS (Sang et al., 2021). Another CMS system, inap CMS, was obtained via backcross of the somatic hybrid of Isatis indigotica (Chinese woad) and B. napus, with B. napus as the recurrent parent (Kang et al., 2017). A restorer line carrying a dominant Rf gene was successfully developed for inap CMS (Li et al., 2019). Further research is required to isolate their restorer genes and use them in rapeseed breeding. In B. napus, there are a variety of CMS systems, such as pol, ogu, nap, hau, Nsa, and inap. Among them, the pol and ogu CMS lines are the most widely used to produce hybrid seeds. Regarding pol CMS, a chimeric ORF, orf224, is co- transcribed with atp6 and causes male sterility (Singh and Brown, 1993; L’Homme et al., 1997; Liu et al., 2016). Its restorer gene, Rfp, encodes a mitochondria-targeted PPR protein that downregulates orf224 (Liu et al., 2016). Ogu CMS originated in Raphanus sativus and was introduced into B. napus. Regarding ogu CMS, orf138 is considered to be the gene that causes male sterility (Bonhomme et al., 1992). Its restorer gene Rfo from Anther and pollen development are extremely complex processes involving many genes and pathways. Therefore, transcriptome sequencing has been applied in many crops as an eﬀective tool to study global transcription networks to elucidate underlying mechanisms. Citation: Yang Q, Nong X, Xu J, Huang F, Wang F, Wu J, Zhang C and Liu C (2021) Unraveling the Genetic Basis of Fertility Restoration for Cytoplasmic Male Sterile Line WNJ01A Originated From Brassica juncea in Brassica napus. Front. Plant Sci. 12:721980. doi: 10.3389/fpls.2021.721980 August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 1 Yang et al. Fertility Restoration for WNJ01A CMS (9 and 252 in sterile and fertile buds, respectively). Regarding the fertile bud-speciﬁc upregulated DEGs, the ubiquitin–proteasome pathway was enriched. The top four hub genes in the protein–protein interaction network (BnaA09g56400D, BnaA10g18210D, BnaA10g18220D, and BnaC09g41740D) encode RAD23d proteins, which deliver ubiquitinated substrates to the 26S proteasome. These ﬁndings provide evidence on the pathways regulated by Rfw and improve our understanding of fertility restoration. Keywords: Brassica napus, cytoplasmic male sterility, fertility restorer, map-based cloning, candidate gene, transcriptome Keywords: Brassica napus, cytoplasmic male sterility, fertility restorer, map-based cloning, candidate gene, transcriptome Frontiers in Plant Science \| www.frontiersin.org Whole-Genome Resequencing of Bulked DNA Pools Currently, many hybrid rapeseed cultivars are bred using the pol CMS system in China. Thus, it is necessary to develop new stable CMS systems (with restorer lines) to solve the issue of cytoplasmic simpliﬁcation (i.e., most of the sterile cytoplasm of current commercial hybrids belong to a single type) to avoid risks to the rapeseed industry. We previously developed a novel type of CMS, WNJ01A, from a natural male-sterile B. juncea mutant. It was introduced into B. napus and displayed a stable and complete pollen-abortion phenotype. Recently, its restorer line, Hui01, was obtained through distant hybridization between WNJ01A CMS and B. rapa. Here, we aimed to ﬁnely map and isolate the candidate restorer gene, designated Restorer of fertility for WNJ01A (Rfw). We also aimed to conduct a transcriptomic analysis to identify the genes and pathways involved in male sterility and fertility restoration. The results will facilitate understanding of the molecular mechanisms of WNJ01A CMS and its fertility restoration, and they will also provide a foundation for its use in breeding. To determine the genomic region containing Rfw, 37 extremely fertile plants and 38 extremely sterile plants were selected from the F2 population. Genomic DNA was extracted from 100 mg fresh leaves from each plant and puriﬁed using a DNA Secure Plant Kit (DP320; TIANGEN, China). A NanoDrop 2000 spectrophotometer (Thermo Fisher Scientiﬁc, United States) was used to determine the DNA concentration and quality. Next, 80 ng DNA per plant were mixed to construct the F- and S-pools. A HiSeq 3000 PE150 sequencing platform (Illumina, United States) was used to perform high- throughput library construction and sequencing for the S- and F-pools. Bulked segregant analysis (BSA)-seq, involving the abovementioned clean genome resequencing data and using the reference genome B. napus cv. Darmor-bzh (Chalhoub et al., 2014), was conducted as previously described (Tang Q. et al., 2017; Su et al., 2020). To locate the Rfw gene, we calculated the proportion of reads in the F- and S-pools with single-nucleotide polymorphism (SNP)/insertion/deletion (InDel) sites that were distinct from the of Darmor-bzh reference reads (i.e., SNP/InDel-index) and the diﬀerence in this index between the F- and S-pools [i.e., 1(SNP/InDel-index)]. To eliminate low-quality SNP/InDels, we discarded all loci with an SNP/InDel-index < 0.3 (Takagi et al., 2013). Using the sliding window method (with a 1 Mb window size and 100 kb increment), the average index values of loci were calculated. Cytological Characterization of Fertility Restoration for WNJ01A Fresh fertile and sterile ﬂoral buds (0–8 mm) at various developmental stages from F2 plants were collected and ﬁxed in FAA ﬁxative (3.7% formaldehyde, 50% ethanol, and 5.0% acetic acid) and processed as previously described (Ning et al., 2019). Paraﬃn-embedded sections were assessed using the Arabidopsis thaliana classiﬁcation criteria for the anther development period (Chang F. et al., 2011). Whole-Genome Resequencing of Bulked DNA Pools The SNP/InDel-index of the F- and S-pools and the corresponding 1(SNP/InDel-index) in the sliding window were used to construct SNP/InDel-index plots. Map-Based Cloning of Rfw To develop the set of molecular markers, the InDel polymorphism sites detected during genome resequencing in the candidate interval, simple sequence repeats (SSR) sites in this interval, and publicly available molecular markers in this interval were used. The molecular markers were veriﬁed in a mixed DNA pool from fertile and sterile plants, with a small population being used to verify the linkage between the INTRODUCTION To elucidate the genome- wide molecular mechanisms underlying Brassica CMS and fertility restoration, transcriptomic proﬁling using RNA-seq has been used. For example, transcriptomic analysis of pol CMS indicated that the energy deﬁciency caused by orf224/atp6 may downregulate many genes involved in pollen development August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 2 Fertility Restoration for WNJ01A CMS Yang et al. via nuclear–mitochondrial interactions (An et al., 2014). Through multi-omics joint analysis and yeast two-hybrid assay, seven Rfp interacting proteins related to RNA editing, anther and tapetum development, and ﬁve orf224 interacting proteins participating in the electronic respiratory transmission chain, anther development and oxidative phosphorylation were obtained and further veriﬁed (Wang B. et al., 2021). In ogu CMS, sterility may be caused by delayed tapetum degradation, as genes related to tapetum programmed cell death were downregulated compared to in the maintainer line. Additionally, sporopollenin biosynthesis and its key genes were both inhibited in ogu CMS (Xing et al., 2018; Lin et al., 2019). Another transcriptomics study on ogu CMS, pol CMS, and their shared maintainer line reported that most of the DEGs that were downregulated in both CMS lines were involved in pollen development, carbon metabolism, lipid metabolism, the tricarboxylic acid (TCA) cycle, and oxidative phosphorylation (Wei et al., 2020). population was generated via self-pollination of F1 plants. The fertility of each plant was determined by assessing at least ﬁve ﬂowers per plant a minimum of three diﬀerence times during the ﬂowering period, following a previously described method (Liu et al., 2012). Plant Materials We previously developed the WJS1A CMS line from a natural B. juncea mutant. To obtain WNJ01A CMS, the sterile cytoplasm was introduced into B. napus cv. Huashuang4 by backcrossing for more than 10 generations. Subsequently, we crossed WNJ01A CMS (female parent) with B. rapa ssp. pekinensis via distant hybridization. We then searched for fertile plants in their progenies after continuous selﬁng, the fertile progeny lines were obtained and it was further conﬁrmed that they could reverse the sterility in WNJ01A CMS by test crossing. One restorer line with the best ability of restoring fertility was designated Hui01. Identiﬁcation of DEGs and Enrichment Analyses Diﬀerentially expressed genes (DEGs) between fertile and sterile buds were identiﬁed with the DESeq R package (1.10.1), using the Benjamini–Hochberg procedure to adjust the generated P-values to control the false discovery rate (FDR). DEGs were identiﬁed based on adjusted P < 0.05, Fragments Per Kilobase of transcript per Million mapped read (FPKM) fold change ≥2, and FDR < 0.01. Four reference genomes containing A subgenomes (B. rapa cv. chiifu-401: AA; B. napus cv. Darmor-bzh and Zhongshuang11: AACC; B. juncea var. tumida: AABB) were used to conduct a collinearity analysis of the candidate Rfw genes in the ﬁnely mapped Rfw region. The genes in the candidate interval of Hui01 were analyzed and predicted through the Fgenesh gene- ﬁnder1. The gene coding sequence (CDS) was compared and analyzed with ClustalX. Next, we used the GOseq R package based on the Wallenius non-central hypergeometric distribution to perform Gene Ontology (GO) enrichment analysis on the DEGs (Young et al., 2010), which accounts for gene length bias in DEGs. We also performed a Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis using KOBAS software (Mao et al., 2005). The KEGG and GO analyses were performed via the Bioinformatics platform2. Genetic Analysis of Fertility Restoration for WNJ01A WNJ01A (female parent) was crossed with restorer line Hui01 (male parent) to produce the F1 population and then the F2 August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 3 Fertility Restoration for WNJ01A CMS Yang et al. V (ATP synthase, atp), cytochrome c biogenesis, ribosome small subunit, ribosome large subunit, twin-arginine translocation (tatC), and intron maturase (matR). molecular markers and fertility restoration. All primer sequences of linked molecular markers are given in Supplementary Table 1. Next, 4454 sterile F2 plants were subjected to map-based cloning experiments to ﬁnely map Rfw, based on the numbers of recombinant plants and the known physical locations of a set of molecular markers. Transcriptional levels in the fertile and sterile buds were assessed using RT-qPCR. All genes and primer sequences are detailed in Supplementary Table 2. A RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientiﬁc, United States) was used to reverse transcribe the total RNA following the manufacturer’s protocol. A Hieﬀ R⃝qPCR SYBR Green Master Mix (No Rox) kit (YEASEN, China) and a CFX384 Touch Real-Time PCR Detection System (Bio-Rad, United States) were used for analysis of the transcription levels. Bna.actin7 was used for normalization in RT-qPCR (for normalization corresponding to total RNA levels). 2http://www.bioinformatics.com.cn/ 3http://string-db.org/ Transcriptomic Sequencing Floral buds (0–2 mm) were carefully harvested from fertile and sterile F2 plants. Three biological replicates (involving ﬁve buds each) were used for the fertile-pool (F-pool; Rfw1, Rfw2, and Rfw3) and three for the sterile-pool (S-pool; rfw1, rfw2, and rfw3). The buds were frozen in liquid nitrogen at once and then stored at –80◦C until further processing. Total RNA was extracted using TRIzol (Invitrogen, United States) according to the manufacturer’s instructions. A HiSeq X Ten platform (Illumina, United States) was used for transcriptome sequencing (Biomarker Technologies, China). 1http://www.softberry.com/ Bacterial Artiﬁcial Chromosome (BAC) Screening, Sequencing, and Candidate Gene Prediction To obtain the candidate region sequence, we constructed a BAC library (using CopyControlTM pCC1BACTM Vector; EPICENTRE, United States) for the restorer line Hui01. HindIII restriction enzyme was used to digest the genomic DNA and then 90–120 kb DNA fragments were selected by pulse- ﬁeld electrophoresis and puriﬁed. The DNA fragments were ligated into the CopyControlTM pCCIBACTM Vectors in the library kit. The library contained about 40,000 clones with 90– 120 kb DNA insert size, representing about 5 × B. napus genome. After batch transformation of EPI300 E. coli, the bacterial suspension was evenly smeared onto LB medium containing 12.5 µg/mL chloramphenicol. Thereafter, the strains were grown and then eluted in a sterile centrifuge tube containing chloramphenicol before further propagation. We mixed sets of 1000 single clones to create 40 pools. Next, the BAC plasmids in each pool were extracted using a Large-Construct Kit (QIAGEN, United States). A molecular marker (TY21) that co-segregated with fertility restoration was used to screen for pools containing a positive clone. Finally, the molecular markers ﬂanking the ﬁnely mapped candidate interval were used to analyze the physical location of the interval in the positive clones, which underwent high- throughput sequencing using Illumina NovaSeq (Illumina, United States) and Oxford Nanopore Technologies (ONT; United Kingdom). Morphological Characterization and Genetic Analysis of CMS and Restorer Lines The cytoplasm of the CMS line WNJ01A originated from a natural male-sterile B. juncea mutant. We developed its restorer line Hui01 via distant hybridization between WNJ01A CMS and B. rapa. The sepals, petals, and pistils of both the WNJ01A and Hui01 lines exhibited normal development (Figure 1). However, the albino WNJ01A anthers were signiﬁcantly smaller and their ﬁlaments were dramatically shorter (Figures 1A,C) compared to the Hui01 anthers (Figures 1B,D). Across more than 20 generations over 11 years (2010–2020) under various ecological conditions (in spring and winter) in China, WNJ01A maintained complete pollen abortion without being aﬀected by temperature or photoperiod. In the remaining developmental stages (early mononuclear, mononuclear, dinuclear, and mature microspore stages) of the fertile anthers, microspores were released from the tetrads and then many mature pollen grains were found in the anther chambers. The tapetums disintegrated and gradually disappeared, and only the epidermis and ﬁber layer remained in the pollen sac wall (Figures 2A5–A8). However, in the sterile anthers, the tapetum cells degraded prematurely. Numerous pollen grains became vacuolated and eventually completely degraded, emptying the anther chamber (Figures 2B5–B8). To elucidate the genetic characteristics of fertility restoration of WNJ01A CMS by Hui01, the fertility of the F1 and F2 populations of a WNJ01A CMS × Hui01 cross was evaluated. All F1 plants produced normally fertile pollen, indicating that Hui01 can completely restore the fertility of WNJ01A CMS. In the F2 population, there were 1606 fertile and 555 sterile plants, which ﬁtted the expected 3:1 ratio (χc2 = 0.5012, P = 0.46). Thus, the fertility restoration of WNJ01A CMS is controlled by a dominant gene. It was designated Restorer of fertility for WNJ01A (Rfw). These results indicate that the sporogenous cell stage is critical for anther development, based on the abnormal WNJ01A CMS plants. Equally, this stage is an important stage for Rfw to begin to prevent pollen abortion. Furthermore, the anther sterility characteristics of WNJ01A CMS were obviously diﬀerent from the previously studied pol, ogu, Nsa, hau, and inap CMS Brassica systems (Wan et al., 2008; An et al., 2014; Kang et al., 2017; Ding et al., 2018). Cytological Characterization of Anther Development To accurately characterize pollen abortion, paraﬃn-embedded sections of the buds of fertile and sterile F2 plants were observed. FIGURE 1 \| Phenotypic characterization of ﬂoral organs. (A,C) Sterile ﬂower of CMS line WNJ01A. (B,D) Fertile ﬂower of restorer line Hui01. Bars = 0.5 cm. RESULTS At the anther primordia division stage, the cells of anther primordia were dividing, with almost no diﬀerence between fertile and sterile anthers (Figures 2A1,B1). However, at the sporogenous cell stage, major sporogenous cells displayed loose and irregular arrangements in sterile anthers, unlike the regular three layers of cells in fertile anthers. The cells in the exothecium, endothecium, and middle layer of the sterile anther walls began to shrink and degenerate. The tapetum cells started to detach from the anther walls and gradually degrade (Figures 2A2,B2). At the microspore mother cell and tetrad stages, microspore mother cells developed normally and then formed numerous tetrads via meiosis in the fertile anthers, and the tapetum cells were regularly arranged. In contrast, the microspore mother cells of the sterile anthers were more scattered, the tetrads were not formed, and the tapetum cells were increasingly abnormal (Figures 2A3,A4,B3,B4). Interaction Analysis g Annotation information of the B. juncea var. tumida and jiangpu mitochondrial genomes (accession nos. KJ461445 and JF920288, respectively) were obtained from the National Center for Biotechnology Information (NCBI) (Chang S. et al., 2011; Zhao et al., 2016). We selected 32 known mitochondrial genes, including genes related to mtETC complex I (NADH dehydrogenase, nad), complex III (ubiquinol–cytochrome c reductase, cob), complex IV (cytochrome c oxidase, cox), complex To predict the protein–protein interactions (PPIs) of the DEGs, the DEG sequences were used in a BLASTx search against the genomes of related species that had PPI information in the STRING database3. The PPI network (conﬁdence score > 0.7), with KEGG annotations, was visualized using Cytoscape (Shannon et al., 2003). The hub genes were identiﬁed using the Cytoscape plugin CytoHubba (Song et al., 2020; Zuo et al., 2020). August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 4 Fertility Restoration for WNJ01A CMS Yang et al. Mapping of Rfw Gene To identify the Rfw region, F- and S-pools were constructed by subjecting extremely fertile and sterile F2 plants to high- throughput sequencing. A total of 229,832,439 and 167,518,595 reads were obtained from the F- and S-pool, respectively, with a sequencing coverage of 92.14% regarding the B. napus cv. Darmor-bzh reference genome, and a GC content of 37%. After trimming the raw reads, 82.21 and 60.14 Gb of clean data, respectively, were generated. Based on comparison with the reference genome, there were 58,610 SNPs and 120,651 InDels in the F- and S-pools. The low-quality SNPs/InDels (SNP/InDel index < 0.3) in both pools were deleted. The indices related to the remaining SNPs/InDels calculated for each pool were plotted onto the 19 B. napus chromosomes. Additionally, 1(SNPs/InDel-index) values (calculated using the sliding window method) were plotted on the 19 B. napus chromosomes. The phenotypic diﬀerence between the fertile and sterile plants resulted in 1(SNPs/InDel-index) ≥0.5 in a speciﬁc genomic locus, which indicates that this region contains the locus underlying the phenotypic diﬀerence (i.e., the Rfw locus). The signiﬁcant signal region was only observed on chromosome A09 (not on FIGURE 1 \| Phenotypic characterization of ﬂoral organs. (A,C) Sterile ﬂower of CMS line WNJ01A. (B,D) Fertile ﬂower of restorer line Hui01. Bars = 0.5 cm. August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 5 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 2 \| Microstructure of fertile (A1–A8) and sterile (B1–B8) anthers from F2 plants. (A1,B1) Anther primordium stage; (A2,B2) sporoblast stage; (A3,B3) microspore mother cell stage; (A4,B4) tetrad stage; (A5,B5) early mononuclear stage; (A6,B6) mononuclear stage; (A7,B7) dinuclear stage; (A8) mature microspores; and (B8) empty pollen sac. AP, anther primordium; Sp., sporogenous cell; E, epidermis; En, endothecium; ML, middle layer; T, tapetum; MMC, microspore mother cells; Ms, microspore; MP, mature pollen; PG, pollen grain. Bars = 100 µm. FIGURE 2 \| Microstructure of fertile (A1–A8) and sterile (B1–B8) anthers from F2 plants. (A1,B1) Anther primordium stage; (A2,B2) sporoblast stage; (A3,B3) microspore mother cell stage; (A4,B4) tetrad stage; (A5,B5) early mononuclear stage; (A6,B6) mononuclear stage; (A7,B7) dinuclear stage; (A8) mature microspores; and (B8) empty pollen sac. AP, anther primordium; Sp., sporogenous cell; E, epidermis; En, endothecium; ML, middle layer; T, tapetum; MMC, microspore mother cells; Ms, microspore; MP, mature pollen; PG, pollen grain. Bars = 100 µm. the other 18 chromosomes) (Figures 3A,B). Mapping of Rfw Gene More speciﬁcally, the 1(SNP/InDel-index) analysis preliminarily mapped Rfw to a 27.62–33.87 Mb region (6.25 Mb) on chromosome A09 (Figures 3C,D). exhibit polymorphism between reconstructed F- and S-pools. We identiﬁed 4 and 3 recombinant plants using ID-595A and SR30, respectively, which were the closest markers ﬂanking Rfw. Thus, Rfw was located within a genetic distance of 0.05 cM in B. napus. Additionally, 3 markers (SR10, TY21, and SR24) co-segregated with the Rfw locus in the assayed plants (no recombinant plants) (Figure 4). To verify and narrow down the Rfw region, 555 sterile F2 plants were used for further mapping. We selected 203 InDel markers in the candidate interval, and 13 of them were found to exhibit polymorphisms between reconstructed F- and S-pools from other fertile and sterile plants, respectively. These 13 markers (Figure 4) were used to assay the 555 sterile plants. We identiﬁed 4 and 2 recombinant plants using the markers ID-551A and IDC-705A, respectively, which were the closest markers ﬂanking Rfw. Thus, Rfw was further mapped to a 30.88–31.98 Mb region on chromosome A09 (Figure 4). Frontiers in Plant Science \| www.frontiersin.org Candidate Rfw Genes To isolate the candidate Rfw gene, all the genes (i.e., 26 genes, BnaA09g45870D–BnaA09g46120D) in the candidate region of the B. napus genome (Darmor-bzh and Zhongshuang11) were analyzed. One of these genes is a PPR gene (BnaA09g46030D). As the restorer line Hui01 originated from crossing WNJ01A CMS and B. rapa, and WNJ01A CMS originated from crossing B. juncea and B. napus, we also analyzed the candidate interval of chromosome A09 in B. rapa and B. juncea. There were 28 (Bra026863–Bra026890) and 26 (BjuA044103–BjuA044078) genes, including two PPR genes (Bra026882/Bra026884 and BjuA044085/BjuA044087) in the A To ﬁnely map Rfw, 3899 sterile F2 plants were further analyzed using the ID-551A and IDC-705A markers, and another 16 and 8 recombinant plants were identiﬁed, respectively. Next, we selected 130 markers (81 InDels, 24 SSRs, and 25 publicly available molecular markers) in the 1.1 Mb candidate region, and 16 dominant markers (Figure 4) were found to August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 6 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 3 \| Distributions of 1(SNPs/InDels-index) across B. napus chromosomes. (A) 1(InDels-index) plot obtained by subtraction of F-pool InDel-index from S-pool InDel-index for F2 population; (B) 1(SNPs-index) plot obtained by subtraction of F-pool SNP-index from S-pool SNP-index for F2 population; (C) 1(InDels-index) on chromosome A09; (D) 1(SNPs-index) on chromosome A09. The x-axes show the B. napus genome position. The y-axes show the index values; red lines in (A,B) represent index = \| 0.5\| . Each spot corresponds to an index value. The ﬁtted curves indicate the average values of 1(SNPs-index) or 1(InDels-index) based on sliding window analysis. The black arrows indicate the candidate regions. FIGURE 3 \| Distributions of 1(SNPs/InDels-index) across B. napus chromosomes. (A) 1(InDels-index) plot obtained by subtraction of F-pool InDel-index from S-pool InDel-index for F2 population; (B) 1(SNPs-index) plot obtained by subtraction of F-pool SNP-index from S-pool SNP-index for F2 population; (C) 1(InDels-index) on chromosome A09; (D) 1(SNPs-index) on chromosome A09. The x-axes show the B. napus genome position. The y-axes show the index values; red lines in (A,B) represent index = \| 0.5\| . Each spot corresponds to an index value. The ﬁtted curves indicate the average values of 1(SNPs-index) or 1(InDels-index) based on sliding window analysis. The black arrows indicate the candidate regions. genome of B. rapa cv. chiifu-401 and the AB genome of B. juncea var. tumida, respectively. Frontiers in Plant Science \| www.frontiersin.org Candidate Rfw Genes genes in this candidate region, there were two PPR genes [PPR-Nd1:(ORF19) and PPR-Nd2:(ORF20)], which were both only identiﬁed in Hui01 (not in WNJ01A CMS). PPR-Nd1 corresponded to BnaA09g46030D, Bra026882 and BjuA044087 while PPR-Nd2 corresponded to Bra026884 and BjuA044085 in the reference genomes. Because the majority of the cloned restorer genes belong to PPR family and the other genes except for PPR-Nd1 and PPR-Nd2 in this candidate interval have not been veriﬁed to be responsible for restoring fertility previously (Supplementary Table 3), the two PPR genes were further analyzed. To determine the candidate region sequence, we constructed a BAC library for Hui01. Of the nine positive pools, three were randomly selected, and four positive clones were then identiﬁed in these pools. The previously identiﬁed closest ﬂanking markers (ID-595A and SR30) were used to further conﬁrm the physical location of the inserted fragment in these four positive clones. One clone (M36A7E6) contained the ID-595A locus, and three clones (M16A6E7, M8A5D1, and M8A1H10) contained the SR30 locus (Supplementary Figure 1). Among them, the M36A7E6 and M16A6E7 clones were selected for Illumina HiSeq and PacBio sequencing to obtain the candidate region sequence in Hui01. The CDS length of PPR-Nd1 in Hui01 is 1923 bp (with no intron). In comparison, the CDS length is 1794 bp (with a 110-bp intron) in the homolog BnaA09g46030D of Darmor-bzh, 1887 bp (with a 29-bp intron) in BjuA044087 of B. juncea var. tumida, and 1815 bp (with a 100-bp intron) in Bra026882 of B. rapa cv. chiifu- 40. In the 5′ non-PPR domain region of PPR-Nd1, there was one InDel (–3 bp) and 24 SNPs compared to its Darmor-bzh homolog BnaA09g46030D, along with many InDels and SNPs in the PPR domain region (Supplementary Figure 2). Amino acid sequence The DNA insert size of M16A6E7 and M36A7E6 were 99,119 and 101,041 bp and the GC content were 35 and 37%, respectively. The interval between the closest ﬂanking markers ID-595A and SR30 was 99.75 kb, with a 15.87 kb diﬀerence between it and the corresponding region (115.62 kb) in the reference genome of Darmor-bzh. Among the 28 predicted August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 7 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 4 \| Fine mapping of the Rfw locus in Brassica napus. Candidate Rfw Genes The number below each marker indicates the number of recombinant plants (with respect to each given marker and the Rfw locus). Black arrows represent predicted genes in the 115.62 kb target region on chromosome A09 of B. napus. Red arrows indicate candidate Rfw genes. FIGURE 4 \| Fine mapping of the Rfw locus in Brassica napus. The number below each marker indicates the number of recombinant plants (with respect to each given marker and the Rfw locus). Black arrows represent predicted genes in the 115.62 kb target region on chromosome A09 of B. napus. Red arrows indicate candidate Rfw genes. genes or other ORFs (Chen and Liu, 2014). During the fertility restoration of CMS lines, the restorer genes regulate mitochondrial transcription, especially regarding the sterility genes (Li et al., 2018). To identify the mitochondrial genes regulated by Rfw, we selected 32 known protein-coding genes (Supplementary Table 2) based on two mitochondrial genomes (NCBI accession no. KJ461445 and JF920288) of B. juncea var. tumida and jiangpu, respectively (Chang S. et al., 2011; Zhao et al., 2016) and assessed their transcriptional levels in fertile and sterile F2 buds by RT-qPCR. alignment showed that there was an amino acid deletion and 10 amino acid substitutions in the 5′ non-PPR domain region that were speciﬁc to PPR-Nd1. Additionally, there were diﬀerences in motifs I and II of the PPR domain. The motif I sequence of PPR- Nd1 was identical to that of BjuA044087, but BnaA09g46030D and Bra026882 had long-fragment deletions and substitutions. The motif II sequence of PPR-Nd1 was identical to that of BnaA09g46030D and almost identical to that of Bra026882, while BjuA044087 had two deletions and one long-fragment substitution (Supplementary Figure 3). Most of the 32 known mitochondrial genes showed signiﬁcant diﬀerences in transcription, except for eight genes (Figure 5). Regarding mtETC complexes I, III, IV, and V, nine nad genes, three cox genes, and four atp genes, ccmC, and ccmFN2 were signiﬁcantly upregulated in fertile buds (and atp4, ccmB, ccmFC, and ccmFN1 were non-signiﬁcantly upregulated while cob was non-signiﬁcantly downregulated). Regarding the ribosome small subunit family, rps4, rps14, and rps3 were signiﬁcantly upregulated (and rps12 was non-signiﬁcantly upregulated). Regarding the ribosome large subunit family, only rpl16 was signiﬁcantly upregulated (and rpl2 and rpl5 were non-signiﬁcantly upregulated). Both tatC and matR were also signiﬁcantly upregulated. These diﬀerentially expressed genes may participate in fertility restoration after regulation by Rfw. Candidate Rfw Genes PPR-Nd2 corresponds to homologs Bra026884 of B. rapa and BjuA044085 of B. juncea (none of them have introns). There were a variety of InDels/SNPs in the 5′ region (Supplementary Figure 4) and multiple length reductions and SNPs in the 5′ terminal amino acid sequence (Supplementary Figure 5). As PPR-Nd2 is not present in the Darmor-bzh genome, PCR ampliﬁcation from 18 varieties of B. napus was carried out, and the gene was found to be present in some of them, such as Yangguang2009 (YG2009). The gene sequence in YG2009 was identical to that in Hui01. However, YG2009 appears not to be capable of restoring the fertility of WNJ01A CMS (data not shown), which means that this gene is not a candidate Rfw gene. DEGs Analysis Cytoplasmic male sterility in plants is generally caused by the formation of chimeric ORFs via mitochondrial genome rearrangement, which are usually co-transcribed with known To explore the molecular mechanism of Rfw regulation of fertility restoration, the transcriptomes of three fertile (Rfw1, Rfw2, and August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 8 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 5 \| Mitochondrial gene transcriptional levels in fertile and sterile buds (0–2 mm) based on RT-qPCR. The x-axis shows the 32 known protein-coding genes from B. juncea mitochondrial genomes (NCBI accession nos. KJ461445 and JF920288). Bna.Actin7 was used as the internal control. Error bars indicate standard deviation (n = 3). P < 0.05. FIGURE 5 \| Mitochondrial gene transcriptional levels in fertile and sterile buds (0–2 mm) based on RT-qPCR. The x-axis shows the 32 known protein-coding genes from B. juncea mitochondrial genomes (NCBI accession nos. KJ461445 and JF920288). Bna.Actin7 was used as the internal control. Error bars indicate standard deviation (n = 3). P < 0.05. FIGURE 5 \| Mitochondrial gene transcriptional levels in fertile and sterile buds (0–2 mm) based on RT-qPCR. The x-axis shows the 32 known protein-coding genes from B. juncea mitochondrial genomes (NCBI accession nos. KJ461445 and JF920288). Bna.Actin7 was used as the internal control. Error bars indicate standard deviation (n = 3). *P < 0.05. TABLE 1 \| Sequencing and mapped reads data. TABLE 1 \| Sequencing and mapped reads data. Samples Replicate Clean bases Clean reads GC content ≥Q30% Mapped reads Unique mapped reads Multiple map reads RFW RFW1 15,707,725,164 52,560,649 47.14% 91.28% 78,813,689 (74.97%) 74,676,697 (71.04%) 4,136,992 (3.94%) RFW2 18,482,309,220 61,894,692 47.24% 91.40% 92,665,358 (74.86%) 86,600,145 (69.96%) 6,065,213 (4.90%) RFW3 11,663,399,684 39,052,902 47.07% 90.65% 58,058,204 (74.33%) 54,879,219 (70.26%) 3,178,985 (4.07%) Rfw rfw1 13,347,442,234 44,679,819 47.26% 90.36% 66,133,402 (74.01%) 62,560,755 (70.01%) 3,572,647 (4.00%) rfw2 12,497,974,338 41,820,296 47.32% 90.33% 61,937,213 (74.05%) 58,653,225 (70.13%) 3,283,988 (3.93%) rfw3 14,728,819,074 49,231,140 47.06% 90.04% 72,921,088 (74.06%) 69,142,107 (70.22%) 3,778,981 (3.84%) 10, and 69 signiﬁcantly enriched DEGs (Kolmogorov–Smirnov P-value < 0.05) in the GO biological process (BP), cellular component (CC), and molecular function (MF) GO categories, respectively (Supplementary Table 4), comprising 47 functional groups (Figure 6). Rfw3) and three sterile (rfw1, rfw2, and rfw3) replicates from the F2 population were sequenced. After trimming the raw reads, the number of clean reads in the six samples ranged from 39,052,902 to 61,894,692. DEGs Analysis A total of 86.43 Gb clean bases was obtained, with each sample having > 11.66 Gb. The GC content was >47% and the Q30 percentage was >90%. Pearson’s correlation coeﬃcients regarding the transcriptional levels among the three sterile bud replicates were ≥0.935 and those among the three fertile bud replicates were ≥0.889 (Supplementary Figure 6). Therefore, the RNA-seq data were suitable for further analysis. g In the BP category, the dominant groups were metabolic process (63.04%), cellular process (61.73%), and single- organism process (65.02%). Regarding the most signiﬁcant term, speciﬁcation of organ position (GO:0010159), ﬁve downregulated DEGs belong to the plant-speciﬁc TF YABBY family (Supplementary Figure 7A), which is involved in the feedback regulation of gibberellin biosynthesis in rice (Dai et al., 2007). In the CC category, the dominant groups were cell (83.33%) and cell part (83.33%). Regarding the plant-type cell wall term (GO:0009505), seven aspartic proteinase (AP) and nine pectin methyl-esterase (PME) DEGs were enriched in this term (Supplementary Figure 7A). Aspartic protease participates in stamen development and cell death in rice (Niu et al., 2013; Ko et al., 2014) and PME genes are involved in pollen development and pollen tube growth in B. campestris (Xiong et al., 2019). In the MF category, the dominant groups were binding (49%) and catalytic activity (44%). Regarding the sugar transmembrane transporter activity term (GO:0051119), six upregulated SWEET DEGs in fertile buds were enriched in this term (Supplementary Figure 7A). SWEET1/8, which are highly expressed in Arabidopsis stamens, play important roles in glucose transport for pollen tube growth and starch accumulation in pollen (Chen et al., 2010). Next, >74% of the clean reads (>70% unique mapped reads and 3.84–4.90% multiple mapped reads) were mapped to the B. napus cv. Darmor-bzh reference genome (Table 1). We detected 104,348 genes in the RNA-seq data from the six replicates (3229 genes were novel, based on comparison to the reference genome). Subjecting the genes to BLASTx searches against COG, GO, KEGG, Swiss-Prot, eggNOG, and NR databases resulted in 102,843 annotated genes, of which 29253 (28.44%), 78382 (76.22%), 32214 (31.32%), 65560 (63.75%), 2305 (2.24%), and 102796 (99.95%) genes were aligned to the proteins in the six databases, respectively. There were 1679 DEGs (1023 upregulated and 656 downregulated) between the fertile and sterile buds. The number of upregulated DEGs in the fertile buds was much higher than the number of downregulated DEGs. GO and KEGG Analyses of DEGs GO and KEGG Analyses of DEGs y Of the 1518 GO annotated DEGs (905 upregulated and 613 downregulated DEGs) in the fertile buds, there were 130, The 581 KEGG-annotated DEGs (382 upregulated and 199 downregulated) were assigned to 107 KEGG pathways. August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 9 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 6 \| GO annotation of differentially expressed genes (DEGs) between fertile and sterile buds. The x-axis shows the terms in the three GO categories, the left y-axis shows the percentage of DEGs enriched in a speciﬁc term (out of all the DEGs in each category), and the right y-axis shows the number of DEGs enriched in a speciﬁc term. FIGURE 6 \| GO annotation of differentially expressed genes (DEGs) between fertile and sterile buds. The x-axis shows the terms in the three GO categories, the left y-axis shows the percentage of DEGs enriched in a speciﬁc term (out of all the DEGs in each category), and the right y-axis shows the number of DEGs enriched in a speciﬁc term buds may be associated with CMS. Among the DEGs, 261 speciﬁcally expressed genes were identiﬁed (9 and 252 in sterile and fertile buds, respectively). GO analysis of the fertile bud-speciﬁc DEGs showed that the enriched BP terms included pollen exine formation (GO:0010584), carbohydrate metabolic process (GO:0005975), and SKP1-CUL-F-box (SCF) complex assembly (GO:0010265). The enriched CC terms included integral component of plasma membrane (GO:0005887), cullin-RING ubiquitin ligase complex (GO:0031461), and condensed nuclear chromosome (GO:0000794). Lastly, the enriched MF terms included sugar transmembrane transporter activity (GO:0051119), protein-lysine N-methyltransferase activity (GO:0016279), glucan endo-1,3-beta-D-glucosidase activity (GO:0042973), cation binding (GO:0043169), and oxidoreductase activity (acting on the aldehyde or oxo group of donors, NAD or NADP as acceptor, GO:0016620) (Figure 8A). The ﬁve most enriched categories were cellular processes (2.6%), environmental information processing (5.3%), genetic information processing (14.5%), metabolism (76.1%), and organismal systems (1.5%) (Supplementary Table 5). Pollen abortion and fertility restoration were shown to involve transport and catabolism, signal transduction, amino acid metabolism, carbohydrate metabolism, energy metabolism, and metabolism of terpenoids and polyketides, etc. (Supplementary Table 5). More speciﬁcally, the upregulated DEGs were signiﬁcantly enriched in 12 pathways, like lysine degradation (ko00310) and phenylalanine metabolism (ko00360) (Figure 7A). The downregulated DEGs were signiﬁcantly enriched in 18 pathways, including cutin, suberine, and wax biosynthesis (ko00073), pentose and glucuronate interconversions (ko00040), and fatty acid metabolism (ko01212) (Figure 7B). GO and KEGG Analyses of DEGs Enriched (A) GO terms and (B) KEGG pathways of DEGs FIGURE 8 \| GO and KEGG analyses of differentially expressed genes (DEGs) speciﬁc to fertile buds. Enriched (A) GO terms and (B) KEGG pathways of DEGs speciﬁc to fertile buds FIGURE 8 \| GO and KEGG analyses of differentially expressed genes (DEGs) speciﬁc to fertile buds. Enriched (A) GO terms and (B) KEGG pathways of DEGs speciﬁc to fertile buds. nucleotide excision repair (ko03420), and base excision repair (ko03410) (Figure 8B). photomorphogenesis by interacting with the red/far-red light photoreceptor phytochrome B and regulates the cytokinin response pathway (Mira-Rodado et al., 2007). After using The Arabidopsis Information Resource (TAIR) database to annotate the DEGs, we discovered that 44 DEGs (35 upregulated and 9 downregulated in fertile buds) were related to anther and pollen development (Supplementary Table 6). The upregulated genes included ALA6, UGE3, DL1C, COX11, PSS1, ARID1, P5CS1, MAZ1, ABCG6, INP1, SPS2F, KOM, UPEX1, BHLH010, BHLH089, MYB108, CALS5, PRX9, PRX40, TAP35/TAP44, AMS, SWEET8, and PIRL1. Among them, MAZ1, ABCG6, SPS2F, KOM, and UPEX1 are involved in pollen wall (including pollen exine) formation. Four INP1 genes play important roles in pollen surface aperture formation. Two CALS5 genes (which encode callose synthase 5) are responsible for the synthesis of callose that is deposited at the primary cell wall of meiocytes, tetrads, and microspores. TAP35/TAP44 and AMS GO and KEGG Analyses of DEGs The ﬁve most enriched categories were cellular processes (2.6%), environmental information processing (5.3%), genetic information processing (14.5%), metabolism (76.1%), and organismal systems (1.5%) (Supplementary Table 5). Pollen abortion and fertility restoration were shown to involve transport and catabolism, signal transduction, amino acid metabolism, carbohydrate metabolism, energy metabolism, and metabolism of terpenoids and polyketides, etc. (Supplementary Table 5). More speciﬁcally, the upregulated DEGs were signiﬁcantly enriched in 12 pathways, like lysine degradation (ko00310) and phenylalanine metabolism (ko00360) (Figure 7A). The downregulated DEGs were signiﬁcantly enriched in 18 pathways, including cutin, suberine, and wax biosynthesis (ko00073), pentose and glucuronate interconversions (ko00040), and fatty acid metabolism (ko01212) (Figure 7B). Additionally, 55 fertile bud-speciﬁc DEGs were annotated with KEGG pathways. The four most signiﬁcant enriched pathways were protein processing in endoplasmic reticulum (ko04141), ubiquitin-mediated proteolysis (ko04120), Additionally, 55 fertile bud-speciﬁc DEGs were annotated with KEGG pathways. The four most signiﬁcant enriched pathways were protein processing in endoplasmic reticulum (ko04141), ubiquitin-mediated proteolysis (ko04120), DEGs Speciﬁc to Fertile and Sterile Buds Diﬀerentially expressed genes speciﬁc to fertile buds may be involved in fertility restoration, while DEGs speciﬁc to sterile August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 10 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 7 \| KEGG pathway annotation of differentially expressed genes (DEGs). Top 20 enriched KEGG pathways for (A) upregulated and (B) downregulated DEGs between fertile and sterile buds. The rich factor is the ratio of the number of DEGs to the total number of genes in each pathway. The color and size of the dots indicate the P-value and the number of DEGs mapped to the pathway, respectively. FIGURE 7 \| KEGG pathway annotation of differentially expressed genes (DEGs). Top 20 enriched KEGG pathways for (A) upregulated and (B) downregulated DEGs between fertile and sterile buds. The rich factor is the ratio of the number of DEGs to the total number of genes in each pathway. The color and size of the dots indicate the P-value and the number of DEGs mapped to the pathway, respectively. FIGURE 8 \| GO and KEGG analyses of differentially expressed genes (DEGs) speciﬁc to fertile buds. Enriched (A) GO terms and (B) KEGG pathways of DEGs speciﬁc to fertile buds. yses of differentially expressed genes (DEGs) speciﬁc to fertile buds. DEGs Related to Hormones and Anther/Pollen Development Stamen development is often modulated by hormone biosynthesis and signaling pathways, and certain hormone- related mutants often display male sterility (Song et al., 2013). We identiﬁed 31 hormone-related DEGs. Of these, 3, 9, 7, 4, 1, 2, and 5 were related to gibberellin, auxin, abscisic acid (ABA), cytokinin, ethylene, jasmonic acid (JA) and salicylic acid, respectively (Supplementary Figure 7B). All DEGs related to JA and salicylic acid were upregulated in the fertile compared to the sterile buds. Additionally, BnaA09g48160D, which encodes the homologous protein Arabidopsis response regulator 4 (ARR4), was speciﬁcally expressed in fertile buds. ARR4 modulates August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 11 Fertility Restoration for WNJ01A CMS Yang et al. participate in tapetal cell development. SWEET8 and PIRL1 function in microspore development. Lastly, ﬁve BHLH and MYB TF DEGs (four of which were upregulated) are involved in anther development and dehiscence. and accuracy of the RNA-seq and the pathway enrichment results (Figure 10). DISCUSSION The downregulated genes included POE1, IPE1, and BHLH010. Among them, all seven DEGs of the POE1 family (which encode Pollen Ole e 1 allergen and extensin family proteins) were downregulated in fertile buds. They can be epigenetically controlled via histone H3 lysine 27 trimethylation (H3K27me3) (Hu et al., 2014). Mapping of Rfw Gene Rfw was preliminarily mapped to 27.62–33.87 Mb on chromosome 9 by BSA-seq. Interestingly, there are many PPR genes in this region, including two restorer genes, Rfp (identiﬁed in Polima CMS; locus: 31.64–31.67 Mb) (Liu et al., 2012) and Rfn (identiﬁed in nap CMS; locus: 31.79–31.81 Mb) (Liu et al., 2017). Recent research has revealed that a highly dense cluster of restorer-of-fertility-like (RFL) genes exists in this region on chromosome A09 and on a collinear region on chromosome C8 in B. napus (Ning et al., 2020). A similar PPR cluster is present in the collinear region of Arabidopsis chromosome A1. The main driving force for evolution of PPI Network The PPI network contained 6077 interactions among 658 proteins, encoded by 423 upregulated, and 235 downregulated DEGs. There were ten modules, which were composed of 610, 18, 9, 5, 3, 3, 3, 3, 2, and 2 DEGs, respectively (Supplementary Figures 8A–J). The 10 hub genes with the top degree scores, were BnaA09g56400D, BnaA10g18210D, BnaA10g18220D, BnaC09g41740D, BnaC04g03080D, BnaA05g25330D, BnaC03g51080D, BnaC09g01600D, BnaA01g10430D, and BnaCnng18680D. The ﬁrst 4 of these had the highest degree scores and were all upregulated in fertile buds (Supplementary Figure 8K). They all encode RAD23d (ubiquitin receptor radiation sensitive23d), which plays essential roles in the cell cycle, morphology, and fertility of higher plants by delivering ubiquitinated substrates to the 26S proteasome (Farmer et al., 2010). Rfw may upregulate these RAD23d hub genes so that they could modulate the expressions of their interacting genes, thereby mediating fertility restoration. The remaining six hub genes were annotated as MPK6, HAOX2, bt1, and GSO1 (three times), respectively. Our RT-qPCR results regarding 32 known protein-coding mitochondrial genes indicate that most of these genes were signiﬁcantly upregulated in fertile buds. For instance, genes related to mtETC complexes I, III, IV, and V were markedly upregulated in fertile buds. This implies that the mtETC was disrupted (with downregulated expression) in sterile buds. In contrast, in fertile plants, this disruption was prevented, restoring normal pollen development. Certain CMS types also display the male-sterile phenotype due to mtETC disruption (Fujii et al., 2010), such as sugar beet CMS-G (Ducos et al., 2001), pepper CMS-Peterson (Ji et al., 2013), rice CMS-HL (Peng et al., 2010; Wang et al., 2013), sunﬂower CMS-PET1 (Sabar et al., 2003), and pearl millet CMS plants (Kale and Munjal, 2005). Only one mtETC-related gene (cob) was downregulated (non-signiﬁcantly) in fertile buds. The DEGs that interacted with the hub genes were subjected to GO and KEGG analyses. Pathways related to carbohydrate metabolism (such as N-glycan biosynthesis, glycolysis/gluconeogenesis, the TCA cycle, and starch and sucrose metabolism) tended to be upregulated. The JA response pathway was also upregulated. Other pathways potentially related to male fertility were also dramatically upregulated, such as ubiquitin-mediated proteolysis, purine metabolism, meiotic DNA double-strand break formation, protein processing in endoplasmic reticulum, and endocytosis. In addition, RNA- related pathways (spliceosome, RNA transport, and aminoacyl- tRNA biosynthesis) were upregulated. In contrast, pentose and glucuronate interconversions, fatty acid metabolism, and cutin, suberine, and wax biosynthesis pathways were highly downregulated (Figure 9). Cytological Characteristics and Mitochondrial Gene Transcription in Sterile and Fertile Buds The WNJ01A CMS line displayed a complete pollen abortion phenotype, and its fertility could be restored by the restorer line Hui01. By observing paraﬃn-embedded sections of sterile and fertile buds, we found that the sporogenous cell stage was the critical stage regarding pollen degeneration, with major sporogenous cells exhibiting loose and irregular arrangements in sterile anthers. The pollen abortion stage of WNJ01A CMS is earlier than in pol CMS and ogu CMS. Regarding pol CMS, the pollen abortion stage is the polarization stage of the archespore, in which sporogenous cells do not diﬀerentiate and normal tetrads are not produced in the sterile anthers (An et al., 2014). Regarding ogu CMS, the pollen abortion stage involves the transition from the tetrad to single-nucleus pollen. The pollen development is impaired due to premature death of tapetal cells at the vacuolate microspore stage (Gonzalez-Melendi et al., 2008). Additionally, the pollen-abortion stage of WNJ01A CMS is later than that of hau CMS; hau CMS stamen primordia lack normal polarization and form petal primordia at the diﬀerentiation stage of stamen primordia (Wan et al., 2008). RT-qPCR Validation To validate the RNA-seq results, 14 DEGs in enriched pathways and related to fertility restoration were subjected to RT-qPCR (Supplementary Table 7). Of these DEGs, 10 were upregulated in fertile buds, comprising LKR, ALDH2B4, PRX40, PRX9, SWEET8, CALS5, CUL1, SKP1, RAD23A, and RAD23D. The remaining four DEGs were downregulated in fertile buds, comprising CYP86A4, CER1, HTH, and POE1. The RT-qPCR results showed that their expression tendencies were consistent with those from RNA-seq analysis, indicating the reliability August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 12 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 9 \| Protein–protein interaction (PPI) network related to the regulator Rfw. Red and green indicate upregulated and downregulated differentially expressed genes [DEGs; the shade reﬂects log2(fold change)]. FIGURE 9 \| Protein–protein interaction (PPI) network related to the regulator Rfw. Red and green indicate upregulated and downregulated differentially expressed genes [DEGs; the shade reﬂects log2(fold change)]. FIGURE 9 \| Protein–protein interaction (PPI) network related to the regulator Rfw. Red and green indicate upregulated and down genes [DEGs; the shade reﬂects log2(fold change)]. raction (PPI) network related to the regulator Rfw. Red and green indicate upregulated and downregulated differentially expressed log2(fold change)] FIGURE 10 \| RT-qPCR validation. S, sterile sample; F, fertile sample. FIGURE 10 \| RT-qPCR validation. S, sterile sample; F, fertile sample. FIGURE 10 \| RT-qPCR validation. S, sterile sample; F, fertile sample. sequence variation) related to fertility restoration in relation to various CMS systems. this region may relate to the segmental duplication and retrotransposition processes of An subgenomes (Gaborieau and Brown, 2016). PPR genes are probably derived from a few common ancestors and have conserved regulatory roles regarding mitochondrial genes. During evolution, the number of PPR genes may have gradually increased and then they may have developed functional speciﬁcity (due to The PPR gene PPR-Nd1 in the Hui01 line (which is a homolog of BnaA09g46030D in Darmor-bzh) is a candidate Rfw gene in the ﬁnely mapped Rfw interval. BnaA09g46030D is known to contain mitochondrial localization signals and is highly upregulated in buds, indicating that it can be regulated and functions in The PPR gene PPR-Nd1 in the Hui01 line (which is a homolog of BnaA09g46030D in Darmor-bzh) is a candidate Rfw gene in the ﬁnely mapped Rfw interval. Frontiers in Plant Science \| www.frontiersin.org RT-qPCR Validation BnaA09g46030D is known to contain mitochondrial localization signals and is highly upregulated in buds, indicating that it can be regulated and functions in August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org Frontiers in Plant Science \| www.frontiersin.org 13 Fertility Restoration for WNJ01A CMS Yang et al. FIGURE 11 \| Mechanistic model of the role of the sterile gene of WNJ01A CMS and the role of Rfw in fertility restoration. In the heat map, the left-hand side represents sterile buds and the right-hand side represents fertile buds. FIGURE 11 \| Mechanistic model of the role of the sterile gene of WNJ01A CMS and the role of Rfw in fertility restoration. In the heat map, the left-hand side represents sterile buds and the right-hand side represents fertile buds. model of the role of the sterile gene of WNJ01A CMS and the role of Rfw in fertility restoration. In the heat map, the left-hand side the right-hand side represents fertile buds. Frontiers in Plant Science \| www.frontiersin.org August 2021 \| Volume 12 \| Article 721980 14 Fertility Restoration for WNJ01A CMS Yang et al. pollen cell wall formation during fertility restoration of WNJ01A CMS. mitochondria (Wang H. et al., 2021). However, comparing the CDS sequences of PPR-Nd1 and Rfh (BnaA09g46030D, which restores the sterile phenotype of hau CMS, Wang H. et al., 2021), there are multiple SNPs and one InDel (–3 bp) diﬀerence, and there are also diﬀerences in the amino acid sequences. Critically, whether PPR-Nd1 can restore the fertility of WNJ01A CMS, and the degree of pollen fertility after restoration, remain unknown. In addition, the maintaining and restoring relationship between WNJ01A CMS and Hau CMS is required to be further analyzed. Our results also showed that the downregulated DEGs were enriched in cutin, suberine, and wax biosynthesis (ko00073) and this pathway is involved in normal mature pollen development in B. napus (Shi et al., 2020). The DEGs included genes in the CYP86A, CER1-like, and glucose-methanol-choline (GMC) families. CYP86A are highly upregulated in mature Arabidopsis ﬂowers, and they function as fatty acid ω-hydroxylases (Duan and Schuler, 2005) and may play indispensable roles in pollen tube growth (Koiwai and Matsuzaki, 1988; Wolters- Arts et al., 1998). Cuticular wax is composed of very-long- chain fatty acids (VLCFAs) and VLC alkane biosynthesis is regulated by CER1-like1 speciﬁc cofactors (Preuss et al., 1993; Lee and Suh, 2013; Wu et al., 2019). DEGs Related to Pollen and Anther Development During the fertility restoration of WNJ01A by restorer line Hui01, 44 DEGs were conﬁrmed to be involved in pollen and anther development. In fertile buds, multiple genes that play predominant roles in tapetum, pollen wall, and callose development were upregulated. These DEGs included two bHLH TFs (BHLH089 and BHLH010), two extensin peroxidases (PRX9 and PRX40), tapetum-speciﬁc TAP35/TAP44, and two CALS5 genes. During pollen formation, tapetum cells play central roles in callose degradation, pollen exine formation, and the provision of various nutrients for pollen development (Han et al., 2021). Our cytological characterization of the abortive anthers showed that the tapetum cells (which surround the microspores and provide crucial enzymes and nutrients for microsporogenesis and pollen wall development, Ferguson et al., 2017) were degraded. This likely explains why the microspores were degraded in the sterile buds. Previous studies have highlighted that, during the microspore stage, two AMS TFs (Xu et al., 2014), two SWEET8 (also called RPG1) genes (Guan et al., 2008; Chen et al., 2010), PSS1 (Forsthoefel et al., 2010; Yamaoka et al., 2011), and P5CS1 (Mattioli et al., 2018) are essential for pollen development. These genes were all upregulated in the fertile buds. In fact, most of the DEGs related to pollen and anther development were upregulated in the fertile buds, indicating that they were modulated by Rfw to achieve normal pollen and anther development. p y LKR/SDH encode the ﬁrst two key enzymes in the lysine degradation pathway (Stepansky et al., 2006). They are upregulated in ﬂowers and strongly upregulated by ABA and JA treatment in Arabidopsis. ALDHs catalyze oxidation of α-aminoadipic semialdehyde during lysine degradation. Additionally, they are upregulated in Arabidopsis anthers and can eliminate lesions induced by toxic aldehydes and reactive oxygen species (Shin et al., 2009; Brocker et al., 2010; Shen et al., 2012; Hou and Bartels, 2015; Zhao et al., 2018). In maize T-CMS, the restorer gene RF2 encodes an ALDH that can complement the lack of ALDH activity in its mitochondria to reverse the male sterility caused by the CMS protein URF13 (Liu et al., 2001). ASH/SDG genes are negative regulators of H3K27me3 (which is known to downregulate genes involved in cellular fate maintenance during development in plants and animals) (Cartagena et al., 2008; Lee et al., 2015; Crevillen, 2020; Cai et al., 2021). ASH/SDG is located upstream of ALDH in the lysine degradation pathway (ko00310). KEGG Analysis of DEGs In plants, high lysine levels can be toxic to cells (Arruda et al., 2000). For instance, Arabidopsis seeds with superabundant lysine exhibited delayed germination and seedling establishment (Zhu and Galili, 2003). Our KEGG analysis of the upregulated DEGs showed that the lysine degradation pathway (ko00310) was enriched, which involved DEGs in the lysine-ketoglutarate reductase/saccharopine dehydrogenase (LKR/SDH), aldehyde dehydrogenase (ALDH), and histone- lysine N-methyltransferase/SET domain group (ASH/SDG) families. These families balance Lys levels in plants (Tang et al., 2000; Stepansky and Galili, 2003). RT-qPCR Validation The GMC protein HTH regulates cutin biosynthesis and postgenital organ fusion during ﬂower development (Krolikowski et al., 2003; Xu et al., 2017). Frontiers in Plant Science \| www.frontiersin.org DEGs Related to Pollen and Anther Development Hence, ASH/SDG probably mediate fertility restoration by aﬀecting the H3K27me3 of the related genes. Our results indicate that upregulation of the lysine degradation pathway is very important to maintain the normal lysine level for anther development in plants with restored fertility. Among the DEGs related to pollen and anther development, three types of genes were signiﬁcantly downregulated, comprising seven POE1 (Pollen Ole e 1 allergen and extensin), one IPE1 (irregular pollen exine1) gene, and one bHLH10 gene. Recent research has suggested that at least 13 AtPOE1 gene loci were modiﬁed by H3K27me3 in Arabidopsis (Hu et al., 2014). Importantly, POE1 protein may participate in pollen tube emergence and guidance (Tang et al., 2002; Alche et al., 2004; Hamman-Khalifa et al., 2008), and H3K27me3 is involved in pollen sporophyte Another signiﬁcantly enriched pathway, phenylalanine metabolism (ko00360) involves the peroxidase (PRX) and copper amine oxidase (CuAO) families. PRX family members, such as PRX9 and PRX40, are essential extensin peroxidases that maintain the integrity of the tapetum and microspore cell walls during anther development in Arabidopsis (Jacobowitz et al., 2019). CuAOs mediate plant cell wall formation, maturation, and programmed cell death (Planas- Portell et al., 2013; Tavladoraki et al., 2016). Our results indicate that the phenylalanine metabolic pathway mediates Frontiers in Plant Science \| www.frontiersin.org August 2021 \| Volume 12 \| Article 721980 15 Fertility Restoration for WNJ01A CMS Yang et al. we propose a model of the mechanism of fertility restoration regulated by the restorer gene Rfw (Figure 11). development (Hoﬀmann and Palmgren, 2013). Regarding IPE1, the putative oxidative pathway of ω-hydroxy fatty acids, which depends on IPE1, plays signiﬁcant roles in anther cuticle and pollen exine formation in maize (Chen et al., 2017). In mitochondria, the male sterility gene downregulates mtETC genes, which causes energy deﬁciency and results in the formation of ﬂower buds exhibiting male sterility. The nuclear genome possesses an Rfw restorer gene, which can inhibit the expression of the male sterility gene at the level of transcription or translation or by another mechanism. The Rfw gene ensures that the function of the mtETC genes and mitochondrial energy production return to normal. FUNDING This study was supported by the National Key Research and Development Program of China (grant 2016YFD0100202). This study was supported by the National Key Research and Development Program of China (grant 2016YFD0100202). Role of RAD23d in PPI Network In the PPI network, the hub genes with the highest degree scores were all RAD23d genes. The RAD23 family provides an essential connection between ubiquitylated proteins and the 26S proteasome in Arabidopsis (Lahari et al., 2017). In addition, RAD23 has been identiﬁed as a ubiquitin-like (UBL)/ubiquitin-associated (UBA) protein, which is key for ubiquitin-mediated protein degradation (Andersson et al., 2005; Zuo and Mahajan, 2005; Dantuma et al., 2009). The UPS is extremely important in plant growth and development as it regulates proteasome-dependent protein turnover (Lowe et al., 2006; Wade and Auble, 2010). During the development of anthers with abnormally active meiosis, DNA repair is essential. Rad protein is involved in recognition of and binding to damaged DNA to correct DNA lesions (Gallego et al., 2000). Rad23 can also eﬃciently initiate the cell cycle and nucleotide excision repair-like repair pathway (Lahari et al., 2017; Okeke et al., 2020). Interestingly, the UPS not only removes incorrectly folded proteins, but also ensures appropriate spatial and temporal protein distributions (Medina et al., 2012; Liang et al., 2014; Chen and Walters, 2016). For example, RAD23B is key in pollen development as it controls the turnover of the key cell cycle protein KIP- related protein 1 (KRP1) (Li et al., 2020). In summary, RAD23d proteins may mediate fertility restoration of WNJ01A, regulated by Rfw, by repairing damaged DNA and degrading sterility-related proteins via the UPS, along with potentially regulating the degradation of other proteins and modulating hormone signals. DEGs Speciﬁc to Fertile Buds Among the KEGG pathways enriched in the DEGs speciﬁc to fertile buds, the ubiquitin–proteasome pathway was identiﬁed, with three SKP1-like genes, three Cullin 1 (CUL1) genes, and three RAD23D genes being speciﬁcally detected in fertile buds. In lily (Lilium longiﬂorum), three LSK1–3 (SKP1- like) genes play critical roles in regulating pollen tube elongation (Chang et al., 2009). In Arabidopsis, AtCUL1 plays an important role in JA signaling (Ren et al., 2005). Furthermore, in Arabidopsis, SCF complexes regulate JA- responsive genes related to pollen development (Devoto et al., 2002; Xu et al., 2002). Thus, our results indicate that fertility restoration of WNJ01A CMS relies on the ubiquitin-proteasome system (UPS). p As a result of the eﬀect of Rfw, RAD23 might be initially up- regulated in fertile buds. Then, the genes LKR/SDH, ALDH7B4, CUL1, SKP1 and their related pathways of lysine degradation, protein processing in endoplasmic reticulum are upregulated. In contrast, the genes CYP86A, HTH and their related pathways of cutin, suberine, and wax biosynthesis are downregulated. Later, the genes related to tapetum, pollen exine wall, microspore, and callose development, such as SWEET8, CALS5, TAP35/44, PRX9, PRX40, AMS, etc. are upregulated and POE1, IPE1, etc. are downregulated. Ultimately, the biological processes of anther and pollen development exhibit normal function and the ﬂower buds restored fertility. DATA AVAILABILITY STATEMENT The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI BioProject, PRJNA739480. AUTHOR CONTRIBUTIONS CL and QY conceived the experiment. FW, CZ, and JW provided advice on the experimental design. QY performed most of the experiments and analyzed the data. XN participated in paraﬃn-embedded sections experiments. JX participated in phenotypic investigate. FH participated in data analysis. CL and QY wrote the manuscript. All the authors reviewed and approved this submission. REFERENCES Chang, F., Wang, Y., Wang, S., and Ma, H. (2011). Molecular control of microsporogenesis in Arabidopsis. Curr. Opin. Plant Biol. 14, 66–73. doi: 10. 1016/j.pbi.2010.11.001 Alche, J. D., M’rani-Alaoui, M., Castro, A. J., and Rodriguez-Garcia, M. I. (2004). Ole e 1, the major allergen from olive (Olea europaea L.) pollen, increases its expression and is released to the culture medium during in vitro germination. Plant Cell Physiol. 45, 1149–1157. doi: 10.1093/pcp/pch127 Chang, L. C., Guo, C. L., Lin, Y. S., Fu, H., Wang, C. S., and Jauh, G. Y. (2009). Pollen-speciﬁc SKP1-like proteins are components of functional scf complexes and essential for lily pollen tube elongation. Plant Cell Physiol. 50, 1558–1572. doi: 10.1093/pcp/pcp100 Plant Cell Physiol. 45, 1149–1157. doi: 10.1093/pcp/pch127 An, H., Yang, Z., Yi, B., Wen, J., Shen, J., Tu, J., et al. (2014). Comparative transcript proﬁling of the fertile and sterile ﬂower buds of pol CMS in B. napus. BMC Genomics 15:258. doi: 10.1186/1471-2164-15-258 Chang, S., Yang, T., Du, T., Huang, Y., Chen, J., Yan, J., et al. (2011). Mitochondrial genome sequencing helps show the evolutionary mechanism of mitochondrial genome formation in Brassica. BMC Genomics 12:497. doi: 10.1186/1471-2164- 12-497 Andersson, H. A., Passeri, M. F., and Barry, M. A. (2005). Rad23 as a reciprocal agent for stimulating or repressing immune responses. Hum. Gene Ther. 16, 634–641. doi: 10.1089/hum.2005.16.634 Chen, L., and Liu, Y. G. (2014). Male sterility and fertility restoration in crops. Annu. Rev. Plant Biol. 65, 579–606. doi: 10.1146/annurev-arplant-050213- 040119 Arruda, P., Kemper, E. L., Papes, F., and Leite, A. (2000). Regulation of lysine catabolism in higher plants. Trends Plant Sci. 5, 324–330. doi: 10.1016/s1360- 1385(00)01688-5 Chen, L. Q., Hou, B. H., Lalonde, S., Takanaga, H., Hartung, M. L., Qu, X.-Q., et al. (2010). Sugar transporters for intercellular exchange and nutrition of pathogens. Nature 468, 527–532. doi: 10.1038/nature09606 Bohra, A., Jha, U. C., Adhimoolam, P., Bisht, D., and Singh, N. P. (2016). Cytoplasmic male sterility (CMS) in hybrid breeding in ﬁeld crops. Plant Cell Rep. 35, 967–993. doi: 10.1007/s00299-016-1949-3 Chen, X., and Walters, K. J. (2016). H-1, N-15, C-13 resonance assignments for Saccharomyces cerevisiae Rad23 UBL domain. Biomol. NMR Assign. 10, 291–295. doi: 10.1007/s12104-016-9686-7 Bonhomme, S., Budar, F., Lancelin, D., Small, I., Defrance, M. C., and Pelletier, G. (1992). Sequence and transcript analysis of the Nco2.5 Ogura-speciﬁc fragment correlated with cytoplasmic male sterility in Brassica cybrids. Mol. Gen. Genet. 235, 340–348. doi: 10.1007/BF00279379 Chen, X. Mechanistic Model We thank, Qinyong Yang (Huazhong Agricultural University) for his help in BSA-seq analysis, Yongtai Yin (Huazhong University of Science and Technology) for his help in PPI analysis, Qinghua Zhang (Huazhong Agricultural University) for his help in Illumina sequencing, and After transcriptome sequencing of sterile and fertile ﬂower buds, the fertile/sterile bud-speciﬁc DEGs, pollen/anther development- and hormone-related DEGs, and DEGs in the PPI network were subjected to GO and KEGG analysis. Based on the results, August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 16 Fertility Restoration for WNJ01A CMS Yang et al. Yue Jia (Huazhong Agricultural University) for his help in transcriptome analysis. Yue Jia (Huazhong Agricultural University) for his help in transcriptome analysis. Yue Jia (Huazhong Agricultural University) for his help in transcriptome analysis. Supplementary Figure 5 \| Full-length protein region alignment between PPR-Nd2, BjuA044085 and Bra026884. Black background: consensus sequences; red line: PPR motif region. Supplementary Figure 6 \| Pearson correlation analysis of the transcript levels in three fertile plant buds (RFW1, RFW2, and RFW3) and three sterile plant buds (rfw1, rfw2, and rfw3). SUPPLEMENTARY MATERIAL The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2021. 721980/full#supplementary-material The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2021. 721980/full#supplementary-material Supplementary Figure 7 \| Heatmap of DEGs. (A) YABBY, PEM, APs, and SWEET family differentially expressed genes (between fertile and sterile buds) that were enriched in the top ten GO terms. BP, biological process; CC, cellular component; MF, molecular function. (B) Plant hormone-related genes that were differentially expressed between fertile and sterile buds. GA, gibberellin; IAA, auxin; ABA, abscisic acid; CTK, cytokinin; ET, ethylene; JA, jasmonic acid; SA, salicylic acid. Supplementary Figure 1 \| Two sequenced BAC clones (four replicates each) veriﬁed using molecular markers. The closest ﬂanking molecular markers (ID-595A and SR30) and a co-segregating marker (TY21) in the ﬁnely mapped candidate interval were identiﬁed by PCR ampliﬁcation. PCR products were detected using 1.2% agarose gel. Supplementary Figure 1 \| Two sequenced BAC clones (four replicates each) veriﬁed using molecular markers. The closest ﬂanking molecular markers (ID-595A and SR30) and a co-segregating marker (TY21) in the ﬁnely mapped candidate interval were identiﬁed by PCR ampliﬁcation. PCR products were detected using 1.2% agarose gel. Supplementary Figure 8 \| Protein–protein interaction (PPI) network analysis of DEGs. (A–J) Ten modules in PPI network. (K) Heatmap of four hub genes. Supplementary Figure 2 \| Full-length CDS region alignment between PPR-Nd1, BnaA09g46030D, BjuA044087, and Bra026882. Black background: consensus sequences; red line: PPR motif region. Supplementary Table 1 \| Primers used in this study. Supplementary Table 2 \| Primers for 32 known protein-coding genes. Supplementary Table 3 \| Gene annotation of 99.75 kb DNA fragment. Supplementary Figure 3 \| Full-length protein region alignment between PPR-Nd1, BnaA09g46030D, BjuA044087, and Bra026882. Black background: consensus sequences; red line: PPR motif region; purple line: motif I; blue line: motif II. Supplementary Table 4 \| GO annotation of DEGs. Supplementary Table 5 \| KEGG pathway annotation of DEGs. Supplementary Figure 4 \| Full-length CDS region alignment between PPR-Nd2, BjuA044085, and Bra026884. Black background: consensus sequences; red line: PPR motif region. Supplementary Table 6 \| DEGs related to pollen and anther development. Supplementary Table 7 \| DEGs selected for RT-qPCR veriﬁcation. REFERENCES RUPTURED POLLEN GRAIN1, a member of the MtN3/saliva gene family, is crucial for exine pattern formation and cell integrity of microspores in Arabidopsis. Plant Physiol. 147, 852–863. doi: 10.1104/pp.108.118026 L’Homme, Y., Stahl, R. J., Li, X. Q., Hameed, A., and Brown, G. G. (1997). Brassica nap cytoplasmic male sterility is associated with expression of a mtDNA region containing a chimeric gene similar to the pol CMS-associated orf224 gene. Curr. Genet. 31, 325–335. doi: 10.1007/s002940050212 Hamman-Khalifa, A., Castro, A. J., Jimenez-Lopez, J. C., Rodriguez-Garcia, M. I., and Alche, J. D. (2008). Olive cultivar origin is a major cause of polymorphism for Ole e 1 pollen allergen. BMC Plant Biol. 8:10. doi: 10.1186/1471-2229-8-10 Li, H., Li, J., Zhao, B., Wang, J., Yi, L., Liu, C., et al. (2015). Generation and characterization of tribenuron-methyl herbicide-resistant rapeseed (Brasscia napus) for hybrid seed production using chemically induced male sterility. Theor. Appl. Genet. 128, 107–118. doi: 10.1007/s00122-014-2415-7 Han, Y., Zhou, S. D., Fan, J. J., Zhou, L., Shi, Q. S., Zhang, Y. F., et al. (2021). OsMS188 is a key regulator of tapetum development and sporopollenin synthesis in rice. Rice 14:4. doi: 10.1186/s12284-020-00451-y Li, L., Li, B., Xie, C., Zhang, T., Borassi, C., Estevez, J. M., et al. (2020). Arabidopsis RAD23B regulates pollen development by mediating degradation of KRP1. J. Exp. Bot. 71, 4010–4019. doi: 10.1093/jxb/eraa167 Heng, S., Gao, J., Wei, C., Chen, F., Li, X., Wen, J., et al. (2018). Transcript levels of orf288 are associated with the hau cytoplasmic male sterility system and altered nuclear gene expression in Brassica juncea. J. Exp. Bot. 69, 455–466. doi: 10.1093/jxb/erx443 Li, P., Kang, L., Wang, A., Cui, C., Jiang, L., Guo, S., et al. (2019). Development of a fertility restorer for inap CMS (Isatis indigotica) Brassica napus through genetic introgression of one alien addition. Front. Plant Sci. 10:257. doi: 10.3389/fpls. 2019.00257 Hoﬀmann, R. D., and Palmgren, M. G. (2013). Epigenetic repression of male gametophyte-speciﬁc genes in the Arabidopsis sporophyte. Mol. Plant 6, 1176– 1186. doi: 10.1093/mp/sst100 Li, S., Chen, Z., Zhao, N., Wang, Y., Nie, H., and Hua, J. (2018). The comparison of four mitochondrial genomes reveals cytoplasmic male sterility candidate genes in cotton. BMC Genom. 19:775. doi: 10.1186/s12864-018-5122-y Hou, Q. C., and Bartels, D. (2015). Comparative study of the aldehyde dehydrogenase (ALDH) gene superfamily in the glycophyte Arabidopsis thaliana and Eutrema halophytes. Ann. Bot. 115, 465–479. doi: 10.1093/aob/ mcu152 Liang, R. Y., Chen, L., Ko, B. REFERENCES Plant Physiol. 137, 1067–1081. doi: 10.1104/pp. 104.055715 Ji, J. J., Huang, W., Yin, C. C., and Gong, Z. H. (2013). Mitochondrial cytochrome c oxidase and F1Fo-ATPase dysfunction in peppers (Capsicum annuum L.) with cytoplasmic male sterility and its association with orf507 and psi atp6-2 genes. Int. J. Mol. Sci. 14, 1050–1068. doi: 10.3390/ijms14011050 Ducos, E., Touzet, P., and Boutry, M. (2001). The male sterile G cytoplasm of wild beet displays modiﬁed mitochondrial respiratory complexes. Plant J. 26, 171–180. doi: 10.1046/j.1365-313x.2001.01017.x Int. J. Mol. Sci. 14, 1050–1068. doi: 10.3390/ijms14011050 Jing, B., Heng, S., Tong, D., Wan, Z., Fu, T., Tu, J., et al. (2012). A male sterility- associated cytotoxic protein ORF288 in Brassica juncea causes aborted pollen development. J. Exp. Bot. 63, 1285–1295. doi: 10.1093/jxb/err355 Farmer, L. M., Book, A. J., Lee, K. H., Lin, Y. L., Fu, H., and Vierstra, R. D. (2010). The RAD23 family provides an essential connection between the 26S proteasome and ubiquitylated proteins in Arabidopsis. Plant Cell 22, 124–142. doi: 10.1105/tpc.109.072660 Kale, A. A., and Munjal, S. V. (2005). Mitochondrial respiration associated with cytoplasmic male sterility in pearl millet. J. Plant Biochem. Biotechnol. 14, 161–165. doi: 10.1007/BF03263245 Ferguson, A. C., Pearce, S., Band, L. R., Yang, C., Ferjentsikova, I., King, J., et al. (2017). Biphasic regulation of the transcription factor ABORTED MICROSPORES (AMS) is essential for tapetum and pollen development in Arabidopsis. New Phytol. 213, 778–790. doi: 10.1111/nph.14200 Kang, L., Li, P., Wang, A., Ge, X., and Li, Z. (2017). A novel cytoplasmic male sterility in Brassica napus (inap CMS) with carpelloid stamens via protoplast fusion with chinese woad. Front. Plant. Sci. 8:529. doi: 10.3389/fpls.2017. 00529 Forsthoefel, N. R., Dao, T. P., and Vernon, D. M. (2010). PIRL1 and PIRL9, encoding members of a novel plant-speciﬁc family of leucine-rich repeat proteins, are essential for diﬀerentiation of microspores into pollen. Planta 232, 1101–1114. doi: 10.1007/s00425-010-1242-6 Kitazaki, K., Arakawa, T., Matsunaga, M., Yui-Kurino, R., Matsuhira, H., Mikami, T., et al. (2015). Post-translational mechanisms are associated with fertility restoration of cytoplasmic male sterility in sugar beet (Beta vulgaris). Plant J. 83, 290–299. doi: 10.1111/tpj.12888 p 1101–1114. doi: 10.1007/s00425-010-1242-6 Fujii, S., and Toriyama, K. (2009). Suppressed expression of Retrograde-Regulated Male Sterility restores pollen fertility in cytoplasmic male sterile rice plants. Proc. Natl. Acad. Sci. U.S.A. 106, 9513–9518. doi: 10.1073/pnas.0901860106 Ko, S. S., Li, M. J., Sun-Ben, K. M., Ho, Y. C., Lin, Y. J., Chuang, M. H., et al. (2014). REFERENCES The bHLH142 transcription factor coordinates with TDR1 to modulate the expression of EAT1 and regulate pollen development in rice. Plant Cell 26, 2486–2504. doi: 10.1105/tpc.114.126292 Fujii, S., Yamada, M., Fujita, M., Itabashi, E., Hamada, K., Yano, K., et al. (2010). Cytoplasmic-nuclear genomic barriers in rice pollen development revealed by comparison of global gene expression proﬁles among ﬁve independent cytoplasmic male sterile lines. Plant Cell Physiol. 51, 610–620. doi: 10.1093/pcp/ pcq026 Koiwai, A., and Matsuzaki, T. (1988). Hydroxy and normal fatty-acid distribution in stigmas of Nicotiana and other plants. Phytochemistry 27, 2827–2830. doi: 10.1016/0031-9422(88)80671-X Krolikowski, K. A., Victor, J. L., Wagler, T. N., Lolle, S. J., and Pruitt, R. E. (2003). Isolation and characterization of the Arabidopsis organ fusion gene HOTHEAD. Plant J. 35, 501–511. doi: 10.1046/j.1365-313x.2003.01824.x Gaborieau, L., and Brown, G. G. (2016). Comparative genomic analysis of the compound Brassica napus Rf locus. BMC Genomics 17:834. doi: 10.1186/ s12864-016-3117-0 Kubo, T., Arakawa, T., Honma, Y., and Kitazaki, K. (2020). What does the molecular genetics of diﬀerent types of restorer-of-fertility genes imply? Plants 9:361. doi: 10.3390/plants9030361 Gaborieau, L., Brown, G. G., and Mireau, H. (2016). The propensity of pentatricopeptide repeat genes to evolve into restorers of cytoplasmic male sterility. Front. Plant Sci. 7:1816. doi: 10.3389/fpls.2016.01816 Lahari, T., Lazaro, J., and Schroeder, D. F. (2017). RAD4 and RAD23/HMR contribute to Arabidopsis UV tolerance. Genes 9:8. doi: 10.3390/genes90 10008 Gallego, F., Fleck, O., Li, A., Wyrzykowska, J., and Tinland, B. (2000). AtRAD1, a plant homologue of human and yeast nucleotide excision repair endonucleases, is involved in dark repair of UV damages and recombination. Plant J. 21, 507–518. doi: 10.1046/j.1365-313x.2000.00694.x Lee, J., Yun, J. Y., Zhao, W., Shen, W. H., and Amasino, R. M. (2015). A methyltransferase required for proper timing of the vernalization response in Arabidopsis. Proc. Natl. Acad. Sci. U.S.A. 112, 2269–2274. doi: 10.1073/pnas. 1423585112 Gonzalez-Melendi, P., Uyttewaal, M., Morcillo, C. N., Hernandez, M. J., Fajardo, S., Budar, F., et al. (2008). A light and electron microscopy analysis of the events leading to male sterility in Ogu-INRA CMS of rapeseed (Brassica napus). J. Exp. Bot. 59, 827–838. doi: 10.1093/jxb/erm365 Lee, S. B., and Suh, M. C. (2013). Recent advances in cuticular wax biosynthesis and its regulation in Arabidopsis. Mol. Plant 6, 246–249. doi: 10.1093/mp/sss159 Guan, Y. F., Huang, X. Y., Zhu, J., Gao, J. F., Zhang, H. X., and Yang, Z. N. (2008). REFERENCES Y., Zhang, H., Sun, H., Luo, H., Zhao, L., Dong, Z., et al. (2017). Irregular pollen exine1 is a novel factor in anther cuticle and pollen exine formation. Plant Physiol. 173, 307–325. doi: 10.1104/pp.16.00629 Brocker, C., Lassen, N., Estey, T., Pappa, A., Cantore, M., Orlova, V. V., et al. (2010). Aldehyde dehydrogenase 7A1 (ALDH7A1) is a novel enzyme involved in cellular defense against hyperosmotic stress. J. Biol. Chem. 285, 18452–18463. doi: 10.1074/jbc.M109.077925 Crevillen, P. (2020). Histone demethylases as counterbalance to H3K27me3 silencing in plants. IScience 23:101715. doi: 10.1016/j.isci.2020.101715 Dai, M., Zhao, Y., Ma, Q., Hu, Y., Hedden, P., Zhang, Q., et al. (2007). The rice YABBY1 gene is involved in the feedback regulation of gibberellin metabolism. Plant Physiol. 144, 121–133. doi: 10.1104/pp.107.096586 Cai, Y., Zhang, Y., Loh, Y. P., Tng, J. Q., Lim, M. C., Cao, Z., et al. (2021). H3K27me3-rich genomic regions can function as silencers to repress gene expression via chromatin interactions. Nat. Commun. 12:719. doi: 10.1038/ s41467-021-20940-y Dantuma, N. P., Heinen, C., and Hoogstraten, D. (2009). The ubiquitin receptor Rad23: at the crossroads of nucleotide excision repair and proteasomal degradation. DNA Repair 8, 449–460. doi: 10.1016/j.dnarep.2009.01.005 Cartagena, J. A., Matsunaga, S., Seki, M., Kurihara, D., Yokoyama, M., Shinozaki, K., et al. (2008). The Arabidopsis SDG4 contributes to the regulation of pollen tube growth by methylation of histone H3 lysines 4 and 36 in mature pollen. Dev. Biol. 315, 355–368. doi: 10.1016/j.ydbio.2007.12.016 Devoto, A., Nieto-Rostro, M., Xie, D., Ellis, C., Harmston, R., Patrick, E., et al. (2002). COI1 links jasmonate signalling and fertility to the SCF ubiquitin-ligase complex in Arabidopsis. Plant J. 32, 457–466. doi: 10.1046/j.1365-313x.2002. 01432.x Chalhoub, B., Denoeud, F., Liu, S., Parkin, I. A., Tang, H., Wang, X., et al. (2014). Early allopolyploid evolution in the post-Neolithic Brassica napus oilseed genome. Science 345, 950–953. doi: 10.1126/science.1253435 Ding, B., Hao, M., Mei, D., Zaman, Q. U., Sang, S., Wang, H., et al. (2018). Transcriptome and hormone comparison of three cytoplasmic male sterile August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 17 Fertility Restoration for WNJ01A CMS Yang et al. wall integrity during Arabidopsis anther development. Plant Cell 31, 848–861. doi: 10.1105/tpc.18.00907 systems in Brassica napus. Int. J. Mol. Sci. 19:4022. doi: 10.3390/ijms191 24022 systems in Brassica napus. Int. J. Mol. Sci. 19:4022. doi: 10.3390/ijms191 24022 Duan, H., and Schuler, M. A. (2005). Diﬀerential expression and evolution of the Arabidopsis CYP86A subfamily. REFERENCES U., Zhang, H. Y., Hao, M. Y., et al. (2019). Organelle genome composition and candidate gene identiﬁcation for Nsa cytoplasmic male sterility in Brassica napus. BMC Genomics 20:813. doi: 10.1186/s12864-019-6187-y Schmitz-Linneweber, C., and Small, I. (2008). Pentatricopeptide repeat proteins: a socket set for organelle gene expression. Trends Plant Sci. 13, 663–670. doi: 10.1016/j.tplants.2008.10.001 Liu, Z., Yang, Z., Wang, X., Li, K., An, H., Liu, J., et al. (2016). A mitochondria- targeted PPR protein restores pol cytoplasmic male sterility by reducing orf224 transcript levels in oilseed rape. Mol. Plant 9, 1082–1084. doi: 10.1016/j.molp. 2016.04.004 Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., et al. (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504. doi: 10.1101/ gr.1239303 Lowe, E. D., Hasan, N., Trempe, J. F., Fonso, L., Noble, M. E., Endicott, J. A., et al. (2006). Structures of the Dsk2 UBL and UBA domains and their complex. Acta Crystallogr. D Biol. Crystallogr. 62, 177–188. doi: 10.1107/S09074449050 37777 Shen, Y., Zhang, Y., Yang, C., Lan, Y., Liu, L., Liu, S., et al. (2012). Mutation of OsALDH7 causes a yellow-colored endosperm associated with accumulation of oryzamutaic acid A in rice. Planta 235, 433–441. doi: 10.1007/s00425-011- 1477-x Mao, X. Z., Cai, T., Olyarchuk, J. G., and Wei, L. P. (2005). Automated genome annotation and pathway identiﬁcation using the KEGG Orthology (KO) as a controlled vocabulary. Bioinformatics 21, 3787–3793. doi: 10.1093/ bioinformatics/bti430 Shen, Y. F., Sun, S., Hua, S., Shen, E., Ye, C. Y., Cai, D., et al. (2017). Analysis of transcriptional and epigenetic changes in hybrid vigor of allopolyploid Brassica napus uncovers key roles for small RNAs. Plant J. 91, 874–893. doi: 10.1111/tpj. 13605 Mattioli, R., Biancucci, M., El, S. A., Mosca, L., Costantino, P., Funck, D., et al. (2018). Proline synthesis in developing microspores is required for pollen development and fertility. BMC Plant Biol. 18:356. doi: 10.1186/s12870-018- 1571-3 Shi, H., Yu, Y., Gu, R., Feng, C., Fu, Y., Fu, Y., et al. (2020). Male sterile 305 mutation leads the misregulation of anther cuticle formation by disrupting lipid metabolism in maize. Int. J. Mol. Sci. 21:8. doi: 10.3390/ijms21072500 Medina, B., Paraskevopoulos, K., Boehringer, J., Sznajder, A., Robertson, M., Endicott, J., et al. (2012). The ubiquitin-associated (UBA) 1 domain of Schizosaccharomyces pombe Rhp23 is essential for the recognition of ubiquitin- proteasome system substrates both in vitro and in vivo. J. Biol. Chem. REFERENCES 287, 42344–42351. doi: 10.1074/jbc.M112.419838 Shin, J. H., Kim, S. R., and An, G. (2009). Rice aldehyde dehydrogenase7 is needed for seed maturation and viability. Plant Physiol. 149, 905–915. doi: 10.1104/pp. 108.130716 Mira-Rodado, V., Sweere, U., Grefen, C., Kunkel, T., Fejes, E., Nagy, F., et al. (2007). Functional cross-talk between two-component and phytochrome B signal transduction in Arabidopsis. J. Exp. Bot. 58, 2595–2607. doi: 10.1093/jxb/ erm087 Singh, M., and Brown, G. G. (1993). Characterization of expression of a mitochondrial gene region associated with the Brassica “Polima” CMS: developmental inﬂuences. Curr. Genet. 24, 316–322. doi: 10.1007/BF003 36783 Morrison, M. J., Harker, K. N., Blackshaw, R. E., Holzapfel, C. J., and O’donovan, J. T. (2016). Canola yield improvement on the Canadian Prairies from 2000 to 2013. Crop Pasture Sci. 67, 245–252. doi: 10.1071/CP15348 Song, D., Tian, J., Hu, Y., Wei, Y., Lu, H., Wang, Y., et al. (2020). Identiﬁcation of biomarkers associated with diagnosis and prognosis of gastroesophageal junction adenocarcinoma-a study based on integrated bioinformatics analysis in GEO and TCGA database. Medicine 99:e23605. doi: 10.1097/MD. Ning, L., Wang, H., Li, D., Li, Y., Chen, K., Zhao, W., et al. (2019). Transcriptomic and proteomic analysis of Shaan2A cytoplasmic male sterility and its maintainer line in Brassica napus. Front. Plant Sci. 10:252. doi: 10.3389/fpls.2019.00252 in GEO and TCGA database. Medicine 99:e23605. doi: 10.1097/MD. 0000000000023605 Song, S. S., Qi, T. C., Huang, H., and Xie, D. X. (2013). Regulation of stamen development by coordinated actions of jasmonate, auxin, and gibberellin in Arabidopsis. Mol. Plant 6, 1065–1073. doi: 10.1093/mp/sst054 Ning, L., Wang, H., Li, D., Lin, Z., Li, Y., Zhao, W., et al. (2020). Genome-wide identiﬁcation of the restorer-of-fertility-like (RFL) gene family in Brassica napus and expression analysis in Shaan2A cytoplasmic male sterility. BMC Genomics 21:765. doi: 10.1186/s12864-020-07163-z Stepansky, A., and Galili, G. (2003). Synthesis of the Arabidopsis bifunctional lysine-ketoglutarate reductase/saccharopine dehydrogenase enzyme of lysine catabolism is concertedly regulated by metabolic and stress-associated signals. Plant Physiol. 133, 1407–1415. doi: 10.1104/pp.103.026294 Niu, N., Liang, W., Yang, X., Jin, W., Wilson, Z. A., Hu, J., et al. (2013). EAT1 promotes tapetal cell death by regulating aspartic proteases during male reproductive development in rice. Nat. Commun. 4:1445. doi: 10.1038/ ncomms2396 Plant Physiol. 133, 1407–1415. doi: 10.1104/pp.103.026294 Stepansky, A., Less, H., Angelovici, R., Aharon, R., Zhu, X., and Galili, G. (2006). Lysine catabolism, an eﬀective versatile regulator of lysine level in plants. Amino Acids 30, 121–125. REFERENCES T., Shen, Y. H., Li, Y. T., Chen, B. R., et al. (2014). Rad23 interaction with the proteasome is regulated by phosphorylation of its ubiquitin-like (UbL) domain. J. Mol. Biol. 426, 4049–4060. doi: 10.1016/j.jmb. 2014.10.004 Hu, B., Liu, B., Liu, L., Liu, C., Xu, L., and Ruan, Y. (2014). Epigenetic control of Pollen Ole e 1 allergen and extensin family gene expression in Arabidopsis thaliana. Acta. Physiol. Plant 36, 2203–2209. doi: 10.1007/s11738-014-1597-6 Lin, S., Miao, Y., Su, S., Xu, J., Jin, L., Sun, D., et al. (2019). Comprehensive analysis of Ogura cytoplasmic male sterility-related genes in turnip (Brassica rapa ssp. rapifera) using RNA sequencing analysis and bioinformatics. PLoS One 14:e218029. doi: 10.1371/journal.pone.0218029 Jacobowitz, J. R., Doyle, W. C., and Weng, J. K. (2019). PRX9 and PRX40 are extensin peroxidases essential for maintaining tapetum and microspore cell August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 18 Fertility Restoration for WNJ01A CMS Yang et al. Ren, C., Pan, J., Peng, W., Genschik, P., Hobbie, L., Hellmann, H., et al. (2005). Point mutations in Arabidopsis Cullin1 reveal its essential role in jasmonate response. Plant J. 42, 514–524. doi: 10.1111/j.1365-313X.2005.02394.x Liu, F., Cui, X. Q., Horner, H. T., Weiner, H., and Schnable, P. S. (2001). Mitochondrial aldehyde dehydrogenase activity is required for male fertility in maize. Plant Cell 13, 1063–1078. doi: 10.1105/tpc.13.5.1063 Liu, F., and Schnable, P. S. (2002). Functional specialization of maize mitochondrial aldehyde dehydrogenases. Plant Physiol. 130, 1657–1674. doi: 10.1104/pp. 012336 Sabar, M., Gagliardi, D., Balk, J., and Leaver, C. J. (2003). ORFB is a subunit of F1FO-ATP synthase: insight into the basis of cytoplasmic male sterility in sunﬂower. EMBO Rep. 4, 381–386. doi: 10.1038/sj.embor.embor800 Liu, Z., Dong, F., Wang, X., Wang, T., Su, R., Hong, D., et al. (2017). A pentatricopeptide repeat protein restores nap cytoplasmic male sterility in Brassica napus. J. Exp. Bot. 68, 4115–4123. doi: 10.1093/jxb/erx239 Sang, S., Cheng, H., Hao, M., Ding, B., Mei, D., Wang, H., et al. (2021). Mitochondrial localization of ORF346 causes pollen abortion in alloplasmic male sterility. Crop J. [Epub ahead of print]. doi: 10.1016/j.cj.2021.01.008 Liu, Z., Liu, P., Long, F., Hong, D., He, Q., and Yang, G. (2012). Fine mapping and candidate gene analysis of the nuclear restorer gene Rfp for pol CMS in rapeseed (Brassica napus L.). Theor. Appl. Genet. 125, 773–779. doi: 10.1007/s00122-012- 1870-2 Sang, S. F., Mei, D. S., Liu, J., Zaman, Q. REFERENCES doi: 10.1105/tpc.114.122986 Xu, L., Liu, F., Lechner, E., Genschik, P., Crosby, W. L., Ma, H., et al. (2002). The SCFCOl1 ubiquitin-ligase complexes are required for jasmonate response Xu, L., Liu, F., Lechner, E., Genschik, P., Crosby, W. L., Ma, H., et al. (2002). The SCFCOl1 ubiquitin-ligase complexes are required for jasmonate response in Arabidopsis. Plant Cell 14, 1919–1935. doi: 10.1105/tpc.003368 Tang, W. H., Ezcurra, I., Muschietti, J., and Mccormick, S. (2002). A cysteine- rich extracellular protein, LAT52, interacts with the extracellular domain of the pollen receptor kinase LePRK2. Plant Cell 14, 2277–2287. doi: 10.1105/tpc. 003103 Xu, Y., Liu, S., Liu, Y., Ling, S., Chen, C., and Yao, J. (2017). HOTHEAD-Like HTH1 is involved in anther cutin biosynthesis and is required for pollen fertility in rice. Plant Cell Physiol. 58, 1238–1248. doi: 10.1093/pcp/pcx063 Tavladoraki, P., Cona, A., and Angelini, R. (2016). Copper-containing amine oxidases and FAD-dependent polyamine oxidases are key players in plant tissue diﬀerentiation and organ development. Front. Plant Sci. 7:824. doi: 10.3389/fpls. 2016.00824 Yamaoka, Y., Yu, Y., Mizoi, J., Fujiki, Y., Saito, K., and Nishijima, M. (2011). PHOSPHATIDYLSERINE SYNTHASE1 is required for microspore development in Arabidopsis thaliana. Plant J. 67, 648–661. doi: 10.1111/j.1365- 313X.2011.04624.x Uyttewaal, M., Arnal, N., Quadrado, M., Martin-Canadell, A., Vrielynck, N., Hiard, S., et al. (2008). Characterization of raphanus sativus pentatricopeptide repeat proteins encoded by the fertility restorer locus for ogura cytoplasmic male sterility. Plant Cell 20, 3331–3345. doi: 10.1105/tpc.107.057208 Young, M. D., Wakeﬁeld, M. J., Smyth, G. K., and Oshlack, A. (2010). Gene ontology analysis for RNA-seq: accounting for selection bias. Genome Biol. 11:R14. doi: 10.1186/gb-2010-11-2-r14 Wade, S. L., and Auble, D. T. (2010). The Rad23 ubiquitin receptor, the proteasome and functional speciﬁcity in transcriptional control. Transcription 1, 22–26. doi: 10.4161/trns.1.1.12201 Zhao, J. Y., Missihoun, T. D., and Bartels, D. (2018). The ATAF1 transcription factor is a key regulator of aldehyde dehydrogenase 7B4 (ALDH7B4) gene expression in Arabidopsis thaliana. Planta 248, 1017–1027. doi: 10.1007/ s00425-018-2955-1 Wan, Z., Jing, B., Tu, J., Ma, C., Shen, J., Yi, B., et al. (2008). Genetic characterization of a new cytoplasmic male sterility system (hau) in Brassica juncea and its transfer to B-napus. Theor. Appl. Genet. 116, 355–362. doi: 10.1007/s00122-007- 0673-3 Zhao, N., Xu, X., Wamboldt, Y., Mackenzie, S. A., Yang, X., Hu, Z., et al. (2016). MutS HOMOLOG1 silencing mediates ORF220 substoichiometric shifting and causes male sterility in Brassica juncea. J. Exp. Bot. 67, 435–444. REFERENCES doi: 10.1093/ jxb/erv480 Wang, B., Farooq, Z., Chu, L., Liu, J., Wang, H., Guo, J., et al. (2021). High-generation near-isogenic lines combined with multi-omics to study the mechanism of polima cytoplasmic male sterility. BMC Plant Biol. 21:130. doi: 10.1186/s12870-021-02852-7 Zhu, X. H., and Galili, G. (2003). Increased lysine synthesis coupled with a knockout of its catabolism synergistically boosts lysine content and also transregulates the metabolism of other amino acids in Arabidopsis seeds. Plant Cell 15, 845–853. doi: 10.1105/tpc.009647 Wang, H., Xiao, Q., Wei, C., Chen, H., Chen, X., Dai, C., et al. (2021). A mitochondria-localized pentatricopeptide repeat protein is required to restore hau cytoplasmic male sterility in Brassica napus. Theor. Appl. Genet. 134, 1377–1386. doi: 10.1007/s00122-021-03777-3 Zuo, B., Zhu, J., Xiao, F., Wang, C., Shen, Y., and Chen, X. (2020). Identiﬁcation of novel biomarkers and candidate small molecule drugs in rheumatoid arthritis and osteoarthritis based on bioinformatics analysis of high-throughput data. Biosci. Rep. 40:BSR20193823. doi: 10.1042/BSR20193823 Wang, K., Gao, F., Ji, Y., Liu, Y., Dan, Z., Yang, P., et al. (2013). ORFH79 impairs mitochondrial function via interaction with a subunit of electron transport chain complex III in Honglian cytoplasmic male sterile rice. New Phytol. 198, 408–418. doi: 10.1111/nph.12180 Zuo, Z., and Mahajan, P. B. (2005). Recombinant expression of maize nucleotide excision repair protein Rad23 in Escherichia coli. Protein Expr. Purif. 41, 287–297. doi: 10.1016/j.pep.2005.02.023 Wei, W., Li, Y., Wang, L., Liu, S., Yan, X., Mei, D., et al. (2010). Development of a novel Sinapis arvensis disomic addition line in Brassica napus containing the restorer gene for Nsa CMS and improved resistance to Sclerotinia sclerotiorum and pod shattering. Theor. Appl. Genet. 120, 1089–1097. doi: 10.1007/s00122- 009-1236-6 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. The reviewer ZL declared a shared aﬃliation, with no collaboration, with the authors to the handling editor at the time of the review. The reviewer ZL declared a shared aﬃliation, with no collaboration, with the authors to the handling editor at the time of the review. Wei, X., Lv, Y., Zhao, Y., Nath, U. K., Yuan, Y., Wang, Z., et al. (2020). Comparative transcriptome analysis in Chinese cabbage (Brassica rapa ssp. pekinesis) for DEGs of Ogura-, Polima-CMS and their shared maintainer. Physiol. Mol. Biol. Plants 26, 719–731. REFERENCES doi: 10.1007/s00726-005-0246-1 Okeke, E., Chen, L., and Madura, K. (2020). The cellular location of Rad23, a polyubiquitin chain-binding protein, plays a key role in its interaction with substrates of the proteasome. J. Mol. Biol. 432, 2388–2404. doi: 10.1016/j.jmb. 2020.03.001 Su, W., Tao, R., Liu, W., Yu, C., Yue, Z., He, S., et al. (2020). Characterization of four polymorphic genes controlling red leaf colour in lettuce that have undergone disruptive selection since domestication. Plant Biotechnol. J. 18, 479–490. doi: 10.1111/pbi.13213 Peng, X., Wang, K., Hu, C., Zhu, Y., Wang, T., Yang, J., et al. (2010). The mitochondrial gene orfH79 plays a critical role in impairing both male gametophyte development and root growth in CMS-Honglian rice. BMC Plant Biol. 10:125. doi: 10.1186/1471-2229-10-125 Takagi, H., Abe, A., Yoshida, K., Kosugi, S., Natsume, S., Mitsuoka, C., et al. (2013). QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant J. 74, 174–183. doi: 10.1111/tpj.12105 Planas-Portell, J., Gallart, M., Tiburcio, A. F., and Altabella, T. (2013). Copper- containing amine oxidases contribute to terminal polyamine oxidation in peroxisomes and apoplast of Arabidopsis thaliana. BMC Plant Biol. 13:109. doi: 10.1186/1471-2229-13-109 Tang, G. L., Zhu, X. H., Tang, X. H., and Galili, G. (2000). A novel composite locus of Arabidopsis encoding two polypeptides with metabolically related but distinct functions in lysine catabolism. Plant J. 23, 195–203. doi: 10.1046/j.1365-313x. 2000.00770.x Preuss, D., Lemieux, B., Yen, G., and Davis, R. W. (1993). A conditional sterile mutation eliminates surface components from Arabidopsis pollen and disrupts cell signaling during fertilization. Genes Dev. 7, 974–985. doi: 10.1101/gad.7. 6.974 Tang, H., Xie, Y., Liu, Y. G., and Chen, L. (2017). Advances in understanding the molecular mechanisms of cytoplasmic male sterility and restoration in rice. Plant Reprod. 30, 179–184. doi: 10.1007/s00497-017-0308-z August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 19 Fertility Restoration for WNJ01A CMS Yang et al. Tang, Q., Tian, M., An, G., Zhang, W., Chen, J., Yan, C., et al. (2017). Rapid identiﬁcation of the purple stem (Ps) gene of Chinese kale (Brassica oleracea var. alboglabra) in a segregation distortion population by bulked segregant analysis and RNA sequencing. Mol. Breeding 37:153. doi: 10.1007/s11032-017-0752-3 Xu, J., Ding, Z., Vizcay-Barrena, G., Shi, J., Liang, W., Yuan, Z., et al. (2014). ABORTED Microspores acts as a master regulator of pollen wall formation in Arabidopsis. Plant Cell 26, 1544–1556. REFERENCES doi: 10.1007/s12298-020-00775-5 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. Wolters-Arts, M., Lush, W. M., and Mariani, C. (1998). Lipids are required for directional pollen-tube growth. Nature 392, 818–821. doi: 10.1038/33929 Wu, H., Shi, S., Lu, X., Li, T., Wang, J., Liu, T., et al. (2019). Expression analysis and functional characterization of CER1 family genes involved in very-long-chain alkanes biosynthesis in Brachypodium distachyon. Front. Plant Sci. 10:1389. doi: 10.3389/fpls.2019.01389 Copyright © 2021 Yang, Nong, Xu, Huang, Wang, Wu, Zhang and Liu. 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. Xing, M., Sun, C., Li, H., Hu, S., Lei, L., and Kang, J. (2018). Integrated analysis of transcriptome and proteome changes related to the Ogura cytoplasmic male sterility in cabbage. PLoS One 13:e0193462. doi: 10.1371/journal.pone.0193462 Xiong, X., Zhou, D., Xu, L., Liu, T., Yue, X., Liu, W., et al. (2019). BcPME37c is involved in pollen intine formation in Brassica campestris. Biochem. Biophys. Res. Commun. 517, 63–68. doi: 10.1016/j.bbrc.2019.07.009 August 2021 \| Volume 12 \| Article 721980 Frontiers in Plant Science \| www.frontiersin.org 20
W4281992067.txt	https://zenodo.org/records/6609466/files/IJET-V8I3P23.pdf	en		CERTAIN INVESTIGATION OF MONITORING SYSTEM IN TRANSPORTS USING IOT	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	2,807	International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 RESEARCH ARTICLE OPEN ACCESS CERTAIN INVESTIGATION OF MONITORING SYSTEM IN TRANSPORTS USING IOT Dr.P.Ramesh kumar,M.Tech,Ph.D1, Brindha G2,Priyadharshini K3,Yuvarani M4 1(ECE, Sengunthar Engineering College(Autonomous), Tiruchengode Email: dean.research@scteng.co.in) 2 (ECE, Sengunthar Engineering College(Autonomous), Tiruchengode Email: brindha2001g@gmail.com) 3 (ECE, Sengunthar Engineering College(Autonomous), Tiruchengode Email: kumarpriya172001@gmail.com ) 4 (ECE, Sengunthar Engineering College(Autonomous), Tiruchengode Email: yuvaranim0@gmail.com) Abstract: Innovations and technology will make our life became much easier now a days. We are working on developing a college bus monitoring system using RFID (Radio Frequency Identification technology). Our project is about controller and RFID based displaying and updating bus management system based on IoT and embedded system, to achieve automatically display and manage the bus database details with bus name or number, entry and exit time of the specific vehicle alongside date without the need for manual operation. It is part of intelligent transportation, at the same time the use of auto data base management, to reduce the work load of the bus management system, microcontroller was used to receive and send the data, then the data is communicated to the IoT module. For every vehicle RFID tag will be given. In gate the RFID readers are located. Using reader tag, information is retrieved. The serial number of each tag is associated with each transport’s database. With the help of the system RFID and the Internet of Things, the transports monitoring will be easy to access; through the accuracy and reliability of the data, the system gives accurate information to the administrator of transports. The proposed system can inform in charge of transports whether the bus is arriving on time, early or late. Keywords — RFID Technology, IoT, Transport Monitoring. I INTRODUCTION Nowadays, travel time information of transports becomes a major component of Advanced Traveler Information System (ATIS). These travel time of transports depending on external parameters such as accidents, stuck in traffic. Most bus station follows fixed schedules, and don’t use intelligent systems for vehicle tracking and control. Many supervisors are deployed at the station to control the entrance and the exit of buses and prepare the trip sheets containing the schedules manually which is time consuming and inaccurate. Manual control can be used in offices, laboratories and libraries where it is essential to keep a record of the people entering and exiting. At present every work should be done manually and because of this performance is degraded. This problem can be solved by using latest technologies like Radio Frequency Identification (RFID). RFID is the wireless non-contact frequency electromagnetic fields to transfer data, for the purposes of ISSN: 2395-1303 automatically identifying and tracking tags attached to objects. In this system the individual RFID tags and readers are viably utilized for observing transports. There is a feature provided in this system to generate daily reports monitoring of transports can be done automatically. In addition, Arduino UNO and Esp8266 (NODEMCU) are utilized in this task. These both communicate with one another and the information's are spread through WIFI - Device NODEMCU to the cloud. Whenever transport entered into the organization the RFID reader peruses the transport number, transport entry timing will be captured. With the assistance of Esp8266 (NODEMCU) all the subtleties are sent to the college transport office through the cloud with the assistance of Esp8266. These subtleties if not matches with the predefined boarded database of the bus, then use MQTT (Message Queuing Telemetry Transport) server module will send the message to the person who is in charge of transportation system of college. Then the monitoring of transport reports can be done easily to the administrator without human force. http://www.ijetjournal.org Page 178 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 II LITERATURE SURVEY Radio Frequency Identification (RFID) is introducing, and it’s bringing a streamlined revolution in this world. When dealing with the tracking device, Radio Frequency Identification (RFID) is the latest phase in the decades that can be used as an efficient tracker. Radio Frequency Identification (RFID) Technology used to develop tracking system is quite new but something that promising. This is used to gives solution RFID technique for monitoring entry and exit of employees with their official assets (E.g., laptops). This system is actually based on external database system that will provide the prerecorded information about the reader. Since the reader detected by the database, then the tracking system will process the data and will show the result of subject tracking. [1] In the era of embedded systems efficiency and time are the matter of priority. RFID (Radio Frequency Identification) one of the converging technologies and transportation plays an important role in urbanization, RFID is one of the key catalysts playing a significant role in it. RFID plays major role in auto ID applications like RFID contact fewer smart cards used by bus riders, in Super market, Textiles and logistics chain management. This is mainly focused to understand the benefits of RFID technology and possibilities to reduce the accidents on Indian roads. The GSM (Global System for Mobile Communications) has been a great success to providing both voice and low speed data services. In GSM one of the major evolutionary steps to serve real-time high-speed data services is to Enhanced Circuit Switched Data (ECSD). [2] The basic concept of connectivity is IoT can be integrated into traditional communication network to reduce many problems. Vehicle-to-Infrastructure (V2I) technology is one of the aspect of IoT enabling intelligent transport systems. In V2I Vehicular monitoring is part which to helps minimize the problem caused by vehicles in the city, like traffic violation and road accidents, congestion. In this system observed an instance of Vehicle-toinfrastructure communication model realizing data transmission between traffic light and vehicle is to be regulated. A On-Board Units (OBU) prototype and Road Side Units (RSU) prototypes are developed. Vehicles send Identity, speed and location messages to the traffic controller fully based on Zigbee wireless technology. The message is analyzed to check speed violation. From the RSU information related to user’s driving is transmitted to a monitoring server to charge the offender of traffic rules. ISSN: 2395-1303 Therefore, the modules based on Controller Area Network (CAN) bus for in- vehicle communication, OBU details for Over speeding in highways, RSU information for Real time data collection and E-mail notification service based on violation rules are developed [3] The lifecycle for industrial applications is becoming shorter, the application complexity increases, performance is too low, fault tolerance is required, reuse of components is desired, and the developer require strong verification tools to cut down the verification phase. These problems are increases with based on longer development time and requirements of higher quality from the customer, its more important to examine flexible and scalable parallel processing for complex real-time systems. This is the motivation forerunning the research project SARA (Scalable Architecture for Real-Time Applications). The first SARA system is now running with vision system connected to an industrial robot (ABB Robot). The systembusses are important resources in computer-system. Today there are no methods to monitor busload during runtime; in this project discuss a simple method of how to do this.[4] In this work present to integrate RFID (Radio Frequency Identification) in WSN (Wireless sensor network). It is used to support Radio Frequency identification process by the read range of an RFID system extended. Besides, we can monitor the environment of an object and optimize RFID reader’s performance and energy by the use of wireless sensor networks. Then the methodology to integrate RFID technology, WSN forms an intelligent bus tracking application is studied. The proposed system can monitoring bus traffic inside spacious bus stations and used to inform the administrators to whether the bus is arriving on time, early or late. This information is then displayed on the different wireless displays inside and outside the bus station. [5] III EXISTING SYSTEM In existing method, have to monitor the transports related data by the use of manual system to get a information in institution. No automatic system was implemented to get information automatically which will lead work load and manual errors may occur in that system.it is a long process to monitoring and maintaining the database by the use ledger, it may give error to database of the transports. There is no storing method to handle the real time daily databases in previous method. Overcoming this method, we can choose RFID methodologies to store and give the daily databases of the transports in all organizations. http://www.ijetjournal.org Page 179 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 IV This proposed method mainly focused on the RFID technology which will overcoming the existing method of monitoring the database these technologies will give better solution to monitoring the database automatically. Radio-frequency identification (RFID) - It is a programmed identification technique; this will be depending on Storing the data and recovering information remotely by using RFID labels or transponders. RFID Reader: It will send the electromagnetic waves which is carrying a signal to identify objects. This will ensure that finally the reader receives the information returned back by these kinds of objects. RFID tag: Unique identity tag will attach to the objects; it will react to receiving the signal from a RFID reader and sent the required database in order to forwarding to it the requested information. Computer/database: It will be used to storing and processing the information collected by the RFID reader. PROPOSED SYSTEM Fig.1 Block Diagram for RFID Monitoring System Based Transport Fig.2 RFID Technology This project proposes a system for tracking college bus using advance techniques which are capable to deliver results and the information in a rapid and efficient way. In this system will propose the bus database monitoring on Android Application which is used MQTT server. It will modifies the advance and latest techniques for the existing System. The system will replace IOT, the traditional way of monitoring the bus by using IOT system. The data will give the exact timing of bus in and out and co-ordinates of the Bus. The RFID will be connected to the microcontroller. The Arduino Uno has web connectivity using IOT module. The RFID is placed in a bus. The software will also stores the daily information monitoring of Bus like, details of Bus, identity number, etc. In and out gate time which will be shown on an LCD display and as well as stored in software if it is exceeded in the predefined time of institution then the message will automatically send to the administrator who is in charge of the transportation. ISSN: 2395-1303 A. ARDUINO UNO MICROCONTROLLER In this system using Arduino uno controller. It is a main part of this system because it internally stores the alltransport related database programmed inside it by the use of embedded programming. It is used to get information from the RFID reader and send it to the WIFI gadget (Node mcu). LCD is likewise associated with it so as to see the passage entry and leave time status, bus number along with date. http://www.ijetjournal.org Page 180 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 Fig.5 NODE MCU (ESP2866) Fig.3 Arduino UNO B. RTC MODULE RTC means Real Time Clock. RTC modules are simply TIME and DATE remembering systems which contains battery setup in the absence of external power keeps the module running. In this system Entry and exit time of the buses are generated by using RTC and the Arrival time is noted and displayed in LCD. Fig.4 RTC Module Fig 6. Circuit diagram for Transport Monitoring using IoT C. NODE MCU V WORKIG PRINCIPLE Node Microcontroller Unit is simply called as NODE MCU it is a open-source software and hardware development environment built around an inexpensive System-on-a-Chip (SoC) called the ESP8266. WIFI gadget utilized for sending the subtleties caught to the cloud (MQTT client).It cooperates with Arduino Uno and gets the subtleties from it and sends to cloud. Node red represents the buses updated on entry in real time database cloud with the help of NODEMCU. If the predefined database will not match the subtleties then the SMS notification and buzzer indication send through it. ISSN: 2395-1303 There are two sections in this project one is transmitter section which contains a RFID tag and it is placed on the bus and another section is receiver section which contains a RFID reader placed on college gate. Bus is consistently monitoring by the use of RFID Technologies. The Reader continuously producing a radio waves. These waves are used to passively identified tagged object by the use of passive RFID tags. The Radio frequency ranges are different in frequencies. These frequency ranges mostly give the RF ranges of the tags from low frequency tag ranges http://www.ijetjournal.org Page 181 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 from 3m to 5m, middle frequency tag ranges from 5m to 17m and high frequency tag ranges from 5ft to 90ft.When bus comes in the range of the college gate, with the help of wireless networks. This can constantly maintain the bus data base with the help of controller. Finally the database will stores inside the software called node red and the data will be monitored by the authority of transport remotely. punctual to the transport schedules that have been established, resulting in a more efficient transport circulation system in the institution. Generating reports with the help of internet browsers is convenient and flexible for printing the History of SMS and Time-in timeout. In future expect that the system will inspire the designers and developers to develop transport monitoring with enhancement of features in future. VII SYSTEM OUTCOMES Fig.7 Architecture diagram VI RESULTS AND CONCLUSION As the RFID technology evolves, various sophisticated applications will use the capability of RFID to send, receive, store and forward the data to a remotely to the sink source. Each area needs to be exclusively programmed. Maintenance is bit costly than traditional system. As the tag is read by the reader it will give relevant information to the system. The reader will accept the card if the tag information is already stored in the database. This framework can be stretched out for full-time monitoring and college transports that will be useful for Transport organization at least expense. This system gives time saving, easy control and reliability and also reduces manpower. Bus drivers will ensure to be more ISSN: 2395-1303 http://www.ijetjournal.org Page 182 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 VIII REFERENCES [1] Ran Hee Jeong, and Laurence R. Rilett (2004) “The Prediction of Bus Arrival Time Using AVL Data”, Transportation Research Board 83rd Annual Meeting, Washington D.C. [2] H. Isredza Rahmi, Comparative study on RFID, hotspot and car plate scanning method : for intelligent parking management system. masters thesis, University Technology MARA. 2005. [3] Urachada Ketprom, Chaichana Mitrpant, Putchapan Lowjun, “Closing Digital Gap on RFID Usage for Better Farm Management”, PICMET 2007, 5-9 August 07. [4] QIN, K., XING, J., CHEN, G., WANG, L., QIN, J. “The design of Intelligent Bus Movement Monitoring and Station Reporting System”, In Proceedings of the IEEE International Conference on Automation and Logistics, Qingdao, China, September 2008, p. 2822-2827. [5] Longer O. (2009),“Implementation of Student Attendance System using RFID Technology”, B. Tech Project Report, Ladoke Akintola University of Technology, Ogbomosho, Nigeria. [6] Mohamed A.B, Abdel-Hamid A and Mohammed K.Y.,(2009), ”Implementation of an Improved secure system detection for E passport by using EPC RFID tags”, World Academy of Science. [7] Ben Ammar Hatem, Hamam Habib, “Bus Management System Using RFID in WSN”, European and Mediterranean Conference on Information Systems. [2010]. [8] A. HANNAN, A. M. MUSTAPHA, A. HUSSAIN and H. BASRI, “Intelligent Bus Monitoring and Management System”, CECS 2012, October 24-26, 2012, San Francisco, USA. [9] Kumar,Chaturvedula.U.P,M.Tech,EmbeddedSystems, RFID Based Embedded System for Vehicle Tracking and Prevention of Road Accidents ,International Journal of Engineering Research&Technology (IJERT)Vol. 1 Issue 6, August-2012. [10] Anwar Al-Lawati, Shaikha Al-Jahdami, Asia Al-Belushi, Dalal Al-Adawi, Medhat Awadalla and Dawood Al-Abri. “RFID- based System for School Children Transportation Safety Enhancement”. [2015]. [11] Akshay S. Kyatam, P. A. M. M. A. A. K. [2015]. “Tracking and Scheduling of State Transport Bus using RFID”. International Journal of Engineering and Computer Science, 4(05). Retrieved from http://www.ijecs.in/index.php/ijecs/article/view/3599. [12] International Journal of computer Application (0975-8887) Volume 168-No.3, June 2017 Intelligent Bus Monitoring System Neha Shinde & Saniya Ansari. [13] M. K. Sangole and Sakshi S. Kedar and Pallavi B Phad and Sanket V Salunke. “RFID BASED TRANSPORT MONITERING SYSTEMS: A Technical Review”, [2018]. [14] PRARTHANA. S, NIVETHA.R, K. POORNIMATHI. [2020]. “IoT Based Bus Entry Monitoring System Using RFID”. International Journal of Computer Science and Mobile Computing - IJCSMC (Vol.9, No. 5). ISSN: 2395-1303 http://www.ijetjournal.org Page 183 International Journal of Engineering and Techniques - Volume 8 Issue 3, June 2022 ISSN: 2395-1303 http://www.ijetjournal.org Page 184
https://openalex.org/W4250664141	https://www.qeios.com/read/UDP49S/pdf	English	null	PBTL CD19CAR-28/CD137/zeta	Definitions	2,020	cc-by	190	Qeios · Definition, February 2, 2020 Open Peer Review on Qeios Open Peer Review on Qeios Qeios ID: UDP49S · https://doi.org/10.32388/UDP49S PBTL CD19CAR-28/CD137/zeta National Cancer Institute National Cancer Institute Source National Cancer Institute. PBTL CD19CAR-28/CD137/zeta. NCI Thesaurus. Code C106429. Peripheral blood T-lymphocytes (PBTLs) transduced with a retroviral vector expressing a chimeric antigen receptor (CAR) consisting of an anti-CD19 scFv (single chain variable fragment) coupled to the costimulatory signaling domain CD28, the signaling domain of 4-1BB (CD137), and the zeta chain of the T-cell receptor (TCR), with potential immunomodulating and antineoplastic activities. Upon transfusion, PBTL CD19CAR- 28/CD137zeta directs the T-lymphocytes to CD19-expressing tumor cells and induces selective toxicity in CD19-expressing tumor cells. CD28, a T-cell surface-associated co- stimulatory molecule, is required for full T-cell activation, proliferation, and survival. The 4-1BB co-stimulatory molecule signaling domain enhances activation and signaling after recognition of CD19. Furthermore, inclusion of the 4-1BB signaling domain may increase the antitumor activity compared to the inclusion of the CD28 costimulatory domain and TCR zeta chain alone. CD19 antigen is a B-cell specific cell surface antigen, which is expressed in all B-cell lineage malignancies. Qeios ID: UDP49S · https://doi.org/10.32388/UDP49S 1/1
https://openalex.org/W3157736485	https://www.researchsquare.com/article/rs-441596/latest.pdf	English	null	Wearable magnetic induction-based approach toward 3D motion tracking	Scientific reports	2,021	cc-by	7,210	Introduction Over the past decade, monitoring and recognition of human activities have embraced a growing number of practical usages in a broad range of domains such as healthcare, rehabilitation, sports training, virtual reality (VR) gaming, human-computer interface (HCI) systems, ﬁnger tracking, daily life-logging, child and elderly care, and assistance for people with cognitive disorders or chronic conditions [1–7]. Tracking and reconstructing limb movements in 3D space facilitates more detailed evaluation, and it is crucial for the analysis and clinical understanding of complex functional movements. Studying biomechanics of human motion has application in human performance assessment, gesture/posture monitoring, behavioral recognition, gait analysis, and patients’ functionality and improvement evaluation during the rehabilitation period [8–10]. There are many solutions for tracking body movement using different monitoring sources [11]. Computer vision-based methods, such as Kinect or optical motion capture (MoCap) system, are the most commonly used techniques that allow users to interact with them and collect data on the user’s motion using depth sensors, color, and infrared cameras [12, 13]. However, they inherit computer vision restrictions such as light dependency, coverage limitation, and high computational cost [14, 15]. The MoCap systems require an expensive setup of infrared cameras for tracking reﬂective markers on an individual’s body, which makes them only applicable to the laboratory environment and restricted in physical space. Besides, the markers placement and soft tissue artifacts have a considerable effect on the system accuracy [16, 17]. The RF-based solutions are another motion tracking method capturing data based on wireless signal changes (e.g., Doppler frequency shift and signal amplitude ﬂuctuation) [18]. These methods also suffer from environmental dependency and limitation in the number of detectable gestures due to the high cost of training data collection and the lack of multi-user identiﬁcation capabilities [19]. Wearable-based solutions are an alternative, cost-effective solution for applications where the optical-based methods are unsuitable. This approach tracks the user’s movement based on the sensors readings placed around the human body [20]. The advancement of sensing technologies, miniaturization, embedded systems, and wireless communication systems combined with predictive models for data analysis and detection have made it possible to develop wearable devices working around the human body for continuous physical activity monitoring. Smart devices like smartphones, smartwatches, and ﬁtness bands are becoming widespread for providing valuable insights about an individual’s performance and health status. These wearables have multiple embedded physiological, inertial, and ambient sensors that enable multi-modal sensing [1]. Research Article License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Wearable Magnetic Induction-based Approach Toward 3D Motion Tracking Negar Golestani1,* and Mahta Moghaddam1 1Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, 90089, USA. golestani.negar@gmail.com 1Department of Electrical and Computer Engineering, University of Southern California, L golestani.negar@gmail.com ABSTRACT Activity recognition using wearable sensors has gained popularity due to its wide range of applications, including healthcare, rehabilitation, sports, and senior monitoring. Tracking the body movement in 3D space facilitates behavior recognition in different scenarios. Wearable systems have limited battery capacity, and many critical challenges have to be addressed to gain a trade-off among power consumption, computational complexity, minimizing the effects of environmental interference, and achieving higher tracking accuracy. This work presents a motion tracking system based on magnetic induction (MI) to tackle the challenges and limitations inherent in designing a wireless monitoring system. We integrated a realistic prototype of an MI sensor with machine learning techniques and investigated one-sensor and two-sensor conﬁguration setups for motion reconstruction. This approach is successfully evaluated using measured and synthesized datasets generated by the analytical model of the MI system. The system has an average distance root-mean-squared error (RMSE) error of 3 cm compared to the ground-truth real-world measured data with Kinect. Introduction Many studies have exploited commercial inertial measurement unit (IMU) devices comprised of accelerometers, gyroscopes, and magnetic sensors, for motion tracking based on wearable sensors. An IMU can be attached to a body segment to estimate its movement in space. By combining multiple of them on adjacent body segments, the kinematics of activities can be determined [20]. For example, [21] presents the development of a smart wearable jumpsuit with multiple built-in IMU sensors for automatic posture and movement tracking of infants. The work in [22] investigates the reliability and validity of IMUs for clinical movement analysis, and [23] presents a single wrist-worn IMU sensor for high-resolution motor state detection in Parkinson’s disease. Inertial (a) Single sensor setting (b) Orthogonal setting (c) Parallel setting Fig. 1. Data-driven location tracking across all settings for a predeﬁned target point. (b) Orthogonal setting (a) Single sensor setting (b) Orthogonal setting Fig. 1. Data-driven location tracking across all settings for a predeﬁned target point. (a) Single sensor setting (c) Parallel setting (c) Parallel setting (c) Parallel setting Fig. 1. Data-driven location tracking across all settings for a predeﬁned target point. sensing can track limb movements by integrating over sensor measurements, though it is subject to drift since the estimation errors caused by the intrinsic noise can grow unbounded with time [1]. sensing can track limb movements by integrating over sensor measurements, though it is subject to drift since the estimation errors caused by the intrinsic noise can grow unbounded with time [1]. In some applications, it is possible to achieve improved accuracy and more speciﬁc inferences by fusing the subsets of data collected from those sensors compared to single sensor modalities [24, 25]. Although these devices provide a solution for physiological health monitoring, condition assessment, and medical diagnosis, they still might face challenges. These single-node devices restrict biosensors’ placement, while optimizing their position can increase the system’s accuracy and robustness in monitoring vital signs (e.g., body temperature and heart rate) [1, 26]. Furthermore, in systems relying on data from a single device, variations in position can signiﬁcantly affect the performance. In motion tracking applications, a single node wearable is not able to cover the entire body. Therefore it cannot get detailed information about the mobility of an individual’s limbs. For example, a smartwatch’s inertial sensor cannot capture the movement of the user’s legs, limiting the system’s ability in classifying activities [1]. Operating Principle The MI-based communication system is a short-range wireless physical layer that transmits signals by inductive coupling between the wire coils rather than radiating as is done in conventional methods [27, 28]. The transmitter node uses a coil to produce an oscillating magnetic ﬁeld at a speciﬁc frequency. Each sensor node’s (receiver) main component is a coil to capture the transmitter’s generated magnetic ﬁeld. According to Faraday’s law, the time-varying magnetic ﬁeld induces a voltage in sensor nodes proportional to the rate of magnetic ﬂux change through their coils. For a predeﬁned coil geometry and operating frequency below 30 MHz, where the environmental effects are negligible, the ﬂux change rate is a function of the sensor coils’ position, and orientation relative to the transmitter [1, 27]. The relationship function from spatial data into induced voltage is non-linear and surjective, and the tracking problem objective is to estimate the sensors’ positions given the induced voltage measurements. Introduction A network of distributed wearable devices operating around the human body is an approach that can address these issues. One of the biggest challenges in wearable-based motion tracking systems is to ﬁnd the optimum type and number of non-invasive sensors with minimal power consumption to achieve acceptable accuracy and satisfy guidelines and constraints. In [1], we introduced a wireless system based on magnetic induction combined with machine learning techniques to detect a wide range of human activities. We showed that this system can address challenges in terms of power consumption, accuracy, coverage, privacy, and cost. Here, we investigate the capability of the magnetic induction (MI) system in 3D motion tracking and evaluate a prototype device in estimating its motion using trained machine learning-based regressors. We use a calibrated MI model presented in our previous works [27, 27] to generate synthetic MI motion data and train regressors without the need for any measured data. We then evaluate the trained model on real-world measured data. System Architecture y We used an analytical model of the MI system presented in [1, 27, 29] to calculate the induced voltage at each sensor coil given its position and orientation. This forms the basis of the data-driven backward estimation algorithm that retrieves a node’s position using its observed data. It helps assess the system performance under different conﬁgurations, such as 2/10 (a) Single sensor setting (b) Two-sensor orthogonal setting (c) Two-sensor parallel setting Fig. 2. Schematic representation of conﬁguration settings used in the experiments. (b) Two-sensor orthogonal setting (c) Two-sensor parallel setting (a) Single sensor setting Fig. 2. Schematic representation of conﬁguration settings used in the experiments. changing the number or arrangement of sensor coils to ﬁnd the near-optimal setup with acceptable tracking accuracy. Since the model is a function of relative distance and alignment of coils to the transmitter, we transform the coordinate system to locate the new coordinate system’s origin at the center of the transmitter coil, with the coil’s surface normal oriented in the z-direction. Given the sensors’ spatial data, we compute the coordinate transformation matrix and calculate each coil’s position and orientation in the new coordinate frame. We explored the node’s position p = (x,y,z) with the resolution of 1 cm, and alignment ˆn = (sinθ cosφ, sinθ sinφ, cosθ) with the resolution of 5◦as these resolutions are expected to satisfy the accuracy requirements for motion tracking applications [30, 31]. It also provides enough data points within the search domain for comprehensive performance analysis of the system with different conﬁguration settings. The possible solutions, which is a unique single-point in an optimal conﬁguration, are retrieved for a given set of observed data. The domain of the search space is deﬁned as follows to represent the average ranges of distances where sensors can be placed for both male and female subjects relative to an on-body central node on their torso: x ∈[−20 cm, 20 cm] y ∈[−20 cm, 20 cm] z ∈[−60 cm, −10 cm] male and female subjects relative to an on-body central node on their torso: x ∈[ 20 cm, 20 cm] y ∈[−20 cm, 20 cm] z ∈[−60 cm −10 cm] male and female subjects relative to an on-body central node on their torso: θ ∈[0◦, 60◦] φ ∈[0◦, 360◦] (1) (1) We studied the performance of an MI sensor (single sensor setting), where the coil can be aligned in any direction. System Architecture We also adopted two-sensor conﬁgurations and investigated different alignment setups. Among these setups, we present the performance analysis of setups where coils’ surface normal are aligned in the same direction (parallel setting) or perpendicular to each other (orthogonal setting). Fig. 2 depicts the conﬁguration of sensors in each described setting. In these experiments, the induced voltage measured at the coils is used as input for location estimation. Fig. 1 shows an example result of the data-driven backward estimation algorithm, where each point represents a possible node’s position with at least one alignment that can produce the given set of inputs for the deﬁned setting. As the results display, there are many possible solutions for a single sensor setup, and this number reduces by adding another sensor. The sensor voltage data are assumed to be measured with 1 mv accuracy and given as inputs to the algorithm. A comparison between the two-sensor conﬁgurations shows that the parallel setting outperforms the orthogonal setting. Although a unique solution cannot be returned as an output, results suggest that the regression methods with proper constraints can meet the minimum required accuracy for the position tracking. Evaluation We deployed machine learning regression algorithms to solve the inverse problem of estimating a node’s 3D position (x,y,z) from its sensors’ measurements in meters. The performance of several regression models, including extra trees (ET), random forest (RF), K-nearest neighbors (KNN), light gradient boosting machine (LightGBM), multi-layer perceptrons (MLP), decision trees (DT), and linear regression (LR) is compared using PyCaret [32], an open-source machine learning library in Python. The models are trained on 70% of synthetic data and then scored on the remaining data using the 10-fold cross-validation method. The metrics used for comparison are RMSE, mean absolute percentage error (MAPE), and R-squared (R2). Before fetching data into the regressors, each feature is standardized individually, and the missing values are substituted with previous non-missing values. The processed data are then divided into ﬁxed-length segments of 2 seconds using the sliding window technique with a 0.1 second step size. Table 1 summarizes the performance results of all models on the synthetic data for different settings. As the results show, the moving node’s distance and position in the z-direction with respect to the transmitter coordinate frame can be tracked with competing accuracy compare to other methods using wearable sensors (e.g., accelerometer) for motion tracking [20, 31]. Adding another transmitter in a different direction to the presented MI-based system can enable the node’s motion tracking in the new direction and allows 3D positional tracking. The work in [33], for example, has reported results on tracking subjects’ arm motion using smartwatch IMU data. The results show that the system can achieve the highest accuracy when the torso is static with a median error of 8.8 cm. Moreover, [34] presents a framework for reconstructing human motion with the highest accuracy of 6 cm using four 3D accelerometers attached to the user. The work in [35] has proposed utilization of spinning linearly polarized antennas to track translation of an object attached to a passive radio frequency identiﬁcation (RFID) tag array in 3D and has reported an average error of 13.6 cm. To provide a realistic assessment of real-world performance, we evaluated each of the optimal models’ tracking accuracies on measured data as well. According to the score measures reported on synthetic data, the LightGBM regressor in the single-sensor setting and the extra trees regressor in the two-sensor (orthogonal and parallel) settings outperform other models. Fig. Data Collection We designed and built an MI sensor for 3D motion tracking (see Methods section), representing the movements by variation in the MI signals received from the transmitter instead of measuring spatial data via conventional sensors such as IMUs. To evaluate the capability of the proposed MI sensor, we employed regression algorithms and investigated their performance on MI sensor’s data. Validating and testing machine learning methods is critical and challenging due to the difﬁculty of collecting realistic valid data and the lack of labeled data. One solution is to create synthetic data for training the model, and here, we used a Variational Auto-Encoder (VAE) model to produce time-series motion data. The MI data corresponding to the generated synthetic movements are then produced using an analytical MI system model [1, 27]. The regressors are then trained on these synthesized data, which removes the need for supervised training measured data. A point to consider is that the MI system model must be calibrated only once to scale the synthetic training data to sensor measurements and tune the regression algorithm (see Methods section). The trained machine learning regressors on the synthetic data are then tested on real-world measurements and reported for comparison. 3/10 Table 1. Performance of regression models in motion tracking using synthetic data generated for the different settings. Data Collection Model Distance X Y Z RMSE MAPE R2 RMSE MAPE R2 RMSE MAPE R2 RMSE MAPE R2 Single Sensor ET 0.028 0.043 0.898 0.05 6.548 0.232 0.035 3.877 0.102 0.031 0.05 0.868 RF 0.028 0.042 0.9 0.05 6.872 0.25 0.035 3.88 0.125 0.031 0.049 0.871 KNN 0.029 0.044 0.89 0.052 6.155 0.161 0.037 3.923 0.017 0.033 0.052 0.856 LightGBM 0.028 0.042 0.9 0.048 6.873 0.29 0.034 3.931 0.173 0.031 0.049 0.872 MLP 0.03 0.045 0.883 0.049 6.806 0.278 0.034 3.999 0.147 0.034 0.053 0.846 DT 0.039 0.057 0.807 0.069 10.529 -0.45 0.049 4.05 -0.708 0.043 0.067 0.753 LR 0.04 0.066 0.796 0.052 8.079 0.177 0.035 4.095 0.138 0.041 0.069 0.769 Orthogonal ET 0.022 0.029 0.879 0.04 4.897 0.537 0.011 0.851 0.18 0.005 0.007 0.169 RF 0.023 0.031 0.863 0.043 5.266 0.47 0.031 2.53 0.362 0.026 0.037 0.827 KNN 0.022 0.029 0.875 0.042 5.749 0.487 0.03 2.137 0.387 0.025 0.034 0.84 LightGBM 0.026 0.036 0.826 0.05 5.989 0.294 0.035 3.018 0.158 0.029 0.043 0.779 MLP 0.026 0.037 0.823 0.05 5.239 0.277 0.036 3.191 0.134 0.03 0.043 0.774 DT 0.033 0.041 0.718 0.062 7.04 -0.096 0.044 2.71 -0.313 0.037 0.049 0.642 LR 0.035 0.05 0.688 0.053 6.829 0.184 0.037 3.2 0.103 0.038 0.057 0.625 Parallel ET 0.011 0.018 0.947 0.037 7.539 0.365 0.028 5.184 0.23 0.014 0.025 0.917 RF 0.011 0.019 0.94 0.037 9.49 0.334 0.028 5.471 0.193 0.014 0.026 0.908 KNN 0.012 0.021 0.932 0.037 10.543 0.352 0.028 4.855 0.196 0.015 0.027 0.9 LightGBM 0.012 0.021 0.932 0.04 10.004 0.261 0.03 5.827 0.075 0.015 0.028 0.897 MLP 0.013 0.022 0.924 0.04 9.251 0.231 0.031 6.268 0.022 0.016 0.03 0.881 DT 0.016 0.026 0.882 0.053 14.302 -0.338 0.04 5.478 -0.636 0.02 0.035 0.818 LR 0.021 0.037 0.804 0.041 11.848 0.185 0.031 6.015 0.048 0.023 0.043 0.762 Discussion We demonstrated an MI-based system to accurately track the motion of receivers with respect to the central transmitter node. We employed an HF RFID transmitter module equipped with a loop antenna. A simple integrated circuit is also used as a receiver, which interfaces with the central node, and records received signals sent from the transmitter using an Arduino. The main focus of this work was to provide a proof of concept for the proposed system, which can also be implemented for real-world applications by proper modiﬁcations. For example, the MI coils should be designed to be suitable for wearing on the human wrist, arm, and ankle. Furthermore, a wearable custom-designed central node capable of driving a controlled amount of current at the operating frequency through its coil is required. The receivers should cover the range of about 0.5 m to 1 m with minimum power consumption. The RF output power of the reader we used here is 1 Watt, which can be reduced by designing an efﬁcient sensors capable of sensing lower power signals. For example, [36] presents a transceiver design exploiting the low path loss of Magnetic Human Body Communication (mHBC) communication channels toward ultra-efﬁcient body area networking. The transmitter and receiver respectively require only 7.15 and 4.7 pJ/bit, and their design is a helpful reference for implementing MI transceivers. One implementation approach to reduce the number of nodes with battery is to make the central node serve as both transmitter and receiver. It means that the central unit can broadcast the signal and listen back to the responses reﬂected from the sensors, similar to an RFID system based on passive (battery-less) tags. In an RFID system, the reader sends an interrogation signal to the transponders, which is also used to energize the tag. The tag activates and sends back its unique identiﬁer (UID) if the received power is higher than its sensitivity [37]. A modulation resistance connected in parallel with the tag antenna switches between two different (usually conjugate matching and a short circuit) load impedances at the clock rate of the signal transmitted from the reader to modulate the backscattered signal [38]. Therefore, the central node can communicate with the tags via a secure near-ﬁeld link backscattering from them. Evaluation 3 presents the evaluation measures of optimal models using the measured data for each setting. Representative samples of motion tracking in all settings are also displayed in Fig. 4. Our results indicate that the parallel setting with the optimal regression model outperforms other settings on both measured and synthetic MI data. 4/10 Fig. 3. Tracking performance metrics across all conﬁguration settings on the measured motion and MI data. Fig. 3. Tracking performance metrics across all conﬁguration settings on the measured motion and MI data. Fig. 3. Tracking performance metrics across all conﬁguration settings on the measured motion and MI data. (a) Single Sensor Setting (b) Orthogonal Setting (c) Parallel Setting Fig. 4. Distance and motion tracking in the z-direction. Fig. 3. Tracking performance metrics across all conﬁguration settings on the measured motion and MI data. Fig. 3. Tracking performance metrics across all conﬁguration settings on the measured motion and MI data. (a) Single Sensor Setting (b) Orthogonal Setting (c) Parallel Setting Fig. 4. Distance and motion tracking in the z-direction. (a) Single Sensor Setting (c) Parallel Setting (b) Orthogonal Setting (c) Parallel Setting (c) Parallel Setting (b) Orthogonal Setting (a) Single Sensor Setting Fig. 4. Distance and motion tracking in the z-direction. Discussion The amplitude of the demodulated signal is calculated and reported at the reader side by a value proportional to the received signal’s power level, known as the received signal 5/10 strength indicator (RSSI). A point to consider is that the load modulation is not a practical solution for data transmission in an MI-based motion tracking system. The reason is that the backscattered ﬁeld, and consequently, the voltage signal received by the reader, switches over two values[13]. The average power returned to the reader is no longer a direct function of distance and misalignment between coils since it varies by the number of zeros and ones in the data stream. Therefore, proper modulation and modiﬁcations are required to be able to employ existing RFID protocols. strength indicator (RSSI). A point to consider is that the load modulation is not a practical solution for data transmission in an MI-based motion tracking system. The reason is that the backscattered ﬁeld, and consequently, the voltage signal received by the reader, switches over two values[13]. The average power returned to the reader is no longer a direct function of distance and misalignment between coils since it varies by the number of zeros and ones in the data stream. Therefore, proper modulation and modiﬁcations are required to be able to employ existing RFID protocols. Here we have compared the relationship between RSSI and MI signals with motion data by recording RSSI data of RFID tags in addition to the MI sensors data. We employed custom air-cored, three-layer copper coils with a 5 cm radius and 34 American wire gauge (AWG) wire diameter as the tag antenna attached to STMicroelectronics ST25DV04K RFID tag. We measured motion and RSSI data reported from the reader for 112 experiments. The calculated average R2 and the correlation between RSSI and the tags distance from the reader are respectively 0.22 and 0.39. For an MI sensor, the calculated R2 and correlation averaged over 110 samples are 0.53 and 0.69, respectively. These results indicate that the MI signal has a stronger relationship with its motion compared to a passive tag. Hardware Design The system consists of a transmitter (central) node generating an oscillating signal at 13.56 MHz. We used ISC.LRM1002 long-range RFID reader module [39] attached to ISC.ANT310/310 long-range HF antenna [39] to generate the RF signal. Since we used this setup for RFID measurements presented in the discussion, we used the same transmitter for a better comparison. The receiver node consists of MI sensors. Each sensor includes an air-cored, single-layer copper coil with a 5 cm radius and 10 AWG wire diameter to capture the transmitter’s signal and measure the induced voltage. Resistance and self-inductance of the coil measured by vector network analyzer (VNA) at the resonance frequency are 101mΩ, and 241 nH, respectively. To improve the system efﬁciency, we have employed resonant inductive coupling attached to the coil. The tuning circuit can be as simple as a capacitor to tune the frequency or be a Π or T matching circuit to tune the frequency, control Q-factor, and match input and output impedances for higher power transfer [40]. Here, we used a 560 pF capacitor parallel to a trimmable capacitor with the adjustable range of 3-10 pF to tune the circuit to resonance accurately. The transmitted AC signal attenuates as a function of distance and alignment of the node with respect to the transmitter antenna. To track the signal’s amplitude changes, we used an envelope detector consisting of an IN5817 Schottky diode, a resistor or 1 KΩ, and a capacitor of 1 nF. The envelope detector’s output, which is the resistor’s voltage, is measured by an Arduino Nano (ATmega168) microcontroller. The resolution of ADC (analog pin A1) is 10 bit for a deﬁned measurement range. Fig. 5 depicts an MI sensor components. Measurements We employed a Microsoft Kinect v2 to capture the 3D position and alignment of the transmitter and the MI sensor node. The Kinect sensor consists of a depth camera, an RGB camera, and a microphone array sensor. The RGB camera and depth camera respectively provide a 1920×1080 color image, and 512×424 depth image at 30 frames per second with a resolution of a few millimeters in measure range between 0.5 m to 4.5 m [41]. The depth stream provides the sensor’s distance to every point within its area of coverage. As the cameras have different pixel resolutions and are not perfectly aligned, three coordinate spaces Fig. 5. The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. Fig. 6. Schematic representation of measurement setup. Fig. 6. Schematic representation of measurement setup. Fig. 6. Schematic representation of measurement setup. Fig. 5. The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. Fig. 6. Schematic representation of measurement setup. 6/10 and types are deﬁned: color space point (xc,yc), depth space point (xd,yd), and camera space point (xw,yw,zw), representing a point in the color images, depth images, and real-world, respectively. The software development kit (SDK)’s mapping function can be used to map a point from one coordinate space to another. We used colored markers to facilitate motion tracking of the devices and developed a video processing algorithm analyzing the color frames to locate pixels corresponding to the target color. The transmitter antenna and the MI node are labeled with distinct colored markers and placed in front of a white background. A threshold range is set for each color to extract pixels with the color value within the deﬁned range. The detected pixels are classiﬁed to Nm clusters, where Nm is the number of markers, using K-means clustering methods. Then, the connected neighboring pixels of each cluster are grouped. Since the markers are colored foam balls, the circle with the minimum area enclosing each set is calculated, and the largest region is given as the target circle. The next step is mapping color to camera space to ﬁnd the corresponding spatial location of each extracted color pixel. Synthetic Data A VAE is based on the auto-encoder architecture and is composed of encoder and decoder networks. The encoder compresses the data into a lower-dimensional space called the latent space representation. The decoder decompresses the reduced representation code to reconstruct the original data. The VAE learns the probabilistic interpretation of these networks and generates new samples using different latent variables as input. Consider dataset {x(i) }N i=1 that consists of N i.i.d. samples of some variable x. VAEs assume that the data are generated by a random process with continuous latent variable, and each latent variable z is related to its corresponding observation x through likelihood pθ(x\|z), where pθ is a probability distribution with parameters θ. This probabilistic interpretation of the decoder can decode a latent (hidden) representation code into a distribution over the observation. Similarly, the encoder network returns a latent code sampled from the posterior density distribution pθ(z\|x) given a sample from the data space [42]. While both prior p(z) and likelihood p(x\|z) can be formulated exactly, the posterior p(z\|x) requires an intractable integral over the latent space. Hence, an approximate posterior qφ(z\|x) closest in Kullback-Leibler (KL) divergence to the actual, intractable posterior distribution is considered. The approximate posterior is parameterized by variational parameters φ, and the training objective is a tractable lower bound to the log-likelihood [43]: log p(x) ≥Eqφ (z\|x) h log pθ(x,z) qφ(z\|x) i = L (x;θ,φ) (2) and can be equivalently written as: (3) L (x;θ,φ) = Eqφ (z\|x) [ log pθ (x\|z)] −DKL(qφ(z\|x)\|\|Pθ(z)) On the right-hand side of equation (3), the ﬁrst term, reconstruction error, represents the likelihood of the model reconstruct- ing the input data. The second term, variational regularization term, is the KL divergence and makes the approximate posterior qφ(z\|x) to be close to pθ(z). The L (x;θ,φ) is a lower bound on the log probability of data pθ(x), which is called evidence lower bound (ELBO). Maximizing ELBO with respect to the model parameters θ and variational parameters φ respectively maximizes the marginal probability pθ(x) and minimizes the KL divergence [42]. We trained the model using the sensors’ motion data tracked by the Kinect to produce synthetic time-series samples. After training the model, new time-series data can be generated by sampling from latent space z with normal distribution parametrized by the mean and the variance [43]. Measurements The result is a list of 3D real-world points mapped from the target circle’s pixels, and each marker’s location is computed by taking the median over all the calculated values. This process repeats for each new color frame that Kinect captures. The analytical model requires the center and alignment of the transmitter and receiver coils/antennas as inputs to estimate the induced voltage. To determine a coil’s surface normal, at least three markers (Mi : i ∈{1,...,Nm} with Nm >= 2) are required. Hence, we used four red and three blue markers to track the transmitter antenna and the MI sensor node. The center of each device is calculated by averaging over its markers’ location c = ∑Nm i=1 Mi, and its surface normal is also calculated by the cross product of vectors passing through the markers: ˆn =⃗v1 ×⃗v2 where⃗v1 = M1 −M2,⃗v2 = M1 −M3. We applied the median ﬁlter, a non-linear digital ﬁltering technique, to remove noise and spikes in the extracted location and alignment data. The induced voltage, Vind, at the MI sensors is measured for 30 seconds via Arduino by using a Python script that controls the recording in order to synchronize Kinect’s motion data and Arduino’s measurements. The sampling frequency is 100 Hz, and the reference voltage range is 0 V to 5 V, which results in the quantization interval of 5/1024 V. The data streams of the node’s MI sensors are recorded and used as inputs for the regression model to estimate the device’s location. The sampling rate of motion data recorded by Kinect and the sensors’ data are different. Therefore, all recordings are resampled with a sampling interval of 100 ms, which also handles the missing sample values. The measurement setup of experimental measurements is presented in Fig. 6. Synthetic Data The generated data include the motion of the coils’ center and alignment in 3D space for a predeﬁned sensor setting. We synthesized angular variables θ and φ to calculate the corresponding coil’s surface normal ˆn that 7/10 Fig. 7. The measured and simulated induced voltage at an MI sensor during two arbitrary movements, such that both relative alignment and location of the coil varies. Fig. 7. The measured and simulated induced voltage at an MI sensor during two arbitrary movements, such that both relative alignment and location of the coil varies. can be deﬁned as ˆn = (sinθ cosφ, sinθ sinφ, cosθ), where the variables θ and φ can take values in the range of 0-90 a 0-360 degrees, respectively. can be deﬁned as ˆn = (sinθ cosφ, sinθ sinφ, cosθ), where the variables θ and φ can take values in the range of 0-90 and 0-360 degrees, respectively. To generate training data for the motion tracking algorithm, we synthesized the induced voltage at the MI sensor using the two-port network model of the MI system [1, 27]. The circuit model represents a forward model of the system estimating the sensor measurements given node motion data. To evaluate the circuit model’s accuracy in synthesizing MI data, we fetched the captured motion data by the Kinect as inputs and estimated the corresponding induced voltage at the sensors. The data streams corresponding to MI sensors are simulated for each motion sample. The circuit model is calibrated by ﬁnding the scale and bias of the synthesized data with respect to the measurements. Synthetic Data Considering si and mi as the generated synthetic data and measurements corresponding to a motion sample, the scale a and bias b can be calculated as follows: a = 1 Ns Ns ∑ i=1 σsi σmi (4) b = 1 Ns Ns ∑ i=1 µsi −σsi σmi µmi (5) a = 1 Ns Ns ∑ i=1 σsi σmi (4) b = 1 Ns Ns ∑ i=1 µsi −σsi σmi µmi (5) a = 1 Ns Ns ∑ i=1 σsi σmi (4) b = 1 Ns Ns ∑ i=1 µsi −σsi σmi µmi (5) a = 1 Ns Ns ∑ i=1 σsi σmi b = 1 Ns Ns ∑ i=1 µsi −σsi σmi µmi (4) (5) where µsi , σsi , µmi , σmi represent the mean and standard deviation of synthetic data and measurements corresponding to ith motion sample from Ns samples. p p We have performed the experiment for 220 motions, including spatial translation and rotation (Ns=220). Fig. 7 shows the measured and simulated sensors’ data during their movement, picked from the evaluation dataset after calibrating the model. The average normalized root-mean-squared error (NRMSE) and cross-correlation of the synthesized and measured data for all experiments are 12% and 0.91, respectively. It should be noted that the reported metrics consider not only the MI system model inaccuracy but also the error associated with the Kinect-based marker tracking algorithm and Arduino measurements. The measured motion data samples are also used for training VAE to generate synthetic motion data. References In Advances in Computational Intelligence Systems, 341–371 (Springer, 2017). 13. Zhang, S. et al. A review on human activity recognition using vision-based method. J. healthcare engineering 2017 (2017). 14. Lin, Q. et al. Beyond one-dollar mouse: A battery-free device for 3d human-computer interaction via rﬁd tags. In 2015 IEEE Conference on Computer Communications (INFOCOM), 1661–1669 (IEEE, 2015). 15. Zhang, H.-B. et al. A comprehensive survey of vision-based human action recognition methods. Sensors 19, 1005 (2019). 16. Poppe, R. A survey on vision-based human action recognition. Image Vis. Comput. 28, 976–990 (2010). 15. Zhang, H.-B. et al. A comprehensive survey of vision-based human action recognition methods. Sensors 19, 1005 (2019). 15. Zhang, H.-B. et al. A comprehensive survey of vision-based human action recognition methods. Sensors 19, 1005 (2019). 16. Poppe, R. A survey on vision-based human action recognition. Image Vis. Comput. 28, 976–990 (2010). 16. Poppe, R. A survey on vision-based human action recognition. Image Vis. Comput. 28, 976–990 (2010). 17. Weygers, I. et al. Inertial sensor-based lower limb joint kinematics: A methodological systematic review. Sensors 20, 673 (2020). 18. Zhao, M. et al. Rf-based 3d skeletons. In Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication, 267–281 (2018). 19. Wang, C. et al. Rf-kinect: A wearable rﬁd-based approach towards 3d body movement tracking. Proc. ACM on Interactive, Mobile, Wearable Ubiquitous Technol. 2, 1–28 (2018). 20. Filippeschi, A. et al. Survey of motion tracking methods based on inertial sensors: A focus on upper limb human motion. Sensors 17, 1257 (2017). 21. Airaksinen, M. et al. Automatic posture and movement tracking of infants with wearable movement sensors. Sci. reports 10, 1–13 (2020). 22. Al-Amri, M. et al. Inertial measurement units for clinical movement analysis: reliability and concurrent validity. Sensors 18, 719 (2018). 23. Pﬁster, F. M. et al. High-resolution motor state detection in parkinson’s disease using convolutional neural networks. Sci. reports 10, 1–11 (2020). 24. Garcia-Ceja, E. et al. Mental health monitoring with multimodal sensing and machine learning: A survey. Pervasive Mob. Comput. 51, 1–26 (2018). 25. Tannous, H. et al. A new multi-sensor fusion scheme to improve the accuracy of knee ﬂexion kinematics for functional rehabilitation movements. Sensors 16, 1914 (2016). 26. Ha, S. et al. The placement position optimization of a biosensor array for wearable healthcare systems. J. Mech. Sci. Technol. 33, 3237–3244 (2019). 27. Golestani, N. & Moghaddam, M. References 1. Golestani, N. & Moghaddam, M. Human activity recognition using magnetic induction-based motion signals and deep recurrent neural networks. Nat. Commun. 11, 1–11 (2020). 1. Golestani, N. & Moghaddam, M. Human activity recognition using magnetic induction-based motion signals and deep recurrent neural networks. Nat. Commun. 11, 1–11 (2020). 2. Kim, J. et al. Wearable smart sensor systems integrated on soft contact lenses for wireless ocular diagnostics. Nat. communications 8, 1–8 (2017). 2. Kim, J. et al. Wearable smart sensor systems integrated on soft contact lenses for wireless ocular diagnostics. Nat. communications 8, 1–8 (2017). 2. Kim, J. et al. Wearable smart sensor systems integrated on soft contact lenses for wireless ocular diagnostics. Nat. communications 8, 1–8 (2017). 3. Mukhopadhyay, S. C. Wearable sensors for human activity monitoring: A review. IEEE Sensors J. 15, 1321–1330 (2014). 4. Parizi, F. S., Whitmire, E. & Patel, S. Auraring: Precise electromagnetic ﬁnger tracking. Proc. ACM on Interactive, Mobile, Wearable Ubiquitous Technol. 3, 1–28 (2019). 4. Parizi, F. S., Whitmire, E. & Patel, S. Auraring: Precise electromagnetic ﬁnger tracking. Proc. ACM on Interactive, Mobile, Wearable Ubiquitous Technol. 3, 1–28 (2019). 5. Jovanov, E., Milenkovic, A., Otto, C. & De Groen, P. C. A wireless body area network of intelligent motion sensors for computer assisted physical rehabilitation. J. NeuroEngineering Rehabil. 2, 6 (2005). 6. Jones, S. E. et al. Genetic studies of accelerometer-based sleep measures yield new insights into human sleep behaviour. Nat. Commun. 10, 1–12 (2019). 7. Kim, J., Campbell, A. S., de Ávila, B. E.-F. & Wang, J. Wearable biosensors for healthcare monitoring. Nat. Biotechnol. 37, 389–406 (2019). 8/10 8. Camomilla, V., Bergamini, E., Fantozzi, S. & Vannozzi, G. Trends supporting the in-ﬁeld use of wearable inertial sensors for sport performance evaluation: A systematic review. Sensors 18, 873 (2018). 9. Ghislieri, M., Gastaldi, L., Pastorelli, S., Tadano, S. & Agostini, V. Wearable inertial sensors to assess standing balance: A systematic review. Sensors 19, 4075 (2019). 10. Prakash, C., Kumar, R. & Mittal, N. Recent developments in human gait research: parameters, approaches, applications, machine learning techniques, datasets and challenges. Artif. Intell. Rev. 49, 1–40 (2018). 11. Ramasamy Ramamurthy, S. & Roy, N. Recent trends in machine learning for human activity recognition—a survey. Wiley Interdiscip. Rev. Data Min. Knowl. Discov. 8, e1254 (2018). 12. Bux, A., Angelov, P. & Habib, Z. Vision based human activity recognition: a review. References Laptev, N. Anogen: Deep anomaly generator. In Outlier Detection De-constructed (ODD) Workshop. Available online: https://research. fb. com/publications/anogen-deep-anomaly-generator/(accessed on 20 August 2018) (2018). Acknowledgements The authors would like to thank Dr. Danilo Brizi for his helpful advice on the RFID antenna design discussed in this paper. A th t ib ti The authors would like to thank Dr. Danilo Brizi for his helpful advice on the RFID antenna design discussed in this paper. References Theoretical modeling and analysis of magnetic induction communication in wireless body area networks (wban). IEEE J. Electromagn. RF Microwaves Medicine Biol. 2, 48–55 (2018). 28. Golestani, N. & Moghaddam, M. Communication system design for magnetic induction-based wireless body area network. In USNC-URSI Radio Science Meeting (Joint with AP-S Symposium), 49–50 (IEEE, 2017). 29. Golestani, N. & Moghaddam, M. Gain analysis of in-motion magnetic inductive coupling-based systems. In 2020 IEEE International Symposium on Antennas and Propagation and North American Radio Science Meeting, 1677–1678 (IEEE, 2020). 9/10 30. Schepers, M., Giuberti, M., Bellusci, G. et al. Xsens mvn: Consistent tracking of human motion using inertial sensing. Xsens Technol 1–8 (2018). 31. Roetenberg, D., Luinge, H. & Slycke, P. Xsens mvn: Full 6dof human motion tracking using miniature inertial sensors. Xsens Motion Technol. BV, Tech. Rep 1 (2009). 32. Ali, M. PyCaret: An open source, low-code machine learning library in Python (2020). PyCaret version 2.1. 33. Shen, S., Gowda, M. & Roy Choudhury, R. Closing the gaps in inertial motion tracking. In Proceedings of the 24th Annual International Conference on Mobile Computing and Networking, 429–444 (2018). 34. Tautges, J. et al. Motion reconstruction using sparse accelerometer data. ACM Transactions on Graph. (ToG) 30, 1–12 (2011). 35. Wang, C. et al. Spin-antenna: 3d motion tracking for tag array labeled objects via spinning antenna. In IEEE INFOCOM 2019-IEEE Conference on Computer Communications, 1–9 (IEEE, 2019). 36. Park, J. & Mercier, P. P. A sub-10-pj/bit 5-mb/s magnetic human body communication transceiver. IEEE J. Solid-State Circuits 54, 3031–3042 (2019). 37. Bufﬁ, A., Michel, A., Nepa, P. & Tellini, B. Rssi measurements for rﬁd tag classiﬁcation in smart storage systems. IEEE Transactions on Instrumentation Meas. 67, 894–904 (2018). 38. Finkenzeller, K. Battery powered tags for iso/iec 14443, actively emulating load modulation. In RFID SysTech 2011 7th European Workshop on Smart Objects: Systems, Technologies and Applications, 1–8 (VDE, 2011). s://www.feig-electronics.com/. Accessed: 2020-09-20. 9. Feig company. https://www.feig-electronics.com/. Accessed: 2020-09-20. 39. Feig company. https://www.feig-electronics.com/. Accessed: 2020-09-20. 40. Golestani, N. & Moghaddam, M. Improving the efﬁciency of magnetic induction-based wireless body area network. In IEEE International Microwave Biomedical Conference (IMBioC), 166–168 (IEEE, 2018). 41. Lachat, E., Macher, H., Landes, T. & Grussenmeyer, P. Assessment and calibration of a rgb-d camera (kinect v2 sensor) towards a potential use for close-range 3d modeling. Remote. Sens. 7, 13070–13097 (2015). 42. Kingma, D. P. & Welling, M. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 (2013). 43. Figure 1 Data-driven location tracking across all settings for a prede¦ned target point. Data-driven location tracking across all settings for a prede¦ned target point. Figure 2 Author contributions N. G. was the main contributor to this work and was responsible for developing and implementing the methods, data generation, and analysis; M. M. supervised the research. 10/10 Figures Figures Figures Figure 1 Schematic representation of con¦guration settings used in the experiments. Schematic representation of con¦guration settings used in the experiments. Schematic representation of con¦guration settings used in the experiments. Schematic representation of con¦guration settings used in the experiment Figure 3 Tracking performance metrics across all con¦guration settings on the measured motion and MI data. Figure 4 Distance and motion tracking in the z-direction. Figure 5 The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. Tracking performance metrics across all con¦guration settings on the measured motion and MI data. Figure 4 Distance and motion tracking in the z-direction. Distance and motion tracking in the z-direction. Distance and motion tracking in the z-direction. Figure 5 The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. Figure 5 Figure 5 The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. The MI sensor prototype hardware contains: 1) Variable capacitor for frequency tuning 2) An envelope detector 3) Arduino microcontroller for measurement. Figure 6 Schematic representation of measurement setup. Figure 7 The measured and simulated induced voltage at an MI sensor during two arbitrary movements, such that both relative alignment and location of the coil varies. Figure 6 Figure 6 Schematic representation of measurement setup. Figure 7 The measured and simulated induced voltage at an MI sensor during two arbitrary movements, such that both relative alignment and location of the coil varies. Figure 7 The measured and simulated induced voltage at an MI sensor during two arbitrary movements, such that both relative alignment and location of the coil varies.
https://openalex.org/W2511581392	https://uu.diva-portal.org/smash/get/diva2:1014890/FULLTEXT01	English	null	Engineering Polarons at a Metal Oxide Surface	Physical review letters	2,016	cc-by	5,703	Engineering Polarons at a Metal Oxide Surface C. M. Yim,1 M. B. Watkins,2 M. J. Wolf,3,4 C. L. Pang,1 K. Hermansson,4 and G. Thornton1,* 1Department of Chemistry and London Centre for Nanotechnology, University College London, 20 Gordon Street, London WC1H 0AJ, United Kingdom 2School of Mathematics and Physics, University of Lincoln, Brayford Pool, Lincoln LN6 7TS, United Kingdom 3Department of Physics & Astronomy and London Centre for Nanotechnology, University College London, Gower Street, London WC1E 6BT, United Kingdom 4D t t f Ch i t Å t ö L b t U l U i it B 538 S 751 21 U l S d C. M. Yim,1 M. B. Watkins,2 M. J. Wolf,3,4 C. L. Pang,1 K. Hermansson,4 and G. Thornton1,* 1Department of Chemistry and London Centre for Nanotechnology, University College London, 20 Gordon Street, London WC1H 0AJ, United Kingdom 2S h l f M th ti d Ph i U i it f Li l B f d P l Li l LN6 7TS U it d Ki d C. M. Yim,1 M. B. Watkins,2 M. J. Wolf,3,4 C. L. Pang,1 K. Hermansson,4 and G. Thornton1 1Department of Chemistry and London Centre for Nanotechnology, University College London, p f y f gy, y 20 Gordon Street, London WC1H 0AJ, United Kingdom 2School of Mathematics and Physics, University of Lincoln, Brayford Pool, Lincoln LN6 7TS, United Kingdom 3Department of Physics & Astronomy and London Centre for Nanotechnology, University College London, Gower Street, London WC1E 6BT, United Kingdom 2School of Mathematics and Physics, University of Lincoln, Brayford Pool, Lincoln LN6 7TS, United Kingdom 3Department of Physics & Astronomy and London Centre for Nanotechnology, University College London, Gower Street, London WC1E 6BT, United Kingdom g 4Department of Chemistry, Ångström Laboratory, Uppsala University, Box 538, S-751 21 Uppsala, Sweden (Received 4 February 2016; revised manuscript received 4 August 2016; published 9 September 2016) Polarons in metal oxides are important in processes such as catalysis, high temperature super- conductivity, and dielectric breakdown in nanoscale electronics. Here, we study the behavior of electron small polarons associated with oxygen vacancies at rutile TiO2ð110Þ, using a combination of low temperature scanning tunneling microscopy (STM), density functional theory, and classical molecular dynamics calculations. We find that the electrons are symmetrically distributed around isolated vacancies at 78 K, but as the temperature is reduced, their distributions become increasingly asymmetric, confirming their polaronic nature. Published by the American Physical Society under the terms of the Creative Commons Attribution 3.0 License. Further distri- bution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. week ending 9 SEPTEMBER 2016 week ending 9 SEPTEMBER 2016 P H Y S I C A L R E V I E W L E T T E R S PRL 117, 116402 (2016) DOI: 10.1103/PhysRevLett.117.116402 DOI: 10.1103/PhysRevLett.117.116402 DOI: 10.1103/PhysRevLett.117.116402 demonstrated very clearly in recent years [17–20]. The structure of rutile TiO2ð110Þ is characterized by rows of fivefold coordinated Ti ions that alternate with rows of bridging oxygen ions. When a bridging oxygen vacancy (Ob-vac) is formed via the removal of a neutral oxygen containing species (e.g., 1 2 O2 or H2O), two electrons are left behind in the lattice. While their association with states in the band gap, approximately 0.8 eV below the Fermi level, is well established [21,22], the spatial distributions of these “excess” electrons have been the subject of intense research in recent years, which is still ongoing. The growing con- sensus from theoretical calculations is that they localize on just two Ti ions close to Ob-vacs, reducing the formal charge of the former from þ4 to þ3 [6,23], and may therefore be considered as polarons bound to the Ob-vac [24]. Furthermore, theory predicts that the Ti3þ ions reside preferentially in the subsurface layer, although they may also populate the surface layer at higher temperatures [12,23]. On the other hand, the picture provided by the results of scanning tunneling microscopy (STM) experiments presented in the literature is ambiguous: Minato et al. [25] observed a symmetric four-lobe structure of the excess electrons at 78 K, whereas Papageorgiou et al. [8] found an asymmetric two-lobe distribution at 5 K. The reason for this discrepancy is not clear; it could be due to the different temperatures used in the two studies, or it could be that the high density of Ob-vacs in the latter study caused the excess electrons to interact with each other [8]. Furthermore, while resonant photoelectron diffraction [26] supports the idea demonstrated very clearly in recent years [17–20]. The structure of rutile TiO2ð110Þ is characterized by rows of fivefold coordinated Ti ions that alternate with rows of bridging oxygen ions. When a bridging oxygen vacancy (Ob-vac) is formed via the removal of a neutral oxygen containing species (e.g., 1 2 O2 or H2O), two electrons are left behind in the lattice. While their association with states in the band gap, approximately 0.8 eV below the Fermi level, is well established [21,22], the spatial distributions of these “excess” electrons have been the subject of intense research in recent years, which is still ongoing. Engineering Polarons at a Metal Oxide Surface By manipulating isolated vacancies with the STM tip, we show that particular configurations of polarons are preferred for given locations of the vacancies, which we ascribe to small residual electric fields in the surface. We also form a series of vacancy complexes and manipulate the Ti ions surrounding them, both of which change the associated electronic distributions. Thus, we demonstrate that the configurations of polarons can be engineered, paving the way for the construction of conductive pathways relevant to resistive switching devices. DOI: 10.1103/PhysRevLett.117.116402 Rectangles are drawn around the Ob- vacs and mark the same areas between the empty- and filled-state images in (a),(b). Scan parameters: (a) ðVS; ITÞ ¼ þ0.9 V, 20 pA; (b) −1.1 V, 5 pA; (c) −2 V, 1.6 pA; and (d)–(f) −2 V, 0.3 pA. We believe that the temperature dependence of the electronic distribution around the Ob-vacs can be under- stood only if the excess electrons associated with Ob-vacs are indeed polarons [28,29]. Two of the defining properties of polaronic trapping are the formation of multiple local structural minima associated with their presence at different lattice sites and the fact that their movement occurs via thermally activated hopping; thus, in contrast to charge carriers occupying delocalized states, the diffusion of self- trapped carriers increases with temperature. In the case of a polaron bound to an Ob-vac, the polaron is constrained to hop around the vacancy because of its electrostatic inter- action with the positively charged vacancy; thus, it is both trapped (at the defect) and self-trapped (at one of a number of equivalent sites) [30]. that the excess electrons reside at the subsurface, it is not clear that these would be observable in STM experiments. Here, we employ STM at temperatures between 7 and 78 K to provide evidence of the polaronic character of the excess electrons associated with Ob-vacs on rutile TiO2ð110Þ, and furthermore, to show that they may be manipulated using the STM tip. Previously, an energy barrier of 24 meV was found for electron polaron hopping in rutile TiO2 [31]. Using this value in a simple Arrhenius expression together with an attempt frequency of 1013 Hz (which corresponds to a typical phonon frequency) gives a polaron hopping fre- quency on the order of 10−5 Hz at 7 K and 1011 Hz at 78 K. Thus, given that the time scale of the STM scans is on the order of minutes, we would expect the electronic distribu- tion around an isolated Ob-vac at 78 K with a perfectly isotropic environment to be representative of thermal equilibrium. On the other hand, we would expect the distribution observed at 7 K to arise from kinetic trapping of the polarons in one of a number of symmetrically equivalent, and therefore degenerate, configurations. DOI: 10.1103/PhysRevLett.117.116402 Such defects have long been thought to be the predominant active sites at oxide surfaces, and in the case of the (110) surface of the model oxide rutile TiO2, this has been 116402-1 Published by the American Physical Society 0031-9007=16=117(11)=116402(5) week ending 9 SEPTEMBER 2016 week ending 9 SEPTEMBER 2016 P H Y S I C A L R E V I E W L E T T E R S PRL 117, 116402 (2016) FIG. 1. 3.2 × 3.2 nm2 representative STM images of isolated Ob-vacs on TiO2ð110Þ recorded using (a) positive (empty states) and (b)–(f) negative (filled states) sample bias at different temperatures: (a),(b) 78 K, (c) 16 K, and (d)–(f) 7 K. Filled-state images in (b)–(f) were recorded in the vicinity of different isolated Ob-vacs. Intersections of the white grids mark the positions of the Ob2−. Rectangles are drawn around the Ob- vacs and mark the same areas between the empty- and filled-state images in (a),(b). Scan parameters: (a) ðVS; ITÞ ¼ þ0.9 V, 20 pA; (b) −1.1 V, 5 pA; (c) −2 V, 1.6 pA; and (d)–(f) −2 V, 0.3 pA. At 16 K [Fig. 1(c)], although the electron distribution around the Ob-vac retains a four-lobe structure, two of the lobes appear much brighter, breaking the original mirror symmetry. This symmetry breaking is common to all the Ob-vacs that we sampled at 16 K, some having three brighter lobes and some having two, as in Fig. 1(c). The asymmetry in the spatial distribution of the excess electrons is even more pronounced at 7 K. Only two bright lobes are present, and these are oriented either directly across the row [Fig. 1(d)], along the row [Fig. 1(e)], or diagonally across it [Fig. 1(f)]. This asymmetry also appeared in spatially resolved scanning tunneling spectroscopy measurements performed at 5 K [8], although we emphasize that in that earlier study, interactions between nearby Ob-vacs could not be ruled out. FIG. 1. 3.2 × 3.2 nm2 representative STM images of isolated Ob-vacs on TiO2ð110Þ recorded using (a) positive (empty states) and (b)–(f) negative (filled states) sample bias at different temperatures: (a),(b) 78 K, (c) 16 K, and (d)–(f) 7 K. Filled-state images in (b)–(f) were recorded in the vicinity of different isolated Ob-vacs. Intersections of the white grids mark the positions of the Ob2−. DOI: 10.1103/PhysRevLett.117.116402 The growing con- sensus from theoretical calculations is that they localize on just two Ti ions close to Ob-vacs, reducing the formal charge of the former from þ4 to þ3 [6,23], and may therefore be considered as polarons bound to the Ob-vac [24]. Furthermore, theory predicts that the Ti3þ ions reside preferentially in the subsurface layer, although they may also populate the surface layer at higher temperatures [12,23]. When an electronic charge carrier is introduced into a polarizable material, it can either be delocalized over a number of symmetrically equivalent sites, or it can become localized at just one of them. In the latter case, the equilibrium positions of the surrounding lattice ions are altered to the extent that a potential well for the charge carrier is formed, in which it becomes strongly bound and immobilized. This self-trapped carrier, together with the surrounding displaced ions, forms a quasiparticle called a small polaron (henceforth referred to simply as a polaron) [1]. Polarons, both in the perfect lattice and bound to defects, are believed to play a central role in the physics and chemistry of many metal oxides, being implicated in phenomena as diverse as photolysis [2], high-Tc super- conductivity [3], and resistive switching [4]. For this reason, numerous studies have been carried out on tech- nologically relevant oxide materials in which self-trapping is believed to occur [5–16]. On the other hand, the picture provided by the results of scanning tunneling microscopy (STM) experiments presented in the literature is ambiguous: Minato et al. [25] observed a symmetric four-lobe structure of the excess electrons at 78 K, whereas Papageorgiou et al. [8] found an asymmetric two-lobe distribution at 5 K. The reason for this discrepancy is not clear; it could be due to the different temperatures used in the two studies, or it could be that the high density of Ob-vacs in the latter study caused the excess electrons to interact with each other [8]. Furthermore, while resonant photoelectron diffraction [26] supports the idea Typical preparation procedures of metal oxide samples under ultra-high vacuum lead to the formation of oxygen vacancies. DOI: 10.1103/PhysRevLett.117.116402 also shown in Fig. 3, indicates that their polaronic character in rutile TiO2 is relatively little affected, even when they are subjected to significant perturbations such as the electro- static field of the Ob-vac. Thus, we believe that the excess electrons associated with Ob-vacs are representative of both bound and unbound polaronic species, an assertion which has also been made recently by other authors [37]. Notwithstanding the long-ranged potential variations which we believe to be responsible for the preference of a given polaron configuration around an otherwise isolated vacancy, by forming aggregates of vacancies, we now go on to show that we can effectively manipulate the excess electrons by changing their local environments. Adding further support to this interpretation, we have also observed reversible changes in the polaron distribution around a vacancy which was manipulated back and forth between two locations. These images are shown in Fig. S2 of the Supplemental Material [27]. By applying a series of þ4 V pulses to several Ob-vacs, we formed Ob-vac dimers and trimers (see Supplemental Movies S2 and S3 [27]), as well as a tetramer, at 7 K. In the empty states [Fig. 4(a)], the dimer is characterized by an elliptical protrusion with a spatial extent of two lattice units along the Ob2−row. In the filled states [Fig. 4(b)], the electron distribution appears to comprise four bright lobes and a central spot appearing less bright than the lobes. The four bright lobes, which are located on the nearest Ti5c neighbors to the Ob-vac dimer, are of unequal brightness, consistent with earlier measurements [36] where Ob-vac dimers were formed in an uncontrolled way during a þ3 V scan on the TiO2ð110Þ surface at temperatures of 80 K and above. This symmetry breaking once again is indicative of the polaronic nature of the excess electrons in contrast with As mentioned above, empty-state imaging of Ob-vacs is well understood [33–35], but filled-state imaging has received significantly less attention [8,25,36]. Thus, the imaging mechanism itself merits some discussion. As shown in Fig. DOI: 10.1103/PhysRevLett.117.116402 If this is the case, manipulation of a particular vacancy at 7 K should lead to a redistribution of the electrons around it, a hypothesis that we now go on to consider. Briefly, the experiments were performed using an ultra- high vacuum, low temperature STM apparatus, configured to take measurements of the current at both positive bias (probing the empty states of the sample) and negative bias (probing the filled states) simultaneously. Values for the negative bias were chosen so as to probe only states within the band gap and not those in the valence band of the material. In order to obtain a sample which is sufficiently conductive at 7 K, yet exhibits Ob-vacs which are well separated, we used a special three-step preparation procedure involving the reaction of Ob-vacs with water and removal of H from the resulting bridging hydroxyl groups. No unusual behavior compared to as-prepared surfaces was observed. See the Supplemental Material for further details [27]. Figure 1 shows typical empty- and filled-state STM images of isolated Ob-vacs recorded at 78, 16, and 7 K. The empty-state images do not change with temperature, so we focus solely on the filled-state images. At 78 K, all the Ob-vacs have an excess electron distribution with a four-lobe structure [Fig. 1(b)]. The lobes are arranged so that the electron distribution exhibits 2mm symmetry about the Ob-vac, with mirror symmetry about both the [001] and ½¯110 axes, which is consistent with earlier measurements by Minato et al. [25]. We use our methodology of precise vacancy manipula- tion (described in the Supplemental Material [27]) to modify the surface in a way that allows us to explore the dependence of the polaron configuration on the local environment. Figure 2 shows three sequential images recorded during manipulation of a single Ob-vac (the full movie is given in Supplemental Movie S1 [27]). It can be seen clearly from the filled-state image in Fig. 2(e) that the 116402-2 week ending 9 SEPTEMBER 2016 P H Y S I C A L R E V I E W L E T T E R S PRL 117, 116402 (2016) FIG. 3. Calculation of individual Ob-vacs on TiO2ð110Þ. (a) Calculated structure of a single Ob-vac and one of its ground-state excess electron configurations on rutile TiO2ð110Þ. Turquoise spheres are sites of the removed bridging oxygen ions (Ob2−). DOI: 10.1103/PhysRevLett.117.116402 Black and pink open circles mark the initial and current positions of the Ob-vac. Intersections of the white grids mark the positions of Ob2−. All images were taken at 7 K. Scan parameters: (a)–(c) ðVS; ITÞ ¼ þ1.0 V, 10 pA and (d)–(f) −2.0 V, 0.3 pA. FIG. 2. STM images of an individual Ob-vac of TiO2ð110Þ taken (a),(d) before manipulation, (b),(e) after the Ob-vac was displaced along the Ob row by a series of þ4 V, 6 s tip pulses, and (c),(f) after the Ob-vac was returned to its initial position by tip pulses. Black and pink open circles mark the initial and current positions of the Ob-vac. Intersections of the white grids mark the positions of Ob2−. All images were taken at 7 K. Scan parameters: (a)–(c) ðVS; ITÞ ¼ þ1.0 V, 10 pA and (d)–(f) −2.0 V, 0.3 pA. FIG. 3. Calculation of individual Ob-vacs on TiO2ð110Þ. (a) Calculated structure of a single Ob-vac and one of its ground-state excess electron configurations on rutile TiO2ð110Þ. Turquoise spheres are sites of the removed bridging oxygen ions (Ob2−). Small pink and red spheres are the Ti and O ions, respectively. Highlighted in blue is an isosurface of the excess electron density at 10−3 electrons=Å3. (b)–(f) Simulated STM images of individual Ob-vacs of TiO2ð110Þ with four nonequivalent polaron configurations. The excess electrons, visualized as red dumbbells in the filled-state images in (c)–(f), are located at different Ti sites beneath the Ob-vacs. Relative energies after ionic relaxation are indicated. Open circles mark the positions of Ob-vacs. Intersections of the white grids mark the positions of the Ob2−. two lobes have moved from their positions in Fig. 2(d) so that their locations relative to the Ob-vac remain unchanged. Our explanation for this is that the observed asymmetric configuration represents the sole ground state. As such, following the manipulation pulses, the polaron distribution relaxes back to this sole ground state rather than being kinetically trapped in one of a number of degenerate ground states. It is well known that oxide surfaces have a nonuniform local potential [32], and we believe that these sorts of variations explain the preferential positioning of the polarons at different locations. To illustrate this point, at 7 K, an energetic preference for a given configuration of just 1 meV would lead to a Boltzmann weight which is approximately 5 times larger than that of the next most stable configuration. DOI: 10.1103/PhysRevLett.117.116402 Small pink and red spheres are the Ti and O ions, respectively. Highlighted in blue is an isosurface of the excess electron density at 10−3 electrons=Å3. (b)–(f) Simulated STM images of individual Ob-vacs of TiO2ð110Þ with four nonequivalent polaron configurations. The excess electrons, visualized as red dumbbells in the filled-state images in (c)–(f), are located at different Ti sites beneath the Ob-vacs. Relative energies after ionic relaxation are indicated. Open circles mark the positions of Ob-vacs. Intersections of the white grids mark the positions of the Ob2−. two lobes have moved from their positions in Fig. 2(d) so that their locations relative to the Ob-vac remain unchanged. Our explanation for this is that the observed asymmetric configuration represents the sole ground state. As such following the manipulation pulses the polaron FIG. 2. STM images of an individual Ob-vac of TiO2ð110Þ taken (a),(d) before manipulation, (b),(e) after the Ob-vac was displaced along the Ob row by a series of þ4 V, 6 s tip pulses, and (c),(f) after the Ob-vac was returned to its initial position by tip pulses. Black and pink open circles mark the initial and current positions of the Ob-vac. Intersections of the white grids mark the positions of Ob2−. All images were taken at 7 K. Scan parameters: (a)–(c) ðVS; ITÞ ¼ þ1.0 V, 10 pA and (d)–(f) −2.0 V, 0.3 pA. FIG. 3. Calculation of individual Ob-vacs on TiO2ð110Þ. (a) Calculated structure of a single Ob-vac and one of its ground-state excess electron configurations on rutile TiO2ð110Þ. Turquoise spheres are sites of the removed bridging oxygen ions (Ob2−). Small pink and red spheres are the Ti and O ions, respectively. Highlighted in blue is an isosurface of the excess electron density at 10−3 electrons=Å3. (b)–(f) Simulated STM images of individual Ob-vacs of TiO2ð110Þ with four nonequivalent polaron configurations. The excess electrons, visualized as red dumbbells in the filled-state images in (c)–(f), are located at different Ti sites beneath the Ob-vacs. Relative energies after ionic relaxation are indicated. Open circles mark the positions of Ob-vacs. Intersections of the white grids mark the positions of the Ob2−. FIG. 2. STM images of an individual Ob-vac of TiO2ð110Þ taken (a),(d) before manipulation, (b),(e) after the Ob-vac was displaced along the Ob row by a series of þ4 V, 6 s tip pulses, and (c),(f) after the Ob-vac was returned to its initial position by tip pulses. DOI: 10.1103/PhysRevLett.117.116402 Scan parameters: (a),(c), (i) ðVS; ITÞ ¼ þ1.0 V, 10 pA, (b),(d),(j) −2.0 V, 0.3 pA, (e),(g) þ0.7 V, 10 pA, and (f),(h) −1.0 V, 1.6 pA. STM images in (a)–(d) and (i) and (j) were taken at 7 K, those in (e)–(h) taken at 78 K. the symmetric, delocalized description of the density functional theory calculations of Cui et al. [36]. most undercoordinated Ti ions at the center of the vacancy complex is relocated to a subsurface octahedral interstitial site, increasing its coordination number from 4 to 6. Each of the two Ti ions can relocate to the left or right of their original positions, accounting for the reflection symmetries observed in the experiments. This gives a total of five structural configurations: four mushrooms and one cross dumbbell (one of the mushroom configurations is illustrated in Supplemental Fig. S3 [27]). According to our calculations, the electrons are redistributed around the vacancy complex following the structural changes but remain well localized in all the configurations considered; thus, in effect, the polarons are manipulated by the STM tip via the induced structural changes. Figures 4(c) and 4(d) show an Ob-vac trimer at 7 K. In the empty-state image [Fig. 4(c)], the trimer is character- ized by a set of two perpendicularly aligned dumbbell shapes, the brighter of which is parallel to the Ob row. In the filled-state image [Fig. 4(d)], the excess electron distribu- tion is characterized by a dumbbell shape, with the excess electrons mainly populating the nearest Ti5c sites and, to a lesser extent, the center of the trimer. By applying an electrical pulse (þ3.75 V, 0.2 s) to a cross-dumbbell trimer [Figs. 4(c) and 4(d)], the structure of the trimer is transformed. In the empty-state image [Fig. 4(e)], the trimer is now characterized by a mushroom shape: it has a stem consisting of a dark, round feature that resides at the bottom-right corner and a bright cap comprising an elliptical part and a round part that are located at the top-right and bottom-left corners of the trimer, respectively. In the filled-state image [Fig. 4(f)], it has a three-lobe structure with a triangular configuration. DOI: 10.1103/PhysRevLett.117.116402 3(a), our density functional theory calcu- lations indicate that, while the major contribution to the excess electron distribution is from dz2 orbitals (z being normal to the surface) of (formally) Ti3þ ions in the second layer of Ti atoms, there are also some contributions from Ti ions at the surface, and it is the electron density decaying into the vacuum above these surface ions that is responsible for the filled-state images. Furthermore, we note that the minor changes in the calculated appearance of the excess electrons upon changes in their proximity to the Ob-vac, 116402-3 116402-3 week ending 9 SEPTEMBER 2016 P H Y S I C A L R E V I E W L E T T E R S PRL 117, 116402 (2016) FIG. 4. Experimental STM images of Ob-vac multimers obtained simultaneously at positive and negative sample biases. The Ob-vac dimer in (a) and (b), cross-dumbbell trimer in (c) and (d), and tetramer in (i) and (j) were formed using a series of þ4 V, 6 s tip pulses at 7 K. The mushroom trimers in (e)–(h) were formed by applying þ3.75 V, 0.2 s tip pulses over a cross-dumbbell trimer at 78 K. The intersections of the white grids mark the positions of Ob2−. Circles mark the removed Ob2−. Scan parameters: (a),(c), (i) ðVS; ITÞ ¼ þ1.0 V, 10 pA, (b),(d),(j) −2.0 V, 0.3 pA, (e),(g) þ0.7 V, 10 pA, and (f),(h) −1.0 V, 1.6 pA. STM images in (a)–(d) and (i) and (j) were taken at 7 K, those in (e)–(h) taken at 78 K. PRL 117, 116402 (2016) P H Y S I C A L R E V I E W L E T T E R S week ending 9 SEPTEMBER 2016 FIG. 4. Experimental STM images of Ob-vac multimers obtained simultaneously at positive and negative sample biases. The Ob-vac dimer in (a) and (b), cross-dumbbell trimer in (c) and (d), and tetramer in (i) and (j) were formed using a series of þ4 V, 6 s tip pulses at 7 K. The mushroom trimers in (e)–(h) were formed by applying þ3.75 V, 0.2 s tip pulses over a cross-dumbbell trimer at 78 K. The intersections of the white grids mark the positions of Ob2−. Circles mark the removed Ob2−. DOI: 10.1103/PhysRevLett.117.116402 White, and Z. Dohnálek, J. Phys. Chem. B 110, 21840 (2006). [20] D. Matthey, J. G. Wang, S. Wendt, J. Matthiesen, R. Schaub, E. Lægsgaard, B. Hammer, and F. Besenbacher, Science 315, 1692 (2007). [21] V. E. Henrich, G. Dresselhaus, and H. J. Zeiger, Phys. Rev. Lett. 36, 1335 (1976). [22] R. L. Kurtz, R. Stock-Bauer, T. E. Madey, E. Román, and J. L. De Segovia, Surf. Sci. 218, 178 (1989). [23] P. M. Kowalski, M. F. Camellone, N. N. Nair, B. Meyer, and D. Marx, Phys. Rev. Lett. 105, 146405 (2010). [24] A. L. Shluger, K. P. McKenna, P. V. Sushko, D. M. Ramo, and A. V. Kimmel, Model. Simul. Mater. Sci. Eng. 17, 084004 (2009). [25] T. Minato, Y. Sainoo, Y. Kim, H. S. Kato, K.-I. Aika, M. Kawai, J. Zhao, H. Petek, T. Huang, W. He, B. Wang, Z. Wang, Y. Zhao, J. Yang, and J. G. Hou, J. Chem. Phys. 130, 124502 (2009). [1] D. Emin, Polarons (Cambridge University Press, New York, 2013). [26] P. Krüger, J. Jupille, S. Bourgeois, B. Domenichini, A. Verdini, L. Floreano, and A. Morgante, Phys. Rev. Lett. 108, 126803 (2012). [2] S. D. Jackson and J. S. J. Hargreaves, Metal Oxide Catalysis (Wiley, Weinheim, 2008). [3] E. Salje, A. S. Alexandrov, and W. Y. Liang, Polarons and Bipolarons in High-Tc Superconductors and Related Ma- terials (Cambridge University Press, Cambridge, England, 2005). [27] See Supplemental Material at http://link.aps.org/ supplemental/10.1103/PhysRevLett.117.116402 for exper- imental details, calculation details, and supplemental figures and movies. [4] M. Wang, C. Bi, L. Li, S. Long, Q. Liu, H. Lv, N. Lu, P. Sun, and M. Liu, Nat. Commun. 5, 4598 (2014). [28] N. A. Deskins, R. Rousseau, and M. Dupuis, J. Phys. Chem. C 115, 7562 (2011). [5] P. Deák, B. Aradi, and T. Frauenheim, Phys. Rev. B 86, 195206 (2012). [29] T. Shibuya, K. Yasuoka, S. Mirbt, and B. Sanyal, J. Phys. Chem. C 118, 9429 (2014). [6] M. Setvin, C. Franchini, X. Hao, M. Schmid, A. Janotti, M. Kaltak, C. G. Van de Walle, G. Kresse, and U. Diebold, Phys. Rev. Lett. 113, 086402 (2014). [30] A. M. Stoneham, J. Gavartin, A. L. Shluger, A. V. Kimmel, D. M. Ramo, H. M. Rønnow, G. Aeppli, and C. Renner, J. Phys. Condens. Matter 19, 255208 (2007). [7] S. Moser, L. Moreschini, J. Jaćimović, O. S. Barišić, H. Berger, A. Magrez, Y. J. Chang, K. S. Kim, A. DOI: 10.1103/PhysRevLett.117.116402 around a vacancy trimer can be made to adopt a number of configurations, which can be selected by an STM tip pulse. Thus, we demonstrate that it is possible to engineer the configurations of polarons, which allows for the study of their interactions and which may allow the construction of directional conduction pathways. This could be important, for example, in memristive devices. around a vacancy trimer can be made to adopt a number of configurations, which can be selected by an STM tip pulse. Thus, we demonstrate that it is possible to engineer the configurations of polarons, which allows for the study of their interactions and which may allow the construction of directional conduction pathways. This could be important, for example, in memristive devices. [12] H. M. Rønnow, C. Renner, G. Aeppli, T. Kimura, and Y. Tokura, Nature (London) 440, 1025 (2006). [13] S. Chrétien and H. Metiu, J. Phys. Chem. C 115, 4696 (2011). [14] N. A. Deskins and M. Dupuis, Phys. Rev. B 75, 195212 (2007). [15] H. Sezen, M. Buchholz, A. Nefedov, C. Natzeck, S. Heissler, C. Di Valentin, and C. Wöll, Sci. Rep. 4, 3808 (2014). We acknowledge useful discussions with Professor Alex Shluger. This work was supported by the European Research Council Advanced Grant ENERGYSURF (G. T.), the Engineering and Physical Sciences Research Council (EPSRC) (UK), European Cooperation in Science and Technology Action CM1104, the Royal Society (UK), and Alexander von Humboldt Stiftung (Germany). M. J. W. andK. H. acknowledge funding from the Swedish Research Council (VR) and the Swedish Foundation for International Cooperation in Research and Higher Education (STINT). Through our membership of the UK’s High Performance Computing Materials Chemistry Consortium, which is funded by the EPSRC (EP/L000202), this work made use of the facilities of High-End Computing Terascale Resource (HECToR) and Advanced Research Computing High-End Resource (ARCHER), the UK’s national high- performance computing service, which is funded through the EPSRC’s High End Computing Programme. [16] M. Setvin, M. Schmid, and U. Diebold, Phys. Rev. B 91, 195403 (2015). [17] S. Wendt, R. Schaub, J. Matthiesen, E. K. Vestergaard, E. Wahlström, M. D. Rasmussen, P. Thostrup, L. M. Molina, E. Lægsgaard, I. Stensgaard, B. Hammer, and F. Besenbacher, Surf. Sci. 598, 226 (2005). [18] O. Bikondoa, C. L. Pang, R. Ithnin, C. A. Muryn, H. Onishi, and G. Thornton, Nat. Mater. 5, 189 (2006). [19] Z. Zhang, O. Bondarchuk, B. D. Kay, J. M. DOI: 10.1103/PhysRevLett.117.116402 While the cross-dumbbell and mushroom configurations are essentially equal in energy to within the accuracy of the calculations, the energy barrier for the transformation between the two structures was calculated to be ∼0.3 eV, which should prevent thermal conversion at the low temperatures used in our experiments. This is consistent with the fact that we do not observe any change to the structure of a trimer without applying a tip pulse. We examine whether or not this transformation is revers- ible by applying additional electrical pulses to the mushroom trimer in Figs. 4(e) and 4(f). Following an electrical pulse, the mushroom transforms into a structure which is equivalent but reflected across the ½¯110 axis, as shown in Figs. 4(g) and 4(h). Another pulse causes the trimer to revert to the original cross-dumbbell shape. We emphasize that each transforma- tion is accompanied by a corresponding change in the excess electron distribution; thus, pulsing of the trimer alters the distribution of the excess electrons in a controlled and reversible way. Simulated empty- and filled-state STM images for the Ob-vac dimer, two types of trimers, and a tetramer are shown in Supplemental Fig. S3 [27]. Finally, we note that the susceptibility of the vacancy complexes to such distortions appears to be a fairly general feature. The nonlinear appearance of the tetramer in the experiment [Figs. 4(i) and 4(j)] is due to a similar displacement of a Ti ion that enables the system to gain ∼0.5 eV in energy (see Supplemental Fig. S3 [27]). To summarize, we have shown that the excess electrons associated with Ob-vacs on TiO2ð110Þ are polaronic in nature, but that their precise distribution around isolated vacancies at low temperatures is governed by electric fields in the surface. However, their distribution can be controlled by manipulating the Ob-vacs into complexes of several adjacent vacancies. We have shown that the electrons Our simulations show that when a cross-dumbbell trimer transforms into the mushroom structure, one of the two 116402-4 week ending 9 SEPTEMBER 2016 week ending 9 SEPTEMBER 2016 P H Y S I C A L R E V I E W L E T T E R S PRL 117, 116402 (2016) [11] K. P. McKenna, M. J. Wolf, A. L. Shluger, S. Lany, and A. Zunger, Phys. Rev. Lett. 108, 116403 (2012). DOI: 10.1103/PhysRevLett.117.116402 Bostwick, E. Rotenberg, L. Forró, and M. Grioni, Phys. Rev. Lett. 110, 196403 (2013). [31] S. Yang, A. T. Brant, N. C. Giles, and L. E. Halliburton, Phys. Rev. B 87, 125201 (2013). [32] C. Barth and C. R. Henry, Nanotechnology 17, S155 (2006). [33] C. L. Pang, R. Lindsay, and G. Thornton, Chem. Rev. 113, 3887 (2013). [8] A. C. Papageorgiou, N. S. Beglitis, C. L. Pang, G. Teobaldi, G. Cabailh, Q. Chen, A. J. Fisher, W. A. Hofer, and G. Thornton, Proc. Natl. Acad. Sci. U.S.A. 107, 2391 (2010). [34] U. Diebold, Surf. Sci. Rep. 48, 53 (2003). 5] M. A. Henderson, Surf. Sci. Rep. 66, 185 (2011 [9] C. Di Valentin, G. Pacchioni, and A. Selloni, Phys. Rev. Lett. 97, 166803 (2006). [36] X. Cui, B. Wang, Z. Wang, T. Huang, Y. Zhao, J. Yang, and J. G. Hou, J. Chem. Phys. 129, 044703 (2008). [10] J.-F. Jerratsch, X. Shao, N. Nilius, H.-J. Freund, C. Popa, M. V. Ganduglia-Pirovano, A. M. Burow, and J. Sauer, Phys. Rev. Lett. 106, 246801 (2011). [37] P. G. Moses, A. Janotti, C. Franchini, G. Kresse, and C. Van de Walle, J. Appl. Phys. 119, 181503 (2016). 116402-5 116402-5
W4383909022.txt	https://journals.wlb-stuttgart.de/ojs/index.php/zwlg/article/download/6822/6709	de		Rezension von: Gassert, Philipp; Nieß, Ulrich; Stockert, Harald (Hrsg.), Zusammenleben in Vielfalt	Zeitschrift für württembergische Landesgeschichte	2,023	cc-by	1,544	Städte und Orte 561 Philipp Gassert / Ulrich Nieß / Harald Stockert (Hg.), Zusammenleben in Vielfalt. Zuwanderung nach Mannheim von 1607 bis heute (Veröffentlichungen zur Mannheimer Migrationsgeschichte, Bd. 1). Ubstadt-Weiher: verlag regionalkultur 2021. 319 S., zahlr. Abb. und Grafiken. ISBN 978-3-95505-311-6. Brosch. € 24,80 „Seit ihrer Gründung ist Mannheim eine Einwanderungsstadt.“ Das knappe Statement im Geleitwort (S. 3) füllen 13 Beiträge mit historisch-sozialem Leben von 1607 bis 2020. Auch die Autoren und Autorinnen, alle von den Mannheimer Geschichtsinstitutionen (Historisches Institut der Universität, Altertumsverein, MARCHIVUM, TECHNOSEUM), betonen, dass Zuwanderung hier „schon immer eine außergewöhnliche Rolle gespielt hat – und immer noch spielt“ (S. 9), bei „Menschen aus rund 170 Nationen“, die „im globalisierten 21. Jahrhundert“ zusammenleben (S. 7). Doch sei sie stets eine Herausforderung: „Die Begegnung mit dem Fremden kann Interesse und Neugier, aber auch Ängste und Ablehnung hervorrufen, kann als Bereicherung wie als Bedrohung wahrgenommen werden“ (S. 3). Beides zieht sich wie ein roter Faden durch die Epochen. „Freyheiten“ in vier Sprachen versprechend, wurden 1607 qualifizierte „reformierte Refugianten“ aus Frankreich und der Wallonie in die Neugründung gelockt. Sie trafen „auf eine pfälzische Mehrheitsbevölkerung, die den Fremden gegenüber keineswegs immer freundlich gesinnt war“ (H. Wiegand, S. 12 – 29). Auch zum Wiederaufbau seit 1649 waren „Leut von allen Nationen“ nötig, weshalb „einzigartige“ Privilegien ausgeschrieben wurden – ein „multinationales und -religiöses Experiment“ ohne „religiöse Abschottung“, auch nicht der Juden. Zur Versöhnung der christlichen Konfessionen entstand 1677 – 80 die Concordienkirche, doch war „die religiöse Vielfalt ein Wunsch des Kurfürsten, weniger der Einwohnerinnen selbst“. Ein „wertschätzendes Miteinander“ kam nicht zustande, und die Zerstörung 1689 war auch das „gewalt same Ende des Mannheimer Experiments“ (S. Schlösser, S. 30 – 49). Für die „Auferstehung aus Ruinen“ nach 1697 war wieder Zuwanderung der „elementare Baustein“, vor allem aus der Region. Zur Multinationalität trugen Spezialisten aus Wallonien, Frankreich und Italien bei. Erneut traten konfessionelle Animositäten zutage, so durch Sonderrechte wie konfessionelle Zünfte, die sich „argwöhnisch beäugten“. „Nicht erwünscht“ waren Sinti und „Betteljuden“, bedingt geduldet Mennoniten und Pietisten; Letztere wanderten seit 1709 nach Nordamerika aus. Wegen ihrer Wirtschaftskraft wurden Juden begünstigt, was Konfliktstoff mit dem christlichen Umfeld barg. So konnte auch jetzt „von einer auf Toleranz basierenden Stadtgesellschaft schwerlich die Rede sein“, und trotz notwendiger Zuwanderung bestand eine „weitgehend klare Abgrenzung“ (U. Nieß, S. 50 – 71). Seit 1720 Residenz, veränderte sich Mannheim zu einer „internationalen, multikulturellen Stadt“, die Menschen anzog. Hofangehörige, Militärs und Klerus hatten einen Sonderstatus und pflegten die französische Hofkultur. Mannheim wurde zur „Konsumentenstadt“, was unter anderem italienische Händler für Luxuswaren nutzten. Den Zuzug anderer „Fremder“ handhabte man jedoch restriktiv und schrieb für eine bürgerliche Existenz Mindestvermögen und -alter vor: „Eine sozial differenzierte Bevölkerungspolitik, die stark auf das ökonomische Potenzial der Einwanderungswilligen setzte“ (H. Kümper, S. 72 – 93). Die „Migrationssteuerung“ blieb, bewirkt auch durch den Verlust des Hofes 1778 und die Revolutionskriege, die Mannheim verarmen ließen. 1803 an Baden gediehen, war den meisten Personen „ein Recht auf Mobilität und Migration“ weiterhin versagt. „Unerwünschte“ ließ die Polizeiordnung ausweisen, das Bürger- und Beisassenrecht war restriktiv, die Ansiedlung Wohlhabender jedoch gewünscht (S. Pister, S. 94 – 109). Die Frühindustrialisierung 1815 – 1860 sah den Konflikt zwischen staatlichen und kommunalen Interessen: Freizügig- Zeitschrift für Württembergische Landesgeschichte 82 (2023) © Kommission für geschichtliche Landeskunde in Baden-Württemberg und Württembergischer Geschichts- und Altertumsverein e.V. ISSN 0044-3786 (Print) und 2749-1277 (Online) 562 Buchbesprechungen keit und Gewerbefreiheit, „dem gesamtökonomischen Fortschritt dienlich“, standen gegen den Schutz des Handwerks, die Furcht vor Übervölkerung und Überlastung des Sozialsystems. Doch waren ausländische Unternehmer und Fachkräfte „gerne gesehen“. So kamen „maßgebliche Impulse zur Entwicklung aus dem Ausland“ und „läuteten den Aufstieg zur Industriemetropole ein“: Italienische Kaufleute (Giulini), britische Rentiers und Ingenieure (Newhouse), das französische Personal der „Spiegelfabrik“ (E. Veyel, S. 110 – 129). Im Gefolge der „neuen Ära“ in Baden seit 1860 mit liberalen Reformen wie Handels- und Gewerbefreiheit, Aufhebung der Zünfte, bedingter Niederlassungsfreiheit und Emanzipa tion der Juden entwickelte sich Mannheim zur Handelsmetropole und Industriestadt. In die Gänge kamen auch Mobilität und Wachstum der Bevölkerung, die sich 1861 – 1890 verdreifachte, mit dem „Bauboom der expandierenden Großstadt und den aus dem Boden schießenden Industriebetrieben“ (W. Kreutz, S. 130 – 147). Bewegungsfreiheit zwischen den deutschen Staaten und ein moderates Einbürgerungsrecht erleichterten 1890 – 1914 die Arbeitsmigration vor allem „aus dem Reich“, die Hälfte aus Baden, dessen arme Bevöl kerung, etwa aus dem Odenwald, „in die verheißungsvolle Metropole strömte“. „Trans nationale Migration“ kam aus Österreich-Ungarn, der Schweiz, Italien und Russland – im Industriegebiet Sandhofen war „ein Gewirr von Sprachen und Trachten“. Doch sah man auch „unverhohlene Anfeindungen“ und „rassistische Einordnungen“ seitens der Einheimischen, die Kriminalität und Verwahrlosung befürchteten, während die Zuwanderer unter den Arbeitsbedingungen und Wohnverhältnissen litten (A. Gillen, S. 148 – 167). 1914 wurden viele zu „feindlichen Ausländern“, zu denen sich Gefangene und Zivilarbeiter gesellten, während Vertriebene und Evakuierte in die Stadt strömten (D. Kelimes, S. 168 – 189). Nach dem 1. Weltkrieg drängten Flüchtlinge und heimkehrende Soldaten herein, Zuwanderung schien unnötig, und der „Inländervorrang“ ließ die Zuzugsanträge ab lehnen. Die „Verhinderung von Arbeitsmigration“ ging einher mit einer „Misstrauenskultur gegenüber Ausländern“. Sie verschärfte das NS-Regime durch „eine rassistische Ausweisungs- und Abschiebepolitik“. Zuwanderung war nur saisonal, so aus Italien; 1938 kamen erste „Fremdarbeiterinnen“ (H. Stockert, S. 190 – 207). Die Kriegs- und Nachkriegsjahre sahen dann „gewaltige Fluchtbewegungen“ und „erzwungene Ab- und Zuwanderungen“. Als Kategorien finden sich: Kriegsgefangene, „Zwangsarbeiterinnen“, Häftlinge, De portierte, Evakuierte, Displaced Persons, Heimatvertriebene, Kriegsheimkehrer, jüdische DPs und Rückwanderer, Ostzonenflüchtlinge, „Spätheimkehrerinnen“ (A. Borgstedt, S. 208 – 225). Die Jahre 1955 – 1980 standen im Zeichen der „Gastarbeiter“: Ihr „Willkommen nur als Werktätige“, die Jobs im Niedriglohnsektor und als „Konjunkturpuffer“, Diskriminierung, behelfsmäßige Unterkünfte, Argwohn der Bevölkerung, Benachteiligung in der Bildung, wogegen Deutsche mehr Chancen für Aufstieg bekamen. Hatten 1983 schon 14,4 % der Einwohner einen Migrationshintergrund, so beinhaltete dies die Erklärung der Stadt zum „überlasteten Siedlungsgebiet“, Notstände durch plötzliche Arbeitslosigkeit und steigende Kosten im Sozial- und Bildungssektor. Langsam kamen auf kommunaler Ebene Maßnahmen zur Integration, so 1974 ein „Ausländer-Beauftragter“, während es bundespolitisch hieß: „Die BRD ist kein Einwanderungsland.“ Und „im öffentlichen Diskurs festigte sich das Vorurteil, ‚Ausländer‘ als ‚Problem‘ und ‚Bedrohung‘ wahrzunehmen“ (J. Scialpi, S. 226 – 259). In die Aktualität „einer widerwilligen Einwanderungsgesellschaft“ führt „1980 – 2020: Eine ‚City of Immigrants‘“. Ausgangspunkt ist die „Erklärung für Toleranz und Offenheit“ von 1989, die für ein „Miteinander in Vielfalt“ plädierte. Mannheim entdeckte sich „affir Zeitschrift für Württembergische Landesgeschichte 82 (2023) © Kommission für geschichtliche Landeskunde in Baden-Württemberg und Württembergischer Geschichts- und Altertumsverein e.V. ISSN 0044-3786 (Print) und 2749-1277 (Online) Städte und Orte 563 mativ als Einwanderungsstadt“, trotz der „Sprengkraft migrationspolitischer Fragen“ und alltäglichem Fremdenhass. Symbol ist der „hart umkämpfte, heute akzeptierte Bau der Yavuz-Sultan-Selim-Moschee“ (1993 – 1995). „Befreiungsschlag“ war das Staatsangehörigkeitsgesetz von 1998 mit dem Doppelpass, während die Stadt die lokale Migrationspolitik professionalisierte und „Multikulti“ zur Leitidee wurde. Neue Abwehrhaltungen kamen mit den Asylanten, ebenso die Erkenntnis, dass „die Probleme herkunftsgesteuerter Vielfalt niemals vollständig gelöst werden können“, da sie „Teil gesellschaftlicher Dynamiken sind“. Die Integration von Eingewanderten ist, wie Sozial-, Verkehrs- und Bildungspolitik, „eine Dauerbaustelle“ – und „Einwanderungsstadt zu sein, heißt nicht, dass Fremdenfeindlichkeit verschwindet“. Doch wird für 2020 festgehalten, dass hier „die Stadt der Eingewanderten Normalität ist“ und man „für das Zusammenleben in Metropolen“ sogar „Vorbild“ sein will (Ph. Gassert, S. 260 – 295). Mit Sicherheit haben die „Expertinnen und Experten“ (S. 7) ihre jeweilige Zielsetzung für die Darstellung der Mannheimer Zuwanderungsgeschichte erreicht, vorbildlich auch für andere Kommunen. Dazu tragen die kompakten, übersichtlichen Texte ebenso bei wie die reiche Bebilderung und die Schaubilder. Von allgemeinem Wert sind die Erkenntnisse über „die Fremdheit“, die weniger „kulturell“ als „sozial“ verursacht ist: durch „mangelnde Zugehörigkeit“ (S. 89 f.) bzw. „Nichtzugehörigkeit“ (S. 247). Betroffen macht, dass trotz aller Initiativen und Kampagnen der Rassismus nicht zu verdrängen ist: „Er ist in die Gesellschaft tief eingeschrieben“ (S. 287). Ein Problem des Rezensenten am Rande: Das ausgiebige „Gendern“, das dem „Vergnügen an der Lektüre“ (S. 5) doch zusetzte. Neben Paarformen („Exulantinnen und Exulanten“) häufen sich die Gendersternchen: „Erst als Bürgerinnen bzw. Beisassinnen waren Migrantinnen […] integriert.“ Kunstwörter wie „Zeitgenossinnen“, „Dienstbotinnen“, „Türkinnen“, „Polinnen“, „Russinnen“ u. v. a. bestehen die „Weglassprobe“ nicht. Wäre in ihrem Kontext das generische Maskulinum nicht sprachlich eindeutig? Wen meinen „Refugiées“, und sind „feindliche Ausländerinnen“ oder „Spätheimkehrerinnen“ wirklich Hans Harter aus den Quellen zitiert? Lea Oberländer, Mannheims verdrängte Opfer. Porträt einer Stadt im System der NS-„Euthanasie“. Ubstadt-Weiher: verlag regionalkultur 2022. 312 S., 27 Abb. ISBN 978-3-95505-318-5. € 29,80 In den letzten Jahren sind mehrere dezidiert lokal- und regionalgeschichtliche Studien zum Themenkomplex Zwangssterilisation und NS-„Euthanasie“ veröffentlicht worden. Während der Fokus zahlreicher historischer Studien bisher auf den Tatorten, also den einzelnen Anstalten, Heimen und Kliniken lag, verschiebt sich in jüngerer Zeit der Fokus mehr auf das soziale Eingebundensein der Institutionen und ihre Interaktionen mit der breiteren Gesellschaft. Die detaillierte Rekonstruktion der verschiedenen Mordaktionen ist noch immer nicht überall abgeschlossen. Dennoch rücken nun verstärkt breitere, sozialhistorische Fragestellungen in den Mittelpunkt. Welche gesamtgesellschaftlichen Konstellationen ermöglichten die Patientenmorde, und wie wirkten sich diese wiederum auf die Sozialstrukturen im Nationalsozialismus und in der frühen Bundesrepublik aus? Diesem Problemkomplex widmet sich auch Lea Oberländer. Oberländers Studie ist die überarbeitete Fassung ihrer Dissertation, die sie im Sommer 2020 an der Universität Mannheim eingereicht hat. Über etwas mehr als 300 Seiten verfolgt Zeitschrift für Württembergische Landesgeschichte 82 (2023) © Kommission für geschichtliche Landeskunde in Baden-Württemberg und Württembergischer Geschichts- und Altertumsverein e.V. ISSN 0044-3786 (Print) und 2749-1277 (Online)
https://openalex.org/W3112088655	https://clinicalepigeneticsjournal.biomedcentral.com/track/pdf/10.1186/s13148-020-00986-3	English	null	Reproductive technologies, female infertility, and the risk of imprinting-related disorders	Clinical epigenetics	2,020	cc-by	7,547	Abstract Background: Epidemiological studies suggest that singletons born from assisted reproductive technologies (ART) have a high risk of adverse perinatal outcomes, specifically for imprinting disorders. Because ART processes take place at times when epigenetic reprogramming/imprinting are occurring, there is concern that ART can affect genomic imprints. However, little is currently known about the risk of imprinting defects according to the type of ART or the type of underlying female infertility. From the French national health database, a cohort of 3,501,495 singletons born over a 5-year period (2013–2017) following fresh embryo or frozen embryo transfers (fresh-ET or FET from in vitro fertilization), intrauterine insemina‑ tion, or natural conception was followed up to early childhood. Based on clinical features, several syndromes/diseases involving imprinted genes were monitored. The effects of ART conception and the underlying cause of female infertil‑ ity were assessed. Results: Compared with infants conceived naturally, children born after fresh-ET had a higher prevalence of imprint‑ ing-related diseases, with an aOR of 1.43 [95% CI 1.13–1.81, p = 0.003]. Namely, we observed an increased risk of neonatal diabetes mellitus (1.96 aOR [95% CI 1.43–2.70], p < 0.001). There was an overall independent increase in risk of imprinting diseases for children with mothers diagnosed with endometriosis (1.38 aOR [95% CI 1.06–1.80], p = 0.02). Young and advanced maternal age, primiparity, obesity, smoking, and history of high blood pressure or diabetes were also associated with high global risk. Conclusions: This prospective epidemiological study showed that the risk of clinically diagnosed imprinting-related diseases is increased in children conceived after fresh embryo transfers or from mothers with endometriosis. The increased perturbations in genomic imprinting could be caused by controlled ovarian hyperstimulation and poten‑ tially endometriosis through the impairment of endometrial receptivity and placentation, leading to epigenetic feto- placental changes. Further studies are now needed to improve understanding of the underlying molecular mecha‑ nisms (i.e. genetic or epigenetic causes). Keywords: Epigenetic diseases, Female infertility, Assisted reproductive technologies, Children, Singletons 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://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Reproductive technologies, female infertility, and the risk of imprinting‑related disorders Patricia Fauque1* , Jacques De Mouzon2, Aviva Devaux3, Sylvie Epelboin4, Marie‑José Gervoise‑Boyer5, Rachel Levy6, Morgane Valentin7, Géraldine Viot8, Arianne Bergère9, Claire De Vienne9, Philippe Jonveaux9 and Fabienne Pessione9 Background Assisted reproductive technologies (ART), mostly intrau- terine insemination (IUI) and in vitro fertilization (IVF), have helped many couples to overcome infertility. World- wide, millions of children have been born via ART, and they now account for > 4% of births in some European countries [1]. Correspondence: patricia.fauque@chu‑dijon.fr 1 CHU Dijon Bourgogne, Laboratoire de Biologie de la Reproduction - CECOS - Université Bourgogne Franche-Comté - INSERM UMR1231, Dijon, France Correspondence: patricia.fauque@chu‑dijon.fr 1 CHU Dijon Bourgogne, Laboratoire de Biologie de la Reproduction - CECOS - Université Bourgogne Franche-Comté - INSERM UMR1231, Dijon, France 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 Fauque et al. Clin Epigenet (2020) 12:191 https://doi.org/10.1186/s13148-020-00986-3 Fauque et al. Clin Epigenet (2020) 12:191 https://doi.org/10.1186/s13148-020-00986-3 © 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://creativeco mmons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Fauque et al. Clin Epigenet (2020) 12:191 Fauque et al. Clin Epigenet (2020) 12:191 Fauque et al. Clin Epigenet (2020) 12:191 Page 2 of 9 Even though most of these children are considered healthy, there is increasing awareness about the poten- tial consequences of ART on a number of complications potentially linked to epigenetic deregulation [2]. of epigenetic marks in imprinting control regions (close to 95% of children with BWS born after IVF/ICSI vs 50% in the general population) [11]. However, the infertility/ subfertility status of the parents may also play a role in the increased incidence of these disorders, as underlined in a Dutch study performed in families with a child with BWS, PWS, or AS [9]. Interestingly, there is increasing evidence that some female infertility syndromes such as polycystic ovary syndrome (PCOS) [16] or endometriosis [17] are associated with epigenetic alterations. Epidemiological studies suggest that singletons born following the use of ART have an increased risk of adverse perinatal outcomes (e.g. low birth weight after fresh embryo transfers and abnormal placentation) [3, 4]. Furthermore, ART-conceived offspring also have an increased risk of rare imprinting disorders, such as Beckwith–Wiedemann (BWS), Russell–Silver (SRS), Angelman (AS) and Prader–Willi (PWS) syndromes [5–10]. However, there are considerable differences in the reported relative risks (Additional file 1: Table S1), which can mostly be explained by the substantial meth- odological heterogeneity (findings mostly obtained from voluntary registries or relatively small populations). The systematic reviews on this field have nevertheless made it possible to assess a substantial number of children. In a review, Vermeiden and Bernardus estimated that the birth of a child with BWS is significantly associated with ART, with a pooled relative risk of 5.2 [95% CI 1.6–7.4] [11]. Similarly, a systematic review demonstrated that the combined odds ratio of any imprinting syndromes in children conceived by ART is 3.67 [95% CI 1.39–9.74] when compared with naturally conceived children [12]. In a recent meta-analysis, a positive association was still found between conception after ART and four imprint- ing conditions (BWS, SRS, AS, PWS), among which BWS had a summary odds ratio of 5.8 [95% CI 3.1–11.1] [13]. It is difficult to draw any conclusions concerning SRS because it is extremely rare [11], but a positive asso- ciation with ART treatment is likely [13]. No significant associations were found between the incidence of AS or PWS and IVF treatments [11]. Fertility problems could be involved in these two last syndromes in ART chil- dren [11], but the results of any meta-analysis must be interpreted in light of the limitations of the contributing studies. However, so far, the risk of imprinting-related diseases in relation to specific types of ART or underlying causes of female infertility has not been assessed. Therefore, the first aim of this extensive national cohort study was to compare the prevalence of imprinting-related disor- ders in singletons born after fresh (fresh-ET) or frozen (FET) embryo transfers, intrauterine insemination (IUI), or following natural conception (NC). Our second aim was to study the role of the three major types of female infertility (i.e. endometriosis, PCOS and primary ovarian insufficiency [POI]) on the prevalence of these imprint- ing-related diseases. Influence of the mode of conception According to the mode of conception, the prevalence of imprinting-related diseases was 0.10% after NC (N = 3478), 0.16% (N = 72) after fresh-ET, 0.12% (N = 23) after FET, and 0.11% (N = 22) after IUI (for a total of 0.14% after ART). Compared with NC infants, children born after fresh-ET had a significantly higher prevalence of imprinting diseases, with an aOR of 1.43 [95% CI 1.13– 1.81], p = 0.003, whereas no relevant differences were observed for the IUI and the FET groups. After taking into account female infertility status in multivariate anal- ysis, the risk remained significant in the fresh-ET group (aOR of 1.32 [95% CI 1.04–1.69], p = 0.02) (Table 1). More specifically, in univariate analysis, we observed a significant increase in the number of cases of neonatal diabetes mellitus (NDM) in the fresh-ET group (N = 38, 0.08%) compared with the NC group (N = 1438, 0.04%, p < 0.001) (Table 2). The increased risk of NDM was con- firmed in multivariate analysis (1.96 aOR [95% CI 1.43– 2.70], p < 0.001), even when the type of female infertility was taken into account (1.58 aOR [95% CI 1.13–2.21], p = 0.008) (Table 3). The increased risk of NDM also remained significant after adjustment for gestational dia- betes (data not shown). In addition, when analyses were performed on a subset of NDM newborns who needed medication or were hospitalized for diabetes within one year after birth (n = 130, corresponding to incidence of 3.7/100,000), the fresh-ET group remained at a higher risk than the NC group (Additional file 1: Table S2). During the periconceptional period, genome-wide epigenetic reprogramming occurs [14, 15]. This repro- gramming includes imprinting, which is crucial for the proper development and future health of offspring. The many manipulations and processes of ART (e.g. hor- monal stimulation, embryo manipulation, culture and cryopreservation) are concurrent with epigenetic repro- gramming and imprinting (i.e. during female gametogen- esis and preimplantation embryo development), leading to concerns that the ART themselves could negatively affect epigenetics and the establishment/maintenance of genomic imprints. Importantly, contrary to the molecular aetiologies of BWS children conceived naturally, almost all ART-conceived BWS typically occurred though loss Fauque et al. Discussionh This unique and large cohort follow-up study made it possible to determine that the risk of imprinting-related disorders was 43% higher in children conceived through IVF or ICSI cycles after fresh embryo transfers than for naturally conceived children. Even though our find- ings are mostly consistent with the results of a previous review and meta-analysis, we found that the risks are much lower [11, 13]. In addition, we found that female infertility resulting from endometriosis could play a role in the increased risk. Overall, the risks of imprinting- related diseases were mainly observed after treatments involving controlled ovarian hyperstimulation (COH). According to the type of female infertility, the prevalence of imprinting-related diseases was 0.16% (N = 59), 0.17% (N = 12), and 0.35% (N = 5) for singletons conceived with mothers diagnosed with endometriosis, PCOS and POI, respectively. There was an overall independent increase in the risk of imprinting-related diseases for the children of moth- ers with endometriosis (1.38 aOR [95% CI 1.06–1.80], p = 0.02) (Table 2). The prevalence of imprinting-related diseases was 0.26% in the fresh-ET group with endo- metriosis vs 0.14% in the group without endometrio- sis (Fig. 1). The children of women with POI tended to have higher risk than those from women without POI (2.24 aOR [95% CI 0.99–5.91], p = 0.05), but PCOS did not appear to influence the risk of imprinting disorders (Table 1). Specifically, except in the IUI group, the preva- lence of NDM was higher in the children whose mothers had endometriosis (Fig. 1). In multivariate analysis, the risk of NDM tended to be higher for children conceived by women with endometriosis (1.47 aOR [95% CI 1.00– 2.16], p = 0.05) (Table 3). Furthermore, the incidence of children requiring diabetic treatment/hospitalization within one year after birth was higher for women with endometriosis or POI (Additional file 1: Table S2). Similar to our observations for imprinting diseases, previous studies have reported a reduced risk of sev- eral birth defects (e.g. cardiovascular, musculoskel- etal, and urogenital abnormalities) following embryo cryopreservation compared with fresh-embryo trans- fer, either significantly [18] or as a trend [19, 20]. One possible explanation for the reduced risk of imprinting defects following cryopreservation is that the endome- trium is not exposed to high levels of gonadotrophins, which may potentially impair trophoblastic invasion, notably through epigenetic changes, and contribute to placental dysfunction and the pathophysiology of dis- ease [3, 21, 22]. Influence of the mode of conception Clin Epigenet (2020) 12:191 Page 3 of 9 Table 1 Risk of imprinting-related diseases according to mode of conception and type of female infertility in multivariate analysis CI confidence interval, Fresh-ET fresh embryo transfer, FET frozen embryo transfer, IUI intrauterine insemination, NC natural conception, OR odds ratio, PCOS polycystic ovary syndrome, POI primary ovarian insufficiency a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure or diabetes, and newborn sex Total children Children with at least one imprinting-related disease Adjusted ORa 95% CI p N N % Mode of conception NC 3,417,089 3478 0.10 1 Fresh-ET 45,303 72 0.16 1.32 1.04 1.69 0.02 FET 18,885 23 0.12 1.04 0.69 1.57 0.86 IUI 20,218 22 0.11 0.97 0.63 1.47 0.87 Female infertility Endometriosis 37,398 59 0.16 1.38 1.06 1.80 0.02 PCOS 6977 12 0.17 1.35 0.76 2.39 0.30 POI 1411 5 0.35 2.42 0.99 5.91 0.05 Table 1 Risk of imprinting-related diseases according to mode of conception and type of female infertility in multivariate analysis d diseases according to mode of conception and type of female infertility in multivariate CI confidence interval, Fresh-ET fresh embryo transfer, FET frozen embryo transfer, IUI intrauterine insemination, NC natural conception, OR odds ratio, PCOS polycystic ovary syndrome, POI primary ovarian insufficiency a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure or diabetes, and newborn sex a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure or diabetes, and ne Discussionh Furthermore, no risk of imprinting dis- eases was associated with intrauterine insemination in our cohort of children. However, by showing that the overall prevalence of congenital malformations was increased with increasing time to naturally-obtained pregnancy, Zhu and colleagues were the first to signal a possible association between infertility and congenital malformations [23]. The notion that infertility itself is i Moreover, the other tested maternal factors (i.e. primi- parity, obesity, smoking, and history of HBP or diabetes, with significant differences between groups of children, Additional file 1: Table S3) were significantly associated with a higher overall risk of imprinting-related diseases in the multivariate analysis (Table 4). The youngest and advanced-maternal-age women were also at high risk (Table 4). Page 4 of 9 Fauque et al. Clin Epigenet (2020) 12:191 Table 2 Comparison of observed to expected cases for each clinically diagnosed imprinting-related disease according to the type of reproductive technology using univariate analyses (Poisson regression model) AS Angelman syndrome, BWS Beckwith–Wiedemann syndrome, Fresh-ET fresh embryo transfer, FET frozen embryo transfer, IUI intrauterine insemination, NDM neonatal diabetes mellitus, PWS Prader–Willi syndrome, SRS Silver–Russell syndrome, Upd uniparental disomy a Diseases based on clinical features and associated to the corresponding ICD-10 code. Some children could have more than one disease diagnosed b Expected cases from NC data ICD-10 codes Imprinting-related diseasesa Fresh-ET FET IUI Observed cases Expectedb cases p Observed cases Expected cases p Observed cases Expected cases p Q93.5 AS 8 10 0.62 4 4 0.99 4 4 0.90 Q87.3 BWS 8 5 0.18 0 2 2 0.88 P70.2 NDM 38 19 < 0.001 10 8 0.50 11 9 0.42 C74.9 Neuroblastoma 2 3 0.57 1 1 0.82 0 E20.1 Pseudohypoparathyroidism 0 0 1 0 0.01 C69.2 Retinoblastoma 3 3 0.86 1 1 0.90 1 1 0.85 Q87.1 SRS/PWS 18 14 0.33 10 6 0.10 4 6 0.35 Q99.8 upd(14) 24 17 0.07 8 7 0.68 10 7 0.34 Fauque et al. Discussionh Clin Epigenet (2020) 12:191 Page 5 of 9 Table 3 Risk of NDM after ART and according to the type of female infertility in multivariate analysis CI confidence interval, Fresh-ET fresh embryo transfer, FET frozen embryo transfer, IUI intrauterine insemination, NC natural conception, OR odds ratio, PCOS polycystic ovary syndrome, POI primary ovarian insufficiency a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure, or diabetes Total children Children with NDM Adjusted ORa 95% CI p N N % Mode of conception NC 3,417,089 1438 0.04 1 Fresh-ET 45,303 38 0.08 1.58 1.13 2.21 0.008 FET 18,885 10 0.05 1.04 0.56 1.96 0.89 IUI 20,218 11 0.05 1.10 0.61 2.00 0.76 Female infertility Endometriosis 37,398 28 0.07 1.47 1.00 2.16 0.05 PCOS 6977 6 0.09 1.39 0.62 3.13 0.42 POI 1411 2 0.14 1.93 0.47 7.86 0.36 Table 3 Risk of NDM after ART and according to the type of female infertility in Table 3 Risk of NDM after ART and according to the type of female infertility in multivariate analysis fter ART and according to the type of female infertility in multivariate analysis CI confidence interval, Fresh-ET fresh embryo transfer, FET frozen embryo transfer, IUI intrauterine insemination, NC natural conception, OR odds ratio, PCOS polycystic ovary syndrome, POI primary ovarian insufficiency y y , p yfi y a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure, or diabetes y y yfi y a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure, or diabetes a Analyses were adjusted for maternal parameters: age, primiparity, smoking, obesity, history of high blood pressure, or diabetes 0.1 0.14 0.14 0.26 0 0.05 0.1 0.15 0.2 0.25 0.3 NC Fresh-ET % of imprinting-related diseases 0.04 0.08 0.07 0.10 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 NC Fresh-ET % of NDM without endometriosis with endometriosis a b Fig. 1 Prevalence of clinically diagnosed imprinting-related diseases (a) and NDM (b) for children with mothers diagnosed with endometriosis or not according to the mode of conception. Discussionh The data does not in any way correspond to a molecular diagnosis, which is the only way to identify the molecular mechanism involved in the aetiology of the disorder. However, it is unlikely that our findings are due to biases in reporting or follow-up, which are completely independent from the mode of conception. In view of the results obtained in this comprehen- sive cohort follow-up study, we propose that underlying maternal infertility, particularly due to endometriosis, could also contribute to the increased risk of imprinting- related disorders associated with ART. We determined that this risk was 38% higher in children whose moth- ers were diagnosed with endometriosis than in children whose mothers did not have endometriosis. Further research is needed to evaluate whether the classification of endometriosis or its association with adenomyosis has a specific effect on this risk. Interestingly, some evidence suggests that epigenetics plays a role not only in the development of endometriosis–adenomyosis lesions but also in the heritability of the condition [28]. These statements reinforce the concept that the mater- nal environment can affect foetal development, likely through the effect on the regulation of imprinted genes, primarily in the placenta [2]. We can speculate that the control of these maternal parameters could preserve the appropriate early regulation of imprinting, which could potentially reduce diseases. y Interestingly, NDM was the condition that was most strongly associated with fresh embryo transfer and endo- metriosis in our results. NDM, which is characterized by hyperglycaemia in the neonatal period, associates growth retardation and hypoglycaemia. It is called transient (TNDM) when it remits during infancy (in about half of children affected by hyperglycaemia), and recovery occurs before 18 months in approximately 50% of these infants. However, infants with TNDM are more prone to developing type 2 diabetes later in life. The dysregulation of two imprinted genes (PLAGL1 and HYMAI) involved in most of the cases is a probable mechanism. These genes are located at the 6q24 region. PLAGL1 encodes a zinc finger protein which controls cell cycle and apopto- sis and enhances insulin secretion, functioning as a regu- lator of pancreatic β-cells [38, 39]. Furthermore, PLAGL1 and HYMAI are also described as the main regulators of the imprinted network involved in cellular growth and metabolism [40, 41]. TNDM results from a ‘double dose’ Epigenetic processes control numerous major cellu- lar functions that occur during development, includ- ing changes in gene expression. Discussionh Other novel findings reported in the current study are that maternal features (i.e. smoking, obesity, history of HBP or diabe- tes) were independently associated with an increased risk of imprinting-related diseases. Remarkably, primipar- ity and young and advanced maternal age also appeared to increase risk. In accordance with our observations, septo-optic dysplasia, which is rarely linked to a genetic cause, has also been associated with primiparity, young maternal age and prenatal exposure to smoking [32, 33]. The underlying aetiology of this condition is likely nutri- tional or environmental, suggesting the potential role of epigenetic factors [34]. Another hypothesis for this increased risk in the children of young mothers is that young women may be less likely to take vitamins during the periconceptional period, especially folic acid supple- mentation which known to be fundamental in cellular biosynthesis and DNA methylation pathways [35, 36]. Concerning advanced maternal age, Hara-Isono et al. recently reported that in Japan ART performed on moth- ers aged ≥ 30 is likely to facilitate epimutation [37]. How- ever, they could not statistically determine whether the effect was caused by ART or the confounding effect of advanced maternal age at conception. registries from a relatively small number of ART chil- dren (6052 children born after IVF/ICSI) [27].h Therefore, even though our study presents some weak- nesses, such as the inability to take into account male infertility, the type of in vitro fertilization method, and the embryo stage at transfer, it also has several strengths. It includes a large cohort of singleton children, it is not restricted to the four imprinting disorders, and it takes into account multiple potential maternal confounding factors. Another strength of the current study is the use of a single national population registry with ascertain- ment of defects from birth to a child’s first birthday or later. However, for some extremely rare syndromes, the number of cases may be too small reliably conclude that there was no difference between groups. Likewise, for childhood cancers involving imprinted genes (i.e. ret- inoblastoma or neuroblastoma) even though our find- ings were in line with the literature (Additional file 1: Table S1), longer follow-up is needed to be certain of the absence of risk after ART. Finally, the recorded disorders in our national health databases are based on diagnostic features. Discussionh Fresh-ET fresh embryo transfer, FET frozen embryo transfer, NC natural conception, NDM neonatal diabetes mellitus 0.1 0.14 0.14 0.26 0 0.05 0.1 0.15 0.2 0.25 0.3 NC Fresh-ET % of imprinting-related diseases a Table 4 Risk of clinically diagnosed imprinting-related diseases for adjustment variables in multivariate analysis 95% CI confidence interval, HBP high blood pressure, OR odds ratio, y years a Analyses were adjusted for infertility status and mode of conception Adjusted ORa 95% CI p Maternal age (y) < 20 1.25 1.03 1.52 0.03 20–29 0.96 0.90 1.03 0.29 30–39 1 ≥ 40 1.58 1.38 1.82 < 0.001 Primiparity 1.28 1.19 1.37 < 0.001 Obesity 1.40 1.23 1.60 < 0.001 Maternal smoking 1.24 1.07 1.43 0.005 History of HBP 4.45 3.69 5.36 < 0.001 History of diabetes 1.68 1.28 2.20 < 0.001 Sex of the newborn Female 1 Male 1.13 1.06 1.20 < 0.001 Table 4 Risk of clinically diagnosed imprinting-related diseases for adjustment variables in multivariate analysis 0.04 0.08 0.07 0.10 0 0.02 0.04 0.06 0.08 0.1 0.12 NC Fresh-ET % of NDM ith t d t i i ith endometriosis a risk factor for increased incidence of birth defects has also been put forward by other researchers [18, 24]. Previous epidemiologic studies have shown that ART- conceived offspring could be at increased risk for rare imprinting disorders, for instance Beckwith–Wiede- mann syndrome, with an estimated relative risk of 5.2 [95% CI 1.6–7.4] [11]. However, for most of the popu- lation-based studies exploring BWS, the risk was estab- lished from voluntary registries [5, 7, 8, 25] and few were based on complete national registries [6, 26, 27] (Additional file 1: Table S1). In addition, among nation- wide registries, two were based on imprinting disor- ders registries (i.e. Gicquel et al. in France [6]; Kallen et al. in Sweden [26]), and one was based on the Danish Fig. 1 Prevalence of clinically diagnosed imprinting-related diseases (a) and NDM (b) for children with mothers diagnosed with endometriosis or not according to the mode of conception. Fresh-ET fresh embryo transfer, FET frozen embryo transfer, NC natural conception, NDM neonatal diabetes mellitus Fauque et al. Clin Epigenet (2020) 12:191 Page 6 of 9 Page 6 of 9 for small number of human genes, exert crucial roles in embryonic/foetal development including placentation [30], postnatal metabolic pathways, and behaviour asso- ciated with the control of resources [31]. Discussionh Most autosomal genes are expressed from both alleles, but imprinted genes are preferentially expressed from alleles inherited from mothers or fathers. The parent-specific expression of imprinted genes is directed by epigenetic marks (mainly DNA methylation), acquired in a sex-specific manner in gametes on regulatory sequences [29]. Then, after fertili- zation, these parental imprinting marks are protected to avoid global embryo epigenetic reprogramming/demeth- ylation [14]. Imprinted genes, even though they account Fauque et al. Clin Epigenet (2020) 12:191 Page 7 of 9 Fauque et al. Clin Epigenet (2020) 12:191 data for this cohort were available until early childhood (mean 2.5 years old [1–5]). During this period, a total of 3,501,495 singleton births were identified (including 20,218 IUI, 45,303 fresh-ET, and 18,885 FET). of the paternal epigenotype (a twofold overexpression of PLAGL1/HYMAI). In the general population, pater- nal uniparental disomy (patUPD6) is involved in 40% of cases, paternal chromosome duplication in 32% of cases, and hypomethylation at the maternal allele of 6q24 region in 28% [39]. Data available in the hospitalization database were par- ity, plurality, maternal age, active smoking during preg- nancy, obesity, maternal history of diabetes (type-1 or 2) and high blood pressure, the diagnosis of endometriosis, PCOS and POI, the mode of conception, the term, weight and sex of the newborn.h To our knowledge, the underlying molecular mecha- nism of NDM after ART has not been specifically stud- ied, but, like for BWS, the loss of methylation might be more frequent after ART. Further studies are needed to establish whether the deregulation of imprinted genes that occurs in ART children with NDM is caused by epi- genetic or genetic mechanisms. All of the imprinting- related disorders studied here could potentially be linked to genetic abnormalities, but only the results of such fur- ther studies could shed light on the underlying molecular mechanisms, making it possible to develop clinical pre- vention strategies. This database contains anonymized patient informa- tion and its access was legally approved. Children defects involving imprinted genes Data were extracted from the national health database. Malformations were classified according to the WHO ICD-10 codes. We monitored several syndromes and diseases diagnosed on clinical features by paediatricians (and not genetic testing) but for which altered expres- sion of imprinted genes are associated: Beckwith–Wiede- mann (BWS), Silver–Russell (SRS), Prader–Willi (PWS), and Angelman (AS) syndromes, pseudohypoparathy- roidism type Ib, neonatal diabetes mellitus (NDM), syn- dromes affecting the imprinted region of chromosome 14q32 (Temple and Kagami-Ogata syndromes), neuro- blastoma and retinoblastoma (description of phenotypes and related imprinted genes is available in the Additional file 1: Table S1). In the following manuscript, we will refer to the syndromes by their names rather than the ICD-10 codes linked to them. Some children could have more than one disease diagnosed. For NDM newborns (charac- terized by hyperglycaemia in the neonatal period requir- ing insulin therapy and persisting for more than 2 weeks), we identified a subset of children who were prescribed medication or hospitalized for diabetes within one year after birth. Overall, this large population-based study shows that the risk of imprinting diseases, especially neonatal diabe- tes mellitus, is increased in children conceived after fresh embryo transfers or from mothers with endometriosis. One speculative explanation for this observation is that controlled ovarian hyperstimulation, and thus poten- tially endometriosis, induces perturbations in genomic imprinting by impairing endometrial receptivity and pla- centation, leading to epigenetic feto-placental changes. Vast evidence suggests that the gestational environment has an impact on epigenetic mechanisms including imprinting, which can lead to harmful outcomes in the offspring [2].hi f These novel findings highlight the importance of tak- ing into account the mode/process of conception and the type of infertility. References 1. De Geyter C, Calhaz-Jorge C, Kupka MS, Wyns C, Mocanu E, Motrenko T, et al. Art in Europe, 2014: results generated from European Registries by Eshre: the European IVF-Monitoring Consortium (EIM) for the European Society of Human Reproduction and Embryology (ESHRE). Hum Reprod. 2018;33:1586–601. 2. Argyraki M, Damdimopoulou P, Chatzimeletiou K, Grimbizis GF, Tarlatzis BC, Syrrou M, et al. In-utero stress and mode of conception: impact on regulation of imprinted genes, fetal development and future health. Hum Reprod Update. 2019;25:777–801. 2. Argyraki M, Damdimopoulou P, Chatzimeletiou K, Grimbizis GF, Tarlatzis BC, Syrrou M, et al. In-utero stress and mode of conception: impact on regulation of imprinted genes, fetal development and future health. Hum Reprod Update. 2019;25:777–801. Statistical analyses d The interactions between variables were tested, and we considered that a p-value < 0.05 provided evidence of a possible interaction. Adjusted odds ratios (aOR) and their 95% confidence intervals (CI) were estimated. Statistical analyses were performed using SAS 9.4 (SAS Institute Inc.). The significance level was set at a two- tailed p-value of < 0.05. smoking, obesity, history of diabetes (type 1 or 2), and maternal high blood pressure (HBP). Secondly, in our multivariate model, we included the type of female infer- tility (endometriosis, PCOS and POI), which provides estimates of the respective risk of the mode of conception and the type of female infertility adjusted for maternal parameters (age, primiparity, smoking, obesity, history of high blood pressure or diabetes), and newborn sex. Pre- maturity can potentially increase the incidence of NDM. However, our sensitivity analysis showed no noticeable differences in the magnitude of the risk estimates when premature infants were excluded from the analyses (data not shown), so all data were entered into a single data- base. The interactions between variables were tested, and we considered that a p-value < 0.05 provided evidence of a possible interaction. Adjusted odds ratios (aOR) and their 95% confidence intervals (CI) were estimated. Statistical analyses were performed using SAS 9.4 (SAS Institute Inc.). The significance level was set at a two- tailed p-value of < 0.05. 2 Unilabs, direction médicale, Clichy‑La‑Garenne, France. 3 Centre d’assistance medicale à la procreation, biologie de la reproduction, CHU Amiens, Ami‑ ens, France. 4 Centre d’assistance medicale à la procreation, gynécologie obstétrique, médecine de la reproduction, Université Paris 7 Diderot, groupe hospitalier Bichat Claude-Bernard, AP–HP, Paris, France. 5 Service de medecine et biologie de la reproduction, Hôpital Saint-Joseph, Marseille 8, France. 6 Sor‑ bonne Université, Saint Antoine Research Center, INSERM équipe Lipodystro‑ phies genetiques et acquises, Service de biologie de la reproduction-CECOS, Hôpital Tenon, AP-HP, 75012 Paris, France. 7 Diagnostic antenatal, gynécologie obstétrique, Université Paris 7 Diderot, groupe hospitalier Bichat Claude-Ber‑ nard, AP–HP, Paris, France. 8 Unité de Génétique Clinique de La Muette, 50 rue Nicolo, 75116 Paris, France. 9 Agence de la Biomédecine, 1 avenue du stade de France, 93212 La Plaine Saint Denis, France. Received: 31 August 2020 Accepted: 23 November 2020 Received: 31 August 2020 Accepted: 23 November 2020 Acknowledgments We thank Suzanne Rankin for proofreading the manuscript and Irène Netchine for her help with the discussion. 10. Hiura H, Okae H, Miyauchi N, Sato F, Sato A, Van De Pette M, et al. Characterization of DNA methylation errors in patients with imprinting disorders conceived by assisted reproduction technologies. Hum Reprod. 2012;27:2541–8. Supplementary Information 3. Choux C, Carmignac V, Bruno C, Sagot P, Vaiman D, Fauque P. The placenta: phenotypic and epigenetic modifications induced by assisted reproductive technologies throughout pregnancy. Clin Epigenet. 2015;7:87. The online version contains supplementary material available at https://doi. org/10.1186/s13148-020-00986-3. The online version contains supplementary material available at https://doi. org/10.1186/s13148-020-00986-3. 4. Qin J, Liu X, Sheng X, Wang H, Gao S. Assisted reproductive technology and the risk of pregnancy-related complications and adverse pregnancy outcomes in singleton pregnancies: a meta-analysis of cohort studies. Fertil Steril. 2016;105(73–85):e1-6. Additional file 1: Table S1. Syndromes and diseases involving imprinted genes including general features, molecular changes, cohort studies. Table S2: Risk of NDM according to mode of conception and type of female infertility in a subset of NDM newborns, i.e. infants who needed medications or were hospitalised for diabetes within 1 year after their birth. Table S3: Maternal characteristics according to the mode of conception. 5. DeBaun MR, Niemitz EL, Feinberg AP. Association of in vitro fertilization with Beckwith–Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet. 2003;72:156–60. 5. DeBaun MR, Niemitz EL, Feinberg AP. Association of in vitro fertilization with Beckwith–Wiedemann syndrome and epigenetic alterations of LIT1 and H19. Am J Hum Genet. 2003;72:156–60. 6. Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, Le Bouc Y. In vitro fertilization may increase the risk of Beckwith–Wiedemann syndrome related to the abnormal imprinting of the KCN1OT gene. Am J Hum Genet. 2003;72:1338–41. 6. Gicquel C, Gaston V, Mandelbaum J, Siffroi JP, Flahault A, Le Bouc Y. In vitro fertilization may increase the risk of Beckwith–Wiedemann syndrome related to the abnormal imprinting of the KCN1OT gene. Am J Hum Genet. 2003;72:1338–41. Abbreviations ART: Assisted reproductive technologies; AS: Angelman syndrome; BWS: Beckwith–Wiedemann syndrome; FET: Frozen embryo transfers; Fresh-ET: Fresh embryo transfers; HBP: High blood pressure; ICSI: Intra-cytoplasmic injection (IVF with sperm injection); IUI: Intrauterine insemination; IVF: In vitro fertilization; NC: Natural conception; NDM: Neonatal diabetes mellitus; PCOS: Polycystic ovary syndrome; POI: Primary ovarian insufficiency; PWS: Prader– Willi syndrome; SRS: Russell–Silver syndrome; TNDM: Transient NDM. 7. Maher ER, Brueton LA, Bowdin SC, Luharia A, Cooper W, Cole TR, et al. Beckwith–Wiedemann syndrome and assisted reproduction technology (ART). J Med Genet. 2003;40:62–4. 7. Maher ER, Brueton LA, Bowdin SC, Luharia A, Cooper W, Cole TR, et al. Beckwith–Wiedemann syndrome and assisted reproduction technology (ART). J Med Genet. 2003;40:62–4. 8. Sutcliffe AG, Peters CJ, Bowdin S, Temple K, Reardon W, Wilson L, et al. Assisted reproductive therapies and imprinting disorders—a preliminary british survey. Hum Reprod. 2006;21:1009–11. 9. Doornbos ME, Maas SM, McDonnell J, Vermeiden JP, Hennekam RC. Infer‑ tility, assisted reproduction technologies and imprinting disturbances: a Dutch study. Hum Reprod. 2007;22:2476–80. Funding g This work was supported by the Biomedicine National Agency. Authors’ contributions PF and FP take primary responsibility for the paper. FP collected the data. FP did the statistical analyses. PF and PF coordinated the research. JDM, AD, SE, MJGB, RL, MV, GV, MB, CDV, PJ participated in the data analyses. PF and FP drafted the manuscript. All authors read and approved the final manuscript. 11. Vermeiden JP, Bernardus RE. Are imprinting disorders more prevalent after human in vitro fertilization or intracytoplasmic sperm injection? Fertil Steril. 2013;99:642–51. 12. Lazaraviciute G, Kauser M, Bhattacharya S, Haggarty P. A systematic review and meta-analysis of DNA methylation levels and imprinting disorders in children conceived by IVF/ICSI Compared with children conceived spontaneously. Hum Reprod Update. 2014;20:840–52. Competing interest 13. Cortessis VK, Azadian M, Buxbaum J, Sanogo F, Song AY, Sriprasert I, et al. Comprehensive meta-analysis reveals association between multiple imprinting disorders and conception by assisted reproductive technol‑ ogy. J Assist Reprod Genet. 2018;35:943–52. p g The authors declare no conflict of interest related to this study. The authors declare no conflict of interest related to this study. Statistical analyses d This study was piloted by the ‘ART women and children health’ working group from the National Agency of Bio- medicine. Data were extracted from the French National Health System database (Système National des Données de Santé—SNDS) (> 99% of national deliveries). Mothers’ records were merged anonymously with those of their newborns and with previous pregnancy-related hospi- talizations. Deliveries for which the mother and child’s data were not linked and multiple births were excluded. The final cohort included 93% of all French deliveries. We thus conducted a comparative analysis of the cohort of singleton births (deliveries ≥ 22 weeks of gestation and/or > 500 g of birthweight) which occurred in France over a 5-year period (2013–2017) and resulting from fresh embryo or frozen embryo transfers (fresh-ET or FET from IVF and ICSI cycles), IUI and NC. Follow-up Univariate and multivariate analyses were performed using logistic regression models to analyse the effect of ART conception and type of female infertility on the risk of imprinting disorders. Univariate analysis was used to compare the number of observed and expected cases (established from data of the NC group). Standardized incidence ratios [SIR] estimated from the control group of spontaneous conception according to the reproduc- tive method were performed using a Poisson regression model. In multivariate analyses, we first assessed the effects of the mode of conception (ART [IUI/fresh-ET/ FET] or NC) and other factors available in the hospitali- zation database that could potentially affect epigenetic risk. These factors were parity, maternal age (catego- rized in 10-year age groups), sex of the newborn, active Fauque et al. Clin Epigenet (2020) 12:191 Page 8 of 9 smoking, obesity, history of diabetes (type 1 or 2), and maternal high blood pressure (HBP). Secondly, in our multivariate model, we included the type of female infer- tility (endometriosis, PCOS and POI), which provides estimates of the respective risk of the mode of conception and the type of female infertility adjusted for maternal parameters (age, primiparity, smoking, obesity, history of high blood pressure or diabetes), and newborn sex. Pre- maturity can potentially increase the incidence of NDM. However, our sensitivity analysis showed no noticeable differences in the magnitude of the risk estimates when premature infants were excluded from the analyses (data not shown), so all data were entered into a single data- base. Fauque et al. Clin Epigenet (2020) 12:191 Page 9 of 9 Fauque et al. Clin Epigenet (2020) 12:191 15. Smith ZD, Chan MM, Humm KC, Karnik R, Mekhoubad S, Regev A, et al. DNA methylation dynamics of the human preimplantation embryo. Nature. 2014;511:611–5. 30. Tunster SJ, Jensen AB, John RM. Imprinted genes in mouse placental development and the regulation of fetal energy stores. Reproduction. 2013;145:R117–37. 31. Constancia M, Kelsey G, Reik W. Resourceful imprinting. Nature. 2004;432:53–7. 16. Vázquez-Martínez ER, Gomez-Viais YI, Garcia-Gomez E, Reyes-Mayoral C, Reyes-Munoz E, Camacho-Arroyo I, et al. DNA methylation in the patho‑ genesis of polycystic ovary syndrome. Reproduction. 2019;158:R27–40. 32. Garne E, Rissmann A, Addor MC, Barisic I, Bergman J, Braz P, et al. Epide‑ miology of septo-optic dysplasia with focus on prevalence and maternal age—a EUROCAT Study. Eur J Med Genet. 2018;61:483–8. 17. Koninckx PR, Ussia A, Adamyan L, Wattiez A, Gomel V, Martin DC. Patho‑ genesis of endometriosis: the genetic/epigenetic theory. Fertil Steril. 2019;111:327–40. 33. McCabe MJ, Alatzoglou KS, Dattani MT. Septo-optic dysplasia and other midline defects: the role of transcription factors: HESX1 and beyond. Best Pract Res Clin Endocrinol Metab. 2011;25:115–24. 18. Davies MJ, Moore VM, Willson KJ, Van Essen P, Priest K, Scott H, et al. Reproductive technologies and the risk of birth defects. N Engl J Med. 2012;366:1803–13. 34. Khaper T, Bunge M, Clark I, Rafay MF, Mhanni A, Kirouac N, et al. Increasing incidence of optic nerve hypoplasia/septo-optic dysplasia spectrum: geographic clustering in Northern Canada. Paediatr Child Health. 2017;22:445–53. 19. Pinborg A, Loft A, Aaris Henningsen AK, Rasmussen S, Andersen AN. Infant outcome of 957 singletons born after frozen embryo replace‑ ment: the Danish National Cohort Study 1995–2006. Fertil Steril. 2010;94:1320–7. 35. Li D, Huang L, Yang W, Qi C, Shang L, Xin J, et al. Knowledge, attitude and practice level of women at the periconceptional period: a cross-sectional study in Shaanxi China. BMC Pregnancy Childbirth. 2019;19:326. 20. Halliday JL, Ukoumunne OC, Baker HW, Breheny S, Jaques AM, Garrett C, et al. Increased risk of blastogenesis birth defects, arising in the first 4 weeks of pregnancy, after assisted reproductive technologies. Hum Reprod. 2010;25:59–65. 36. Irwin RE, Pentieva K, Cassidy T, Lees-Murdock DJ, McLaughlin M, Prasad G, et al. The Interplay between DNA methylation, folate and neurocognitive development. Epigenomics. 2016;8:863–79. 21. Fauque P. Placental epigenetics and outcomes in children encyclopedia of endocrine diseases. 2nd ed. Amsterdam: Elsevier; 2019. 37. Fauque et al. Clin Epigenet (2020) 12:191 Hara-Isono K, Matsubara K, Mikami M, Arima T, Ogata T, Fukami M, et al. Assisted reproductive technology represents a possible risk factor for development of epimutation-mediated imprinting disorders for mothers aged ≥ 30 years. Clin Epigenet. 2020;12:111. 22. Fauque P. Ovulation induction and epigenetic anomalies. Fertil Steril. 2013;99:616–23. 23. Zhu JL, Basso O, Obel C, Bille C, Olsen J. Infertility, infertility treatment, and congenital malformations: Danish National Birth Cohort. BMJ. 2006;333:679. 38. Kamiya M, Judson H, Okazaki Y, Kusakabe M, Muramatsu M, Takada S, et al. The cell cycle control gene ZAC/PLAGL1 is imprinted—a strong candidate gene for transient neonatal diabetes. Hum Mol Genet. 2000;9:453–60. 24. Rimm AA, Katayama AC, Katayama KP. A meta-analysis of the impact of IVF and ICSI on major malformations after adjusting for the effect of subfertility. J Assist Reprod Genet. 2011;28:699–705. 39. Mackay DJ, Temple IK. Transient neonatal diabetes mellitus type 1. Am J Med Genet C Semin Med Genet. 2010;154C:335–42. 25. Bowdin S, Allen C, Kirby G, Brueton L, Afnan M, Barratt C, et al. A Survey of assisted reproductive technology births and imprinting disorders. Hum Reprod. 2007;22:3237–40. 40. Iglesias-Platas I, Martin-Trujillo A, Petazzi P, Guillaumet-Adkins A, Esteller M, Monk D. Altered expression of the imprinted transcription factor PLAGL1 deregulates a network of genes in the human IUGR placenta. Hum Mol Genet. 2014;23:6275–85. 26. Kallen B, Finnstrom O, Nygren KG, Otterblad Olausson P, Wennerholm UB. In vitro fertilisation in Sweden: obstetric characteristics, maternal morbid‑ ity and mortality. BJOG Int J Obstet Gynaecol. 2005;112:1529–35. 41. Varrault A, Gueydan C, Delalbre A, Bellmann A, Houssami S, Aknin C, et al. Zac1 regulates an imprinted gene network critically involved in the control of embryonic growth. Dev Cell. 2006;11:711–22. 27. Lidegaard O, Pinborg A, Andersen AN. Imprinting diseases and IVF: Dan‑ ish National IVF Cohort Study. Hum Reprod. 2005;20:950–4. 28. Lagana AS, Garzon S, Gotte M, Vigano P, Franchi M, Ghezzi F, et al. The pathogenesis of endometriosis: molecular and cell biology insights. Int J Mol Sci. 2019;20:5615. Author details 14. Guo H, Zhu P, Yan L, Li R, Hu B, Lian Y, et al. The DNA methylation land‑ scape of human early embryos. Nature. 2014;511:606–10. 14. Guo H, Zhu P, Yan L, Li R, Hu B, Lian Y, et al. The DNA methylation land‑ scape of human early embryos. Nature. 2014;511:606–10. 1 CHU Dijon Bourgogne, Laboratoire de Biologie de la Reproduction - CECOS - Université Bourgogne Franche-Comté - INSERM UMR1231, Dijon, France. Fauque et al. Clin Epigenet (2020) 12:191 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. 29. Reik W, Walter J. Genomic imprinting: parental influence on the genome. Nat Rev Genet. 2001;2:21–32. 29. Reik W, Walter J. Genomic imprinting: parental influence on the genome. Nat Rev Genet. 2001;2:21–32. • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your ﬁeld • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year • At BMC, research is always in progress. Learn more biomedcentral.com/submissions Ready to submit your research Ready to submit your research ? Choose BMC and benefit from: ? Choose BMC and benefit from:
https://openalex.org/W4292743759	https://zenodo.org/records/6362002/files/736-739.pdf	English	null	Influence of Pesticides on the Trace Element Status of Soils (BHC on Manganese and Iron)	Zenodo (CERN European Organization for Nuclear Research)	1,980	cc-by	2,996	Influence of Pesticides on the Trace Element Status of Soils (BHC on Manganese and Iron) 0. P. BANSAL and R.N. GOYAL 0. P. BANSAL and R.N. GOYAL Chemical Laboratory, D. S. College, AJigarh-202 001 0. P. BANSAL and R.N. GOYAL Chemical Laboratory, D. S. College, AJigarh-202 001 Manuscript received 21 September 1978, revised 24 January 1980, accepted 9 April 1980 Manuscript received 21 September 1978, revised 24 January 1980, accepted 9 April 1980 The eft"ects of dift"erent doses of hexachlorocyclohexane (BHC), a well known insecticide, on the Mo and Fe status of a saline sodic soil were examined. The results revealed that 0.25 g of BHC per kg of soil after an interval of 30 days produt'ed optimum eft"ect on trace element status. At this"dose and time interval water soluble MD, exchangeable Mn, active Mo, exchangeable and available Fe significantly increased while reducible Mn decreased. The eft"ect of the chemical started declining after a period of 4 weeks of application. The responses have been explained on the basis of chemical and microbial activity in soil. The experiments on the influence of pesticide on the trace elements were done in earthenware pots of 20 x 20 em size. They were cleaned and fitted with polythene sheets on the wall of pots towards the inner side to prevent absorption of water. One kg of the soil was then placed in each of the twenty one pots. Seven doses of BHC (0.0, 0,1, 0.2. 0.25, 0.3, 0.4, 0.5 g) were then added and mixed with 1 kg of soil in three replications without growing any crop. All the samples were then watered upto the water holding capacity with disti1Ied water every week throughout the period of experiment. The treated samples were sampled at intervals offifteen days and were analysed for exchangeable, water, soluble and reducible Mn and for exchangeable and available Fe. PESTICIDES are widely used in agricultural pro- duction. They are used as insecticides, herbici- des, fungicides and nematocides. The impor· tance of pesticides in crop production is due to their accumulation in soils and their effect on soil proper- ties. Several workers 1 - 8 have studied this subject in considerable detail, but information on the influence of pesticides on the trace element status is lacking. P Mnll+ exists in soil as a water soluble and as exchangeable ion and hence in this state is available to plants. MnH exists as unavailable form but is reducible to available form. Influence of Pesticides on the Trace Element Status of Soils (BHC on Manganese and Iron) Manganese along with iron assists in formation of chlorophy11 and control many complicated systems of oxidation-reduction in plants and soils. Iron exists as Fe~~+ and Fe 8 + and its availability is governed by oxidation-reduction potential of the soil. Micro-organisms are mainly responsible for iron transformation. The pH and electrical conductivity were measured in 1 : 5 soil water suspension at 25±1°, Organic matter was estimated as per Walkley and Black's' method. Hexachlorocyclohexane (BHC) is widely used as an insecticide to control insect pests. In view of the great importance of the trace elements in plants metabolism and the increasing use of pesticides for high crop yields, it was considered that an examina- tion of the effects of BHC on manganese and iron transformation in a saline sodic soil would be useful. Water soluble, exchangeable and reducible Mn were estimated colorimetricaUy according to Jack· son 8 using Leeper's method as modified by Sherman and Harmer 9 with normal ammonium acetate (pH-1) as extractant and 85% orthophosphoric acid and potassium periodate as colour reagents. A 2%. hydroquinone solution of normal ammonium acetate was used as extractant for reducible manganese. The absorbance was recorded -using a Bausch and Lomb spectronic '20' reading at 525 nm. Exchangeable and available iron were determined spectrophotometrica11y by Saywe1J's10 method using normal ammonium acetate (pH-7) and ammonium acetate (pH-4.8) as extractant respectively ; and hydroxylamine hydrochloride and orthophenanthro· line as colour reagents. Absorbance was recorded at 490 nm. Blanks were used in aU the cases. .1. Indian Cbem. Soc., Vol. LVII, July 1980, pp.736-739 .1. Indian Cbem. Soc., Vol. LVII, July 1980, pp.736-739 Results and Discussion The data on the water soluble, exchangeable, C ON WATER SOLUBLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATION AND TIME ABLE I-EFFECT OF BHC ON WATER SOLUBLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATIO Doses Water soluble manganese in ppm at various days Average in g per kg of soil 1 15 30 45 60 75 90 0.00 3.26 4.43 5.68 5.00 4.54 3.95 3.35 4.32 0.10 3.23 5.69 7.50 6 82 6.25 5.56 5.00 5.72 0.20 3.22 6.87 8.75 7.89 7.14 6.58 6.00 6.64 0.25 3.26 8.06 10.71 10.00 9.09 7.89 7.30 8.05 0.30 3.20 6.82 8.33 7.50 6.97 6.25 5.70 6.39 0.40 3.23 5.98 7.50 6.67 6.10 5.71 5.10 5.75 0.50 3.22 5.00 6.58 5.98 5.55 4.76 4.20 5.04 Average 3.23 6.12 7.86 7.12 6.52 5.81 5.26 Variance ratio and L.S.D. for days= 61.84, 0.292. Variance ratio and L.S.D. for dosage=4l.46, 0.292. CT OF BHC ON EXCHANGEABLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATION AND TIME fABLE 2 -EFFECT OF BHC ON EXCHANGEABLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRAT Doses Exchangeable manganese in~ppm at various days in g per kg of soil 1 15 30 45 60 75 90 0.00 8.90 13.75 16.00 14.38 13.75 13.00 12.50 0.10 8.85 15.00 19.38 16.50 15.79 15.00 14.00 0.20 8.94 19.38 21.25 20.00 19.23 18.46 16.70 0.25 8.90 22.50 25.00 22.00 21.25 20.64 19.80 0.30 8.96 20.00 22.00 19.38 18.44 17.84 17.00 0.40 8.98 18.75 20.00 18.00 17.07 16.32 15.60 0.50 8.92 14.32 17.50 15.00 14.26 13.52 12.90 Average 8.92 17.68 20.16 17.89 17.11 16.39 15.50 Variance ratio and L.S.D. for days=54.75, 0.721. Variance ratio and L.S D. for dosage =25.90, 0.721. Experimental The soil used in the study (depth 0-30 em) was collected from a representative area of AJigarh district. It was dried, crushed and seived. The general physicochemical characteristics of the soil were determined and yielded the following data : pH= 8.80, electrical conductivity = 4.5 x IO-• mhos cm-1 , organic matter= 0.41%, water soluble Mn = 3.2 ppm, exchangeable Mn = 8.8 ppm, reducible Mn = 61.4 ppm, exchangeable Fe = 11.9 ppm and available Fe = 38.5 ppm. 736 BANSAL & GOYAL : INFLUENCE OF PESTICIDES ON THE TRACE ELEMENT ETC. reducible and active Mn in presence of BHC with different doses at various time intervals ot time are represented in Tables 1 to 4. Results for exchang- eable and available Fe are recorded in Tables 5 and 6. Each datum in the table is a mean of triplicate values. The data were subjected to statistical analysis of variance according to the two way classification with one observation per cell, the null hypothesis being the equality of the effects of different doses and different time intervals. Correlation coefficients were also calculated. The results show that with the application of BHC there was no significant change in pH and electrical conductivity of the soil. J. INDIAN CHEM. SOC., VOL. LVII, JULY 1980 An examination of Table 1 showed that compared to control, water soluble Mn increased significantly. Maximum value reached with 0.25 g of BHC at 30 days and then declined. The increase in water soluble Mn was 10.7 ppm from3.2 ppm. The influence of the doses was in the order 0.25>0. 2 > 0.3 > 0.4> 0.1 >0.5>0.0 g of BHC per kg of soil. Similar results with variable intensity were observed in exchangeable and active Mn (Tables 2 and 4). With the 0.25 g dose of BHC the increase in exchangeable Mn was 25.0 from 8.9 ppm and in active Mn from 74.0 to 84.5 ppm at an interval of 30 days. The influence of the doses for exchangeable Mn was in the order 0.25>0.3>0.2>0.4>0.1 >0 5>0.0 g of BHC per kg of soil and 0.25>0.2>0.3>0.1 >0.4>0.5> 0.0 g for active Mn. Wensley16 and Martin et al. 15 , who reported an enhanced microbial activity in soils as a result of fumigation ; with the work of Klemmer16 who noti- ced that lower doses of certain eradicant type chemi· cals produced greater, reinvasion potentials in soils; and with the results of Smith11 who found an in· crease in water soluble and available Mn2 +- on fumi· gation. With the toxic effect of higher doses of the chemical on the soil micro-organisms, the stimulated effect declined after a certain period, and the equi· Jibrium shifted towards Mnu resulting in a decrease of Mn 2 + form (Table 3). The decline in water soluble and exchangeable Mn with passage of time appeared, may be due to the dissipation of the organic chemical with re·establishment of pesticide sensitive Mn oxidising bacteria such as Bacillus, Actinomycetes, Aerobactor etc. The results were in agreement with the findings of Charles and PauJH. Various workers11 have reported the mechanisms by which an increase in availability of water soluble, available and active Mn occurred in fumigated and unamended soil. The fact that all the three forms of Mn simultaneously increased and then declined showed an effect of both biological and chemical factors on the Mn status of the soil under study. Aligarh soils have been found to con- tain hard nodular concretions of manganiferous deposits12• Correlation coefficient data also supportS the above conclusions. Thus while exchangeable Mn was positively correlated to water soluble form (r2=0.973), there was a negative correlation between exchangeable and reducible Mn (r2 = -0.971). J. INDIAN CHEM. SOC., VOL. LVII, JULY 1980 An examination of Tables 5 and 6 showed that exchangeable and available Fe significantly increa- sed, reached a maximum value with 0.25 g dose of BHC at 30 days of its application and then decrea- sed significantly. The effect of the doses on the increase of exchangeable Fe was 0.25>0.2>0.3> 0 4>0.1 >0 5>0.0 g BHC per kg soil and 0.25:> 0.3>0.2>0.1>0.4>0.5>0.0 g for available Fe. The maximum increase for exchangeable Fe was 25.0 ppm from 12.2 ppm (Table 5) and 58.9 ppm from 39.4 ppm for available Fe (Table 6). The increase in water soluble, available and active Mn occurred, may be due to the solubilising effects caused by soil fungi Aspergillus niger1 8 and/ or bacteria Thiobacil/us; the liberation of Mn from organic matter by the stimulated activity of fumigant resistant bacteria which led to decomposition of organic matter as the soil was wetted with water. These results found support, with the work of TABLE 5-EFFECT OF BHC ON EXCHANGEABLE IRON OF SOIL WITH VARIATION IN CONCENTRATION AND TIME Doses Exchangeable iron in ppm at various days in g per kg of soil 1 15 30 45 60 75 90 0.00 12.14 14.29 16.42 15.22 14.65 13.57 12.90 0.10 12.22 17.14 19.28 18.02 17.49 16.37 15.70 0.20 12.35 19.99 22.50 21.42 20.71 19.29 18.50 0.25 12.24 21.89 25.00 23.57 22.14 20.72 20.00 0.30 12.14 19.46 22.14 21.04 19.19 18.92 18.10 0.40 12.55 18.03 20.71 19.73 18.75 17.35 16.50 0.50 12.34 16.60 19.11 18.03 17.14 15.73 15.00 Average 12.28 18.20 20.74 19.58 18.90 17.42 16.57 Variance ratio and L.S.D. for days-42.50, 0.640. Variance ratio and L.S.D. for dosage=26.37, 0,640. Average 14.17 16.60 19.25 20.79 18.83 17.66 16.28 OF BHC ON EXCHANGEABLE IRON OF SOIL WITH VARIATION IN CONCENTRATION AND TIME TABLE 5-EFFECT OF BHC ON EXCHANGEABLE IRON OF SOIL WITH VARIATION IN CONCENT TABLE 6-BFFECT Ol' BHC ON AVAILABLE IRON OF SOIL WITH VARIATION IN CONCENTRATION AND TIMB Doses Available iron in ppm at various days in g per kg of ~oil 1 15 30 45 0.00 39.40 47.14 50.00 48.58 0.10 39.34 49.27 52.13 50.00 0.20 39.78 51.40 54.97 52.84 0.25 39.42 54.95 58.92 54.66 0.30 39.57 52.09 55.37 52.23 0.40 39.34 49.96 52.09 50.67 0.50 39.42 42.73 49.96 48.54 Average 39.47 50.36 53.35 51.07 Var!ance ratio and L.S.D. for days=501.25, 0.300. Vanance ratio and L.S.D. for dosage=124.94, 0.300. Results and Discussion Average 13.18 14.93 17.71 20.01 17.66 16.39 13.76 ECT OF BHC ON RBDUCIBLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATION AND TIME TABLE 3-EFFECT OF BHC ON RBDUCIBLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATI TABLE 3-EFFECT OF BHC ON RBDUCIBLE MANGANESE OF SOIL WITH VARIATION IN CONCENTRATION AND TIME Doses Reducible manganese in ppm at various days in g per kg of soil 1 15 30 45 60 75 90 0 00 61.74 56.82 55.52 57.26 57.91 58.05 58.1S 0.10 61.94 56.19 54.37 57.20 57.40 57.82 58.80 0.20 61.53 53.68 52.33 53.67 54.59 54 87 56.30 0 25 61.90 51.08 48.83 50.89 51.69 52.00 52.40 0.30 61.71 52.42 51.70 53.37 53.82 54.15 54.30 0.40 6~.85 53.21 53.38 55.26 55.87 56.09 56.80 0.50 61.89 56.91 54.16 56.70 57.27 58.22 58.40 Average 61.79 54.33 52.90 54.91 55.52 55.89 56.45 Variance ratio and L.S.D. for days=25.21, 0.836. Variance ratio and L.S.D. for dosage= 10.50, 0.836. Average 57.92 57.69 55.28 52.68 54.50 56.07 57.65 TABLE 4-EFFECT OF BHC ON ACTIVE MANGANESE OP SOIL WITH VARIATION IN CONCENTRATION AND TIME Dose~ Active manganese in ppm at various days in g per kg of soil 15 30 45 0.00 73.90 75.00 77.20 76.64 0.10 74.02 76.88 81.25 80.52 0.20 73.69 79.93 82.33 81.56 0.25 74.06 81.64 84.54 82.89 0.30 73.87 79.24 82.03 80.25 0.40 74.06 77.94 80.88 79.93 0.50 74.03 76.29 78.24 77.68 Average 73.95 78.13 80.92 79.92 Variance ratio and L.S.D. for days=22.24, 0.744. Variance ratio and L.S.D. for dosage= !0.80, 0. 7 44. 60 76.20 79.53 80.96 82.03 79.23 79.04 77.08 79.16 75 75.00 79.38 79.93 80.52 78.24 78.12 76.50 78.24 90 74.00 77.80 79.00 79.50 77.00 76.50 75.50 77.04 Average 75.42 78.48 79.63 80.74 78.55 78.07 76.45 737 J. INDIAN CHEM. SOC., VOL. LVII, JULY 1980 BANSAL & GOYAL : INFLUENCE OF PESTICIDES ON THE TRACE ELEMENT ETC. A. WALKLEY and I. A. BLACK, Soil Sci., 1947, 63, 251. 8. M. L. JACKSON, ''Soil Chemical Analysis'', Prentice- Hall, Englewood Cliffs, 1958. 9. G. D. SHERMAN and P. M. HARMER, Soil Sci. Soc. Am. Proc., 1942, 7, 398. 10. L. G. SAYWELL and B. B. CUNNINGHAM, Ind. Eng. Chern., 1937, 9, 67. 11. H. D. SMITH, Soil Sci. Soc. Am. Ptoc., 1963, 27, 538. 12. J.P. SINGHAL and MOHO, AKHTAR, Ag•·ochemica, 1973,. 17,424. 13. K. R. BUTALIN, J. R. POSTGATE, B. A. FRY and J. L. PEEL, "Autotrophic Micro-organism'', Cambridge Univ. Press, 1941, p. 271. 14. R. M. WENSLEY, Canad. J. Bioi., 1953, 31, 227. 15. J.P. MARTIN, R. B. BAINES and J. 0. EBVIN, Soil Sci. Soc. Am. Proc., 1957, 21, 163. KLEMMER~ 16. H. W. KLEMMER~ Amer. Bacterial. Proc., 1957, 57, 12. 16. H. W. KLEMMER~ Amer. Bacterial. Proc., 195 17. F. E. CHARLES and H. E. PAUL, Soil Sci. Soc. Amer. Proc., 1958, 22, 235. BANSAL & GOYAL : INFLUENCE OF PESTICIDES ON THE TRACE ELEMENT ETC. TABLE 7-CORRELATION COEFFICIENTS BETWEEN DIFFERENT FORMS OF MANGANESE AND IRON Sl. Form Water No· soluble Mn 1. Exchangeable Mn 0.973 2. Water soluble Mn 3. Reducible Mn 4. Exchangeable Fe Reducible Exchangeable Available Mn Fe Fe -0.971 0.976 0.978 -0.930 -0.973 -0.971 -0.932 -0.956 0.936 TABLE 7-CORRELATION COEFFICIENTS BETWEEN DIFFERENT FORMS OF MANGANESE AND IRON References An examination of Table 7 showed that there existed significant positive correlation between exchangeable Mn and exchangeable and available Fe (r2=0.976 and 0.978 respectively) as well as between water soluble Mn and exchangeable al1d available Fe (r2 = 0.973 and 0.971 respectively). There was also a positive correlation between exchangeable and available Fe (r2 = 0.936). The exchangeable and available Fe, however, were negatively and signifi- cantly correlated to reducible Mn (r2 = -0.932 and -0.956 respectively). 1. D. G. ALDRICH and J, P. MARTIN, Soil Sci., 1952, 73._ 149. 1. D. G. ALDRICH and J, P. MARTIN, Soil Sci., 1952, 73._ 149. 2. W. J. HANSON and R. W. NEX, Soil Sci., 1953, 76, 209. 2. W. J. HANSON and R. W. NEX, Soil Sci., 1953, 76, 209. 3. M. ALEXANDER, "Introduction to soil microbiology .. John Wiley and Sons Inc., New York and London,. 1961, p. 387 and 404. 4. J. P. SINGHAL and C. P. SINGH, Soil Sci., Pl. Nutrition,. 1974, 20, 413. 5. J.P. SINGHAL, S. U. KHAN and G. K. GUPTA, J. Indian Chern. Soc., 1975, 52, 193. The above results pointed to the fact that condi- tions affecting Fe and Mn availability in soils were of similar type, a view in accordance with results of Smith11 • It appeared that, reducible character of organic matter in soil and its microbial decomposi- tion with liberation of acids, the drop in oxidation- reduction potential in moist conditions of the soil, stimulation in the activity of micro-organism by pesti- cides extending upto a period of about 30 days and favourable pH 1 9 all contributed towards an increase in exchangeable and available Fe. The decline in available 'and exchangeable Fe with passage of time appeared, may be due to the fact that with dissi- pation of the organic chemical, the stimulatory effects decline ; and Fe bacteria take over both the forms. Same results were also observed by Pring- sheim a,zo and Alexanders. 6. J.P. SINGHAL, S. U. KHAN and 0. P. BANSAL, J. Indian Chern. Soc., 1976, 53, 313. 7. J. INDIAN CHEM. SOC., VOL. LVII, JULY 1980 60 47.17 48.58 50.11 53.53 50.81 48.54 46.31 49.38 75 46.34 47.62 49.96 53.44 49.88 47.62 45.60 48.49 90 45.50 47.80 49.00 51.60 49.00 46.80 44.80 47.78 Average 46.30 47.82 49.81 52.22 49.85 47.86 46.05 ECT Ol' BHC ON AVAILABLE IRON OF SOIL WITH VARIATION IN CONCENTRATION AND TIMB TABLE 6-BFFECT Ol' BHC ON AVAILABLE IRON OF SOIL WITH VARIATION IN CONCENT 738 AL & GOYAL : INFLUENCE OF PESTICIDES ON THE TRACE ELEMENT ETC. Acknowledgement 18. R. L. STARKEY, Producers lvfonthly, 1958,22, 12. Authors are thankful to Dr. S. S. Gupta, Princi- pal and Dr. H. Singh, Dept. of Chemistry, for encoura- gement, cooperation and helpful suggestions. 19. E. G. PRINGSHEIM, Phil. Trans. Royal Soc. London,. 1946,232, 311. 20. E. G. PRINGSHEIM, Bioi. Re~ .• 1949, 24, 200. 73~ JICS-10
https://openalex.org/W2972239447	https://pure.mpg.de/pubman/item/item_3170758_2/component/file_3170759/shh2440.pdf	English	null	Sourcepredict: Prediction of metagenomic sample sources using dimension reduction followed by machine learning classification	Journal of open source software	2,019	cc-by	1,620	License Authors of papers retain copyright and release the work under a Creative Commons Attribution 4.0 International License (CC-BY). In cases where the origin of a metagenomic sample, its source, is unknown, it is often part of the research question to predict and/or confirm the source. For example, in microbial archaelogy, it is sometimes necessary to rely on metagenomics to validate the source of paleofaeces. Using samples of known sources, a reference dataset can be established with the taxonomic composition of the samples, i.e., the organisms identified in the samples as features, and the sources of the samples as class labels. With this reference dataset, a machine learning algorithm can be trained to predict the source of unknown samples (sinks) from their taxonomic composition. Other tools used to perform the prediction of a sample source already exist, such as Source- Tracker (Knights et al., 2011), which employs Gibbs sampling. However, the Sourcepredict results are more easily interpreted since the samples are embedded in a human observable low-dimensional space. This embedding is performed by a dimension reduction algorithm followed by K-Nearest-Neighbours (KNN) classification. DOI: 10.21105/joss.01540 O 0 05/joss 0 5 0 Software • Review • Repository • Archive Submitted: 28 June 2019 Published: 04 September 2019 License Authors of papers retain copyright and release the work under a Creative Commons Attribution 4.0 International License (CC-BY). /j Software • Review • Repository • Archive Submitted: 28 June 2019 Published: 04 September 2019 License Authors of papers retain copyright and release the work under a Creative Commons Attribution 4.0 International License (CC-BY). Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimensio Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimension reduction followed by machine learning classification. Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 1 Summary SourcePredict is a Python package distributed through Conda, to classify and predict the origin of metagenomic samples, given a reference dataset of known origins, a problem also known as source tracking. Submitted: 28 June 2019 Published: 04 September 2019 DNA shotgun sequencing of human, animal, and environmental samples has opened up new doors to explore the diversity of life in these different environments, a field known as metage- nomics (Hugenholtz & Tyson, 2008). One aspect of metagenomics is investigating the com- munity composition of organisms within a sequencing sample with tools known as taxonomic classifiers, such as Kraken (Wood & Salzberg, 2014). Sourcepredict: Prediction of metagenomic sample sources using dimension reduction followed by machine learning classification Maxime Borry1 1 Department of Archaeogenetics, Max Planck Institute for the Science of Human History, Jena, 07745, Germany Prediction of the proportion of unknown sources Let Si ∈{S1, .., Sn} be a sample from the normalized sinks dataset Dsink, o i j ∈{o i 1 , .., o i n i o } an organism in Si, and n i o the total number of organisms in Si, with o i j ∈Z+. Let m be the mean number of samples per source in the reference dataset, such that m = 1 O ∑O i=1 Si. For each Si sample, I define \|\|m\|\| derivative samples U Si k ∈{U Si 1 , .., U Si \|\|m\|\|} to add to the reference dataset to account for the unknown source proportion in a test sample. Separately for each Si, a proportion denoted α ∈[0, 1] (default = 0.1) of each o i j organism of Si is added to each U Si k sample such that U Si k (o i j ) = α · xi j , where xi j is sampled from a Gaussian distribution N ( Si(o i j ), 0.01). The \|\|m\|\| U Si k samples are then added to the reference dataset Dref, and labeled as unknown, to create a new reference dataset denoted unkDref. To predict the proportion of unknown sources, a Bray-Curtis (Bray & Curtis, 1957) pairwise dissimilarity matrix of all Si and U Si k samples is computed using scikit-bio (Rideout et al., 2018). This distance matrix is then embedded in two dimensions (default) with the scikit-bio implementation of PCoA. This sample embedding is divided into three subsets: unkDtrain (64%), unkDtest (20%), and unkDvalidation(16%). The scikit-learn (Pedregosa et al., 2011) implementation of KNN algorithm is then trained on unkDtrain, and the training accuracy is computed with unkDtest. This trained KNN model is then corrected for probability estimation of the unknown proportion using the scikit-learn implementation of Platt’s scaling method (Platt & others, 1999) with unkDvalidation. The proportion of unknown sources in Si, pu ∈[0, 1] is then estimated using this trained and corrected KNN model. Ultimately, this process is repeated independently for each sink sample Si of Dsink. Prediction of the proportion of known sources First, only organism TAXIDs corresponding to the species taxonomic level are retained us- ing the ETE toolkit (Huerta-Cepas, Serra, & Bork, 2016). A weighted Unifrac (default) (Lozupone, Hamady, Kelley, & Knight, 2007) pairwise distance matrix is then computed on the merged and normalized training dataset Dref and test dataset Dsink with scikit-bio, using the NCBI taxonomy as a reference tree. This distance matrix is then embedded in two dimen- sions (default) using the scikit-learn implementation of t-SNE (Maaten & Hinton, 2008). The 2-dimensional embedding is then split back to training tsneDref and testing dataset tsneDsink. The KNN algorithm is then trained on the train subset, with a five (default) cross validation to look for the optimum number of K-neighbors. The training dataset tsneDref is further divided into three subsets: tsneDtrain (64%), tsneDtest (20%), and tsneDvalidation (16%). The training accuracy is then computed with tsneDtest. Finally, this second trained KNN model is also corrected for source proportion estimation using the scikit-learn implementation of the Platt’s method with tsneDvalidation. The proportion pcs ∈[0, 1] of each of the ns sources cs ∈{c1, .., cns} in each sample Si is then estimated using this second trained and corrected KNN model. Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimensio Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimension reduction followed by machine learning classification. Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 2 Method Starting with a numerical organism count matrix (samples as columns, organisms as rows, obtained by a taxonomic classifier) of merged references and sinks datasets, samples are first normalized relative to each other, to correct for uneven sequencing depth using the geometric mean of pairwise ratios (GMPR) method (default) (L. Chen et al., 2018). After normalization, Sourcepredict performs a two-step prediction algorithm. First, it predicts the proportion of unknown sources, i.e., which are not represented in the reference dataset. Second, it predicts the proportion of each known source of the reference dataset in the sink samples. Organisms are represented by their taxonomic identifiers (TAXID). Acknowledgements Thanks to Dr. Christina Warinner, Dr. Alexander Herbig, Dr. AB Rohrlach, and Alexander Hübner for their valuable comments and for proofreading this manuscript. This work was funded by the Max Planck Society and the Deutsche Forschungsgemeinschaft, project code: EXC 2051 #390713860. Thanks to Dr. Christina Warinner, Dr. Alexander Herbig, Dr. AB Rohrlach, and Alexander Hübner for their valuable comments and for proofreading this manuscript. This work was funded by the Max Planck Society and the Deutsche Forschungsgemeinschaft, project code: EXC 2051 #390713860. Combining unknown and source proportions For each sample Si of the test dataset Dsink, the predicted unknown proportion pu is then combined with the predicted proportion pcs for each of the ns sources cs of the training dataset such that ∑ns cs=1 sc + pu = 1 where sc = pcs · pu. Finally, a summary table gathering the estimated sources proportions is returned as a csv file, as well as the t-SNE embedding sample coordinates. Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimensi Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 Borry, (2019). Sourcepredict: Prediction of metagenomic sample sources using dimension reduction followed by machine learning classification. Journal of Open Source Software, 4(41), 1540. https://doi.org/10.21105/joss.01540 3 References Bray, J. R., & Curtis, J. T. (1957). An ordination of the upland forest communities of southern Wisconsin. Ecological monographs, 27(4), 325–349. doi:10.2307/1942268 Chen, L., Reeve, J., Zhang, L., Huang, S., Wang, X., & Chen, J. (2018). GMPR: A robust normalization method for zero-inflated count data with application to microbiome sequencing data. PeerJ, 6, e4600. doi:10.7717/peerj.4600 Huerta-Cepas, J., Serra, F., & Bork, P. (2016). ETE 3: Reconstruction, analysis, and visualization of phylogenomic data. Molecular Biology and Evolution, 33(6), 1635–1638. doi:10.1093/molbev/msw046 Hugenholtz, P., & Tyson, G. W. (2008). Microbiology: Metagenomics. Nature, 455(7212), 481. doi:10.1038/455481a Knights, D., Kuczynski, J., Charlson, E. S., Zaneveld, J., Mozer, M. C., Collman, R. G., Bushman, F. D., et al. (2011). Bayesian community-wide culture-independent microbial source tracking. Nature methods, 8(9), 761. doi:10.1038/nmeth.1650 Lozupone, C. A., Hamady, M., Kelley, S. T., & Knight, R. (2007). Quantitative and quali- tative beta diversity measures lead to different insights into factors that structure microbial communities. Appl. Environ. Microbiol., 73(5), 1576–1585. doi:10.1128/AEM.01996-06 Maaten, L. van der, & Hinton, G. (2008). Visualizing data using t-SNE. Journal of Machine Learning Research, 9(Nov), 2579–2605. Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., et al. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12(Oct), 2825–2830. Platt, J., & others. (1999). Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in Large Margin Classifiers, 10(3), 61–74. Rideout, J. R., Caporaso, G., Bolyen, E., McDonald, D., Baeza, Y. V., Alastuey, J. C., Pitman, A., et al. (2018, December). biocore/scikit-bio: scikit-bio 0.5.5: More compositional methods added. doi:10.5281/zenodo.2254379 Wood, D. E., & Salzberg, S. L. (2014). Kraken: Ultrafast metagenomic sequence classification using exact alignments. Genome Biology, 15(3), R46. doi:10.1186/gb-2014-15-3-r46
https://openalex.org/W3192158569	https://zenodo.org/record/5106652/files/34740.pdf	English	null	Influence of cranioskeltal trauma, complicated with blood loss, on the bile-forming function of liver during the early period of traumatic disease among rats with different resistance to hypoxia and its correction	Journal of Education, Health and Sport	2,021	cc-by	5,021	8 8 Received: 25.12.2020. Revised: 14.12.2020. Accepted: 29.01.2021. UDC 616.711/.714-001.3-005.1-036.4:616.36-008.8]-085-092.9 UDC 616.711/.714-001.3-005.1-036.4:616.36-008.8]-085-092.9 Sikirynska D. O., Hudyma A. A., Pokhodun K. A., Antonyshyn I. V. Influence of cranioskeltal trauma, complicated with blood loss, on the bile-forming function of liver during the early period of traumatic disease among rats with different resistance to hypoxia and its correction. Journal of Education, Health and Sport. 2021;11(1):326-337. eISSN 2391-8306. DOI http://dx.doi.org/10.12775/JEHS.2021.11.1.032 https://apcz.umk.pl/czasopisma/index.php/JEHS/article/view/JEHS.2021.11.1.032 htt // d / d/5106652 8 The journal has had 5 points in Ministry of Science and Higher Education parametric evaluation. § 8. 2) and § 12. 1. 2) 22.02.2019. © The Authors 2021; This article is published with open access at Licensee Open Journal Systems of Nicolaus Copernicus University in Torun, Poland Open Access. This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author (s) and source are credited. This is an open access article licensed under the terms of the Creative Commons Attribution Non commercial license Share alike. (http://creativecommons.org/licenses/by-nc-sa/4.0/) which permits unrestricted, non commercial use, distribution and reproduction in any medium, provided the work is properly cited. The authors declare that there is no conflict of interests regarding the publication of this paper. Abstract Abstract Introduction. The development of multiorgan dysfunction and insufficiency is one of the serious complications of severe multiple and combined lesions. In their pathogenesis, a key role is played by secondary lesions of organs remote from the site of direct injury, which is caused by the systemic response of the body to inflammation and is a characteristic feature of traumatic disease. As a model of the development of organ damage in experimental polytrauma, the biliary function of liver is often investigated. However, the features of systemic disorders, in particular, the biliary function of liver, in conditions of severe skeletal injury, depending on the resistance to hypoxia, have not been studied enough. Purpose: to determine the dynamics of biliary function of liver in case of cranioskeletal trauma complicated with blood loss among rats with different resistance to hypoxia during the early period of traumatic disease and to evaluate the effectiveness of thiocetam in the correction of identified disorders. 326 Materials and methods. 108 nonlinear white male rats weighing 180-200 g were used in the experiments. Previously, individual resistance to hypoxia was determined, rats were divided into two groups: with high- and low-resistance (HR and LR) animals to hypoxia. Subsequently, HR and LR rats were divided into 4 groups: one control and three experimental. Under conditions of thiopentale sodium anesthesia (40 mg kg-1) the first experimental group HR and LR of rats was had cranioskeletal trauma, the second - acute blood loss in the amount of 20-22% of circulating blood volume, the third experimental group after application of cranioskeletal injuries and acute blood loss in order to correct HR and LR- rats were thrown Thiocetam intraperitoneally, dose 250 mg kg-1 of body weight 1 time per day. Animals of the control and experimental groups in 1, 3 and 7 days after injury were catheterized the common bile duct, collected bile, in which the content of total bile acids was determined. It was found out that laboratory white rats, which are genetically different in resistance to hypoxia, also differ in the intensity of biliary function of liver. Among HR rats, the bile content of total bile acids was higher than that among LR rats. Under the influence of cranioskeletal injury, the bile content of total bile acids decreased among both HR and LR rats. Abstract Despite the lower initial content of total bile acids in the bile of LR rats, under the influence of cranioskeletal trauma in this group, the degree of the studied indicator reduction was significantly greater than among HR rats, indicating greater sensitivity of LR rats to pathogens of traumatic disease, which lead to damage to the microsomal monooxygenase system of hepatocytes. Under the conditions of acute blood loss additional modeling, the disorders were more visible, but the result was statistically significant only among LR rats in 3 days of the experiment. In addition, LR rats had a greater degree of reduction of the studied indicator in 3 and 7 days of the experiment. Under the influence of seven-day-usage of Thiocetam in the groups of LR and HR rats with cranioskeletal trauma, the violation of total bile acids content in the bile became significantly lower compared to animals without correction. A similar result was found under conditions of additional acute blood loss. Under these conditions, the degree of increase of total bile acids content in the bile was greater among LR rats than among HR rats, which indicates a higher efficiency of Thiocetam among LR rats. Conclusions: 1. The content of total bile acids in the bile among intact HR-rats is significantly higher than among LR-rats. Under the influence of cranioskeletal trauma, the content of total bile acids in the bile is reduced compared to the control, but the degree of reduction is significantly greater among LR rats after 7 days of the experiment. Complication 327 of cranioskeletal injury with acute blood loss causes a significantly greater reduction in the bile content of total bile acids among LR rats compared with HR rats after 3 and 7 days of the experiment. The use of Thiocetam is accompanied by a decrease of liver biliary dysfunction, which is more visible among LR rats with acute blood loss. Key words: cranioskeletal trauma; blood loss; resistance to hypoxia; bile-forming function of liver; Thiocetam. Introduction. The development of multiorgan dysfunction and insufficiency is one of serious complications of severe multiple and combined lesions. In their pathogenesis, a key role is played by secondary lesions of organs remote from the site of direct injury, which is caused by the systemic response of the body to inflammation and is a characteristic feature of traumatic disease [1]. Abstract As a model of the development of organ damage in experimental polytrauma, the biliary function of liver is often studied, in particular the concentration of total bile acids and conjugated bilirubin in bile [2], which are organ-specific features of the functional state of liver and are related to the functional capacity of the microsomal hepatocyte system, in which the main synthetic and detoxification processes take place [3]. In particular, [4] the work shows that in the acute period and the period of early manifestations of traumatic disease in the case of cranioskeletal trauma there is a violation of biliary function of liver, manifested by decreased bile secretion and excretion of total bile acids, cholesterol, total and direct bilirubin with a minimum 7 days after injury. The revealed disturbances deepen after additional acute blood loss. In the pathogenesis of the identified disorders, the authors perceive the influence of liver hypoxia, which occurs due to impaired microcirculation and decreased blood supply of the organ with the activation of lipid peroxidation and degradation of endoplasmic membranes. It is known that one of the constitutional features of the mammalian body is resistance to hypoxia. However, the features of systemic disorders, in particular biliary function of liver, under conditions of severe skeletal injury, depending on the resistance to hypoxia, have not been studied enough. Giving the leading role to hypoxia in the pathogenesis of polytrauma, we can assume a different course of traumatic disease among animals with high- and low- resistance to hypoxia. Purpose: to determine the dynamics of liver biliary function in the case of cranioskeletal trauma complicated with blood loss among rats with different resistance to 328 hypoxia during the early period of traumatic disease and to evaluate the effectiveness of thiocetam in the correction of the identified disorders. Materials and methods. 108 nonlinear white male rats weighing 180-200 g were used in the experiments. All animals were on a standard vivarium diet. Previously, according to the method described in the work [5], rats were exemined for individual resistance to hypoxia and two groups were identified: high- and low-resistant (HR and LR) to hypoxia animals. In 14 days, HR and LR rats were divided into 4 groups: one control (6 intact hypoxia-resistant rats), and three experimental. Abstract Under thiopental sodium anesthesia (40 mg kg-1) in the first experimental group HR and LR rats (36 animals) simulated cranioskeletal trauma by sequential application of a dosed stroke to the skull with the achievement of closed craniocerebral trauma of moderate severity and dosed stroke on the thigh by percussion device with a wedge-shaped nozzle to achieve a closed hip fracture [6]. In the second experimental group, HR and LR-rats (36 animals) under conditions of asepsis and antiseptics additionally simulated acute blood loss in the amount of 20-22% of the circulating blood volume by crossing the femoral vein [7]. In the third experimental group after application of cranioskeletal trauma and acute blood loss for the purpose of correction of HR and LR rats (24 animals) were intraperitoneally administered Thiocetam (JSC "Halychpharm", Lviv, Ukraine), dose of 250 mg kg-1 body weight once a day [8]. The biliary function of liver was studied among animals of the experimental groups 1, 3 and 7 days after injury, which corresponded to the acute period and the period of early manifestations of traumatic disease. Under thiopental-sodium anesthesia (60 mg kg-1), the common bile duct was catheterized in animals and bile was collected for 1 hour. The content of total bile acids was determined in the obtained bile [3]. Animals in the control group were only thrown under thiopental sodium anesthesia, bile secretion function was studied after 1 day. In order to neutralize the effect on the result of different content of total bile acids in HR- and LR-rats of the control group, in addition to the absolute values of the studied indicator, the ratio of individual values of indicators in groups to the average value of the control group is calculated. This allowed us to assess the statistical significance of the deviation degree of the studied indicator in the groups of HR and LR. We also calculated the average ratio of individual values of total bile acids in the group of animals with correction to the average value of the animal group without correction to assess the statistical significance of the deviation degree of the studied indicator among animals treated with Thiocetam compared with animals without correction. 1. * - differences relative to the control group are statistically significant (p0,05); 2. # - differences between groups of HR and LR animals are statistically significant (p0,05); otes. Here and in other tables: 3. 1d, 3d - differences in relation to 1 and 3 days of observation are statistically significant (p0,05); g p 4. p1 - the probability of differences in groups of emergencies of LR animals with and without additional blood loss; 5. p2 - the probability of differences in the groups of HR animals with existing and no additional blood loss; Abstract 329 During the work with laboratory animals, international requirements for the humane treatment with animals were complied in accordance with the rules of the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (European Convention, 1984). Euthanasia of rats after bile collection was performed by total bloodletting from the heart. Statistical analysis of the results was performed using the Mann-Whitney test in the software package STATISTICA 10.0 ("StatSoft, Inc.", USA). Research results and their discussion. As can be seen from table 1, in the control group of HR rats, the bile content of total bile acids was statistically significantly higher than that of LR rats (by 30,2%, p <0,05). Table 1 - Dynamics of total bile acids content (g ∙ l-1) in response to cranioskeletal trauma complicated by blood loss among animals with different resistance to hypoxia (Me (LQ; UQ)) - median (lower and upper quartiles) Resistance to hypoxia Control The term of the post-traumatic period 1 day 3 day 7 day Cranioskeletal trauma Low-resistant 2,48 (2,29; 2,55) (n=6) 2,18* (1,99; 2,25) (n=7) 1,881д (1,69; 1,95) (n=7) 1,461д,3д (1,31; 1,50) (n=7) Highly resistant 3,23# (3,15; 3,41) (n=6) 2,78# (2,48; 2,96) (n=8) 2,51# (2,42; 2,78) (n=10) 2,33# (2,14; 2,51) (n=10) Cranioskeletal trauma + blood loss Low-resistant 2,48 (2,29; 2,55) (n=6) 1,95 (1,95; 2,06) (n=6) 1,431д (1,24; 1,50) (n=6) 1,201д (1,20; 1,28) (n=5) Highly resistant 3,23# (3,15; 3,41) (n=6) 2,51# (2,31; 2,66) (n=8) 2,36# (2,27; 2,51) (n=9) 2,03#1д,3д (1,95; 2,21) (n=8) р1 >0,05 <0,05 >0,05 р2 >0,05 >0,05 >0,05 d i h bl Notes. Here and in other tables: g (p , ); 4. p1 - the probability of differences in groups of emergencies of LR animals with and without additional blood loss; 5. p2 - the probability of differences in the groups of HR animals with existing and no additional blood loss; 330 Under the influence of cranioskeletal trauma among LR rats, the bile content of total bile acids decreased: after 1 day - by 12,1%, after 3 days - by 24,2%, after 7 days - by 41,1%. In all terms of the post-traumatic period, the differences were statistically significant compared with the control (p <0,05). Among HR rats, the indicator also decreased statistically significantly compared with the control: by 13,9, 22,2 and 27,9%, accordingly (p <0,05). Abstract Comparison of the studied indicator value between the experimental groups showed that in all periods of the post-traumatic period among HR rats, the absolute value of the studied indicator was significantly higher than among LR rats (27,5, 33,5 and 59,6%, accordingly, p < 0,05). Analysis of the average ratio dynamics of individual bile contents of total bile acids to the average value of the control group showed (Table 2) that after 1 and 3 days the differences between the experimental groups were statistically not significant (p> 0,05). At the same time, after 7 days, the value of the studied indicator in the group of HR-rats was significantly higher (by 22,0 %, p <0,05). Table 2 - Dynamics of the average ratio of individual values of bile flow rate to the average value of the control group in response to cranioskeletal trauma complicated by blood loss among animals with different resistance to hypoxia (Me (LQ; UQ)) - median (lower and upper quartiles) Resistance to hypoxia The term of the post-traumatic period 1 day 3 day 7 day Cranioskeletal trauma Low-resistant 0,88 (0,80; 0,91) (n=7) 0,76 (0,68; 0,79) (n=7) 0,59 (0,53; 0,60) (n=7) Highly resistant 0,86 (0,77; 0,92) (n=8) 0,78 (0,75; 0,86) (n=10) 0,72# (0,66; 0,78) (n=10) Cranioskeletal trauma + blood loss Low-resistant 0,79 (0,79; 0,83) (n=7) 0,57 (0,50; 0,60) (n=7) 0,48 (0,48; 0,51) (n=7) Highly resistant 0,78 (0,71; 0,82) (n=8) 0,73# (0,70; 0,78) (n=10) 0,63# (0,60; 0,68) (n=10) Simulation of additional blood loss among animals with cranioskeletal trauma also led to a statistically significant decrease in the content of total bile acids in bile (Table 1): among LR animals, accordingly, by 21,4, 42,3 and 51,6% (p <0,05) among HR rats - by 22,3, 26,9 331 and 37,2%, accordingly (p <0,05). Comparison of experimental groups showed that during all periods of the post-traumatic period in the group of HR-rats the value of the studied indicator was significantly lower than among LR-rats (accordingly 28,7, 65,0 and 69,2%, p <0,05). Analysis of the average ratio dynamics of individual bile contents of total bile acids to the average value of the control group showed (Table 2) that after 1 day the differences between the experimental groups were statistically not significant (p> 0,05). At the same time, after 3 and 7 days, the value of the studied indicator in the group of HR rats was significantly higher (by 28,1 and 31,2%, p <0,05, accordingly). Abstract Groups comparison of LR rats with cranioskeletal trauma depending on the presence or absence of acute blood loss showed that under conditions of acute blood loss, the bile content of total bile acids became statistically significantly lower only after 3 days of the experiment (by 23,9%, p1 <0,05). Among HR rats with cranioskeletal trauma, additional blood loss also led to a decrease in the value of the studied indicator, but the result during all periods of the post-traumatic period was not statistically significant (p2> 0,05). Under conditions of correction with Thiocetam after 7 days, the content of total bile acids in the bile of animals, which simulated only cranioskeletal trauma (Fig. 1), increased: among LR animals - by 28,7% (p <0,05), among HR- rats - by 19,3% (p <0,05). Figure 1 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙ l-1) days after modeling of cranioskeletal trauma among animals with different resistance to poxia (Here and in Fig. 2: - differences in the group of animals without statistical rrection probable, p <0,05). 0 0.5 1 1.5 2 2.5 3 Low resistant Highly resistant 1.46 2.33 1,88* 2,78* Without correction 0 0.5 1 1.5 2 2.5 3 Low resistant Highly resistant 1.46 2.33 1,88* 2,78* Without correction Low resistant Highly resistant Figure 1 - The effect of Thiocetam on the content of total bile acids in the bile Figure 1 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙ l-1) 7 days after modeling of cranioskeletal trauma among animals with different resistance to hypoxia (Here and in Fig. 2: * - differences in the group of animals without statistical correction probable, p <0,05). Figure 1 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙ l-1) 7 days after modeling of cranioskeletal trauma among animals with different resistance to hypoxia (Here and in Fig. 2: * - differences in the group of animals without statistical correction probable, p <0,05). 332 Animals, which were additionally were simulated acute blood loss, the rate also increased under the influence of Thiocetam, (Fig. 2): among LR animals by 50,0% (p <0,05), among HR rats - by 29,6% <0,05). Animals, which were additionally were simulated acute blood loss, the rate also increased under the influence of Thiocetam, (Fig. Abstract 2): among LR animals by 50,0% (p <0,05), among HR rats - by 29,6% <0,05). Figure 2 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙l-1) 7 days after modeling of cranioskeletal trauma complicated by blood loss among animals with different resistance to hypoxia. 0 0.5 1 1.5 2 2.5 3 Low resistant Highly resistant 1,20 2.03 1,80* 2,63* Without correction Without correction Figure 2 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙l-1) 7 days after modeling of cranioskeletal trauma complicated by blood loss among animals with different resistance to hypoxia. Figure 2 - The effect of Thiocetam on the content of total bile acids in the bile (g ∙l-1) 7 days after modeling of cranioskeletal trauma complicated by blood loss among animals with different resistance to hypoxia. Estimation of the average ratio of individual bile contents of total bile acids in the group of animals with correction to the average value of animals without correction showed (Fig. 3) that the degree of increase in bile content of total bile acids under the influence of Thiocetam among animals with different resistance to hypoxia, which simulated only cranioskeletal trauma, did not differ (p> 0,05), although there was a tendency to a larger value in the group of LR animals. However, under conditions of additional blood loss, the rate was statistically significantly higher among LR rats compared with HR rats (by 16,3%, p <0,05). The obtained results indicate that laboratory white rats, which are genetically different by the degree of resistance to hypoxia, also differ by the intensity of bile-forming function of the liver. Among HR rats, the bile content of total bile acids was higher than that of LR rats. It is known that in the process of bile acids synthesis from cholesterol the leading role is played by the membranes state of the hepatocytes microsomal system. In this regard, it can be argued that the functional activity of hepatocyte microsomes is higher among HR rats. Given the 333 broad substrate specificity of hepatocyte microsome enzymes [9], it can be assumed that the metabolic capacity of hepatocytes is greater among HR rats. The last, in turn, is also genetically determined, which allowed among rats to identify fast and slow metabolizers [10]. Abstract During control these authors also found a higher content of total bile acids among fast- metabolizing rats. Figure 3 - Dynamics of the average ratio of individual values of bile secretion in a group of animals with correction to the average value of animal group without correction under the condition of cranioskeletal trauma modeling complicated by blood loss among high- and low-resistant animals to hypoxia. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Without blood loss Blood loss 1.28 1,50 1.19 1,29* Low resistant 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Without blood loss Blood loss 1.28 1,50 1.19 1,29* Low resistant 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Without blood loss Blood loss 1.28 1,50 1.19 1,29* Low resistant Without blood loss Figure 3 - Dynamics of the average ratio of individual values of bile secretion in a group of animals with correction to the average value of animal group without correction under the condition of cranioskeletal trauma modeling complicated by blood loss among high- and low-resistant animals to hypoxia. Figure 3 - Dynamics of the average ratio of individual values of bile secretion in a group of animals with correction to the average value of animal group without correction under the condition of cranioskeletal trauma modeling complicated by blood loss among high- and low-resistant animals to hypoxia. Under the influence of cranioskeletal trauma, the bile content of total bile acids decreased among both LR rats and HR rats. This is probably due to the damage tof the membranes of the microsomal enzyme hepatocytes by reactive oxygen species and free radicals, as it is shown in the works of other authors [11]. Despite the lower initial content of total bile acids in the bile of LR rats, under the influence of cranioskeletal trauma in this group, the degree of reduction of the studied index was significantly greater than among HR rats, indicating greater sensitivity of LR rats to pathogens of traumatic disease. which lead to the affection of the microsomal monooxygenase system of hepatocytes [12]. Under additional simulation of acute blood loss, the disorders were more visible, but the result was statistically significant only among LR rats after 3 days of the experiment, 334 indicating a greater impact of hypoxemia due to blood loss on the functional state of liver. Abstract An additional confirmation of this was the greater degree of reduction of the studied indicator among LR rats, which was statistically significant after 3 and 7 days of the experiment. It can be assumed that among LR rats, hypoxia is a key pathogenic factor of cranioskeletal trauma and blood loss, which contribute to the slowing of the hepatocytes metabolic function, in particular its bile-forming function. In order to correct the detected disorders, we used the medication Thiocetam, which, as shown by our previous studies [13], proved to be effective in the correction of lipid peroxidation disorders, antioxidant protection and cytolysis. Studies have shown that under the influence of seven-day use of Thiocetam in groups of LR and HR rats with cranioskeletal trauma disorders of the total bile acids in bile became significantly smaller compared to animals without correction. A similar result was found under conditions of additional acute blood loss. Analysis of the deviation degree of the studied indicator among animals injected with Thiocetam, compared with animals without correction, showed that the degree of increase in total bile acids in bile among LR rats was greater than among HR rats, which was statistically significant in terms of additional blood loss. Thus, the efficiency of Thiocetam among LR rats was higher compared to HR rats, which requires further preclinical studies. Conclusions: 1. The content of total bile acids in bile, as a basic indicator of bile- forming function of liver, among intact HR rats is significantly higher than among LR rats. 2. Under the influence of cranioskeletal trauma, the content of total bile acids in bile is probably reduced compared to control in the groups of LR and HR rats, but the degree of reduction is significantly greater among LR rats after 7 days of post-traumatic period. 2. Under the influence of cranioskeletal trauma, the content of total bile acids in bile is probably reduced compared to control in the groups of LR and HR rats, but the degree of reduction is significantly greater among LR rats after 7 days of post-traumatic period. 3. Complication of cranioskeletal trauma by acute blood loss causes a significantly greater decrease in the bile content of total bile acids among LR rats compared with HR rats after 3 and 7 days of the experiment. 4. Abstract Usage of Thiocetam is accompanied by a decrease in the disorders of bile-forming function of liver among LR and HR rats with cranioskeletal trauma, which is more visible among LR rats with acute blood loss. Prospects for further research. In the future, it is necessary to investigate the structural differences of liver among animals with different resistance to hypoxia with cranioskeletal trauma complicated by blood loss. References References 1. Kalinkin O.G. Traumatic disease. Trauma. 2013. T. 14, № 3. URL: http://www.mif-ua.com/archive/article/36559. 1. Kalinkin O.G. Traumatic disease. Trauma. 2013. T. 14, № 3. URL: http://www.mif-ua.com/archive/article/36559. 335 2. Kozak D.V. Peculiarities of bile-forming function of liver in the dynamics of polytrauma in the experiment. Journal of Health Sciences. 2013. Vol. 3, № 12. pp 241—256. 2. Kozak D.V. Peculiarities of bile-forming function of liver in the dynamics of polytrauma in the experiment. Journal of Health Sciences. 2013. Vol. 3, № 12. pp 241—256. 3. Preclinical studies of drugs: guidelines; edited by Corresponding Member Academy of Medical Sciences of Ukraine O.V. Stefanov. К.: Avicenna, 2001. 528 p. 3. Preclinical studies of drugs: guidelines; edited by Corresponding Member Academy of Medical Sciences of Ukraine O.V. Stefanov. К.: Avicenna, 2001. 528 p. 4. Zayets T. A., Gudyma A. A. Peculiarities of liver functional state in the early period of cranioskeletal injury combined with bleeding, and its correction by cell therapy. Journal of Health Sciences. 2014. Vol. 4, No 1. pp. 383––390. 4. Zayets T. A., Gudyma A. A. Peculiarities of liver functional state in the early period of cranioskeletal injury combined with bleeding, and its correction by cell therapy. Journal of Health Sciences. 2014. Vol. 4, No 1. pp. 383––390. 5. Saturska G.S., Bondarenko Iu. I. Peculiarities of connective tissue metabolism in experimental diffuse ischemic-necrotic cardiosclerosis among rats with different resistance to hypoxia. Bulletin of VNMU. 2014. T. 18, № 2. pp. 425––429. 5. Saturska G.S., Bondarenko Iu. I. Peculiarities of connective tissue metabolism in experimental diffuse ischemic-necrotic cardiosclerosis among rats with different resistance to hypoxia. Bulletin of VNMU. 2014. T. 18, № 2. pp. 425––429. 6. Borys R.M., Gozhenko A.I., Hudyma A.A. Dynamics of lipid peroxidation indicators during the period of early and late manifestations of traumatic disease under the conditions of experimental cranioskeletal trauma and its correction with cell therapy. Clinical and experimental pathology. 2013. T. XII, № 2 (44). pp. 31––34. 6. Borys R.M., Gozhenko A.I., Hudyma A.A. Abstract Dynamics of lipid peroxidation indicators during the period of early and late manifestations of traumatic disease under the conditions of experimental cranioskeletal trauma and its correction with cell therapy. Clinical and experimental pathology. 2013. T. XII, № 2 (44). pp. 31––34. 7. Influence of two-hour tourniquets ischemia of limb and acute blood loss on systemic disorders of the body in the reperfusion period (experimental study) / I. I. Horban, A. A. Hudyma, R. V. Maksymiv, I. V. Antonyshyn. Wiadomości Lekarskie. 2020. Vol. LXXIII, Iss. 7. pp. 1330––1333. 7. Influence of two-hour tourniquets ischemia of limb and acute blood loss on systemic disorders of the body in the reperfusion period (experimental study) / I. I. Horban, A. A. Hudyma, R. V. Maksymiv, I. V. Antonyshyn. Wiadomości Lekarskie. 2020. Vol. LXXIII, Iss. 7. pp. 1330––1333. 8. Mykhailiuk I.A., Hudyma A.A. Metabolic disorders of liver in the late period of cranioskeletal trauma and their correction by thiocetam. Medical and clinical chemistry. 2015. T. 17, № 2. pp. 56––58. 8. Mykhailiuk I.A., Hudyma A.A. Metabolic disorders of liver in the late period of cranioskeletal trauma and their correction by thiocetam. Medical and clinical chemistry. 2015. T. 17, № 2. pp. 56––58. 9. Microsomal oxidation in physiological and pathological processes / E. E. Kuznetsova, V.G. Gorokhova, A.G. Gorokhov [etc.]. Bulletin of ESSC SD RAMS. 2007. № 4 (56). pp. 170––180. 9. Microsomal oxidation in physiological and pathological processes / E. E. Kuznetsova, V.G. Gorokhova, A.G. Gorokhov [etc.]. Bulletin of ESSC SD RAMS. 2007. № 4 (56). pp. 170––180. 10. Hudyma A.A., Iarema V.V. Violation of bile formation and bile secretion in the early period of polytrauma among animals with different metabolic capacity of liver. Achievements of clinical and experimental medicine. 2012. № 2 (17). pp. 48––52. 10. Hudyma A.A., Iarema V.V. Violation of bile formation and bile secretion in the early period of polytrauma among animals with different metabolic capacity of liver. Achievements of clinical and experimental medicine. 2012. № 2 (17). pp. 48––52. 11. Zayets T. A., Gudyma A. A. Peculiarities of liver functional state in the early period of cranioskeletal injury combined with bleeding, and its correction by cell therapy. Journal of Health Sciences. 2014. Vol. 4, No 1. pp. 383––390. 12. Kozak D. V. Lipid peroxidation, antioxidant defence, endogenous intoxication and cytolysis variations in early and late manifestations of traumatic disease. Journal of Health Sciences. 2013. Vol. 3, № 4. pp. 108––117. 12. Kozak D. V. Lipid peroxidation, antioxidant defence, endogenous intoxication and cytolysis variations in early and late manifestations of traumatic disease. Journal of Health Sciences. 2013. Vol. 3, № 4. pp. 108––117. 13. Sikirynska D.O. Influence of thiocetam on the course of the early period of cranioskeletal trauma among animals with different resistance to hypoxia. Current issues of pharmacology and pharmacotherapy: materials all-Ukrainian. scientific-practical Conf., Sept. 26-27. 2019. Ternopil: TNMU, 2019. pp. 65––66. Abstract 336 337 337
https://openalex.org/W3196017536	https://www.frontiersin.org/articles/10.3389/fpsyg.2021.717380/pdf	English	null	Into the Unknown: Uncertainty and Procrastination in Students From a Life History Perspective	Frontiers in psychology	2,021	cc-by	8,118	Correspondence: Amy J. Lim Correspondence: Amy J. Lim amy.lim@murdoch.edu.au Into the Unknown: Uncertainty and Procrastination in Students From a Life History Perspective While existing ﬁndings have established an increase in procrastination among students in higher education during COVID-19, they do not elucidate how COVID-19 has effected an increase in procrastination. Drawing upon previous ﬁndings and employing a life history framework, this paper proposed that the increase in procrastination may be attributed to the heightened levels of uncertainty in the pandemic. Additionally, this paper examined life history strategy as the psychological mechanism underlying the relation between uncertainty and procrastination. By collecting data across two school semesters in a university (N = 253), we found that uncertainty and procrastination did not differ between the semester where changes were abruptly imposed to stem the spread of COVID-19 and the following semester. Our ﬁndings also revealed that uncertainty predicted procrastination, and that life history strategy mediated the relation between uncertainty and procrastination. Speciﬁcally, uncertainty was associated with a faster life history strategy, which was positively associated with procrastination. By shedding light on the psychology behind the effect of uncertainty on procrastination, the ﬁndings of this paper hold important implications for the design of procrastination interventions for the uncertain climate during the pandemic and “the new normal” post COVID-19. INTRODUCTION COVID-19 has unequivocally disrupted the academic life of students (Organization for Economic Cooperation and Development, 2020; Chaturvedi et al., 2021). Studies examining the impact of COVID-19 reported decreases in study hours and increases in academic procrastination among students in higher education (Aucejo et al., 2020; Biricik and Sivrikaya, 2020; Jia et al., 2020). Additionally, the nature of online learning from home further encourages procrastination as students not only need to exert higher levels of self-control to overcome isolated learning and the challenges of online learning (Drumm and Jong, 2020; Rasheed et al., 2020; Hong et al., 2021), they must also resist distractions present at home (e.g., television and social media, Meier et al., 2016; Pan, 2020). Taken together, these studies evidenced that procrastination was at higher than average levels during the pandemic. Specialty section: This article was submitted to Educational Psychology, a section of the journal Frontiers in Psychology Received: 30 May 2021 Accepted: 26 July 2021 Published: 18 August 2021 Received: 30 May 2021 Accepted: 26 July 2021 ORIGINAL RESEARCH published: 18 August 2021 doi: 10.3389/fpsyg.2021.717380 Keywords: COVID-19, higher education, uncertainty, procrastination, life history strategy Edited by: Sina Fackler, Leibniz Institute for Educational Trajectories (LG), Germany Reviewed by: Sebastian Franz, Leibniz Institute for Educational Trajectories (LG), Germany Antonio P. Gutierrez de Blume, Georgia Southern University, United States Correspondence: Amy J. Lim amy.lim@murdoch.edu.au Uncertainty in the Air Uncertainty was at an alarmingly elevated level when the COVID-19 pandemic was declared as a global health emergency (Ahir et al., 2019; Rettie and Daniels, 2020). The looming state of uncertainty was also undoubtedly experienced by students in higher education. Students faced concerns of becoming sick and the possibility that they may lack ﬁnancial resources to complete their studies (Biricik and Sivrikaya, 2020; Jenei et al., 2020). Additionally, with the closure of schools and the abrupt shift from physical to online learning, students faced uncertain futures of their academic and professional careers (Jenei et al., 2020). A survey of 1,500 university students revealed that students withdrew from classes, changed their majors, delayed their graduation, and expressed less conﬁdence in ﬁnding a job by graduation as a result of COVID-19 (Aucejo et al., 2020). Within this framework, procrastination, characterized by engaging in an activity that provides immediate hedonic rewards, is conceptualized as a manifestation of a fast life history strategy. As a fast life history strategy, procrastination may be the response to harsh and unpredictable environments where the likelihood of future success is low, hence serving the adaptive function of avoiding the cost of a current eﬀort when there may not be a future in which the payoﬀs can be realized (Chen and Chang, 2016). g Existing data supports this proposition. Procrastination was positively associated with a fast life history orientation (Chen and Chang, 2016). Procrastination was higher when the environment was unpredictable and likelihood of future success was low (Chen and Qu, 2017; Chen and Kruger, 2017). Furthermore, the relation between environmental unpredictability and procrastination was mediated by life history strategy; environmental unpredictability was negatively related to a slower life history strategy, and in turn, slower life history strategy was negatively related to procrastination (Chen and Qu, 2017). Collectively, these ﬁndings suggest that an uncertain environment is likely to result in procrastination by psychologically shifting one’s resource allocation strategy to a faster one that favors present gains. Procrastination, deﬁned as the deliberate delay of a planned course of action, was found to be associated with negative emotional states such as anxiety and stress (Steel, 2007; Hen and Goroshit, 2014). Citation: Lim AJ and Javadpour S (2021) Into the Unknown: Uncertainty and Procrastination in Students From a Life History Perspective. Front. Psychol. 12:717380. doi: 10.3389/fpsyg.2021.717380 With the adverse eﬀects procrastination has on academic performance and the eﬀectiveness of online learning during COVID-19 (Kim and Seo, 2015; Hong et al., 2021), attention has been called to manage and reduce the increased procrastination observed in the pandemic (Ariﬁana et al., 2020). Yet, existing studies do not elucidate how COVID-19 has eﬀected an increase August 2021 \| Volume 12 \| Article 717380 1 Frontiers in Psychology \| www.frontiersin.org Uncertainty and Procrastination Lim and Javadpour in procrastination. Without understanding how COVID-19 has introduced a preponderance of procrastination, eﬀorts to curb procrastination would be ineﬀective. Drawing upon previous ﬁndings and employing a life history framework (Griskevicius et al., 2013; Del Giudice et al., 2015; Chen and Qu, 2017), this paper proposed that the increase in procrastination may be attributed to the heightened levels of uncertainty in the pandemic. life history strategy evolved in favorable and predictable environments; as such, it promotes the preference for eﬀorts directed at building the future, such as growing human and social capital. In contrast, a faster life history strategy evolved in harsh and unpredictable environments where it is optimal to focus on the present; as such, it is characterized by the preference for eﬀorts directed at fulﬁlling immediate goals. Generally, fast life history strategists discount the future in favor of present gains (e.g., Wang et al., 2009; Griskevicius et al., 2011b). A faster life history strategy hedges against an uncertain future as individuals prioritize present survival and accelerated growth over investments for the future (Ellis et al., 2009). Uncertainty in the Air Given that the uncertain climate introduced by the pandemic has resulted in several negative psychological consequences for students, including anxiety and frustration, it is likely that increased procrastination stemmed from the uncertainty experienced by students during the pandemic (Biricik and Sivrikaya, 2020; Brooks et al., 2020; Jenei et al., 2020; Wang and Zhao, 2020; Rahimi and Vallerand, 2021). Existing ﬁndings provide initial support for this prediction as they demonstrated that environmental unpredictability was positively related to procrastination (Chen and Qu, 2017). Frontiers in Psychology \| www.frontiersin.org The Present Research Taking it all together, existing studies established that COVID-19 has resulted in a preponderance of procrastination among students in higher education (Aucejo et al., 2020; Biricik and Sivrikaya, 2020; Jia et al., 2020); yet, they do not elucidate how COVID-19 have eﬀected an increase in procrastination. Drawing upon previous ﬁndings and employing a life history framework (Griskevicius et al., 2013; Del Giudice et al., 2015; Chen and Qu, 2017), we propose that the increase in procrastination may be attributed to the heightened levels of uncertainty in the pandemic. While it is not possible to compare the diﬀerence in uncertainty before and after the occurrence of the pandemic in this paper, we examine the change in uncertainty, and its eﬀect on procrastination, across diﬀerent phases of the pandemic. Speciﬁcally, we predict that students experience elevated levels of uncertainty, and correspondingly, higher levels of procrastination during the initial phase of the pandemic than subsequent phases. To better eﬀect strategies and interventions to reduce procrastination among students in higher education, it is important to also understand the underlying psychological mechanism through which procrastination occurs. Without understanding the psychology behind the eﬀect of uncertainty on procrastination, preventing procrastination in a pandemic would be ineﬀective especially since an uncertain climate is likely to persist as people ﬁgure out “the new normal” post COVID-19. As such, this paper employed the life history theory in examining the underlying psychological mechanism behind uncertainty and procrastination. Procrastination Procrastination on academic tasks was assessed by the Pure Procrastination Scale (Steel, 2010). Participants responded to 12 statements (e.g., “I delay making decision until it’s too late” and “Even after I make a decision I delay acting upon it”) on a Materials Uncertainty The Environmental Unpredictability Scale (Davis and Werre, 2008) was employed to assess participants’ perception of environmental uncertainty. Participants were asked to rate the chances they had of attaining future outcomes in three broad aspects: resource acquisition, oﬀspring survivability, and social rank. Participants responded to items such as, “I will have a job that pays well,” “Life will turn out better for me than it has for my parents,” and “I will have a happy family life” for each aspect, respectively. Participants rated a total of 12 items on a 5-point scale (1 = very low chance, 5 = very high chance). The scores for each item was reversed coded and averaged to form a composite score for environmental uncertainty, where higher scores indicated higher perceived environmental uncertainty (M = 2.72, SD = 0.66, αresourceacquisition = 0.88, αoﬀspringsurvivability = 0.75, αsocialrank = 0.73, αoverall = 0.88, ω = 0.75). Additionally, we further propose life history strategy as the psychological mechanism through which procrastination occurs: uncertainty in the pandemic climate psychologically shifts resource allocation strategies such that it is more optimal to favor present gains, which consequently results in procrastination. Speciﬁcally, we predict that uncertainty is associated with a faster life history strategy, which in turn, is positively related to procrastination. In doing so, we hope to shed light on the psychology behind the eﬀect of uncertainty on procrastination, which is imperative for the design of procrastination interventions for the pandemic and “the new normal” post COVID-19. Life History Theory and Procrastination Collectively, these changes are likely to produce heightened levels of uncertainty among students during this school semester (February to May 2020). SDage = 7.30). One hundred and forty six participants were recruited in Time 1 (118 females, Mage = 24.03, SDage = 7.67) and 107 participants were recruited in Time 2 (83 females, Mage = 22.14, SDage = 6.65). All participants were undergraduate students with the majority in their ﬁrst year (NTime 1 = 52, NTime 2 = 62). Participants were only allowed to participate in this study once; participants who have responded to the survey in Time 1 did not participate in the survey conducted at Time 2. Procedure After providing informed consent, participants responded to a series of questionnaires assessing uncertainty, life history strategy, and procrastination. For the survey in Time 1, participants were instructed to respond to the questions with reference to the period of time between March and May 2020 when COVID-19 cases spiked, and several measures were introduced in Western Australia and the university to curb the spread of COVID-19. For the survey in Time 2, participants were instructed to respond to the questions with reference to the period of time between August and November 2020 when COVID-19 was more managed. Participants provided demographic information before completing the study. COVID-19 restrictions started to ease in Western Australia by June 2020 (Government of Western Australia, 2020b). Social and recreational activities can resume with social distancing measures (limited number of people, two square meter per person capacity). Moreover, before the commencement of the following semester commenced, the university of this sample in which the paper was based on, announced that classes in the following semester were to be the same as the semester before. That is, classes and examinations were to take place online. Given student’s prior experience with online learning, and the improving COVID-19 situation, students are likely to experience less uncertainty in the following semester (August to November 2020) compared to the semester before. MATERIALS AND METHODS The life history strategy adopted by an individual was assessed with the Life History Strategy Short-Form scale (Figueredo et al., 2006). Participants responded to 20 statements (e.g., “I often make plans in advance” and “I avoid taking risks”) using a 7-point Likert scale (1 = “strongly disagree” to 7 = “strongly agree”). The items were reverse-scored and averaged to form a composite score; higher scores indicated the adoption of a faster life history strategy (M = 3.10, SD = 0.71, α = 0.77). To examine the change in uncertainty, and its eﬀect on procrastination, across diﬀerent phases of the pandemic, data was collected across two time periods for this study. The ﬁrst period (Time 1) was during the school semester in 2020, between August and November 2020 when COVID-19 surfaced and several changes were abruptly imposed, and the second period (Time 2) during the following school semester in 2021, between February and May 2021. Life History Theory and Procrastination Life History Theory and Procrastination The life history theory posits that organisms allocate limited resources, energy, and time based on environmental constrains (Del Giudice et al., 2015). Life history strategies exist on a slow-fast continuum and are determined by the most optimal allocation of resources, energy, and time between somatic and reproductive eﬀort in response to environmental conditions (Griskevicius et al., 2013; Del Giudice et al., 2015). A slower For this paper, we examine the perceptions of students from a university based in Western Australia across two semesters. Following a declared state of emergency, social distancing restrictions began in Western Australia in early March 2020 August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 2 Uncertainty and Procrastination Lim and Javadpour (Government of Western Australia, 2020a). With the exception of essential services, public spaces were closed and gatherings were only limited to two people (GardaWorld, 2020a,b). Even though universities were exempt from the restrictions—as they fall under essential services, Western Australian universities prepared for closures of physical campuses and swiftly transited to online teaching and learning services (Pilat, 2020). At the university in which the sample of this paper was based in, these measures were implemented in the middle of a school semester. Classes were suspended for 2 weeks to allow for staﬀand students to prepare to transit to an online learning environment. Timetables had to be rescheduled to accommodate for such changes. However, classes that required the use of labs and clinics were under review. Collectively, these changes are likely to produce heightened levels of uncertainty among students during this school semester (February to May 2020). (Government of Western Australia, 2020a). With the exception of essential services, public spaces were closed and gatherings were only limited to two people (GardaWorld, 2020a,b). Even though universities were exempt from the restrictions—as they fall under essential services, Western Australian universities prepared for closures of physical campuses and swiftly transited to online teaching and learning services (Pilat, 2020). At the university in which the sample of this paper was based in, these measures were implemented in the middle of a school semester. Classes were suspended for 2 weeks to allow for staﬀand students to prepare to transit to an online learning environment. Timetables had to be rescheduled to accommodate for such changes. However, classes that required the use of labs and clinics were under review. 2A similar regression analysis was conducted with the exclusion of univariate and multivariate outliers. The analysis excluding the outliers revealed similar ﬁnding. The model accounted for a signiﬁcant 9.3% of the variability in procrastination, R2 = 0.093, adjusted R2 = 0.074, F(5, 239) = 4.89, p < 0.01. The analysis revealed that environmental uncertainty predicted procrastination, B = 0.33, t(239) = 3.99, p < 0.01, 95% CI [0.17, 0.49]. The signiﬁcance of the eﬀect of uncertainty on procrastination remained. 1The school year participants were in were coded as 1 = ﬁrst years and 0 = non- ﬁrst-years. Participants A total of 253 participants were recruited through an Australian university’s subject pool system (201 females, Mage = 23.23, August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 3 Uncertainty and Procrastination Lim and Javadpour TABLE 1 \| Descriptive statistics of all variables. Variables Time 1 (N = 146) Time 2 (N = 107) M SD Skew Kurtosis M SD Skew Kurtosis Environmental uncertainty 2.77 0.68 0.31 0.05 2.65 0.63 0.46 0.99 Life history strategy 3.00 0.68 0.60 −0.20 3.22 0.73 0.43 −0.39 Procrastination 3.21 0.82 −0.06 −0.88 3.19 0.85 −0.35 −0.15 TABLE 2 \| Intercorrelations of all variables. Variables 1. 2. 3. 1. Environmental uncertainty − 2. Life history strategy 0.32* − 3. Procrastination 0.26** 0.21* − p < 0.01; p < 0.001. TABLE 1 \| Descriptive statistics of all variables. 5-point scale (1 = very seldom true of me, 5 = very often true of me) (M = 3.21, SD = 0.83, α = 0.90). 5-point scale (1 = very seldom true of me, 5 = very often true of me) (M = 3.21, SD = 0.83, α = 0.90). ANALYTICAL STRATEGY Statistical analyses were performed using the software package SPSS Statistics for Windows, version 26.0 (IBM Corporation, 2019). Descriptive statistics were provided for environmental unpredictability, life history strategy, and procrastination. Before proceeding to the main analyses, the assumption of normality was checked. Values for skewness and kurtosis for all variables were between −1 and + 1, which were acceptable standards for a normal distribution (George and Mallery, 2010). This indicates that parametric tests can be employed in the subsequent analyses. Independent t-tests were conducted to compare the diﬀerences in means of environmental uncertainty and procrastination between two phases of the pandemic (i.e., Time 1 versus Time 2). TABLE 2 \| Intercorrelations of all variables. Variables 1. 2. 3. 1. Environmental uncertainty − 2. Life history strategy 0.32* − 3. Procrastination 0.26** 0.21* − p < 0.01; p < 0.001. correlated with procrastination. Additionally, life history strategy was negatively correlated with procrastination. correlated with procrastination. Additionally, life history strategy was negatively correlated with procrastination. A multiple regression analysis was conducted to investigate if environmental uncertainty predicted procrastination. Gender, age, the school year in which participants were in,1 and the time at which the survey was conducted were also included in the regression model as control variables. Assumptions for regression analyses were also evaluated before the interpretation of its results. Inspection of the normal probability plot of standardized residuals and the scatterplot of standardized residuals against standardized predicted values indicated that the assumptions of normality, linearity, and homoscedasticity of residuals were met. Relatively high tolerances of all predictors in the regression model (Tolerance values were between 0.86 and 0.99) indicated that multicollinearity is not an issue. Examination of boxplots indicated the presence of 3 univariate outliers for environmental uncertainty. Mahalanobis distance exceeded the critical χ2 for df = 5 (at α = 0.01) of 15.09 for 6 cases in the data, indicating the presence of multivariate outliers. Multiple regression analyses were conducted with and without these univariate and multivariate outliers. Main Analyses An independent t-test was conducted to examine if environmental uncertainty was higher in Time 1 than in Time 2. Results indicated that environmental uncertainty was higher in Time 1 (M = 2.77, SD = 0.68) than in Time 2 (M = 2.65, SD = 0.63); however, this diﬀerence was not signiﬁcant, t(251) = 1.40, p = 0.16, d = 0.18. An independent samples t-test was conducted to examine if procrastination was higher in Time 1 than in Time 2. The analysis yielded no signiﬁcant diﬀerence in procrastination between the two time periods, t(251) = 0.20, p = 0.84, d = 0.02. It is worth noting that procrastination was higher in Time 1 (M = 3.21, SD = 0.82) than in Time 2 (M = 3.19, SD = 0.85). A multiple regression analysis was conducted to examine the eﬀect of environmental uncertainty on procrastination. Results showed that the model accounted for a signiﬁcant 9.2% of the variability in procrastination, R2 = 0.092, adjusted R2 = 0.074, F(5, 247) = 5.03, p < 0.01, f2 = 0.10. The analysis revealed that environmental uncertainty predicted procrastination, B = 0.32, t(247) = 4.15, p < 0.01, 95% CI [0.17, 0.47].2 Unstandardized (B) and standardized (β) regression coeﬃcients for each predictor are reported in Table 3. A sensitivity analysis conducted using G-Power indicated that given a total sample size of 253, the minimum eﬀect size to detect a power of 0.80 at α = 0.05 is f2 = 0.04 for this study. Finally, a mediation analysis using PROCESS version 3.1 (Hayes, 2018) was performed to examine if life history strategy mediated the relation between environmental uncertainty and procrastination. For mediation to be demonstrated, the bootstrap conﬁdence interval of the indirect eﬀect (path a∗b) must not include zero (bootstrap samples = 5,000) (Hayes, 2018). Similarly, mediation analyses were conducted with and without the univariate and multivariate outliers. To examine if participants’ life history strategy mediated the relationship between perceived uncertainty and procrastination, a mediation analysis using Hayes’ PROCESS model 4 was conducted (Hayes, 2018). Environmental uncertainty was included as the independent variable, procrastination as the RESULTS Descriptive Analyses 3A similar mediation analysis was conducted with the exclusion of univariate and multivariate outliers. The analysis excluding the outliers revealed similar ﬁndings: Eenvironmental uncertainty was positively associated with life history strategy, B = 0.37, p < 0.01, 95% CI [0.22, 0.52]. Life history strategy was positively associated with procrastination, B = 0.19, p = 0.02, 95% CI [0.03, 0.35]. Perceived uncertainty was positively associated with procrastination via participants’ life history strategy, B = 0.07, 95% CI = [0.006, 0.14], indicating that life history strategy mediated the relation between environmental uncertainty and procrastination. Descriptive Analyses Participants with a faster life history strategy were more likely to procrastinate. Finally, results indicated that perceived uncertainty was positively associated with procrastination via participants’ life history strategy, B = 0.07, 95% CI = [0.003, 0.14], thus demonstrating the mediation eﬀect of life history strategy on the relation between perceived uncertainty and procrastination.3 Unstandardized (B) regression coeﬃcients, 95% conﬁdence intervals, and R2 values for the mediation model are presented in Table 4. dependent variable, and life history strategy as the mediator. The time at which the survey was conducted was included as a covariate. Gender, age, and the school year participants were in were also included as covariates. Results revealed that environmental uncertainty was positively associated with life history strategy, B = 0.38, p < 0.01, 95% CI [0.26, 0.51], where higher perceived uncertainty predicted the adoption of faster life history strategies. Next, results also revealed that life history strategy was positively associated with procrastination, B = 0.17, p = 0.03, 95% CI [0.02, 0.32]. Participants with a faster life history strategy were more likely to procrastinate. Finally, results indicated that perceived uncertainty was positively associated with procrastination via participants’ life history strategy, B = 0.07, 95% CI = [0.003, 0.14], thus demonstrating the mediation eﬀect of life history strategy on the relation between perceived uncertainty and procrastination.3 Unstandardized (B) regression coeﬃcients, 95% conﬁdence intervals, and R2 values for the mediation model are presented in Table 4. procrastination were similar between both semesters. We also found environmental uncertainty predicted procrastination. Furthermore, our results revealed that life history orientation mediated the relation between uncertainty and procrastination, suggesting that environmental uncertainty psychologically shifted the resource allocation strategies of students to a faster one such that it was more optimal to favor present gains, which consequently predicted procrastination. q y p p Our ﬁndings showed that uncertainty was higher in Time 1 than in Time 2, even though this diﬀerence was not signiﬁcant, it suggests that perceived uncertainty was especially heightened at the time when several changes were eﬀected to curb the spread of COVID-19. This is consistent with anecdotal evidence that reported the uncertain futures students face relating to their education and professional careers (Aucejo et al., 2020; Jenei et al., 2020). Descriptive Analyses The lack of signiﬁcance in the diﬀerence in perceived uncertainty between the two time periods is likely due to the evolving nature of COVID-19 (e.g., new strain and sudden lockdowns due to new clusters of infected cases). As such, even though students may have accustomed to the changes made to their curriculum (e.g., online learning), which may lower uncertainty, being on a constant lookout for abrupt changes regarding the pandemic may keep uncertainty at relatively high levels for students. Moreover, individuals diﬀer in their sensitivity toward uncertainty. Intolerance to uncertainty is the tendency to perceive and interpret uncertain situations as aversive and stressful (Dugas et al., 2004). Students’ intolerance to uncertainty may have inﬂuenced their perceptions of uncertainty during the pandemic, such that those with lower intolerance to uncertainty would have perceived less uncertainty than those with higher intolerance to uncertainty. Such individual diﬀerence could contribute to the similar levels of uncertainty perceived between the two time periods. Descriptive Analyses Table 1 displays the means, standard deviations, skewness, and kurtosis of all the variables involved in this study. Table 2 displays the intercorrelations of the variables. Environmental uncertainty was negatively correlated with life history strategy and positively August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 4 Uncertainty and Procrastination Lim and Javadpour TABLE 4 \| Mediation model coefﬁcients for environmental uncertainty, life history strategy, gender, age, school year, time, and procrastination (N = 253). Variables B LLCI ULCI se DV = life history strategy (R2 = 0.18, p < 0.01) Constant 1.59 1.03 2.15 0.28 Environmental uncertainty 0.38 0.26 0.51 0.06 Gender 0.34* 0.14 0.54 0.10 Age 0.002 −0.01 0.01 0.01 School year −0.03 −0.20 0.14 0.09 Time 0.26* 0.09 0.43 0.09 DV = procrastination (R2 = 0.11, p < 0.01) Constant 2.07 1.35 2.79 0.37 Environmental uncertainty 0.25* 0.09 0.42 0.08 Life history strategy 0.17 0.02 0.32 0.08 Gender 0.06 −0.19 0.31 0.13 Age −0.004 −0.02 0.01 0.01 School year 0.23 0.02 0.44 0.11 Time −0.09 −0.30 0.12 0.11 p < 0.05. p < 0.01. TABLE 3 \| Unstandardized (B) and standardized (β) regression coefﬁcients for predictors in regression model predicting procrastination. TABLE 4 \| Mediation model coefﬁcients for environmental uncertainty, life history strategy, gender, age, school year, time, and procrastination (N = 253). TABLE 3 \| Unstandardized (B) and standardized (β) regression coefﬁcients for predictors in regression model predicting procrastination. TABLE 3 \| Unstandardized (B) and standardized (β) regression coefﬁcients for predictors in regression model predicting procrastination. Variables 95% CI B LL UL se β Constant 2.34 1.65 3.03 0.35 Environmental uncertainty 0.32 0.17 0.47 0.08 0.26 Gender 0.12 −0.12 0.37 0.12 0.06 Age −0.003 −0.02 0.01 0.01 −0.03 School year 0.23 0.01 0.44 0.11 0.14 Time −0.04 −0.25 0.16 0.11 −0.03 p < 0.05; *p < 0.01. dependent variable, and life history strategy as the mediator. The time at which the survey was conducted was included as a covariate. Gender, age, and the school year participants were in were also included as covariates. Results revealed that environmental uncertainty was positively associated with life history strategy, B = 0.38, p < 0.01, 95% CI [0.26, 0.51], where higher perceived uncertainty predicted the adoption of faster life history strategies. Next, results also revealed that life history strategy was positively associated with procrastination, B = 0.17, p = 0.03, 95% CI [0.02, 0.32]. DISCUSSION This paper sought to elucidate how COVID-19 has eﬀected an increase in procrastination. We proposed that the increase in procrastination may be attributed to the heightened levels of uncertainty in the pandemic. Moreover, we also examined the underlying psychological mechanism for how an uncertain climate drives procrastination. Speciﬁcally, we investigated life history strategy as the psychological mechanism through which procrastination occurs. Data was collected from undergraduate students across two time periods. Data collected at Time 1 assessed the perceptions of students in the semester where changes were abruptly imposed to stem the spread of COVID-19. Data collected at Time 2 assessed the perceptions of students one semester after changes to the curriculum were made. Our ﬁndings showed that environmental uncertainty and Our ﬁndings also demonstrated that procrastination was higher in Time 1 than in Time 2, though the diﬀerence August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 5 Uncertainty and Procrastination Lim and Javadpour in procrastination was not signiﬁcant between the two time periods. While our results cannot conclude that procrastination levels were higher during the pandemic than before (Aucejo et al., 2020; Biricik and Sivrikaya, 2020; Jia et al., 2020), our ﬁnding showed that procrastination was higher during the initial phases of the pandemic than in the subsequent phases, suggesting that sudden changes brought about by COVID-19 played a part in encouraging procrastination. The lack of signiﬁcant diﬀerence in procrastination may be an artifact of the similarity in environmental uncertainty levels. Additionally, as procrastination is often engaged to cope with negative emotional states (Hen and Goroshit, 2014), and elevated negative emotional states were also constantly reported during the pandemic (Biricik and Sivrikaya, 2020; Jenei et al., 2020; Rahimi and Vallerand, 2021), it is likely the aﬀective state of participants contributed to similar levels of procrastination observed between the time two periods. a faster life history strategy tendency as a function of their early life low socioeconomic environments are sensitive to environmental stressors and likely to discount the future in favor of present gains (Boyce and Ellis, 2005; Griskevicius et al., 2011b). As such, it is likely that these individuals will be more sensitive to the uncertain climate during the pandemic and hence, are more responsive in adopting a faster life history strategy than individuals who grew up in high socioeconomic environments (and have developed a tendency to adopt slower life history strategies). Limitations and Future Directions Limitations and Future Directions Our work is far from conclusive and poses questions for future work. As we have discussed previously, childhood socioeconomic status can potentially inﬂuence one’s sensitivity to environment stresses and shifts in life history strategy. Given that we only measured participants’ life history strategy based on their current environment than their early life environment, and childhood socioeconomic status was not assessed in this study, we are not able to ascertain the extent to which participants’ childhood socioeconomic status aﬀected participants’ reaction to heightened levels of uncertainty during the pandemic. Future studies should assess participants’ childhood socioeconomic status and test for a moderated mediation model, where life history strategy mediates the relation between uncertainty and procrastination and childhood socioeconomic status moderates the shift in life history orientation in response to environmental uncertainty. We also found that uncertainty predicted procrastination, and that life history orientation mediated the relation between uncertainty and procrastination, which is consistent with the life history framework (Chen and Qu, 2017; Chen and Kruger, 2017). In line with this theoretical framework, our results demonstrated that environmental uncertainty predicted a faster life history strategy, which signaled that it was more optimal to favor present gains than future ones, consequently predicting procrastination. Our ﬁndings also provided further support to the conceptualization of procrastination as an adaptive response to environmental conditions—speciﬁcally, to environmental unpredictability (Chen and Qu, 2017; Chen and Kruger, 2017). Moreover, consistent with previous ﬁndings, our ﬁndings also showed that procrastination was associated with a faster life strategy, suggesting that procrastination is more likely when present gains are favored (Chen and Chang, 2016; Chen and Qu, 2017). This also corroborates ﬁndings that demonstrated the association procrastination and negative attitudes of academic investment have with reductions in future time perspective and future outlook (Ferrari and Díaz-Morales, 2007; Schechter and Francis, 2010). Additionally, we did not account for personality traits that may inﬂuence one’s tendency to procrastinate. The tendency to procrastinate has also been associated with personality traits, speciﬁcally, conscientiousness and neuroticism. Conscientiousness was inversely correlated to procrastination (Johnson and Bloom, 1995; Van Eerde, 2003; Steel, 2007). Neuroticism was also signiﬁcantly correlated to procrastination (Johnson and Bloom, 1995). With these personality traits predicting procrastination, it would be diﬃcult to tease apart the unique eﬀects of environmental uncertainty from the eﬀects of these traits on procrastination. Frontiers in Psychology \| www.frontiersin.org DISCUSSION This diﬀerence in response to environmental uncertainty could explain the small mediation eﬀect observed. CONCLUSION self-reported measures of environmental uncertainty and life history strategy. This paper demonstrated that levels of uncertainty and procrastination in undergraduate students were similar between the semester where COVID-19 changes are abruptly imposed to stem the spread of COVID-19 and the following semester. Employing an evolutionary life history framework, this paper found that uncertainty predicted procrastination. This paper also provided an underlying explanation for how an uncertain climate drives procrastination. Speciﬁcally, the ﬁndings of this paper showed that uncertainty in the current pandemic prompted students to psychologically shift their life history strategy such that it was more optimal to focus on present gains, which consequently predicted procrastination. This paper demonstrated that levels of uncertainty and procrastination in undergraduate students were similar between the semester where COVID-19 changes are abruptly imposed to stem the spread of COVID-19 and the following semester. Employing an evolutionary life history framework, this paper found that uncertainty predicted procrastination. This paper also provided an underlying explanation for how an uncertain climate drives procrastination. Speciﬁcally, the ﬁndings of this paper showed that uncertainty in the current pandemic prompted students to psychologically shift their life history strategy such that it was more optimal to focus on present gains, which consequently predicted procrastination. Furthermore, as participants often evaluate their behaviors negatively when asked to think about them retrospectively, self-reported measures of procrastination may not accurately reﬂect actual procrastination (Steel et al., 2001; Moon and Illingworth, 2005). Thus, it is likely that procrastination scores were inﬂated—as an artifact of using a self-report measure—and not truly reﬂective of participants’ actual procrastination in this study. To overcome this shortcoming, future studies should consider employing observed measures for assessing actual procrastination. p With the adverse eﬀect procrastination has on academic performance (Kim and Seo, 2015), it is important that procrastination is managed. Given that an emphasis on present gains predicts procrastination, one way to ameliorate procrastination in students would be to shift the emphasis to the future by boosting their perceived value of the future. This can be done by making the future self a salient concept in students as students have found that a future self can motivate action. An event-related fMRI study found that future self-continuity reduced temporal distancing—the extent to which individuals distinguish between the present self and future self; when individuals perceived their future-self more clearly, they made better decisions for their future (Ersner-Hershﬁeld et al., 2009). CONCLUSION Increasing the congruence between the present and future self was found to generate motivation for current action (Lewis and Oyserman, 2015). Self-focused mental imagery can be used as a psychological tool to bridge the present-future gap to reduce procrastination (Blouin-Hudon and Pychyl, 2017). As an uncertain climate is likely to persist, eliminating procrastination by reducing uncertainty may not necessarily be an ineﬀective strategy. Having identiﬁed life history orientation as the psychological mechanism behind uncertainty and procrastination, future studies can examine the feasibility in shifting students’ emphasis on the future to minimize procrastination. ETHICS STATEMENT The present study has been approved by the Murdoch University Research Ethics with the following approval reference number: 2020/145. The patients/participants provided their written informed consent to participate in this study. AUTHOR CONTRIBUTIONS AL and SJ were involved in the conceptualization of the project and formulation the study design. AL acquired and analyzed the data and drafted the manuscript. Both authors read and approved the ﬁnal manuscript, agreed to be accountable for the content of this manuscript. DATA AVAILABILITY STATEMENT The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. Limitations and Future Directions Although we found life history strategy to mediate the association between uncertainty and procrastination, the mediation eﬀect is relatively small. This suggests that there could be factors that inﬂuenced the mediating eﬀect of life history strategy—one factor could be an individual’s childhood socioeconomic status. Within life history theory, an individual’s early life environment determines the life history strategy they adopt (Del Giudice, 2009; Ellis et al., 2009; Griskevicius et al., 2011a). Individuals who grew up in low socioeconomic environments were more likely to be exposed to environmental stressors, such as ﬂuctuating resource availability and changing household memberships, which prompts them to adopt faster life history strategies compared to individuals who grew up in high socioeconomic environments (where environmental stressors were largely absent) (Belsky, 2007). This early exposure to environmental stressors not only aﬀects the life history strategy one adopts during their childhood, it also aﬀects how individuals respond to environmental stressors later in life (Caretta et al., 1995). Individuals who have developed We recognize that our study is limited to self-reported measures, which may limit interpretation of our ﬁndings, especially since the internal consistency of the measures assessing environmental uncertainty (α socialrank = 0.73) and life history strategy (α = 0.77) were relatively lower, though still within acceptable range (Cortina, 1993), than the rest. Given that uncertainty have been associated with physiological changes, such as heart rate (Averill et al., 1972; Monat et al., 1972), future studies can consider measuring for physiological changes on top of self-reported perceived environmental uncertainty. Life history strategy can also be assessed behaviorally by observing how they interact with others—fast life history strategists tend to express criticism and talk with physical animations (Sherman et al., 2013). Hence, future studies can consider employing these other means of assessment to complement the August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 6 Uncertainty and Procrastination Lim and Javadpour REFERENCES Blouin-Hudon, E. M. C., and Pychyl, T. A. (2017). A mental imagery intervention to increase future self-continuity and reduce procrastination. Appl. Psychol. 66, 326–352. doi: 10.1111/apps.12088 Ahir, H., Bloom, N., and Furceri, D. (2019). The World Uncertainty Index. Rochester, NY: SSRN. Boyce, W. T., and Ellis, B. J. (2005). Biological sensitivity to context: I. an evolutionary–developmental theory of the origins and functions of stress reactivity. Dev. Psychopathol. 17, 271–301. doi: 10.1017/S09545794050 50145 Ariﬁana, I. Y., Rahmawati, H., Hanurawan, F., and Eva, N. (2020). Stop academic procrastination during Covid 19: academic procrastination reduces subjective well-being. KnE Soc. Sci. 4, 312–325. Aucejo, E. M., French, J., Araya, M. P. U., and Zafar, B. (2020). The impact of COVID-19 on student experiences and expectations: evidence from a survey. J. Public Econ. 191:104271. doi: 10.1016/j.jpubeco.2020.10 4271 Brooks, S. K., Webster, R. K., Smith, L. E., Woodland, L., Wessely, S., Greenberg, N., et al. (2020). The psychological impact of quarantine and how to reduce it: rapid review of the evidence. Lancet 395, 912–920. doi: 10.1016/S0140-6736(20) 30460-8 Averill, J. R., Olbrich, E., and Lazarus, R. S. (1972). Personality correlates of diﬀerential responsiveness to direct and vicarious threat: a failure to replicate previous ﬁndings. J. Pers. Soc. Psychol. 21, 25–29. doi: 10.1037/h003 1948 Caretta, C. M., Caretta, A., and Cavaggioni, A. (1995). Pheromonally accelerated puberty is enhanced by previous experience of the same stimulus. Physiol. Behav. 57, 901–903. doi: 10.1016/00319384(94)00344-5 Chaturvedi, K., Vishwakarma, D. K., and Singh, N. (2021). COVID-19 and its impact on education, social life and mental health of students: a survey. Child. Youth Serv. Rev. 121:105866. doi: 10.1016/j.childyouth.2020.105866 Belsky, J. (2007). “Childhood experiences and reproductive strategies,” in Oxford Handbook of Evolutionary Psychology, eds L. Barrett and R. Dunbar (New York, NY: Oxford University Press), 237–254. Chen, B. B., and Chang, L. (2016). Procrastination as a fast life history strategy. Evol. Psychol. 14, 1474704916630314. Biricik, Y. S., and Sivrikaya, M. H. (2020). COVID-19 fear in sports sciences students and its eﬀect on academic procrastination behavior. Int. J. Appl. Exerc. Physiol. 9, 50–56. Chen, B. B., and Kruger, D. (2017). Future orientation as a mediator between perceived environmental cues in likelihood of future success and August 2021 \| Volume 12 \| Article 717380 7 Frontiers in Psychology \| www.frontiersin.org Lim and Javadpour Uncertainty and Procrastination Hen, M., and Goroshit, M. (2014). REFERENCES Academic procrastination, emotional intelligence, academic self-eﬃcacy, and GPA: a comparison between students with and without learning disabilities. J. Learn. Disabil. 47, 116–124. doi: 10.1177/002221941243 9325 procrastination. Pers. Individ. Diﬀer. 108, 128–132. doi: 10.1016/j.paid.2016.12. 017 procrastination. Pers. Individ. Diﬀer. 108, 128–132. doi: 10.1016/j.paid.2016.12. 017 Chen, B. B., and Qu, W. (2017). Life history strategies and procrastination: the role of environmental unpredictability. Pers. Individ. Diﬀer. 117, 23–29. doi: 10.1016/j.paid.2017.05.036 Cortina, J. M. (1993). What is coeﬃcient alpha? An examination of theory and applications. J. Appl. Psychol. 78, 98–104. doi: 10.1037/0021-9010.7 8.1.98 Hong, J. C., Lee, Y. F., and Ye, J. H. (2021). Procrastination predicts online self- regulated learning and online learning ineﬀectiveness during the coronavirus lockdown. Pers. Individ. Diﬀer. 174:110673. doi: 10.1016/j.paid.2021.11 0673 Davis, J., and Werre, D. (2008). A longitudinal study of the eﬀects of uncertainty on reproductive behaviors. Hum. Nat. 19, 426–452. doi: 10.1007/s12110-008- 9052-2 IBM Corporation (2019). IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corporation. doi: 10.1016/j.paid.2021.110673 Del Giudice, M. (2009). Sex, attachment, and the development of reproductive strategies. Behav. Brain Sci. 32, 1–67. doi: 10.1017/S0140525X0900 0016 Jenei, K., Cassidy-Matthews, C., Virk, P., Lulie, B., and Closson, K. (2020). Challenges and opportunities for graduate students in public health during the COVID-19 pandemic. Can. J. Public Health 111, 408–409. doi: 10.17269/ s41997-020-00349-8 Del Giudice, M., Gangestad, S. W., and Kaplan, H. S. (2015). “Life history theory and evolutionary psychology,” in The Handbook of Evolutionary Psychology, ed. D. M. Buss (New York, NY: Wiley). doi: 10.1002/9781119125563.evpsy ch102 Jia, J., Jiang, Q., and Lin, X. H. (2020). Academic Anxiety and Self-Handicapping Among Medical Students During the COVID-19 Pandemic: A Moderated Mediation Model. Durham: Research Square. Drumm, B. T., and Jong, A. S. (2020). A semester like no other: a student and lecturer perspective on the impact of Covid-19 on 3rd level academic life. All Ireland J. High. Educ. 12, 1–14. Johnson, J. L., and Bloom, A. M. (1995). An analysis of the contribution of the ﬁve factors of personality to variance in academic procrastination. Pers. Individ. Diﬀer. 18, 127–133. doi: 10.1016/0191-8869(94)00 109-6 Dugas, M. J., Buhr, K., and Ladouceur, R. (2004). “The role of intolerance of uncertainty in etiology and maintenance,” in Generalized Anxiety Disorder: Advances in Research and Practice, eds R. G. Heimberg, C. L. Turk, and D. S. Mennin (New York, NY: Guilford Press), 143–163. Kim, K. R., and Seo, E. H. (2015). REFERENCES The relationship between procrastination and academic performance: a meta-analysis. Pers. Individ. Diﬀer. 82, 26–33. doi: 10.1016/j.paid.2015.02.038 Lewis, N. A. Jr., and Oyserman, D. (2015). When does the future begin? Time metrics matter, connecting present and future selves. Psychol. Sci. 26, 816–825. doi: 10.1177/0956797615572231 Ellis, B. J., Figueredo, A. J., Brumbach, B. H., and Schlomer, G. L. (2009). Fundamental dimensions of environmental risk. Hum. Nat. 20, 204–268. Meier, A., Reinecke, L., and Meltzer, C. E. (2016). “Facebocrastination”? Predictors of using Facebook for procrastination and its eﬀects on students’ well-being. Comput. Hum. Behav. 64, 65–76. doi: 10.1016/j.chb.2016.0 6.011 Ersner-Hershﬁeld, H., Wimmer, G. E., and Knutson, B. (2009). Saving for the future self: neural measures of future self-continuity predict temporal discounting. Soc. Cogn. Aﬀect. Neurosci. 4, 85–92. doi: 10.1093/scan/ns n042 Monat, A., Averill, J. R., and Lazarus, R. S. (1972). Anticipatory stress and coping reactions under various conditions of uncertainty. J. Pers. Soc. Psychol. 24, 237–253. doi: 10.1037/h0033297 Ferrari, J. R., and Díaz-Morales, J. F. (2007). Procrastination: diﬀerent time orientations reﬂect diﬀerent motives. J. Res. Pers. 41, 707–714. doi: 10.1016/ j.jrp.2006.06.006 Moon, S. M., and Illingworth, A. J. (2005). Exploring the dynamic nature of procrastination: a latent growth curve analysis of academic procrastination. Pers. Individ. Diﬀer. 38, 297–309. doi: 10.1016/j.paid.2004.0 4.009 Figueredo, A. J., Vásquez, G., Brumbach, B. H., Schneider, S. M., Sefcek, J. A., Tal, I. R., et al. (2006). Consilience and life history theory: from genes to brain to reproductive strategy. Dev. Rev. 26, 243–275. doi: 10.1016/j.dr.2006.0 2.002 Organization for Economic Cooperation and Development [OECD] (2020). Education at a Glance 2020. Paris: Organization for Economic Cooperation and Development. GardaWorld (2020a). Australia: Government Announces Lockdown Extension April 16. Montreal: GardaWorld. GardaWorld (2020b). Australia: Non-Essential Services Suspended and Public Spaces Close March 23. Montreal: GardaWorld. Pan, H. (2020). A glimpse of university students’ family life amidst the COVID-19 virus. J. Loss Trauma 25, 594–597. doi: 10.1080/15325024.2020.175 0194 George, D., and Mallery, M. (2010). SPSS for Windows Step by Step: A Simple Guide and Reference, 17.0 Update (10th Edn.). Boston, MA: Pearson. Pilat, L. (2020). WA Unis Move to Online Learning as Social Distancing Measures Heighten. East Perth, WA: WAtoday. Government of Western Australia (2020a). Coronavirus: COVID-19 State of Emergency - Message from the State Emergency Coordinator. West Perth, WA: Government of Western Australia. Rahimi, S., and Vallerand, R. J. (2021). REFERENCES The role of passion and emotions in academic procrastination during a pandemic (COVID- 19). Pers. Individ. Diﬀer. 179:110852. doi: 10.1016/j.paid.2021.11 0852 Government of Western Australia (2020b). Phase 3 of COVID-19 Roadmap to Commence from Saturday, June 6. West Perth, WA: Government of Western Australia Rasheed, R. A., Kamsin, A., and Abdullah, N. A. (2020). Challenges in the online component of blended learning: a systematic review. Comput. Educ. 144:103701. doi: 10.1016/j.compedu.2019.10 3701 Griskevicius, V., Ackerman, J. M., Cantú, S. M., Delton, A. W., Robertson, T. E., Simpson, J. A., et al. (2013). When the economy falters, do people spend or save? Responses to resource scarcity depend on childhood environments. Psychol. Sci. 24, 197–205. doi: 10.1177/095679761245 1471 Rettie, H., and Daniels, J. (2020). Coping and tolerance of uncertainty: predictors and mediators of mental health during the COVID- 19 pandemic. Am. Psychol. 76, 427–437. doi: 10.1037/amp000 0710 Griskevicius, V., Delton, A. W., Robertson, T. E., and Tybur, J. M. (2011a). Environmental contingency in life history strategies: the inﬂuence of mortality and socioeconomic status on reproductive timing. J. Pers. Soc. Psychol. 100, 241–254. doi: 10.1037/a0021082 Schechter, D. E., and Francis, C. M. (2010). A life history approach to understanding youth time preference: mechanisms of environmental risk and uncertainty and attitudes toward risk-behavior and education. Hum. Nat. 21, 140–164. doi: 10.1007/s12110-010-9 084-2 Griskevicius, V., Tybur, J. M., Delton, A. W., and Robertson, T. E. (2011b). The inﬂuence of mortality and socioeconomic status on risk and delayed rewards: a life history theory approach. J. Pers. Soc. Psychol. 100, 1015–1026. doi: 10.1037/ a0022403 Sherman, R. A., Figueredo, A. J., and Funder, D. C. (2013). The behavioral correlates of overall and distinctive life history strategy. J. Pers. Soc. Psychol. 105, 873–888. doi: 10.1037/a003 3772 Hayes, A. F. (2018). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach, Second Edn. New York, NY: Guilford Press. August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 8 Lim and Javadpour Uncertainty and Procrastination Steel, P. (2007). The nature of procrastination: a meta-analytic and theoretical review of quintessential self-regulatory failure. Psychol. Bull. 133, 65–94. doi: 10.1037/0033-2909.133.1.65 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. Steel, P. (2010). Arousal, avoidant and decisional procrastinators: do they exist? Pers. Individ. Diﬀer. 48, 926–934. Frontiers in Psychology \| www.frontiersin.org August 2021 \| Volume 12 \| Article 717380 REFERENCES doi: 10.1016/j.paid.2010.02.025 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. Steel, P., Brothen, T., and Wambach, C. (2001). Procrastination and personality, performance, and mood. Pers. Individ. Diﬀer. 30, 95–106. doi: 10.1016/S0191- 8869(00)00013-1 Van Eerde, W. (2003). A meta-analytically derived nomological network of procrastination. Pers. Individ. Diﬀer. 35, 1401–1418. doi: 10.1016/S0191- 8869(02)00358-6 Copyright © 2021 Lim and Javadpour. 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. Wang, C., and Zhao, H. (2020). The impact of COVID-19 on anxiety in Chinese university students. Front. Psychol. 11:1168. doi: 10.3389/fpsyg.2020.01168 Wang, X. T., Kruger, D. J., and Wilke, A. (2009). Life-history variables and risk- taking propensity. Evol. Hum. Behav. 30, 77–84. doi: 10.1016/j.evolhumbehav. 2008.09.006 August 2021 \| Volume 12 \| Article 717380 Frontiers in Psychology \| www.frontiersin.org 9
https://openalex.org/W2567339939	https://www.pure.ed.ac.uk/ws/files/43688396/RubioCampilloEtalA2017BayesianAnalysisAndFreeMarketTrade.pdf	English	null	Bayesian analysis and free market trade within the Roman Empire	Antiquity	2,017	cc-by	6,443	Edinburgh Research Explorer Edinburgh Research Explorer Digital Object Identifier (DOI): 10.15184/aqy.2017.131 Document Version: Publisher's PDF, also known as Version of record 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: Antiquity Published In: Antiquity Bayesian analysis and free market trade within the Roman Empire Citation for published version: Rubio-Campillo, X, Coto-Sarmiento, M, Pérez-Gonzalez, J & Remesal Rodríguez, J 2017, 'Bayesian analysis and free market trade within the Roman Empire', Antiquity, vol. 91, no. 359, pp. 1241-1252. https://doi.org/10.15184/aqy.2017.131 Citation for published version: Rubio-Campillo, X, Coto-Sarmiento, M, Pérez-Gonzalez, J & Remesal Rodríguez, J 2017, 'Bayesian analysis and free market trade within the Roman Empire', Antiquity, vol. 91, no. 359, pp. 1241-1252. https://doi.org/10.15184/aqy.2017.131 Bayesian analysis and free market trade within the Roman Empire Xavier Rubio-Campillo1,2,∗, María Coto-Sarmiento2, Jordi Pérez-Gonzalez3 & José Remesal Rodríguez3 Research Xavier Rubio-Campillo1,2,∗, María Coto-Sarmiento2, Jordi Pérez-Gonzalez3 & José Remesal Rodríguez3 The trade networks of the Roman Empire are among the most intensively researched large-scale market systems in antiquity, yet there is no consensus on the economic structure behind this vast network. The difﬁculty arises from data fragmentation and the lack of formal analytical methods. Here, the authors present a Bayesian analysis quantifying the extent to which four previously proposed hypotheses match the evidence for the market system in Roman olive oil. Results suggest that the size of economic agents involved in this network followed a power-law distribution, strongly indicating the presence of free market structures supplying olive oil to Rome. This new analysis offers an important tool to researchers exploring the impact of trade on the dynamics of past societies. Keywords: Roman Empire, Bayesian analysis, olive oil, trade, model selection 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 1 School of History, Classics and Archaeology, University of Edinburgh, William Robertson Wing, Old Medical School, 4 Teviot Place, Edinburgh EH8 9AG, UK 2 Barcelona Supercomputing Center, Carrer de Jordi Girona 29–31, 08034 Barcelona, Spain 3 CEIPAC, Departament de Prehistòria, Història Antiga y Arqueologia, Universitat de Barcelona, Carrer de Montalegre 6, 08001 Barcelona, Spain ∗ Author for correspondence (Email: xavier.rubio@ed.ac.uk) © Antiquity Publications Ltd, 2017. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re- use, distribution and reproduction in any medium, provided the original work is properly cited. antiquity 91 359 (2017): 1241–1252 doi:10.15184/aqy.2017.131 1241 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C © Antiquity Publications Ltd, 2017. This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re- use, distribution and reproduction in any medium, provided the original work is properly cited. antiquity 91 359 (2017): 1241–1252 doi:10.15184/aqy.2017.131 Introduction Trade is one of the most important social interactions in human societies. Intuition suggests that we should ﬁnd evidence of common patterns in trade activities at different periods and in different territories. Trade activities also show, however, a large degree of diversity that needs to be considered. This diversity is particularly relevant when examining past economies distinct from our own (Oka & Kusimba 2008). Contrasting dynamics generate many research questions: how did trade networks change over time? What role did communication play? What factors promoted the collapse or emergence of trade routes? These challenges are even more relevant for long-range trade requiring complex organisation. The Mediterranean is one of the most heavily studied areas to have developed intensive long-range trade. Benign climatic conditions encouraged maritime mobility, and, as a consequence, cultural exchange ﬂourished between regions thousands of kilometres apart (Bevan 2014). This feedback loop between economic and political processes reached its peak during the Roman Empire (ﬁrst to third centuries AD). Despite the abundance of archaeological data for this activity, some of the dynamics behind this interconnected Xavier Rubio-Campillo et al. Xavier Rubio-Campillo et al. economy are still subject to debate (e.g. Carreras & Soto 2013; Brughmans & Poblome 2016; Orengo & Livarda 2016). economy are still subject to debate (e.g. Carreras & Soto 2013; Brughmans & Poblome 2016; Orengo & Livarda 2016). The archaeological record provides a wealth of evidence, such as the vast quantities of amphorae used to ship liquid products. Assumptions regarding the use of these containers have, however, been recently challenged. Such assumptions include the lack of re-use, or the unambiguous association between amphora types and their contents (Romanus et al. 2009; Foley et al. 2012; Pecci et al. 2017). In addition, proposed patterns of increase or decrease in trade activities based on the frequency distribution of amphorae could also be explained by the use of containers that rarely survive in the archaeological record, such as barrels and skins (Wilson 2009). All of these uncertainties affect the ﬁeld of Roman economy studies (Bowman & Wilson 2009). Rome was one of the largest economies of the past, but was the Roman world really interconnected? What role did the state play? Did political uniﬁcation promote economic growth? The lack of clear archaeological proxies for the structure of Roman markets sparked a long and heated debate between the initial modernist, almost-capitalistic perspective, and the opposed primitivists. https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at © Antiquity Publications Ltd, 2017 Introduction A number of works have recently recognised the diversity of dynamics within Roman trade (e.g. Finley 1999; Scheidel et al. 2007; Scheidel 2012). These recent works highlight the need to apply quantitative methods to study the spectrum of market structures (Bowman & Wilson 2009; Scheidel 2014). This is not an easy task: working hypotheses often overlap and most of them probably explain only a small part of the entire picture. It is also not clear what type of archaeological proxy is required to evaluate the hypotheses, due to the data uncertainty already mentioned. Ultimately, it seems unrealistic to assume that a traditional statistical framework would be able, from existing evidence, to reject competing hypotheses. It is important to note that these challenges are not restricted to Roman studies, nor to study of the past. There is a growing debate concerning the need to develop research frameworks able to quantify the plausibility of different hypotheses addressing existing data (Johnson & Omland 2004; Towner & Luttbeg 2007; Crema et al. 2014; Kandler & Powell 2015; Buck & Meson 2015; Rubio-Campillo 2016). Bayesian inference is a statistical framework particularly well suited to confront this challenge by model selection. It is founded on the premise that knowledge of a given system with its associated uncertainty can gradually be updated through new evidence. First, competing hypotheses are transformed into a set of statistical models and parameters. The possible values for each parameter receive an initial probability (i.e. a prior), following a speciﬁc statistical distribution based on existing knowledge. A likelihood function is used to derive the probabilities of any given result by assessing the value of the input parameters. The updated knowledge (i.e. the posterior distribution) is then computed to infer a) the probability that a given model generated the existing evidence; and b) the relative explanatory power of a set of competing models. The framework of model selection that we have developed is designed to quantify the extent to which different market structures match the evidence for long-range trade within the Roman Empire. The approach has been applied to a dataset of thousands of codes stamped in amphora handles collected from Monte Testaccio in Rome. This evidence was used to infer the market structure of the massive olive-oil trade system that supplied the https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C 1242 1242 Bayesian analysis and free market trade within the Roman Empire capital from the production area of the Guadalquivir River, located in the Hispania Baetica province. Bayesian inference has then been applied to evaluate four previously suggested models by examining their explanatory power in relation to the dataset. Research Archaeological proxies of market structures g p Monte Testaccio is a 45m-high artiﬁcial mound in Rome located near to the east bank of the Tiber. It was created through the deposition of thousands of amphorae following their Figure 1. Dressel 20 amphora with a stamp, PNN, in one of its sides (after Aguilera Martín 2012). shipment to the city via river transport. Most of the excavated materials are iden- tiﬁed as Dressel 20 amphorae, similar to those depicted in Figure 1. This amphora type is typically associated with Baetican olive oil, which was widely distributed from southern Spain throughout the western Empire from the ﬁrst to the third centuries AD (Remesal Rodríguez 1998). A large percentage of Dressel 20 amphorae displayed a stamp identifying an economic agent linked to the container (see Figure 2 for an example of the most frequently found code from Monte Testaccio). It has been suggested that the stamped code identiﬁes the owner of the land where the olives were harvested and the oil was produced (Garlan 1983; Remesal Rodríguez 2013). Other authors suggest that the stamp refers to the workshop that made the amphorae (Funari 1996; Berni Millet 2008). Whichever is the case, the actors identiﬁed in the stamps were part of a common economic activity linked to long-range trade. Monte Testaccio is a 45m-high artiﬁcial mound in Rome located near to the east bank of the Tiber. It was created through the deposition of thousands of amphorae following their shipment to the city via river transport. Most of the excavated materials are iden- tiﬁed as Dressel 20 amphorae, similar to those depicted in Figure 1. This amphora type is typically associated with Baetican olive oil, which was widely distributed from southern Spain throughout the western Empire from the ﬁrst to the third centuries AD (Remesal Rodríguez 1998). Figure 1. Dressel 20 amphora with a stamp, PNN, in one of its sides (after Aguilera Martín 2012). g A large percentage of Dressel 20 amphorae displayed a stamp identifying an economic agent linked to the container (see Figure 2 for an example of the most frequently found code from Monte Testaccio). It has been suggested that the stamped code identiﬁes the owner of the land where the olives were harvested and the oil was produced (Garlan 1983; Remesal Rodríguez 2013). Other authors suggest that the stamp refers to the workshop that made the amphorae (Funari 1996; Berni Millet 2008). https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at ttps://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Camb Model selection framework Four market structures were evaluated against this archaeological proxy. The com- peting models were chosen from previously published studies that could be linked to olive-oil trade within the Roman Empire. Figure 2. Photograph and digitised version of a stamp where PNN can be read. It is the most common code found in the dataset, and its three letters probably identify the initials of a full Roman name or tria nomina. H1: balanced model Demand is fulﬁlled by the aggregation of small-scale production units (Poblome et al. 2013). This would translate into several agents of similar capacity. H2: balanced model with large actors Political connections alter the previous dynamic by creating large-scale producers linked to the state (e.g. Erdkamp 2005: 109, for grain trade). H3: concentration dynamics Farming activities are often characterised by a minimum threshold, below which production is not sustainable. As a consequence, small agents are absorbed by larger agents and land ownership is gradually concentrated (Allanson 1992). Archaeological proxies of market structures Whichever is Figure 1. Dressel 20 amphora with a stamp, PNN, in one of its sides (after Aguilera Martín 2012). the case, the actors identiﬁed in the stamps were part of a common economic activity linked to long-range trade. The number of stamps displaying the same code was chosen as the archaeological proxy of the actor’s weight within the market. This proxy is relative, as it does not deﬁne the absolute production of each actor. A code displayed in 100 stamps will, however, naturally identify an individual with higher economic weight than another represented by a code found in 5 stamps. The aggregated frequency distribution of the codes is used as the proxy for the entire market structure (i.e. the number of agents displaying the same frequency value). Our analysis used a dataset of 7520 stamps collected from 10 different excavations at Monte Testaccio, from 1990–2014 (Blázquez et al. 1994; Blázquez & Remesal Rodríguez 1999, 2001, 2003, 2007, 2010, 2014). The chronological span of these remains covers 117 years (from AD 143–260). A quarter of these stamps were incomplete due to fragmentation or erosion, and a probabilistic model was created to integrate this uncertainty into the © Antiquity Publications Ltd, 2017 © Antiquity Publications Ltd, 2017 1243 08, subject to the Cambridge Core terms of use, available at Xavier Rubio-Campillo et al. analysis by assigning complete codes to incomplete stamps. The method then used Monte Carlo simulation to generate a dataset ready to be analysed (the complete process is described in the online supplementary material (OSM), section 1). Figure 3 shows the frequency distribu- tion of the codes found in the amphora dataset. It can be observed that the shape is dominated by a heavy tail, but further analysis is required to assess the validity of each competing hypothesis. H4: Pareto principle A self-organised and interconnected economic system follows the Pareto principle: 80 per cent of the market is controlled by 20 per cent of its agents (Axtell 2001). This structure would generate a small number of large-scale producers complemented by a large number of small-scale producers. The focus on the code frequency distributions allowed us to translate the hypotheses above into statistical models (as summarised in Table 1). This approach captured the properties of each suggested structure while accepting a certain degree of stochasticity: © Antiquity Publications Ltd, 2017 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C s. https://doi.org/10.15184/aqy.2017.131 dge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at 1244 © Antiquity Publications Ltd, 2017 © Antiquity Publications Ltd, 2017 1244 Bayesian analysis and free market trade within the Roman Empire Fi 3 E pi i al f q di t ib ti f d i ta p Th pl t h th b f d ( a i ) app Figure 3. Empirical frequency distribution of codes in stamps. The plot shows the number of codes (y-axis) appearing in a given number of stamps (x-axis). The red dot depicts the code PNN, which has the largest number of repetitions (253). Both axes are transformed to logarithmic scale. Figure 3. Empirical frequency distribution of codes in stamps. The plot shows the number of codes (y-axis) appearing in a given number of stamps (x-axis). The red dot depicts the code PNN, which has the largest number of repetitions (253). Both axes are transformed to logarithmic scale. Empirical frequency distribution of codes in stamps. The plot shows the number of codes (y-axis) appearing in a ber of stamps (x-axis). The red dot depicts the code PNN, which has the largest number of repetitions (253). Both nsformed to logarithmic scale. Figure 3. Empirical frequency distribution of codes in stamps. The plot shows the number of codes (y-axis) appearing in a given number of stamps (x-axis). The red dot depicts the code PNN, which has the largest number of repetitions (253). Both axes are transformed to logarithmic scale. Table 1. Each hypothesis has been translated into a statistical model based on expected frequency distributions. Table 1. Each hypothesis has been translated into a statistical model based on expected frequency distributions. Model Description Distribution M1 balanced Poisson M2 balanced with large actors negative binomial M3 concentration log-normal M4 power-law Pareto r M1 transformed H1 into a Poisson distribution where parameter λ deﬁnes the average size for the entire set of producers. r M2 used a negative binomial distribution to generate a heavy tail composed of a few large-scale producers, according to H2. r M3 represented the concentration process of H3 as a log-normal distribution. The shape of this model is able to represent: a) a lack of small producers; b) the predominance of mid-sized producers; and c) a heavy tail of large-scale producers. r M3 represented the concentration process of H3 as a log-normal distribution. The shape of this model is able to represent: a) a lack of small producers; b) the predominance of mid-sized producers; and c) a heavy tail of large-scale producers. https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at ttps://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Camb 1245 Xavier Rubio-Campillo et al. Table 2. Parameters for the four examined models. Model Parameter Name Prior M1 λ mean U(1,30) M2 r number of successful trials U(1,100) p probability of success U(0,1) M3 μ mean U(0.1,4) σ standard deviation U(0.1,3) M4 α shape U(0.1,10) Xm scale 1 Table 2. Parameters for the four examined models. r M4 implemented H4 as a Pareto distribution, displaying a large number of small producers complemented by a small number of large producers. r M4 implemented H4 as a Pareto distribution, displaying a large number of small producers complemented by a small number of large producers. The analytical framework chosen for model selection was based on Bayesian inference. The translation of the hypotheses into statistical models allowed us to compute the posterior distributions easily using Markov chain Monte Carlo (MCMC). Uninformed prior beliefs were chosen for all the models, as summarised in Table 2. The OSM, section 2, lists the parameters used in the MCMC execution. The entire framework was implemented using R, Jags and R2Jags (Plummer 2003; Su & Yajima 2015; R Core Team 2016). The ﬁnal step of the method was to quantify the plausibility of the four competing models, taking into account a) the goodness of ﬁt to the evidence; and b) their degree of parsimony. The Deviance Information Criterion (DIC) was chosen for this task, as it is an intuitive information-based measure equivalent to popular indices, such as the Akaike Information Criterion (Spiegelhalter et al. 2002, 2014). Despite some limitations, DIC is a method commonly used in Bayesian inference when the posterior distributions are approximately multivariate-normal. This measure allows the researcher to take into account both the complexity and the deviance from the model’s outcome to existing evidence, thereby quantifying parsimony and goodness of ﬁt within a Bayesian framework. https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at Results All posterior distributions follow unimodal shapes, thus suggesting that the signal in the dataset is strong enough to overcome noise and generate consistent results (as can be seen in the OSM, section 3). The goodness of ﬁt of the models can ﬁrst be explored by comparing the evidence against simulated data obtained from each model. Figure 4 shows the results generated by each model using the mean of each posterior distribution as parameter values. The two balanced models (M1 and M2) generate patterns substantially different from the observations. Models M3 and M4 are closer to the frequency distribution of the collected stamps. The ﬁtness of M3 and M4 suggests the presence of some producers of signiﬁcantly larger size than the average value of the sample. This could have resulted from a concentration process generated by the market. The huge demand for olive oil generated by the city of Rome must surely have put a lot of pressure on the production of this commodity. Consequently, Baetican producers could have generated economies of scale in order to maximise the efﬁciency of the existing infrastructures. © Antiquity Publications Ltd, 2017 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C 1246 1246 Research ey dots display the frequency distributions of codes in stamps, while coloured points show random deviates generated with the Research dots display the frequency distributions of codes in stamps, while coloured points show random deviates generated with the Research red points show random deviates generated with the the frequency distributions of codes in stamps, while c ts displa ots displa Xavier Rubio-Campillo et al. Table 3. DIC measurements for each model. The following measures are provided: a) goodness of ﬁt; b) complexity penalty; c) penalised deviance; and d) distance to best candidate (i.e. M4). Better models have lower values, and differences beyond 100 are typically deﬁned as signiﬁcant. Model Mean Deviance Penalty Penalised deviance DIC M1 18392 1.044 18394 13333 M2 8529 1.038 8530 3649 M3 7206 2.06 7208 2147 M4 5060 1.043 5061 0 Table 3. DIC measurements for each model. The following measures are provided: a) goodness of ﬁt; b) complexity penalty; c) penalised deviance; and d) distance to best candidate (i.e. M4). Better models have lower values, and differences beyond 100 are typically deﬁned as signiﬁcant. Table 3. DIC measurements for each model. The following measures are provided: a) goodness of ﬁt; b) complexity penalty; c) penalised deviance; and d) distance to best candidate (i.e. M4). Better models have lower values, and differences beyond 100 are typically deﬁned as signiﬁcant. Model Mean Deviance Penalty Penalised deviance DIC M1 18392 1.044 18394 13333 M2 8529 1.038 8530 3649 M3 7206 2.06 7208 2147 M4 5060 1.043 5061 0 DIC values for the four models are summarised in Table 3. Comparison of these measures suggests that the best ﬁt for this sample is a power-law distribution, in which most olive- oil producers identiﬁed by stamps were very small, while concentration processes created a signiﬁcant quantity of middle-sized agents and a very small number of large-scale producers. Even though it is not usually possible to identify these large actors, they were probably linked to the Roman political system (i.e. members of the Senate or the imperial family). https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at Discussion The identiﬁcation of a power-law distribution requires consideration of the mechanism causing this distinct signal in the evidence (Stumpf & Porter 2012). This statistical distribution has received considerable attention since its creation, owing to its unique properties. One of the most distinctive characteristics is that power-laws exhibit scale invariance (i.e. the features of the distribution remain constant over the entire range). If both axes of the power-law are transformed to a logarithmic scale, the sample will follow a straight line (this can be observed in Figure 4). g g Power-laws have been identiﬁed in disciplines as varied as physics, ecology, sociology and linguistics (Clauset et al. 2009). Their transdisciplinary relevance is explained by the fact that power-laws are proxies for self-organised optimal systems with some degree of hierarchy (Newman 2005). It would be very difﬁcult to generate a power-law by external causes on account of scale invariance, but a power-law emerges when evolutionary dynamics enforce optimality by selection. Free markets are complex systems fulﬁlling both self-organisation and optimality, and could therefore be the underlying process generating the shape of our dataset. This prediction has been proved elsewhere in contemporary case studies (Axtell 2001). Although we cannot disregard the possibility that the pattern detected here was generated by a different structure, the simplest interpretation for our results is that the Roman olive-oil trade was organised as a free market. This is the outcome that would be expected by researchers who have documented the existence of private enterprise (e.g. Temin 2001, 2013). It is difﬁcult to conceive how a state-organised market could generate a power-law. It would require a structure with thousands of small-sized producers complemented by a few large-sized agents. The creation of this organisation, if at all possible, would have required the investment of a large amount of resources and ﬁne control over the system. Finally, the © Antiquity Publications Ltd, 2017 © Antiquity Publications Ltd, 2017 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C 1248 1248 Bayesian analysis and free market trade within the Roman Empire nature of the dataset also supports the existence of a densely interconnected free market, as the amphora stamps that follow the power-law were made 2000km from Rome. The work presented here has its limitations, as does any other model. Our analysis, for example, cannot conﬁrm whether the stamps identify olive-oil producers or pottery makers. The ﬁrst option seems more plausible, as it could generate the economies of scale needed to produce a power-law; it is feasible that an olive-oil producer owned an area much larger than the competitors, while a pottery workshop could not be hundreds of times bigger than its competitors due to physical constraints. Neither hypothesis can, however, be discarded, unless new data are discovered. This uncertainty does not invalidate our results, as the analysis indicates long-range trade activity exhibiting free market properties independently of the agents identiﬁed by the stamps (olive-oil producers or amphora producers). g Power-law distributions are ubiquitous to datasets deriving from contemporary activities, but they are less commonly observed in archaeological datasets (Bentley & Maschner 2003; Bentley & Shennan 2003; Grove 2011; Haas et al. 2015). Several studies show how their dataset’s shape follows a power-law only in the heavy tail, while the left segment does not match the log-log straight line. These partial ﬁts could, however, also be explained by log- normal distributions, as in our model M3 (for discussion within archaeology, see Bentley 2003; beyond archaeology, see Allen et al. 2001; Eeckhout 2004; Levy 2009). This distinc- tion is important, as the two shapes are generated by critically different social phenomena. Log-normal distributions are essentially produced by additive processes, while power-laws emerge from complex self-organised systems and indicate a higher degree of inequality and interaction amongst their components. Additionally, log-normal distributions display an arithmetic mean value (i.e. there is an average size of producers), while power-laws do not. The analysis presented here has revealed one of the ﬁrst examples of an archaeological dataset that follows a complete power-law distribution. There are probably very few contexts that meet the conditions required for the development of a large-scale integrated market. The subset of these scenarios, where the structure of the market can be found in the archaeological record, is even lower, due to a diversity of taphonomic processes (i.e. survivorship bias). 1248 We have identiﬁed a power-law distribution within an economy quite different from contemporary systems, thus suggesting that the same pattern could be identiﬁed in other case studies. Additional research is needed to understand whether this power-law is an isolated scenario speciﬁc to the Roman Empire, or a general pattern within the trade networks of past societies. Acknowledgements XRC and MCS were funded by the European Research Council Advanced Grant EPNet (340828). We would like to thank the two anonymous reviewers for their comments on previous versions of the manuscript. We would also like to thank Enrico Crema, Albert Díaz-Guilera, Luce Prignano, Ignacio Morer, Jean-Marc Montanier, Simon Carrignon, Joan Rodríguez, Víctor Revilla, Maria Yubero, Jonàs Alcaina, Pau Valdés, Alessandra Pecci and Tom Brughmans for advice on the analysis, access to unpublished data and productive discussions on the methods and results of this work. The dataset is distributed under the terms of a Creative Commons Attribution-Share Alike 4.0 International Licence. The source code of the model is licensed under a GNU General Public Licence; both can be downloaded from https://github.com/xrubio/bayesRome (accessed 9 June 2017). © Antiquity Publications Ltd, 2017 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at References −2010. Estudios sobre el Monte Testaccio (Roma) 5 (Col·lecció Instrumenta 35). Barcelona: Universitat de Barcelona. Aguilera Martín, A. 2012. La normalisation de l’épigraphie amphorique: les tituli picti des amphores Dressel 20, in M.E. Fuchs, R. Sylvestre & C.S. Heidenreich (ed.) Inscriptions mineures: nouveautés et réﬂexions. Actes du premier colloque Ductus: 135–43. Lausanne: Université de Lausanne. −2014. Estudios sobre el Monte Testaccio (Roma) 6 (Col·lecció Instrumenta 47). Barcelona: Universitat de Barcelona. Blázquez, J.M., J. Remesal Rodríguez & E. Rodríguez Almeida (ed.). 1994. Excavaciones arqueológicas en el Monte Testaccio (Roma). Memoria campaña 1989. Madrid: Ministerio de Cultura. Allanson, P. 1992. Farm size structure in England and Wales 1939–89. Journal of Agricultural Economics 43: 137–48. https://doi.org/10.1111/j.1477- 9552.1992.tb00210.x Bowman, A. & A. Wilson. 2009. Quantifying the Roman economy. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/ 9780199562596.001.0001 Allen, A., B.-L. Li & E. Charnov. 2001. Population ﬂuctuations, power laws and mixtures of lognormal distributions. Ecology Letters 4(1): 1–3. https://doi. org/10.1046/j.1461-0248.2001.00194.x Brughmans, T. & J. Poblome. 2016. Roman bazaar or market economy? Explaining tableware distributions through computational modelling. Antiquity 90: 393–408. https://doi.org/10.15184/aqy.2016.35 Axtell, R.L. 2001. Zipf distribution of U.S. ﬁrm sizes. Science 293: 1818–20. https://doi.org/10.1126/science.1062081 Bentley, R.A. 2003. Scale-free network growth and social inequality, in R.A. Bentley & H.D. Maschner (ed.) Complex systems and archaeology: 27–46. Salt Lake City: University of Utah Press. Buck, C.E. & B. Meson. 2015. On being a good Bayesian. World Archaeology 47: 567–84. https://doi.org/10.1080/00438243.2015.1053977 Carreras, C. & P.D. Soto. 2013. The Roman transport network: a precedent for the integration of European mobility. Historical Methods: a Journal of Quantitative and Interdisciplinary History 46(3): 117–33. Bentley, R.A. & H.D. Maschner (ed.). 2003. Complex systems and archaeology. Salt Lake City: University of Utah Press. Bentley, R.A. & S.J. Shennan. 2003. Cultural transmission and stochastic network growth. American Antiquity 68: 459–85. https://doi.org/10.2307/3557104 https://doi.org/10.1080/01615440.2013.803403 Clauset, A., C.R. Shalizi & M.E.J. Newman. 2009. Power-law distributions in empirical data. SIAM Review 51: 661–703. https://doi.org/10.1137/070710111 Berni Millet, P. 2008. Epigrafía anfórica de la Bética. Nuevas formas de análisis (Col·lecció Instrumenta 29). Barcelona: Universitat de Barcelona. Crema, E., K. Edinborough, T. Kerig & S. Shennan. 2014. An approximate Bayesian computation approach for inferring patterns of cultural evolutionary change. Journal of Archaeological Science 50: 160–70. https://doi.org/10.1016/j.jas.2014.07.014 Bevan, A. 2014. Mediterranean containerization. Current Anthropology 55: 387–418. https://doi.org/10.1086/677034 Blázquez, J.M. & J. Remesal Rodríguez (ed.). 1999. Estudios sobre el Monte Testaccio (Roma) 1 (Col·lecció Instrumenta 6). Barcelona: Universitat de Barcelona. Eeckhout, J. 2004. Supplementary material To view supplementary material for this article, please visit https://doi.org/10.15184/aqy. 2017.131 1249 ttps://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Camb https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at Xavier Rubio-Campillo et al. https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C References Gibrat’s law for (all) cities. American Economic Review 94: 1429–51. https://doi.org/10.1257/0002828043052303 −2001. Estudios sobre el Monte Testaccio (Roma) 2 (Col·lecció Instrumenta 10). Barcelona: Universitat de Barcelona. Erdkamp, P. 2005. The grain market in the Roman Empire: a social, political and economic study. Cambridge: Cambridge University Press. −2003. Estudios sobre el Monte Testaccio (Roma) 3 (Col·lecció Instrumenta 14). Barcelona: Universitat de Barcelona. Finley, M.I. 1999. The ancient economy. Berkeley: University of California Press. −2007. Estudios sobre el Monte Testaccio (Roma) 4 (Col·lecció Instrumenta 24). Barcelona: Universitat de Barcelona. © Antiquity Publications Ltd, 2017 1250 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C Bayesian analysis and free market trade within the Roman Empire https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C Bayesian analysis and free market trade within the Roman Empire Foley, B.P., M.C. Hansson, D.P. Kourkoumelis & T.A. Theodoulou. 2012. Aspects of ancient Greek trade re-evaluated with amphora DNA evidence. Journal of Archaeological Science 39: 389–98. https://doi.org/10.1016/j.jas.2011.09.025 Plummer, M. 2003. Jags: a program for analysis of Bayesian graphical models using Gibbs sampling, in K. Hornik, F. Leisch & A. Zeileis (ed.) Proceedings of the 3rd International Workshop on Distributed Statistical Computing: 1–10. Vienna: Technische Universität Wien. Funari, P.P.A. 1996. Dressel 20 inscriptions from Britain and the consumption of Spanish olive oil (British Archaeological Reports international series 250). Oxford: British Archaeological Reports. Poblome, J., P. Bes & R. Willet. 2013. Thoughts on the archaeological residue of networks. A view from the East, in S. Keay (ed.) Rome, Portus and the Mediterranean (Archaeological Monographs of the British School at Rome 21): 393–401. Rome: British School at Rome. Garlan, Y. 1983. Greek amphorae and trade, in P. Garnsey, K. Hopkins & C.R. Whittaker (ed.) Trade in the ancient economy: 27–35. Berkeley: University of California Press. R Core Team. 2016. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing. Grove, M. 2011. An archaeological signature of multi-level social systems: the case of the Irish Bronze Age. Journal of Anthropological Archaeology 30(1): 44–61. https://doi.org/10.1016/j.jaa.2010.12.002 Remesal Rodríguez, J. 1998. Baetican olive oil and the Roman economy. Journal of Roman Archaeology supplementary series 29: 183–200. Haas, W.R., C.J. Klink, G.J. Maggard & M.S. Aldenderfer. 2015. Settlement-size scaling among prehistoric hunter-gatherer settlement systems in the new world. PLoS ONE 10: e0140127. https://doi.org/10.1371/journal.pone.0140127 −2013. El Monte Testaccio. De vertedero a archivo. Atti della Pontiﬁcia Accademia romana di Archeologia (Serie III). Rendiconti LXXXV: 111–28. Romanus, K., J. Baeten, J. Poblome, S. Accardo, P. Degryse, P. Jacobs, D. De Vos & M. Waelkens. 2009. Wine and olive oil permeation in pitched and non-pitched ceramics: relation with results from archaeological amphorae from Sagalassos, Turkey. Journal of Archaeological Science 36: 900–909. https://doi.org/10.1016/j.jas.2008.11.024 Johnson, J.B. & K.S. Omland. 2004. Model selection in ecology and evolution. Trends in Ecology & Evolution 19(2): 101–108. https://doi.org/10.1016/j.tree.2003.10.013 https://doi.org/10.1016/j.tree.2003.10.013 Kandler, A. & A. Powell. 2015. Inferring learning strategies from cultural frequency data, in A. Mesoudi & K. Aoki (ed.) Learning strategies and cultural evolution during the Palaeolithic: 85–101. Tokyo: Springer Japan. Rubio-Campillo, X. 2016. Model selection in historical research using approximate Bayesian computation. PLoS ONE 11: e0146491. https://doi.org/10.1371/journal.pone.0146491 Levy, M. 2009. Gibrat’s law for (all) cities: comment. American Economic Review 99: 1672–75. https://doi.org/10.1257/aer.99.4.1672 Scheidel, W. (ed.). 2012. The Cambridge companion to the Roman economy. Cambridge: Cambridge University Press. Newman, M. 2005. Power laws, pareto distributions and Zipf’s law. Contemporary Physics 46: 323–51. https://doi.org/10.1080/00107510500052444 −2014. The shape of the Roman world: modelling imperial connectivity. Journal of Roman Archaeology 27: 7–32. h //d i /10 1017/S1047759414001147 Oka, R. & C.M. Kusimba. 2008. The archaeology of trading systems, part 1: towards a new trade synthesis. Journal of Archaeological Research 16: 339–95. https://doi.org/10.1007/s10814-008-9023-5 https://doi.org/10.1017/S1047759414001147 Scheidel, W., I. Morris & R.P. Saller (ed.). 2007. The Cambridge economic history of the Greco-Roman world. Cambridge & New York: Cambridge University Press. Orengo, H.A. & A. Livarda. 2016. The seeds of commerce: a network analysis-based approach to the Romano-British transport system. Journal of Archaeological Science 66: 21–35. https://doi.org/10.1016/j.jas.2015.12.003 Spiegelhalter, D.J., N.G. Best, B.P. Carlin & A. van der Linde. 2002. Bayesian measures of model complexity and ﬁt. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 64: 583–639. https://doi.org/10.1111/1467-9868.00353 Pecci, A., J. Clarke, M. Thomas, J. Muslin, I. van der Graaff, L. Toniolo, D. Miriello, G.M. Crisci, M. Buonincontri & G. Di Pasquale. 2017. Use and reuse of amphorae. Wine residues in Dressel 2–4 amphorae from Oplontis Villa B (Torre Annunziata, Italy). Journal of Archaeological Science: Reports 12: 515–21. https://doi.org/10.1016/j.jasrep.2017.02.025 −2014. The deviance information criterion: 12 years on. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 76: 485–93. https://doi.org/10.1111/rssb.12062 © Antiquity Publications Ltd, 2017 1251 Xavier Rubio-Campillo et al. Stumpf, M.P.H. & M.A. Porter. 2012. Critical truths about power laws. Science 335: 665–66. https://doi.org/10.1126/science.1216142 Su, Y.-S. & M. Yajima. 2015. R2jags: using R to run ‘JAGS’. R package version 0.5–7. Available at: https://rdrr.io/cran/R2jags/ (accessed 9 June 2017). Temin, P. 2001. A market economy in the early Roman Empire. Journal of Roman Studies 91: 169–81. https://doi.org/10.1017/S0075435800015902 −2013. The Roman market economy. Princeton (NJ): Princeton University Press. Received: 23 June 2016; Accepted: 23 September 2016; R Towner, M.C. & B. Luttbeg. 2007. Alternative statistical approaches to the use of data as evidence for hypotheses in human behavioral ecology. Evolutionary Anthropology: Issues, News, and Reviews 16(3): 107–18. https://doi.org/10.1002/evan.20134 Stumpf, M.P.H. & M.A. Porter. 2012. Critical truths about power laws. Science 335: 665–66. https://doi.org/10.1126/science.1216142 Su, Y.-S. & M. Yajima. 2015. R2jags: using R to run ‘JAGS’. R package version 0.5–7. Available at: https://rdrr.io/cran/R2jags/ (accessed 9 June 2017). Wilson, A. 2009. Approaches to quantifying Roman trade, in A. Bowman & A. Wilson (ed.) Quantifying the Roman economy: 213–49. Oxford: Oxford University Press. https://doi.org/10.1093/ acprof:oso/9780199562596.003.0009 Wilson, A. 2009. Approaches to quantifying Roman trade, in A. Bowman & A. Wilson (ed.) Quantifying the Roman economy: 213–49. Oxford: Oxford University Press. https://doi.org/10.1093/ acprof:oso/9780199562596.003.0009 Temin, P. 2001. A market economy in the early Roman Empire. Journal of Roman Studies 91: 169–81. https://doi.org/10.1017/S0075435800015902 −2013. The Roman market economy. Princeton (NJ): Princeton University Press. −2013. The Roman market economy. Princeton (NJ): Princeton University Press. https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at 1252 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at https://doi.org/10.1017/S1047759414001147 Received: 23 June 2016; Accepted: 23 September 2016; Revised: 12 October 2016 Received: 23 June 2016; Accepted: 23 September 2016; Revised: 12 October 2016 © Antiquity Publications Ltd, 2017 1252 https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the C https://www.cambridge.org/core/terms. https://doi.org/10.15184/aqy.2017.131 Downloaded from https://www.cambridge.org/core. University of Edinburgh, on 21 Sep 2017 at 08:40:08, subject to the Cambridge Core terms of use, available at
https://openalex.org/W4312172994	https://publications.goettingen-research-online.de/bitstream/2/118960/1/document.pdf	English	null	On the Fabrication and Characterization of Polymer-Based Waveguide Probes for Use in Future Optical Cochlear Implants	Materials	2,022	cc-by	12,485	Article On the Fabrication and Characterization of Polymer-Based Waveguide Probes for Use in Future Optical Cochlear Implants Christian Helke 1,2,* , Markus Reinhardt 1,3,4, Markus Arnold 2, Falk Schwenzer 2, Micha Haase 1,2, Matthias Wachs 3, Christian Goßler 3,4, Jonathan Götz 4, Daniel Keppeler 4, Bettina Wolf 4 , Jannis Schaeper 5,6 , Tim Salditt 5,6, Tobias Moser 4,6 , Ulrich Theodor Schwarz 3 and Danny Reuter 1,2 Christian Helke 1,2,* , Markus Reinhardt 1,3,4, Markus Arnold 2, Falk Schwenzer 2, Micha Haase 1,2, Matthias Wachs 3, Christian Goßler 3,4, Jonathan Götz 4, Daniel Keppeler 4, Bettina Wolf 4 , Jannis Schaeper 5,6 , Tim Salditt 5,6, Tobias Moser 4,6 , Ulrich Theodor Schwarz 3 and Danny Reuter 1,2 1 Fraunhofer Institute for Electronic Nanosystems ENAS, 09126 Chemnitz, Germany 1 Fraunhofer Institute for Electronic Nanosystems ENAS, 09126 Chemnitz, Germany 2 Center for Microtechnologies (ZfM), Technical University of Chemnitz, 09126 Chemnitz, Germany 3 Experimental Sensor Science, Technical University of Chemnitz, 09126 Chemnitz, Germany 4 Institute for Auditory Neuroscience and InnerEarLab, University Medical Center Goettingen, 37075 Goettingen, Germany 1 Fraunhofer Institute for Electronic Nanosystems ENAS, 09126 Chemnitz, Germany 2 Center for Microtechnologies (ZfM), Technical University of Chemnitz, 09126 Chemnitz, Germany 3 Experimental Sensor Science, Technical University of Chemnitz, 09126 Chemnitz, Germany 4 Institute for Auditory Neuroscience and InnerEarLab, University Medical Center Goettingen, 37075 Goettingen, Germany 5 Institute for X-ray Physics, University of Goettingen, 37075 Goettingen, Germany 6 Multiscale Bioimaging Cluster of Excellence, University Medical Center Goettingen, 37075 Goettingen, Germany * Correspondence: christian.helke@enas.fraunhofer.de Fraunhofer Institute for Electronic Nanosystems ENAS, 09126 Chemnitz, Germany 2 Center for Microtechnologies (ZfM), Technical University of Chemnitz, 09126 Chemnitz, Germany 3 E i l S S i T h i l U i i f Ch i 09126 Ch i G 2 Center for Microtechnologies (ZfM), Technical University of Chemnitz, 09126 Chemnitz, Germany 3 Experimental Sensor Science, Technical University of Chemnitz, 09126 Chemnitz, Germany 4 Institute for Auditory Neuroscience and InnerEarLab, University Medical Center Goettingen, 37075 Goettingen, Germany 3 Experimental Sensor Science, Technical University of Chemnitz, 09126 Chemnitz, Germany 4 Institute for Auditory Neuroscience and InnerEarLab, University Medical Center Goettingen, 37075 Goettingen, Germany 5 Institute for X-ray Physics, University of Goettingen, 37075 Goettingen, Germany 6 Multiscale Bioimaging Cluster of Excellence, University Medical Center Goettingen, 37075 Goettingen, Germany * Correspondence: christian.helke@enas.fraunhofer.de Abstract: Improved hearing restoration by cochlear implants (CI) is expected by optical cochlear implants (oCI) exciting optogenetically modiﬁed spiral ganglion neurons (SGNs) via an optical pulse generated outside the cochlea. The pulse is guided to the SGNs inside the cochlea via ﬂexible polymer-based waveguide probes. The fabrication of these waveguide probes is realized by using 6” wafer-level micromachining processes, including lithography processes such as spin-coating cladding layers and a waveguide layer in between and etch processes for structuring the waveguide layer. Citation: Helke, C.; Reinhardt, M.; Arnold, M.; Schwenzer, F.; Haase, M.; Wachs, M.; Goßler, C.; Götz, J.; Keppeler, D.; Wolf, B.; et al. On the Fabrication and Characterization of Polymer-Based Waveguide Probes for Use in Future Optical Cochlear Implants. Materials 2023, 16, 106. https://doi.org/10.3390/ ma16010106 Keywords: SU-8; PMMA; polymer; waveguide; optical cochlear implant; spin coating; etching Article On the Fabrication and Characterization of Polymer-Based Waveguide Probes for Use in Future Optical Cochlear Implants Christian Helke 1,2,* , Markus Reinhardt 1,3,4, Markus Arnold 2, Falk Schwenzer 2, Micha Haase 1,2, Matthias Wachs 3, Christian Goßler 3,4, Jonathan Götz 4, Daniel Keppeler 4, Bettina Wolf 4 , Jannis Schaeper 5,6 , Tim Salditt 5,6, Tobias Moser 4,6 , Ulrich Theodor Schwarz 3 and Danny Reuter 1,2 Further adhesion layers and metal layers for laser diode (LD) bonding and light-outcoupling structures are also integrated in this waveguide process ﬂow. Optical microscope and SEM images revealed that the majority of the waveguides are sufﬁciently smooth to guide light with low intensity loss. By coupling light into the waveguides and detecting the outcoupled light from the waveguide, we distinguished intensity losses caused by bending the waveguide and outcoupling. The probes were used in ﬁrst modules called single-beam guides (SBGs) based on a waveguide probe, a ball lens and an LD. Finally, these SBGs were tested in animal models for proof-of-concept implantation experiments. materials materials 1. Introduction Academic Editor: JaeJong Lee Received: 20 October 2022 Revised: 21 November 2022 Accepted: 22 November 2022 Published: 22 December 2022 With more than a million users currently worldwide [1], electrical cochlear implants (eCI) are most likely the most implanted neuroprosthesis. An eCI stimulates spiral ganglion neurons (SGNs) by means of an electrical pulse, and thus bypasses the defect of lost hair cells of a deaf person. eCIs partially restore hearing in most users, yet its performance is limited due to a broad spread of the electric ﬁeld surrounding the electrode and thus stimulating a large amount of SGNs, which results in an impairment of the frequency resolution [2,3]. In addition, the intensity coding is rather poor concerning a small output dynamic range [4,5]. Based on those limitations, it is difﬁcult for eCI patients to understand speech in background noise [6]. A solution could be the use of light instead of electri- cal pulses, as light can be focused much better. An optical cochlear implant (oCI) that stimulates the SGNs by means of optical pulses could improve hearing signiﬁcantly by stim- ulating fewer SGNs with a single pulse. This leads to a higher-frequency resolution [7–10]. However, SGNs are not light-sensitive per se. The control of cells by light envisioned decades ago [11] by optogenetics combines optics and genetics. To achieve this, light-gated ion channels such as channelrhodopsin-2 (ChR2) are genetically introduced [12], which Received: 20 October 2022 Revised: 21 November 2022 Accepted: 22 November 2022 Published: 22 December 2022 2. Waveguide Design and Fabrication Process 2.1. Design of the Flexible Polymer-Based Waveguides Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). https://www.mdpi.com/journal/materials Materials 2023, 16, 106. https://doi.org/10.3390/ma16010106 Materials 2023, 16, 106 2 of 17 have a maximum sensitivity in the range between 460 and 480 nm [13]. To investigate the feasibility of oCIs, implantable 230-µm-wide ﬂexible µLED probes were fabricated, combining two wafer-level processes and the layer transfer of the LEDs from the sapphire growth substrate to a carrier wafer via metal wafer bonding and laser lift-off [10,14,15]. In this work, another approach is pursued by fabricating ﬂexible waveguide probes, which guide light from an external source to the cochlear. Two polymers with differ- ent refractive indices are needed for ensuring both total internal reﬂection between core and cladding layer and device ﬂexibility [16], in contrast to rigid waveguide probes with polymer core and silicon oxide-based cladding layers [17,18]. Waveguides exhibit bend- ing [19–21] and scattering [22,23] losses, which are experimentally investigated in this paper. Total losses in SU-8 waveguides were reported to be in the range of 0.2 dB/cm up to 6.4 dB/cm [24–27]. A material combination of PMMA/SU-8/PMMA is deposited for the realization of the ﬂexible polymer-based waveguides. To structure the polymer-based layer stack, the PMMA/SU-8/PMMA is patterned by UV lithography and dry etching of 6” silicon wafers. The implants are based on optical waveguides which are designed as bendable and ﬂexible devices. Therefore, the polymer structures have to be released from the substrate by sacriﬁcial layer etching after fabrication. Waveguide probes can be used in a wide ﬁeld of sensor applications. For example, waveguide probes with inscribed Bragg gratings embedded in 3D-printed ABS sensing pads for force measurements were developed [28], as well as sensors for environmental monitoring and drinking water quality control based on a gold-coated tilted ﬁber Bragg grating with immobilized Acinetobacter sp. Bacteria on the gold surface used as receptors [29]. Polymer waveguides are frequently also used as light probes in photonic sensor networks [30] and optogenetics [16–18,27,31,32]. 2.1. Design of the Flexible Polymer-Based Waveguides The layout of the waveguide probes is shown in Figure 1a. It consists of ﬂexible polymer-based waveguides guiding the light from the laser diodes (LD) bonded directly on the polymer material to the light-outcoupling structure that is at the tip of the ﬂexible polymer waveguide probe. A certain length of the probe from the tip is intended for implantation into the cochlea. Different types of these waveguide probes are realized within this development and are arranged quarter-clockwise on a 6” wafer (Figure 1b). Different waveguide probe lengths, a number of waveguide probes and the bending radius of the waveguide probe have been designed to observe the inﬂuence of outcoupling and bending losses as well as the control of the individual waveguide probes within a ten- channel waveguide system. The waveguide probe shown in Figure 1a is characterized by ten individual, directly controllable waveguides, a probe length of 45 mm and 5 µm × 5 µm outcoupling structures. The detailed cross section of the polymer waveguide probe system is explained within its fabrication process in Section 2.2. IEW 3 of 17 Figure 1. (a) Layout of a ten-beam flexible polymer based waveguide probe with details of (left) the LD bond pads at coupling side and (right) the light-outcoupling structures. (b) Layout of the 6” wafer with the different types of waveguide probes. Figure 1. (a) Layout of a ten-beam ﬂexible polymer based waveguide probe with details of (left) the LD bond pads at coupling side and (right) the light-outcoupling structures. (b) Layout of the 6” wafer with the different types of waveguide probes. Figure 1. (a) Layout of a ten-beam flexible polymer based waveguide probe with details of (left) the LD bond pads at coupling side and (right) the light-outcoupling structures. (b) Layout of the 6” wafer with the different types of waveguide probes. Figure 1. (a) Layout of a ten-beam ﬂexible polymer based waveguide probe with details of (left) the LD bond pads at coupling side and (right) the light-outcoupling structures. (b) Layout of the 6” wafer with the different types of waveguide probes. Materials 2023, 16, 106 3 of 17 3 of 17 2.2. 2.1. Design of the Flexible Polymer-Based Waveguides Wafer-Level Fabrication of the Flexible Polymer-Based Waveguides The wafer-level fabrication of the ﬂexible polymer waveguides is shown in detail as schematic in Figure 2 and is separated in three columns by means of the status of the following technology ﬂow as cross-section views of (I) the outcoupling structure, (II) the waveguide probe and (III) the side view of the LD coupling structure. The wafer-level fabrication is realized on 6” Si wafers and starts (Figure 2a) with the deposition of a 100 nm thermal SiO2 and spin coating of a 1 µm thick sacriﬁcial layer LOR 10 B (from Microchemicals GmbH—MC) used to detach the polymer waveguides from the Si substrate at the end of the fabrication. Furthermore, a 5 µm thick PMMA layer (from Micro Resist Technology GmbH—MRT) acting as supporting layer is used, as well as a 200 nm SU-8 2000.5 (MRT) adhesion layer needed for the bottom cladding PMMA layer, which is realized later by spin coating. The subsequent sputter deposition of the metal layer stack starts with an adhesion layer consisting of 20 nm Cr followed by 100 nm Au seed layer. Afterwards, electroplating is performed for 3 µm thick Au bond pads by negative-tone resist AZ125nXT (MC) with 12 µm thickness. After electroplating the bond pads, the negative-tone resist AZ125nXT (MC) is removed. Due to the 3 µm high Au bond, to remove the Au seed layer and Cr adhesion layer, a lithography step is used with spray coating of 7.5 µm AZ4999 (MC), followed by dry etching to form the bond pads and the light-outcoupling structures at the end of the waveguides (Figure 2b). A supporting PMMA layer that acts in addition as the bottom cladding layer underneath the SU-8 waveguide core layer is used in the following step to adjust the LD emission facet to the core of the waveguide. Therefore, a 2.5 µm thick PMMA (MRT) layer is spin-coated on the wafer followed by a SU-8 2000.5 layer of 200 nm to avoid cracks that occurred in the 2.5 µm bottom cladding PMMA when the following SU-8 3005 (MRT) layer is directly processed on top (Figure 2c). The SU-8 3005 waveguide core layer is spin-coated with a thickness of 5 µm. Afterwards, the SU-8 3005 is exposed with the waveguide-layer mask in a mask aligner MA200 (SUSS MicroTec GmbH) with 100 mJ/cm2 for 3 s. 2.1. Design of the Flexible Polymer-Based Waveguides Finally, the SU-83005 is developed in mr-Dev 600 (MRT) topside down for 8 min and rinsed 1 min in isopropanol and DI-water. Thus, the waveguide and the outcoupling structures are realized. A 2.5 µm thick PMMA cladding layer is spin-coated on top of the SU-8 3005 waveguide core layer (Figure 2d). Afterwards, a 50 nm thin Ti hard mask is deposited by sputtering and patterned by wet etching in buffered 0.5% hydroﬂuoric acid using a lithography mask realized by spray-coated AZ4999 resist (Figure 2e). Finally, the polymer layer stack of 2.5 µm top cladding PMMA, 200 nm adhesion layer SU-8 2000.5, 2.5 µm PMMA bottom cladding/supporting pedestal layer, 200 nm adhesion layer SU-8 2000.5 and the 5 µm PMMA substrate layer is patterned by lithography with the waveguide probe system, which is etched using the Ti hard mask in an O2 dry etching process at a Sentech SI 500 etch tool with 50 sccm O2, 60 W bias and 300 W ICP down to the SiO2. Afterwards, the Ti hard mask is removed selectively to the polymer stack in a dry etching process with 36 sccm Cl and 15 sccm Ar at 160 W bias and 500 W ICP (Figure 2f). Finally, the LDs are bonded on the 3 µm thick Au bond pads on the ﬂexible polymer-based waveguide probes (Figure 2g). Figure 3a shows the waveguides and a photo of the light-outcoupling structures as Figure 3a shows the waveguides and a photo of the light-outcoupling structures, as well as a detailed microscope image from the region of interest. Polymer-based waveguide probes and light-outcoupling structures are shown as SEM images in Figure 3b as a cross section. An SEM image as a detailed view of the cross section of the polymer waveguide with its cladding and core layers is given in Figure 3c. 4 of 17 ure 3b a er wave Materials 2023, 16, 106 guide with its cladding and core layers is given in Figure 3c. I II III (a) (b) (c) (d) (e) (f) Materials 2022, 15, x FOR PEER REVIEW 5 of (g) Figure 2. Fabrication sequence of the polymer waveguides as cross views for (I) the light-outco pling structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wa is just shown in (a), but is in every step (b–g) part of the flexible polymer waveguide probe. Figure 2. 2.1. Design of the Flexible Polymer-Based Waveguides Fabrication sequence of the polymer waveguides as cross views for (I) the light-outcoupl structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wafer is j shown in (a), but is in every step (b–g) part of the ﬂexible polymer waveguide probe. g g y g g I II III (a) (b) (c) (d) II III (a) (b) (c) (d) (e) (f) aterials 2022, 15, x FOR PEER REVIEW 5 of 1 (g) Figure 2. Fabrication sequence of the polymer waveguides as cross views for (I) the light-outcou pling structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wafe is just shown in (a), but is in every step (b–g) part of the flexible polymer waveguide probe. Figure 2. Fabrication sequence of the polymer waveguides as cross views for (I) the light-outcouplin structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wafer is jus shown in (a), but is in every step (b–g) part of the ﬂexible polymer waveguide probe. (e) (f) aterials 2022, 15, x FOR PEER REVIEW 5 of (e) (f) (g) Figure 2. Fabrication sequence of the polymer waveguides as cross views for (I) the light-outcou- pling structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wafer is just shown in (a), but is in every step (b–g) part of the flexible polymer waveguide probe. Figure 2. Fabrication sequence of the polymer waveguides as cross views for (I) the light-outcoupling structures, (II) the waveguides and (III) the side view of LD coupling side. The 6” silicon wafer is just shown in (a), but is in every step (b–g) part of the ﬂexible polymer waveguide probe. 5 of 17 ight-out silicon w Materials 2023, 16, 106 igure 3. (a) Microscopy image of a fabricated polymer-based waveguide probe and details of oupling structures. SEM images of the cross section of (b) waveguide probe with outcoupling st ures and (c) waveguide facet after scribe-and-break procedure of silicon wafer. Figure 3. (a) Microscopy image of a fabricated polymer-based waveguide probe and details of outcoupling structures. SEM images of the cross section of (b) waveguide probe with outcoupling structures and (c) waveguide facet after scribe-and-break procedure of silicon wafer. gure 3. (a) Microscopy image of a fabricated polymer-based waveguide probe and details of upling structures. W id Ch i i 3. Waveguide Characterization Waveguide Characterization The fully processed wafers (Figure 4a) have to be investigated concerning roughn duced scattering and failures such as scratches and particles enclosed in the diffe yers or process residues. Each of these failures has the potential to disturb the light g g properties of the waveguides. Hence, different methods, which will be introduce The fully processed wafers (Figure 4a) have to be investigated concerning roughness- induced scattering and failures such as scratches and particles enclosed in the different layers or process residues. Each of these failures has the potential to disturb the light guiding properties of the waveguides. Hence, different methods, which will be introduced in the following, are used to identify such failures and optical loss mechanisms. VIEW 6 of 17 he following, are used to identify such failures and optical loss mechanisms. Figure 4. (a) Fully processed wafer with different waveguide probes and test structures; (b) scheme of experimental setup for scattered light detection. Figure 4. (a) Fully processed wafer with different waveguide probes and test structures; (b) scheme of experimental setup for scattered light detection. lures and optical loss mechanisms. e following, are used to identify such fa Figure 4. (a) Fully processed wafer with different waveguide probes and test structures; (b) scheme of experimental setup for scattered light detection. Figure 4. (a) Fully processed wafer with different waveguide probes and test structures; (b) scheme of experimental setup for scattered light detection. 2.1. Design of the Flexible Polymer-Based Waveguides SEM images of the cross section of (b) waveguide probe with outcoupling st es and (c) waveguide facet after scribe-and-break procedure of silicon wafer. Figure 3. (a) Microscopy image of a fabricated polymer-based waveguide probe and details of outcoupling structures. SEM images of the cross section of (b) waveguide probe with outcoupling structures and (c) waveguide facet after scribe-and-break procedure of silicon wafer. 3.2. Scattered Light Detection Setup 3.2. Scattered Light Detection Setup 3.2. Scattered Light Detection Setup For scattered light detection, the wafers were separated ﬁrst into four quarters by the scribe-and-break method. Thus, ﬂat waveguide facets could be achieved, facilitating the incoupling of focused laser light. The scattering light setup is shown in Figure 4b and comprises both a probe and laser stage and a camera system to couple in laser light and observe the probe’s scattered light from a top-view position based on a LabView program. The collimated laser beam is focused onto the scribe-and-break-generated waveguide facet by a mounted aspherical lens (A220TM-A, Thorlabs) with 11 mm focal length and a numerical aperture (NA) of 0.26. The lens has a working distance of 6.91 mm and the lens surface is AR-coated for the spectral range from 350 nm to 700 nm. Probes without gold bonding pads were used for these experiments. By maximizing the signal at the outcoupling structure by means of using the two stages, and thus positioning the waveguide’s facet precisely into the focus point of the asphere, the coupling into the waveguide can be optimized. Once an optimum is reached, the software stores pictures of the whole probe at different exposure times to enhance the dynamic range of the camera in order to evaluate the bright outcoupling sides and the comparably dark waveguide shank simultaneously. For further analyzing the integration time series of images, ﬁrst, a Mathematica script extracts the relevant waveguide parts. Forming the sum of intensities in one pixel column for all columns in every picture, one-dimensional plots of the intensity along the waveguide shank can be obtained. 3.1. Bright- and Dark-Field Microscopy 3.1. Bright- and Dark-Field Microscopy ig a a ie i o opy Optical microscope images were used as a first method of identification of failures in the waveguide structures. Some failures—such as particles—are well detectable with bright-field microscopy. Other ones, such as interface roughness, are visible in dark-field microscopy in a better way. Thus, bright-field and dark-field microscopy were used sup- Optical microscope images were used as a ﬁrst method of identiﬁcation of failures in the waveguide structures. Some failures—such as particles—are well detectable with bright-ﬁeld microscopy. Other ones, such as interface roughness, are visible in dark- ﬁeld microscopy in a better way. Thus, bright-ﬁeld and dark-ﬁeld microscopy were used Materials 2023, 16, 106 6 of 17 supplementarily. In addition, SEM was used to obtain a more detailed view on speciﬁc surface structures, e.g., etching-related damage. supplementarily. In addition, SEM was used to obtain a more detailed view on speciﬁc surface structures, e.g., etching-related damage. 3.3. Probe Diagnostics Different microscopy images of waveguides and outcoupling structures o probes: (a) Dark-field microscopy of a probe’s tip with waveguides and outcoupling stru Microscopy image of an operating probe tip; (c) dark-field microscopy of a probe tip w outcoupling structures; (d) microscopy image of an operating probe tip with missing o structures; (e) dark-field microscopy of an etched waveguide; (f) SEM image of an unin etched waveguide with outcoupling structure. Figure 5. Different microscopy images of waveguides and outcoupling structures of different probes: (a) Dark-ﬁeld microscopy of a probe’s tip with waveguides and outcoupling structures; (b) Mi- croscopy image of an operating probe tip; (c) dark-ﬁeld microscopy of a probe tip with missing outcoupling structures; (d) microscopy image of an operating probe tip with missing outcoupling structures; (e) dark-ﬁeld microscopy of an etched waveguide; (f) SEM image of an unintentionally etched waveguide with outcoupling structure. EVIEW 8 of 17 Figure 5. Different microscopy images of waveguides and outcoupling structures o probes: (a) Dark-field microscopy of a probe’s tip with waveguides and outcoupling stru Microscopy image of an operating probe tip; (c) dark-field microscopy of a probe tip wi outcoupling structures; (d) microscopy image of an operating probe tip with missing o structures; (e) dark-field microscopy of an etched waveguide; (f) SEM image of an unin etched waveguide with outcoupling structure. Figure 5. Different microscopy images of waveguides and outcoupling structures of different probes: (a) Dark-ﬁeld microscopy of a probe’s tip with waveguides and outcoupling structures; (b) Mi- croscopy image of an operating probe tip; (c) dark-ﬁeld microscopy of a probe tip with missing outcoupling structures; (d) microscopy image of an operating probe tip with missing outcoupling structures; (e) dark-ﬁeld microscopy of an etched waveguide; (f) SEM image of an unintentionally etched waveguide with outcoupling structure. VIEW 8 of 17 Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. The incoupling causes light scattering at the waveguide’s facet, which influences the graph on the first centimeter after the facet. Three regions 1), 2) and 3) are identified for linear fitting the scattered light detection curve and determining scattered light attenuation parameters. 3.4. Scattering Loss Analysis Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. 3.3. Probe Diagnostics Figure 5 shows different images of an ideally processed probe. In Figure 5a, the tip of a probe in dark-ﬁeld microscopy is shown. The outcoupling structures can be clearly seen as bright areas at the end of every waveguide, and the waveguide walls are smooth, which is crucial for guiding light with low losses. In Figure 5b, a top-view image of a probe tip is shown when light is coupled into one waveguide. On the right side of the image, the waveguide can be seen. The guided light is slightly scattered at the waveguide walls, which is detected by a camera. Due to back-reﬂected light and partial crosstalk between waveguides, the outcoupling structures of all waveguides appear bright in the image. The most intensive and overexposed outcoupling structure corresponds to the waveguide where light is coupled in. The detailed analysis of respective light intensities can be found in Figure 6. g The dark-ﬁeld microscopy image in Figure 5c shows a probe’s tip with missing out- coupling structures. The outcoupling structures are made of SU-8 (MRT), which is a negative-tone resist, and the single elements have a small footprint of 5 µm each. The development time had to be chosen carefully in order to prevent undercut and related detachment of the small structures. Figure 5d shows outcoupling sides without outcoupling structures. A waveguide with incoupled light is visible on the right hand side. However, at the probe tip, there is no dominant outcoupling position concerning light intensity. A rather diffuse region of brighter light intensities is visible instead. This is due to the missing outcoupling structures at the end of the waveguide, which leads to a scattering of the light at different sides along towards the probe tip at the ends of the neighboring waveguides. In Figure 5e, a dark-ﬁeld microscopy image of a waveguide and the corresponding outcoupling structures is shown. The upper part of the waveguide appears bright, different to the waveguides in Figure 5a,c. In addition, these waveguides show untypically high intensity losses. A further investigation with an SEM showed that the SU-8 core of these waveguides was unintentionally etched due to nonsufﬁcient thickness of the metal hard mask (Figure 5f). Materials 2023, 16, 106 7 of 17 SU-8 cor of the m Figure 5. 3.4. Scattering Loss Analysis 3.4. Scattering Loss Analysis The graph shown in Figure 6 serves as basis for the evaluation of losses caused by scattering. By applying a ﬁt to the scattered light graph in the region of the shank of one probe’s waveguide (red, violet and orange bars in Figure 6), a slope can be determined. This slope corresponds to a scattered light attenuation parameter S, describing the fraction of the light getting lost per distance traveled through a waveguide. The scattered light curve in Figure 6 reveals higher values at the left side in a region of up to 1 cm after the waveguide facet. Some light of the LD is coupled into the cladding and damped fast in the region close to the entrance facet. On the right side of the curve, ten local maxima are detected, indicating the outcoupling structures of the ten waveguide channels. The most intense one is the ﬁfth one, which corresponds to the position of the outcoupling structure of the addressed waveguide. Accordingly, the other maxima correspond to the outcoupling structures of the other nine waveguides, which is caused partially by crosstalk of the waveguides and back-reﬂection of intentionally outcoupled light. Compared to the other nine maxima, the most intense one is between 12 dB to 22 dB brighter. The addressed outcoupling site is about 33 dB brighter than the middle region of the waveguide shank. The low light intensity in this region indicates that light is guided with low intensity loss. The intensity loss in this straight-waveguide region is exclusively caused by scattering. A ﬁt was applied to the scattered light curve in that region to extract a scattered light attenuation parameter from the curve. However due to surface-quality-related leaps, the curve does not show a broad region along the probe’s shank where the application of one linear ﬁt is applicable. Thus, three narrower regions without leaps were identiﬁed to apply separate linear ﬁts (red, violet and orange bars in Figure 6). In these regions, scattered light attenuation parameters of 1.04 dB/cm, 0.78 dB/cm and 0.30 dB/cm could be derived from the red, violet and green ﬁtting curves, respectively. These parameters are reproducible for other waveguides on the same wafer and represent a range of waveguide qualities. 3.3. Probe Diagnostics The incoupling causes light scattering at the waveguide’s facet, which inﬂuences the graph on the ﬁrst centimeter after the facet. Three regions 1), 2) and 3) are identiﬁed for linear ﬁtting the scattered light detection curve and determining scattered light attenuation parameters. Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. The incoupling causes light scattering at the waveguide’s facet, which influences the graph on the first centimeter after the facet. Three regions 1), 2) and 3) are identified for linear fitting the scattered light detection curve and determining scattered light attenuation parameters. 3 4 S i L A l i Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. The incoupling causes light scattering at the waveguide’s facet, which inﬂuences the graph on the ﬁrst centimeter after the facet. Three regions 1), 2) and 3) are identiﬁed for linear ﬁtting the scattered light detection curve and determining scattered light attenuation parameters. Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. The incoupling causes light scattering at the waveguide’s facet, which influences the graph on the first centimeter after the facet. Three regions 1), 2) and 3) are identified for linear fitting the scattered light detection curve and determining scattered light attenuation parameters. 3 4 Scattering Loss Analysis Figure 6. Scattered light intensity along a waveguide probe shank. High values on the right side of the graph indicate the dedicated outcoupling structures of the waveguide. The incoupling causes light scattering at the waveguide’s facet, which inﬂuences the graph on the ﬁrst centimeter after the facet. Three regions 1), 2) and 3) are identiﬁed for linear ﬁtting the scattered light detection curve and determining scattered light attenuation parameters. Materials 2023, 16, 106 8 of 17 8 of 17 3.5. Bending Loss Analysis Figure 7 shows the measured scatter light for a probe with double bends of 0.1 cm radius each at a longitudinal pitch of 5 mm. A locally increasing detected light intensity is caused by light more efﬁciently redirected into the objective of the measurement camera. An on-average steeper decrease in the intensity in the region of the bends can be observed, indicating additional bending losses, which will be examined in detail later. Optical microscope images were used to calculate the contribution of scattering losses and bending losses to the total loss of a bended waveguide. First, the total length l of the waveguide has to be determined. The length of the waveguide without any bends is d. Depending on the number of bends N and the additional length ∆s caused by one bend, the total length l of a waveguide can be calculated as l = d + N·∆s (1) (1) with with (2) ∆s = s −∆d. (2) The straight distance between the starting point and the end point of the bend is ∆d, which is known from the design of the lithography mask. The distance along one bend s = 4θ 360 · u (3) (3) where u is the periphery of a circle, with bending radius r and θ the angle per bend. 9 of 17 Materials 2023, 16, 106 Figure 7. Scattered light graph along four bends from P1 to P2 with radius 0.1 cm of a waveguide. At points P1 and P2, light intensity and position is recorded for further evaluation. Inset: Wave- guides on a wafer with different bending radii and other parameters needed for the evaluation of intensity losses along the waveguides. To extract the total loss from an intensity graph of a test structure, a linear fit is ap- plied between two different maxima of the graph, as shown in Figure 7. The maxima are d b l t i th t t t t With th k l d f th iti P1 Figure 7. Scattered light graph along four bends from P1 to P2 with radius 0.1 cm of a waveguide. At points P1 and P2, light intensity and position is recorded for further evaluation. Inset: Waveguides on a wafer with different bending radii and other parameters needed for the evaluation of intensity losses along the waveguides. Figure 7. 3.5. Bending Loss Analysis Scattered light graph along four bends from P1 to P2 with radius 0.1 cm of a waveguide. At points P1 and P2, light intensity and position is recorded for further evaluation. Inset: Wave- guides on a wafer with different bending radii and other parameters needed for the evaluation of intensity losses along the waveguides. To extract the total loss from an intensity graph of a test structure, a linear fit is ap- plied between two different maxima of the graph, as shown in Figure 7. The maxima are Figure 7. Scattered light graph along four bends from P1 to P2 with radius 0.1 cm of a waveguide. At points P1 and P2, light intensity and position is recorded for further evaluation. Inset: Waveguides on a wafer with different bending radii and other parameters needed for the evaluation of intensity losses along the waveguides. y g g p and P2 of the maxima and the intensities at these points, which can be read directly from the graph, the straight distance d and the total intensity loss between these two points can be identified. Using Equations (1)–(3), the total distance d that the light has to travel through the waveguide can be calculated. Total intensity losses can be determined by To extract the total loss from an intensity graph of a test structure, a linear ﬁt is applied between two different maxima of the graph, as shown in Figure 7. The maxima are caused by several turnings on the test structures. With the knowledge of the positions P1 and P2 of the maxima and the intensities at these points, which can be read directly from the graph, the straight distance d and the total intensity loss between these two points can be identiﬁed. Using Equations (1)–(3), the total distance d that the light has to travel through the waveguide can be calculated. Total intensity losses can be determined by multiplying this distance with the scattering loss of straight waveguides per length. The remaining losses are bending losses. To make bending losses of different bending radii comparable to each other, the bending losses are related to 360◦. αbend, 360◦= αbend 4N · θ · 360◦ (4) (4) In addition, the scattering losses can be related to a full circle of 360◦. In addition, the scattering losses can be related to a full circle of 360◦. 3.5. Bending Loss Analysis αscat, 360◦= 2π · r · 0.78 dB cm (5) (5) Based on the scattered light attenuation parameters from Section 3.4, the total loss and the contributions of scattering and bending losses can be calculated according to the procedure described above. Structure parameters necessary for this calculation are shown in Figure 7. The results from the calculation can be found in Table 1. The bending losses are expected to increase exponentially with decreasing bending radius as the angle under which light is been reﬂected at the inner walls of the waveguide approaches the angle of total reﬂection, especially for higher-order lateral modes. This effect can clearly be found in the data in Table 1. Bending losses become the dominant path of losses for decreasing bending radius. Due to exponentially increasing bending losses, total losses per 360◦increase with the decreasing bending radius. The determination of three attenuation parameters permits the estimation of the total losses of bent waveguides. The dependencies of the different losses on the bending radius and the corresponding ranges are shown in Figure 8. Bending losses are dominant for bending radii up to 2.5 mm and likely beyond. Therefore, it is advantageous to minimize the bending radius in the tapered section at the initial part of the probe (compare Figure 1a) to reduce overall losses. The measured bending losses are for bends in the plane of the probe. Out-of-plane bending losses due to the curvature of the probe when following the geometry inside the cochlea can be estimated from these data, but should be evaluated separately. Materials 2023, 16, 106 10 of 17 Table 1. Structure parameters and scattered light measurement results for PMMA/SU-8/PMMA waveguides on a silicon wafer. Table 1. Structure parameters and scattered light measurement results for PMMA/SU-8/PMMA waveguides on a silicon wafer. 3.5. Bending Loss Analysis Structure Parameters/Measurement Results Radius 1 Radius 2 Radius 3 2.5 mm 1 mm 0.5 mm Geometrical parameters Bending angle θ [◦] 30 45 45 Distance s along one bend [cm] 5.24 3.14 1.57 Straight distance ∆d [cm] 5.0 3.0 1.5 Difference between straight and bent waveguide ∆s [cm] 0.24 0.14 0.07 Number of bends N 2 4 3 Straight distance between P1 and P2 [cm] 1.81 1.81 1.35 Total intensity loss between P1 and P2 [dB] 3.33 13.50 17.00 Total length between P1 and P2 [cm] 1.86 1.87 1.37 Scattered light attenuation parameter S = 1.04 dB/cm Total scattering losses [dB] 2.97 2.99 2.19 Total bending losses [dB] 0.36 10.51 14.81 αscat, 360◦ h dB 360◦ i 2.51 1.01 0.50 αbend, 360◦ h dB 360◦ i 0.54 5.25 9.87 αtotal, 360◦ h dB 360◦ i 3.73 6.43 10.71 Scattered light attenuation parameter S = 0.78 dB/cm Total scattering losses [dB] 1.45 1.46 1.07 Total bending losses [dB] 1.88 12.04 15.93 αscat, 360◦ h dB 360◦ i 1.22 0.49 0.25 αbend, 360◦ h dB 360◦ i 2.83 6.02 10.62 αtotal, 360◦ h dB 360◦ i 4.05 6.51 10.87 Scattered light attenuation parameter S = 0.30 dB/cm Total scattering losses [dB] 0.56 0.57 0.42 Total bending losses [dB] 2.77 12.93 16.59 αscat, 360◦ h dB 360◦ i 0.48 0.19 0.10 αbend, 360◦ h dB 360◦ i 4.15 6.47 11.06 αtotal, 360◦ h dB 360◦ i 4.63 6.66 11.15 IEW 11 of 17 The bending losses are expected to increase exponentially with decreasing bending radius as the angle under which light is been reflected at the inner walls of the waveguide approaches the angle of total reflection, especially for higher-order lateral modes. This effect can clearly be found in the data in Table 1. Bending losses become the dominant path of losses for decreasing bending radius. Due to exponentially increasing bending losses, total losses per 360 ° increase with the decreasing bending radius. The determina- tion of three attenuation parameters permits the estimation of the total losses of bent waveguides. The dependencies of the different losses on the bending radius and the cor- responding ranges are shown in Figure 8. Bending losses are dominant for bending radii up to 2.5 mm and likely beyond. Therefore, it is advantageous to minimize the bending radius in the tapered section at the initial part of the probe (compare Figure 1a) to reduce overall losses. 4. Application of Waveguide Probes A procedure for the preparation of modules was established the so-called single-beam guides (SBGs) for ﬁrst tests of the waveguides in animal models. They comprise an LD as light source; a 3D-printed ABS (acrylonitrile butadiene styrene) shell with a ball lens ﬁxed inside for focusing the light; another 3D-printed holder with the waveguide probe ﬁxed inside; and the LD electrical connections, including an ESD protection diode. For the preparation of an SBG, two parts have to be set up, which are assembled at a later stage. The preparation of the ﬁrst part, the laser module, starts with the 3D printing of the ball lens shell. For printing, an Ultimaker 3 FDM printer was used. EW A 2 mm diameter sapphire ball lens was ﬁxed with black epoxy (E320, Epoxy Technol- ogy) inside the ball lens shell. As the shell deﬁnes the position of the ball lens in front of the laser facet and thus the image distance, the correct ball lens position is crucial for efﬁcient coupling of the laser light. Therefore, the upper part of the shell is pressed to a smooth surface (e.g., an object slide) to allow for a reproducible positioning of the lens inside the shell (Figure 9). After mounting the lens, the black epoxy was applied and cured for 2 h at 65 ◦C in an oven. inside the shell (Figure 9). After mounting the lens, the for 2 h at 65 °C in an oven. Figure 9 Principle of ball lens positioning inside th Figure 9. Principle of ball lens positioning inside the ball lens shell. Figure 9 Principle of ball lens positionin Figure 9. Principle of ball lens positioning inside the ball lens shell. g p p g After applying and curing the transparent epoxy the socket was screwed in a lens tube (SM05L10, Thor a cap (SM05CP1, Thorlabs) with a drilled hole insid transparent epoxy is applied to the thread mounting prevent relative motions of the thread. The transparen active alignment is necessary up to here. All measure sible reproducibility of the geometry. The waveguid with 3D-printing a waveguide probe holder shell and and encapsulation, respectively. At this point, the wa which serves as substrate However to facilitate an im After applying and curing the transparent epoxy ﬁlling the socket for 2 h at 65 ◦C, the socket was screwed in a lens tube (SM05L10, Thorlabs). 3.5. Bending Loss Analysis The measured bending losses are for bends in the plane of the probe. Out- of-plane bending losses due to the curvature of the probe when following the geometry inside the cochlea can be estimated from these data, but should be evaluated separately. Figure 8. Curvature radius dependencies of bending, scattering and total losses. By means of the determination of three different scattered light attenuation parameters, a range for the losses can be derived. 4. Application of Waveguide Probes 4.1. Single-Channel Module Assembly A procedure for the preparation of modules was established the so-called single- Figure 8. Curvature radius dependencies of bending, scattering and total losses. By means of the determi- nation of three different scattered light attenuation parameters, a range for the losses can be derived. igure 8. Curvature radius dependencies of bending, scattering and total losses. By means of the etermination of three different scattered light attenuation parameters, a range for the losses can be erived. Application of Waveguide Probes Figure 8. Curvature radius dependencies of bending, scattering and total losses. By means of the determi- nation of three different scattered light attenuation parameters, a range for the losses can be derived. Materials 2023, 16, 106 11 of 17 11 of 17 4. Application of Waveguide Probes 4.1. Single-Channel Module Assembly 4. Application of Waveguide Probes 4.1. Single-Channel Module Assembly 4. Application of Waveguide Probes EW 13 of 17 After gluing the Si substrate to the probe holder with transparent epoxy and curing it for 2 h at 65 ◦C, the holder is encapsulated with a 3D-printed light-blocking part. Existing joints are closed with black epoxy, hindering light leakage. After this step, the probe module is ready to be assembled together with the LD module. After gluing the Si substrate to the probe holder with transparent epoxy and curing it for 2 h at 65 °C, the holder is encapsulated with a 3D-printed light-blocking part. Exist- ing joints are closed with black epoxy, hindering light leakage. After this step, the probe module is ready to be assembled together with the LD module. For assembling the LD module and the probe module the two parts were fixed on For assembling the LD module and the probe module, the two parts were ﬁxed on an optical table and a manual 5-axis stage, respectively. The waveguide probe was observed with the top-view camera system introduced in Chapter 3.2. After adjusting the laser beam orientation along the optical axis, the adjustment of the probe module in front of the LD module was performed. Once the highest intensity value at the outcoupling point of the waveguide is detected, the probe stage axis was driven back along the optical axis. The transparent epoxy was stirred for about 30 min on a hot plate at 50 ◦C and 1500 rpm to catalyze the epoxy cross-linking process and increase its viscosity, facilitating the application to the glued surface. Afterwards, the epoxy was applied to the bottom side of the probe module. The adjustment process was ﬁnalized by positioning the probe stage axis again at the predetermined ideal distance in respect to the LD module. All the active alignment steps in the second part of the procedure can in principle be automated to allow for production in larger quantities. A schematic side view of the aligned laser and waveguide probe module can be seen in Figure 11. For assembling the LD module and the probe module, the two parts were fixed on an optical table and a manual 5-axis stage, respectively. The waveguide probe was ob- served with the top-view camera system introduced in Chapter 3.2. After adjusting the laser beam orientation along the optical axis, the adjustment of the probe module in front of the LD module was performed. 4. Application of Waveguide Probes The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 ◦C. VIEW 13 of 17 Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To fix the LD in a defined position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 °C. Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To ﬁx the LD in a deﬁned position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 ◦C. IEW 13 of 17 Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To fix the LD in a defined position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 °C. Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To ﬁx the LD in a deﬁned position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 ◦C. 4. Application of Waveguide Probes The rear side was closed with a cap (SM05CP1, Thorlabs) with a drilled hole inside to mount the SMA jack. Finally, transparent epoxy is applied to the thread mounting the laser socket in the lens tube to prevent relative motions of the thread. The transparent epoxy is cured for 2 h at 65 ◦C. No active alignment is necessary up to here. All measures are made to warrant the best possible reproducibility of the geometry. The waveguide probe module preparation starts with 3D-printing a waveguide probe holder shell and another part for holding the probe and encapsulation, respectively. At this point, the waveguide probe is still on the wafer, which serves as substrate. However, to facilitate an implantation into a cochlea, the probe has to be ﬂexible. Indeed, incoupling tests showed that coupling light into a fully detached waveguide without Si substrate at the incoupling side is very difﬁcult. Incoupling becomes more stable and feasible for probes with remaining Si substrate underneath the waveguide facets. As shown in Figure 10a, the probe is partly immersed upside-down into a developer solution (ma-D 533, MRT), which dissolves the sacriﬁcial layer LOR 10 B (MC) between the probe and the Si substrate, and thus removes the probe from the Si substrate. After dissolving the sacriﬁcial layer, the released part of the Si substrate is removed by the scribe-and-break procedure, scribing the edge of the silicon part at the front side (Figure 10b). Materials 2023, 16, 106 12 of 17 e is re- e front 12 of 17 e is re- e front Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To fix the LD in a defined position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. The ball lens shell is slid over the LD’s cap and glued to the socket with transparent epoxy (E301, Epoxy Technology). In the next step, the epoxy is cured for 2 h at 65 °C. Figure 10. (a) Scheme of a waveguide probe immersed upside down in the developer ma-D 533 and (b) scribe-and-break procedure to remove surplus silicon from the released probe. To ﬁx the LD in a deﬁned position, a special socket (S05LM56, Thorlabs) is used where the LD is screwed in. 4. Application of Waveguide Probes Once the highest intensity value at the outcoupling point of the waveguide is detected, the probe stage axis was driven back along the optical axis. The transparent epoxy was stirred for about 30 min on a hot plate at 50 °C and 1500 rpm to catalyze the epoxy cross-linking process and increase its viscosity, facilitating the application to the glued surface. Afterwards, the epoxy was applied to the bottom side of the probe module. The adjustment process was finalized by positioning the probe stage axis again at the predetermined ideal distance in respect to the LD module. All the active alignment steps in the second part of the procedure can in principle be automated to allow for production in larger quantities. A schematic side view of the aligned laser and wave- guide probe module can be seen in Figure 11. Figure 11. Side view of aligned LD and waveguide probe module. The gap between the two mod- ules is filled with transparent epoxy. Figure 11. Side view of aligned LD and waveguide probe module. The gap between the two modules is ﬁlled with transparent epoxy. Figure 11. Side view of aligned LD and waveguide probe module. The gap between the two mod- ules is filled with transparent epoxy. Figure 11. Side view of aligned LD and waveguide probe module. The gap between the two modules is ﬁlled with transparent epoxy. Materials 2023, 16, 106 13 of 17 After curing the epoxy for 24 h at room temperature, the SBG can be released from the optical alignment setup. An example of an operated SBG module is depicted in Figure 12. The output power of the SBG was measured by inserting the waveguide probe into an integrating sphere. In addition, long-term tests of not electrically driven SBGs were conducted by immersing the polymer waveguide into buffered saline solution at 37 ◦C. At temporal intervals the probe was removed from the liquid and inserted into an integrating sphere to remeasure the output power of the SBG. VIEW 14 of 17 Figure 12. SBG in operation. Via a SMA jack, a source–measure unit (SMU) is connected, which electrically drives the SBG. Figure 12. SBG in operation. Via a SMA jack, a source–measure unit (SMU) is connected, which electrically drives the SBG. Figure 12. SBG in operation. Via a SMA jack, a source–measure unit (SMU) is connected, which electrically drives the SBG. Figure 12. 4. Application of Waveguide Probes SBG in operation. Via a SMA jack, a source–measure unit (SMU) is connected, which electrically drives the SBG. 4.2. Module Degradation Tests 4.2. Module Degradation Tests As described at the end of Section 4.1., the SBGs were subjected a lifetime test where the optical output power of the SBGs was measured frequently. The temporal evolution of the output power of four SBG modules is shown in Figure 13. degradation of the geometrical incoupling from the inﬂuence of the saline solution on the waveguide degradation. As described at the end of Section 4.1., the SBGs were subjected a lifetime test where the optical output power of the SBGs was measured frequently. The temporal evolution of the output power of four SBG modules is shown in Figure 13. Figure 13. Fitted temporal trend of the output power of four SBG modules. Except for SBG013, probes were stored in buffered saline solution without operating the LDs. After temporal intervals, the probes were removed from the solution and the optical output power of the SBGs was measured in an integrating sphere. R² values of the four fits are 0.94628 (SBG006), 0.72605 (SBG008), 0.99142 (SBG013) and 0.96180 (SBG014). Figure 13. Fitted temporal trend of the output power of four SBG modules. Except for SBG013, probes were stored in buffered saline solution without operating the LDs. After temporal intervals, the probes were removed from the solution and the optical output power of the SBGs was measured in an integrating sphere. R2 values of the four ﬁts are 0.94628 (SBG006), 0.72605 (SBG008), 0.99142 (SBG013) and 0.96180 (SBG014). Figure 13. Fitted temporal trend of the output power of four SBG modules. Except for SBG013, probes were stored in buffered saline solution without operating the LDs. After temporal intervals, the probes were removed from the solution and the optical output power of the SBGs was measured in an integrating sphere. R² values of the four fits are 0.94628 (SBG006), 0.72605 (SBG008), 0.99142 (SBG013) and 0.96180 (SBG014). Figure 13. Fitted temporal trend of the output power of four SBG modules. Except for SBG013, probes were stored in buffered saline solution without operating the LDs. After temporal intervals, the probes were removed from the solution and the optical output power of the SBGs was measured in an integrating sphere. R2 values of the four ﬁts are 0.94628 (SBG006), 0.72605 (SBG008), 0.99142 (SBG013) and 0.96180 (SBG014). 4.2. Module Degradation Tests 4.2. Module Degradation Tests As described at the end of Section 4.1., the SBGs were subjected a lifetime test where the optical output power of the SBGs was measured frequently. The temporal evolution of the output power of four SBG modules is shown in Figure 13. As described at the end of Section 4.1., the SBGs were subjected a lifetime test where the optical output power of the SBGs was measured frequently. The temporal evolution of the output power of four SBG modules is shown in Figure 13. p p g p p g With about 13 mW output power, SBG013 exhibits the highest optical output power. At an operating current of 92 mA, the output power of the LD used in an SBG in constant wave mode is 40 mW. Hence, the optical efﬁciency of SBG013 is determined to be 33%. Compared to the other SBGs, SBG013 shows a lower decrease in output power due to minimizing the waveguide aging by not storing the probe in saline solution. Thus, the decrease should be exclusively incoupling-alignment-related. It can be enhanced further by employing more advanced lens-coupling solutions compared to the applied ball lenses, e.g., aspheric lenses correcting spherical aberrations. In addition, the output power of each of the four SBGs shown in Figure 13 decreases over time. The main reason for that should be a decrease in the quality of incoupling, as it is spatially very sensitive. One reason for that could be the creep behavior of the epoxy used for ﬁxing the probe shell with the probe in front of the lens shell. Another reason could be a degradation of the 3D-printed shells themselves due to UV exposure from sun or luminescent tubes, because the ABS used for the shells is not UV-resistant. This could lead to a realignment of the relative positions of the probe and the lens, which inﬂuences the incoupling efﬁciency. In addition, the 3D-printed ABS parts should be replaced by more rigid metal solutions for long-term stable applications, and epoxy gluing should be performed with thin joint distances in the range of 0.1 mm to minimize creeping. Further investigations with and without saline solution are necessary to separate the effects of the Materials 2023, 16, 106 14 of 17 14 of 17 degradation of the geometrical incoupling from the inﬂuence of the saline solution on the waveguide degradation. 5. Cochlear Implantation Waveguides were manually coated with up to four layers of silicon and inserted into cochleae of gerbils, an animal model commonly used in cochlear implant research, in a surgery situs after euthanasia. The waveguide probe itself only exhibited a thickness of 10–15 µm of polymer materials. Previous work showed that a deep implantation was beneﬁcial for covering a greater area of the hearing range, which led to less compression of the frequency information and a sufﬁcient amount of usable electrodes, which lastly correlated with an increase in speech recognition to a certain extent [33]. p g The part of the implant that is inserted into the round window should exhibit long- term mechanical stability as well as ﬂexibility to reduce insertion trauma. Yet, to allow for a safe and reproducible implantation, it also needs to have the necessary stiffness [34], causing an interplay of two parameters that needs to be considered. The stiffness is needed to allow for the insertion of the probe without having it bend away from the insertion force, yet at the same time it causes stronger shearing force against the osseous structures of the cochlea and can also cause the implants position to change from the Scala tympani to the Scala vestibuli by piercing the basilary membrane. In line with known parameters of the gerbil cochlea [35], the mechanical parameters of the probe allowed for successful implantation followed by micro-CT imaging in four out of four probes tested so far, yet the probes varied in insertion depth as well as damage caused in the process of implantation, which will be the subject of further investigation in the near future. The probes are already produced with an intrinsic material memory in a spiral shape due to the fabrication, which is beneﬁcial for an easier implantation but also helps to reduce the distance between the outcoupling structures and the neuronal cells in the modiolus, which consequently also leads to an increased spectral resolution. Materials 2023, 16, 106 15 of 17 The focus of further development will therefore also lie on the mechanical properties of the probe. A graded silicone layering of the waveguide probe with lower stiffness at the probes tip compared to its base, as mentioned by [34], could increase the implantation success. Furthermore, an increase in stiffness perpendicular to the natural bending of the implant could be an option to reduce damage through insertion [36]. 6 6. Conclusions A 6” wafer-level micromachining process for the fabrication of flexible probes, made of PMMA as cladding and SU-8 as core material, was establishe of a scattered light detection setup, as well as different microscopy meth bright- and dark-field optical and SEM microscopy, fabricated waveguide p evaluated, concerning failures and intensity loss in operating mode, i.e., whe coupled into the single waveguides. A profound evaluation of the scattered l ment made it possible to determine the contribution of scattering and bendi the total intensity loss of the waveguides. In this work, the scattering attenua eter was determined to be (0 7 ± 0 3) dB/cm A 6” wafer-level micromachining process for the fabrication of ﬂexible waveguide probes, made of PMMA as cladding and SU-8 as core material, was established. By means of a scattered light detection setup, as well as different microscopy methods such as bright- and dark-ﬁeld optical and SEM microscopy, fabricated waveguide probes were evaluated, concerning failures and intensity loss in operating mode, i.e., when light was coupled into the single waveguides. A profound evaluation of the scattered light experiment made it possible to determine the contribution of scattering and bending losses to the total intensity loss of the waveguides. In this work, the scattering attenuation parameter was determined to be (0.7 ± 0.3) dB/cm. eter was determined to be (0.7 ± 0.3) dB/cm. The probes were further processed to establish a single-channel proto designed for the application in animal models. Those probes exhibit up to efficiency at an output power of up to 13 mW. The probes were further processed to establish a single-channel prototype system designed for the application in animal models. Those probes exhibit up to 33% optical efﬁciency at an output power of up to 13 mW. Author Contributions: Conceptualization of manuscript, C.H. and M.R.; waveguide d rial choice and layout, C.G., M.R. and U.T.S.; process development, M.A., F.S., C.G. and characterization setup, M.W. and U.T.S.; optical module design and implementation, M cochlear implantation, J.G. and B.W.; XPCT scans and analysis, J.S., J.G., D.K. and T.S.; m C.H., M.R., M.H., C.G., U.T.S., J.S., T.S., D.K. and B.W.; software, M.W. and C.G.; for M.R. and C.G.; investigation, M.R., M.A., F.S. and D.R.; resources, M.A., F.S. and M.H original draft preparation, C.H. and M.R.; writing—review and editing, C.G., U.T.S., D B.W.; project administration, C.H., M.A., F.S. and D.R.; funding acquisition, C.H., D.R. and T.S. 5. Cochlear Implantation Figure 14 shows an XPCT scan of implanted waveguide structures in the cochlea with an implanted length of 11.5 mm at a total cochlea length of 13 mm, resulting in an implanted cochlea fraction of 88%. 15, x FOR PEER REVIEW Figure 14. Implanted waveguide probe (marked green) after silicone encapsulation image, (a) 3D reconstruction and (b) cross view. Figure 14. Implanted waveguide probe (marked green) after silicone encapsulation in XPCT scan image, (a) 3D reconstruction and (b) cross view. Figure 14. Implanted waveguide probe (marked green) after silicone encapsulation image, (a) 3D reconstruction and (b) cross view. Figure 14. Implanted waveguide probe (marked green) after silicone encapsulation in XPCT scan image, (a) 3D reconstruction and (b) cross view. References 1. Zeng, F.-G. Celebrating the one millionth cochlear implant. JASA Express Lett. 2022, 2, 077201. [CrossRef] [PubMed] 2. Shannon, R.V. Multichannel electrical stimulation of the auditory nerve in man. II. Channel interaction. Hear. Res. 1983, 12, 1–16. [CrossRef] [PubMed] 3. Kral, A.; Hartmann, R. Spatial resolution of cochlear implants: The electrical ﬁeld and excitation of auditory afferents. Hear. Res. 1998 121 11 28 [CrossRef] [PubMed] 1. Zeng, F.-G. Celebrating the one millionth cochlear implant. JASA Express Lett. 2022, 2, 077201. [CrossRef] [PubMed] 2. Shannon, R.V. Multichannel electrical stimulation of the auditory nerve in man. II. Channel interaction. Hear. Res. 1983, 12, 1–16. [CrossRef] [PubMed] 1. Zeng, F.-G. Celebrating the one millionth cochlear implant. JASA Express Lett. 2022, 2, 077201. [CrossRef] [PubMed] 2. Shannon, R.V. Multichannel electrical stimulation of the auditory nerve in man. II. Channel interaction. Hear. Res. 1983, 12, 1–16. [CrossRef] [PubMed] 3 K l A H t R S ti l l ti f hl i l t Th l t i l ﬁld d it ti f dit ff t H R y [CrossRef] [PubMed] 3. Kral, A.; Hartmann, R. Spatial resolution of cochlear implants: The electrical ﬁeld and excitation of auditory afferents. Hear. Res. 1998, 121, 11–28. [CrossRef] [PubMed] [ ] [ ] 3. Kral, A.; Hartmann, R. Spatial resolution of cochlear implants: The electrical ﬁeld and excitation of auditory afferents. Hear. Res. 1998, 121, 11–28. [CrossRef] [PubMed] 4. Zeng, F.-G. Trends in cochlear implants. Trends Amplif. 2004, 8, 1–34. [CrossRef] [PubMed] 5. Miller, C.A.; Abbas, P.J. Electrical excitation of the acoustically sensitive auditory nerve: Singleﬁber responses to electric pulse train. J. Assoc. Res. Otolaryngol. 2006, 7, 195–210. [CrossRef] y g 6. Caldwell, M.T.; Jiam, N.T. Assessment and improvement of sound quality in cochlear implant users. Laryngoscope Investig. Otolaryngol. 2017, 2, 119–124. [CrossRef] 6. Caldwell, M.T.; Jiam, N.T. Assessment and improvement of sound quality in cochlear implant users. Laryngoscope Investig. Otolaryngol. 2017, 2, 119–124. [CrossRef] y g 7. Richter, C.-P.; Rajguru, S.M. Spread of cochlear excitation during stimulation with pulsed infrared radiation: Inferior colliculus measurements. J. Neural Eng. 2011, 8, 056006. [CrossRef] 7. Richter, C.-P.; Rajguru, S.M. Spread of cochlear excitation during stimulation with pulsed infrared radiation: Inferior colliculus measurements. J. Neural Eng. 2011, 8, 056006. [CrossRef] g 8. Hernandez, V.H.; Gehrt, A. Optogenetic stimulation of the auditory pathway. J. Clin. Investig. 2014, 124, 1114–1129. [CrossRef] 9. Dieter, A.; Duque-Afonso, C.J. 6 6. Conclusions Conﬂicts of Interest: T.M., D.K., C.G. and U.S. are cofounders of the OptoGenTech Company. References Near physiological spectral selectivity of cochlear optogenetics. Nat. Commun. 2019, 10, 1962. [CrossRef] 8. Hernandez, V.H.; Gehrt, A. Optogenetic stimulation of the auditory pathway. J. Clin. Investig. 2014, 124, 1114–1129. [CrossRef] 9. Dieter, A.; Duque-Afonso, C.J. Near physiological spectral selectivity of cochlear optogenetics. Nat. Commun. 2019, 10, 1962. [CrossRef] Hernandez, V.H.; Gehrt, A. Optogenetic stimulation [ ] 10. Dieter, A.; Klein, E. µLED-based optical cochlear implants for spectrally selective activation of the auditory nerve. EMBO Mol. Med. 2020, 12, e12387. [CrossRef] [ ] 10. Dieter, A.; Klein, E. µLED-based optical cochlear implants for spectrally selective activation of the auditory nerve. EMBO Mol. Med. 2020, 12, e12387. [CrossRef] 12. Nagel, G.; Szellas, T. Channelrhodopsin-2, a directly light-gated cation-selective membrane channel. Proc. Natl. Acad. Sci. USA 2003, 100, 13940–13945. [CrossRef] [PubMed] 12. Nagel, G.; Szellas, T. Channelrhodopsin-2, a directly light-gated cation-selective membrane channel. Proc. Natl. Acad. Sci. USA 2003, 100, 13940–13945. [CrossRef] [PubMed] 13. Boyden, E.S.; Zhang, F. Millisecond-timescale, genetically targeted optical control of neural activity. Nat. Neurosci. 2005, 8, 1263–1268. [CrossRef] [PubMed] 14. Goßler, C.; Bierbrauer, C. GaN-based micro-LED arrays on ﬂexible substrates for optical cochlear implants. J. Phys. D Appl. Phys. 2014, 47, 205401. [CrossRef] 15. Klein, E.; Goßler, C. High-density µLED-based optical cochlear implant with improved thermo Neurosci. 2018, 12, 659. [CrossRef] , , [ ] 16. Reddy, J.W.; Lassiter, M. Parylene photonics: A ﬂexible, broadband optical waveguide platform with integrated micromirrors for biointerfaces. Microsyst. Nanoeng. 2020, 6, 85. [CrossRef] 16. Reddy, J.W.; Lassiter, M. Parylene photonics: A ﬂexible, broadband optical waveguide platform biointerfaces. Microsyst. Nanoeng. 2020, 6, 85. [CrossRef] 17. Schwaerzle, M.; Paul, O. Compact silicon-based optrode with integrated laser diode chips, SU-8 waveguides and platinum electrodes for optogenetic applications. J. Micromech. Microeng. 2017, 27, 065004. [CrossRef] 18. Kampasi, K.; English, D.F. Dual color optogenetic control of neural populations using low-noise, multishank optoelectrodes. Microsyst. Nanoeng. 2018, 4, 10. [CrossRef] 19. Marcatili, E.A. Bends in optical dielectric guides. Bell Syst. Tech. J. 1969, 48, 2103–2132. [CrossRef] 20. Marcuse, D. Inﬂuence of curvature on the losses of doubly clad ﬁbers. Appl. Opt. 1982, 21, 4208–4213. [CrossRef] 20. Marcuse, D. Inﬂuence of curvature on the losses of doubly clad ﬁbers. Appl. Opt. 1982, 21, 4 g g p p g g y 22. Lacey, J.P.; Payne, F.P. Radiation loss from planar waveguides with random wall imperfections. IEE Proc. J. (Optoelectron.) 1990, 137, 282–289. [CrossRef] 23. Payne, F.P.; Lacey, J.P. 6 6. Conclusions All authors have read and agreed to the published version of the manuscript. Author Contributions: Conceptualization of manuscript, C.H. and M.R.; waveguide design, material choice and layout, C.G., M.R. and U.T.S.; process development, M.A., F.S., C.G. and D.R.; optics characterization setup, M.W. and U.T.S.; optical module design and implementation, M.R. and C.G.; cochlear implantation, J.G. and B.W.; XPCT scans and analysis, J.S., J.G., D.K. and T.S.; methodology, C.H., M.R., M.H., C.G., U.T.S., J.S., T.S., D.K. and B.W.; software, M.W. and C.G.; formal analysis, M.R. and C.G.; investigation, M.R., M.A., F.S. and D.R.; resources, M.A., F.S. and M.H.; writing—original draft preparation, C.H. and M.R.; writing—review and editing, C.G., U.T.S., D.R., T.M. and B.W.; project administration, C.H., M.A., F.S. and D.R.; funding acquisition, C.H., D.R., U.T.S., T.M. and T.S. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the Photonik Inkubator GmbH and the German Education and Research (BMBF) in the project OptoGenTech (13N14622). J.G. received of the Studienstiftung des Deutschen Volkes. The research was funded by Deutsche F meinschaft (DFG, German Research Foundation) under Germany’s Excellence St 2067/1- 390729940 to T.S. and T.M. as well as by the Leibniz Program of the DFG to search is supported by Fondation Pour l’Audition (FPA RD-2020-10). Funding: This work was supported by the Photonik Inkubator GmbH and the German Ministry for Education and Research (BMBF) in the project OptoGenTech (13N14622). J.G. received a scholarship of the Studienstiftung des Deutschen Volkes. The research was funded by Deutsche Forschungs- gemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy—EXC 2067/1-390729940 to T.S. and T.M. as well as by the Leibniz Program of the DFG to T.M. This research is supported by Fondation Pour l’Audition (FPA RD-2020-10). Materials 2023, 16, 106 16 of 17 Informed Consent Statement: The animal study protocol was approved by the Animal Welfare Ofﬁce of the University Medical Center Goettingen. Informed Consent Statement: The animal study protocol was approved by the Animal Welfare Ofﬁce of the University Medical Center Goettingen. Informed Consent Statement: The animal study protocol was approved by the Animal Welfare Ofﬁce of the University Medical Center Goettingen. Acknowledgments: The authors thank René Krone for technical support and Christoph Robert Meinecke for reviewing and supporting with the article. Acknowledgments: The authors thank René Krone for technical support and Christoph Robert Meinecke for reviewing and supporting with the article. 31. Yamagiwa, S.; Ishida, M. Flexible optrode array: Parylene-ﬁlm waveguide arrays with microelectrodes for optogenetics. In Proceedings of the 2015 Transducers—2015 18th International Conference on Solid-State Sensors, Actuators and Microsystems (TRANSDUCERS), Anchorage, AK, USA, 21–25 June 2015; pp. 277–280. References A theoretical analysis of scattering loss from planar optical waveguides. Opt. Quantum Electron. 1994, 26, 977–986. [CrossRef] [ ] 24. Wang, X.-B.; Sun, J. Thermal UV treatment on SU-8 polymer for integrated optics. Opt. Mater. Express 2014, 4, 509–517. [CrossRef] 25. Sun, X.; Xie, Y. Effect of ﬁlm compatibility on electro-optic properties of dye doped polymer DR1/SU-8. Appl. Surf. Sci. 2013, 285, 469–476. [CrossRef] 24. Wang, X.-B.; Sun, J. Thermal UV treatment on SU-8 polymer for integrated optics. Opt. Mater. Expres Wang, X.-B.; Sun, J. Thermal UV treatment on SU-8 p g p y g p p p 25. Sun, X.; Xie, Y. Effect of ﬁlm compatibility on electro-optic properties of dye doped polymer DR1/SU-8. Appl. Surf. Sci. 2013, 285, 469–476. [CrossRef] [ ] 26. Nordstrom, M.; Zauner, D.A. Single-Mode Waveguides With SU-8 Polymer Core and Cladding for MOEMS Applications. J. Light. Technol. 2017, 25, 1284–1289. [CrossRef] 27. Alt, M.T.; Fiedler, E. Let There Be Light-Optoprobes for Neural Implants. Proc. IEEE 2016, 105, 101–138. [CrossRef] 28. Leal-Junior, A.G.; Marques, C. FBG-Embedded 3-D Printed ABS Sensing Pads: The Impact of Inﬁll Density on Sensitivity and Dynamic Range in Force Sensors. IEEE Sens. J. 2018, 18, 8381–8388. [CrossRef] g p p p [ ] 28. Leal-Junior, A.G.; Marques, C. FBG-Embedded 3-D Printed ABS Sensing Pads: The Impact of Inﬁll Density on Sensitivity and Dynamic Range in Force Sensors. IEEE Sens. J. 2018, 18, 8381–8388. [CrossRef] y g J 29. Cai, S.; Pan, H. Selective detection of cadmium ions using plasmonic optical ﬁber gratings functionalized with bacteria. Opt. Express 2020, 28, 19740–19749. [CrossRef] p 30. Rahlves, M.; Rezem, M.; Günther, A. Towards fabrication and application of polymer based photonics networks and sensors. In Proceedings of the SPIE 10545, MOEMS and Miniaturized Systems XVII, San Francisco, CA, USA, 22 February 2018. Materials 2023, 16, 106 17 of 17 17 of 17 31. Yamagiwa, S.; Ishida, M. Flexible optrode array: Parylene-ﬁlm waveguide arrays with microelectrodes for optogenetics. In Proceedings of the 2015 Transducers—2015 18th International Conference on Solid-State Sensors, Actuators and Microsystems (TRANSDUCERS), Anchorage, AK, USA, 21–25 June 2015; pp. 277–280. 33. Baskent, D.; Shannon, R.V. Interactions between cochlear implant electrode insertion depth and frequency-place mapping. J. Acoust. Soc. Am. 2005, 117, 1405. [CrossRef] 34. Stöver, T.; Lenarz, T. Biomaterials in cochlear implants. GMS Curr. Top. Otorhinolaryngol. Head Neck Surg. 2009, 8, Doc10. [PubMed] 34. Stöver, T.; Lenarz, T. Biomaterials in cochlear implants. GMS Curr. 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 Top. Otorhinolaryngol. Head Neck Surg. 2009, 8, Doc10. [PubMed] [ ] 35. Keppeler, D.; Kampshoff, C.A. Multiscale photonic imaging of the native and implanted cochlea. Proc. Natl. Acad. Sci. USA 2021, 118, e2014472118. [CrossRef] [PubMed] 35. Keppeler, D.; Kampshoff, C.A. Multiscale photonic imaging of the native and implanted cochlea. Proc. Natl. Acad. Sci. USA 2021, 118, e2014472118. [CrossRef] [PubMed] , [ ] [ ] 36. Rebscher, S.J.; Hetherington, A. Considerations for the design of future cochlear implant electrode arrays: Electrode array stiffness, size and depth of insertion. J. Rehabil. Res. Dev. 2008, 45, 731–748. [CrossRef] [PubMed] 36. Rebscher, S.J.; Hetherington, A. Considerations for the design of future cochlear implant electrode arrays: Electrode array stiffness, size and depth of insertion. J. Rehabil. Res. Dev. 2008, 45, 731–748. [CrossRef] [PubMed] 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. 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/W3176521928	https://eprints.ncl.ac.uk/fulltext.aspx?url=275851/E9633832-CA2A-43BB-982A-9869BC5C7577.pdf&pub_id=275851	English	null	EUROlinkCAT protocol for a European population-based data linkage study investigating the survival, morbidity and education of children with congenital anomalies	BMJ open	2,021	cc-by	12,660	To cite: Morris JK, Garne E, Loane M, et al. EUROlinkCAT protocol for a European population-based data linkage study investigating the survival, morbidity and education of children with congenital anomalies. BMJ Open 2021;11:e047859. doi:10.1136/ bmjopen-2020-047859 Strengths and limitations of this study Introduction Congenital anomalies (CAs) are a major cause of infant mortality, childhood morbidity and long- term disability. Over 130 000 children born in Europe every year will have a CA. This paper describes the EUROlinkCAT study, which is investigating the health and educational outcomes of children with CAs for the first 10 years of their lives. ► ►The implementation of a common data model en- ables the same centrally developed syntax script to be run in all registries which is efficient and ensures standardisation of analysis. on July 12, 2021 by guest. Protected by co http://bmjopen.bmj.com/ e 2021. Downloaded from ► ►The use of a reference population allows country differences to be adjusted for and enables more accurate comparisons of the burden of disease attributable to congenital anomalies (CAs) across countries to be made. Methods and analysis EUROCAT is a European network of population-based registries for the epidemiological surveillance of CAs. EUROlinkCAT is using the EUROCAT infrastructure to support 22 EUROCAT registries in 14 countries to link their data on births with CAs to mortality, hospital discharge, prescription and educational databases. Once linked, each registry transforms their case data into a common data model (CDM) format and they are then supplied with common STATA syntax scripts to analyse their data. The resulting aggregate tables and analysis results are submitted to a central results repository (CRR) and meta-analyses are performed to summarise the results across all registries. The CRR currently contains data on 155 594 children with a CA followed up to age 10 from a population of 6 million births from 1995 to 2014. ► ►Prepublication history and supplemental material for this paper is available online. To view these files, please visit the journal online (http://dx.doi. org/10.1136/bmjopen-2020- 047859). ► ►Merging births with CAs to their records in routine healthcare data enables a detailed evaluation of the accuracy of the data and enables improvements to be suggested. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ ► ►Being unable to share individual case data or aggre- gate data that might be disclosive reduces the pow- er of the analysis that can be performed, particularly for rare anomalies. Received 14 December 2020 Accepted 07 May 2021 ► ►Only specific areas in Europe are represented, with a lack of data in particular from Eastern Europe. Protocol Protocol Open access EUROlinkCAT protocol for a European population-based data linkage study investigating the survival, morbidity and education of children with congenital anomalies EUROlinkCAT protocol for a European population-based data linkage study investigating the survival, morbidity and education of children with congenital anomalies BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021 on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Joan K Morris ‍ ‍ ,1 Ester Garne ‍ ‍ ,2 Maria Loane ‍ ‍ ,3 Ingeborg Barisic ‍ ‍ ,4 James Densem,5 Anna Latos-Bieleńska ‍ ‍ ,6 Amanda Neville,7 Anna Pierini ‍ ‍ ,8 Judith Rankin ‍ ‍ ,9 Anke Rissmann ‍ ‍ ,10 Hermien de Walle ‍ ‍ ,11 Joachim Tan ‍ ‍ ,1 Joanne Emma Given ‍ ‍ ,3 Hugh Claridge ‍ ‍ ,1 EUROlinkCAT Consortium Received 14 December 2020 Accepted 07 May 2021 INTRODUCTION Ethics The CA registries have the required ethics permissions for routine surveillance and transmission of anonymised data to the EUROCAT central database. Each registry is responsible for applying for and obtaining additional ethics and other permissions required for their participation in EUROlinkCAT. Over 130 000 children born in Europe every year will have a major congenital anomaly (CA), equivalent to 2.5% of all European births. CAs include structural defects, chro- mosomal anomalies and genetic syndromes. CAs are a leading cause of perinatal and infant mortality, especially in developed countries.1 From 2003 to 2012, CAs were asso- ciated with about 40% of all infant deaths in Sweden and England.2 There is a large vari- ation in child death rates across Europe; in 2013 the child death rates (age 0–14 years) were 60% higher in the UK and Belgium compared with Sweden, with an additional 10 countries being 30% higher than Sweden.3 For numbered affiliations see end of article. Correspondence to Professor Joan K Morris; jmorris@sgul.ac.uk © Author(s) (or their employer(s)) 2021. Re-use permitted under CC BY. Published by BMJ. Dissemination The CDM and associated documentation, including linkage and standardisation procedures, will be available post-EUROlinkCAT thus facilitating future local, national and European-level analyses to improve healthcare. Recommendations to improve the accuracy of routinely collected data will be made. guest. Protected by copyright. Findings will provide evidence to inform parents, health professionals, public health authorities and national treatment guidelines to optimise diagnosis, prevention and treatment for these children with a view to reducing health inequalities in Europe. 1 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Open access on July 12, 2021 by gue http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from To identify potentially preventable and remedial causes it is important to investigate the health inequalities in survival in children with CA across Europe. include 39 registries in 21 countries covering more than 29% of European births (1.7 million) per year.22–24 The main objectives of EUROCAT are to provide essential epidemiological information and surveillance on CAs in Europe, to evaluate the effectiveness of primary preven- tion and to assess the impact of developments in prenatal screening.25 26 Hence, the emphasis is on information collected up to a baby’s first year of life. INTRODUCTION It has been shown that relying on death certificates as a source of information on mortality due to CAs does not provide an accurate assessment of the survival for children with specific CAs. Death certificates state the direct or primary cause of death which may be infection, seizures or others and therefore may not mention the CA.4 Copeland and Kirby4 concluded that the only way to accurately study mortality and survival in children with rare CAs is to pool data across CA registries and link these to death registries. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from p y y The aim of EUROlinkCAT is to investigate the survival, morbidity and educational outcomes of children with specific CAs for the first 10 years of their lives by linking births with CAs in EUROCAT registries to electronic healthcare and education databases. The availability of population-based data on births with CAs across the EUROCAT network will enable survival, morbidity and education to be investigated for specific CAs as well as differences in these outcomes across Europe according to specific risk factors and social inequalities to be explored. Advances in fetal, neonatal and paediatric care have improved outcomes for individuals with some CAs, for example, Down syndrome (DS)5–7 and cardiac anoma- lies. Several studies have shown that children with CAs account for a very high proportion of all hospital admis- sions.8 9 However, there is a lack of information on the length of hospital stays for children with specific CAs, with most studies concerning children with DS, orofacial clefts or congenital heart defects (CHD).10 Often hospital stays are investigated for the first 2 or 3 years of a child’s life.11–14 However, Wehby et al showed that hospital admis- sions for those born with orofacial clefts were increased at all ages up to 60 years of age.15 Rarely has length of hospital stay been related to other factors, such as social class. Two studies (Derrington 2013 in the USA to 3 years of age and Hung 2011 in Taiwan for all ages) both iden- tified other factors such as ethnicity and socioeconomic factors as important influences on the length of inpatient stays in children and adults with DS.12 16 Electronic healthcare data are increasingly being used by researchers to investigate the epidemiology of CAs, rather than using information from CA registries. INTRODUCTION Such healthcare data have often been found to be incom- plete.27–31 A small number of registries will analyse the maternal pregnancy records for women registered as having had a termination of pregnancy for a fetal anomaly (TOPFA) in EUROCAT. This will enable the accuracy of routine information on TOPFAs to be evalu- ated. The accuracy of CA coding in live births will be eval- uated by comparing the EUROCAT data for live births with the CA diagnosis from the electronic healthcare databases covering both inpatient and outpatient visits. The information on death certificates will be compared with the anomalies recorded in EUROCAT. Recommen- dations will be developed to enable the maximum infor- mation from electronic healthcare data to be extracted for research purposes and to quantify the amount of data that cannot be obtained. on July 12, 2021 by guest. Protected by copy http://bmjopen.bmj.com/ 21. Downloaded from y The proportion of children born with a CA surviving beyond infancy is increasing.5 6 How these children are performing in school and their additional educational needs is therefore becoming increasingly important as there may be a growing population of children and young people requiring additional support and resources in the future. However, apart from the more common genetic syndromes, there is a paucity of information about this.17 18 The American Heart Association reviewed the literature on children with CHD and concluded that they are at an increased risk of developmental delay, even once the frequent occurrence of genetic syndromes has been taken into account, particularly for neonates or infants requiring open heart surgery.19 Wehby et al also showed that children with isolated orofacial clefts were at a much greater risk of low achievement at school than their classmates.20 A systematic review of neurocognitive outcomes following general anaesthesia and surgery in children concluded that exposure to general anaesthesia in young children did affect their development in some neurocognitive domains.21 However, the authors recom- mended that the effects of surgery should be considered separately for each specific anomaly. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ This paper describes the design of the study, the methods used to obtain and analyse the linked data and evaluates the first 3 years’ progress of EUROlinkCAT. METHODS AND ANALYSIS Design and setting In 2017, all EUROCAT registries were invited to partic- ipate in the EUROlinkCAT study. Twenty-two registries from 14 countries agreed to participate and to link all live births with a CA registered in their registries and born from 1 January 1995 to 31 December 2014. Almost all EUROCAT registries send anonymised data on CAs occurring in all live births, fetal deaths from 20 weeks gestation and TOPFAs to the EUROCAT central data- base. Comprehensive coding instructions32 and the use of the EUROCAT Data Management Programme (EDMP) to import data into the central database ensure that stan- dard variables, definitions and coding are used by all registries in the network. CAs are coded locally using the WHO International Statistical Classification of Diseases guest. Protected by copyright. EUROCAT (https://eu-rd-platform.jrc.ec.europa.eu/ eurocat) is a European network of population-based CA registries, which started in 1979 and has expanded to EUROCAT (https://eu-rd-platform.jrc.ec.europa.eu/ eurocat) is a European network of population-based CA registries, which started in 1979 and has expanded to 2 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Open access on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from and Related Health Problems (ICD) 9th or 10th Revi- sion with the British Paediatric Association code exten- sion offering more specificity (table 1). Cases with minor anomalies only are excluded (see EUROCAT Guide 1.4, Minor Anomalies for Exclusion (V.14.10.14)). Registries can code up to nine anomalies for each case and provide additional information in the specified text fields. Based on the ICD-9 or ICD-10 codes present, cases are automat- ically assigned by EDMP to defined major CA subgroups in accordance with the EUROCAT Guide 1.4. A case with more than one major anomaly may be assigned to more than one subgroup. Since 2015, the central database has been hosted by the European Commission Joint Research Centre in Ispra (Italy). to link their information on children with CAs to educa- tion records. To evaluate the accuracy of the routine healthcare data, five registries are additionally linking to outpatient data and four will also link to pregnancy infor- mation recorded in the mother’s health records about TOPFAs. The 19 registries survey over 6 million births in the population. For the evaluation of survival and morbidity, each child will be followed up for a maximum of 10 years. Standardisation and common data models Standardisation and common data models EUROCAT registries submit 96 core and non-core vari- ables to the EUROCAT central database providing pseudonymised information on the baby and mother, diagnosis, karyotype (if known), exposure, family history and sociodemographic details. These have already been standardised and table 5 lists the 52 variables of which 34 are core variables and their common coding scheme that are used in the EUROlinkCAT study. In contrast, all the data obtained from linkage have to be standardised to a common format, as the healthcare and educational systems across Europe use different native languages and coding classification schemes. To do this, each registry provided their data dictionaries describing the variables in their local databases including the variable names, format, definitions and coding schemes. METHODS AND ANALYSIS Design and setting This age cut-off has been chosen to enable enough children to be identified and followed up; a longer follow-up would mean fewer children would be eligible as currently national or local electronic healthcare record sources often do not go back more than 10 years. For education the maximum follow-up is until the end of compulsory school age (typically 16 across participating countries), although for some registers data are only available for a shorter period of follow-up. The longer follow-up was chosen, because in Finland there are no national educa- tion tests and national data on education attainment are available at the age of 15–16 (9th grade). p ( y) Tables 1 and 2 provide a list of the 81 EUROlinkCAT CA subgroups which include structural anomalies, genetic syndromes and chromosomal anomalies that will be investigated. There are 60 EUROCAT subgroups (table 1) and an additional 21 new CA subgroups not defined in EUROCAT (table 2). The subgroups have been identified as being reasonably homogeneous to provide meaningful information and also to be preva- lent enough to enable sufficiently precise estimates to be obtained from the analyses. For example, the EUROCAT subgroup ‘Chromosomal’ was not included as it includes all genetic syndromes, but specific syndromes such as DS (a EUROCAT subgroup) and Di George syndrome (a new subgroup) were included. For some analysis, such as mortality, DS children will be analysed according to whether they have a cardiac anomaly and/or a gastro- intestinal anomaly, as these are common and are likely to influence their survival. As outcomes are expected to be more severe for children with multiple and more complex CAs, analyses are also performed separately for children with isolated anomalies or with multiple anom- alies defined according to the methodology by Garne et al.33 Isolated anomalies are defined as a CA in one organ system only or with a known sequence where multiple CAs cascade as a consequence of a single primary anomaly. Multiple anomalies are defined as two or more major structural CAs in different organ systems, where the pattern of anomalies has not been recognised as part of a syndrome or sequence. Reference population Where possible, each registry obtained information from electronic healthcare records, prescription records and education records on children without a CA. The defini- tion of these ‘control’ cohorts will vary according to the registry, ranging from all children in the same population covered by the registry to a 10% random sample of chil- dren stratified by birth year and child’s sex. The use of such a reference population is essential in interpreting differences across countries, as it will provide information on key outcomes, such as duration of hospital stays and medication prescribing, on children without reported anomalies, which is expected to vary by country. Table 4 provides information on the reference populations being identified. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ m Linkage Table 3 provides details of the linkages originally planned by the EUROCAT registries and the current linkages occurring (as of August 2020). The reasons why some registries could not link their data are explored in detail in another paper submitted for publication—they include not being able to obtain the necessary permissions, rele- vant outcomes not being recorded in specific data sources and the time scale for the data supply being after the end of the study’s funding. Currently, 19 registries are linking their data to mortality records, 15 plan to link to hospital in-patient records and 7 to prescription records for the work package that will consider morbidity for children born with a CA. At the time of writing, nine registries plan guest. Protected by copyright. Table 6 shows how the variable identifying the sex (male and female) of the child is coded in the different registries with different variable names, different formats and different coding schemes. For each substudy in Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 3 on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from on July 12, 2021 http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from EUROCAT subgroups ICD-10-BPA ICD-9-BPA Trisomy 18 Q910-Q913 7582 Turner syndrome Q96 75860, 75861, 75862, 75869 Klinefelter syndrome Q980–Q984 7587 Rare structural anomalies with a EUROCAT subgroup Encephalocele Q01 7420 Arhinencephaly/ holoprosencephaly Q041, Q042 74226 Anophthalmos/microphthalmos Q110, Q111, Q112 7430, 7431 Anophthalmos Q110, Q111 7430 Congenital glaucoma Q150 74320 Anotia Q160 74401 Common arterial truncus Q200 74500 Double outlet right ventricle Q201 No code Single ventricle Q204 7453 Triscuspid atresia and stenosis Q224 7461 Ebstein’s anomaly Q225 7462 Pulmonary valve atresia Q220 74600 Hypoplastic right heart Q226 No code Aortic atresia/interrupted aortic arch Q252 74720 Total anomalous pulmonary venous return Q262 74742 Choanal atresia Q300 7480 Hirschsprung’s disease Q431 75130–75133 Atresia of bile ducts Q442 75165 Annular pancreas Q451 75172 Indeterminate sex Q56 7527 Situs inversus Q893 7593 VATER/VACTERL Q8726 759895 All anomalies=all cases of congenital anomaly, excluding cases with only minor anomalies as defined in Section 3.2 in EUROCAT Guide 1.4 for cases born post-2005. Cases with more than one anomaly are only counted once in the ‘all Anomalies’ subgroup. †ICD10 code D1810 (ICD9 code 2281) is the code for cystic hygroma. GA, gestational age; ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. Table 1 Continued on Ju http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from All anomalies=all cases of congenital anomaly, excluding cases with only minor anomalies as defined in Section 3.2 in EUROCAT Guide 1.4 for cases born post-2005. Cases with more than one anomaly are only counted once in the ‘all Anomalies’ subgroup. †ICD10 code D1810 (ICD9 code 2281) is the code for cystic hygroma. GA, gestational age; ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. All anomalies=all cases of congenital anomaly, excluding cases with only minor anomalies as defined in Section 3.2 in EUROCAT Guide 1.4 for cases born post-2005. Cases with more than one anomaly are only counted once in the ‘all Anomalies’ subgroup. †ICD10 code D1810 (ICD9 code 2281) is the code for cystic hygroma. on July 12, 2021 by guest. Prot http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from p Table 1 EUROCAT congenital anomaly subgroups in EUROlinkCAT EUROCAT subgroups ICD-10-BPA ICD-9-BPA All anomalies Q-chapter, D215, D821, D1810†, P350, P351, P371 74, 75, 27910, 2281†, 76076, 76280, 7710, 7711, 77121 Structural anomalies Spina bifida Q05 741 Hydrocephalus Q03 7423 Severe microcephaly Q02 7421 Congenital cataract Q120 74 332 Congenital heart defects (CHD) Q20–Q26 745, 746, 7470–7474 Severe CHD Q200, Q201, Q203, Q204, Q212, Q213, Q220, Q224, Q225, Q226, Q230, Q232, Q233, Q234, Q251, Q252, Q262 74500, 74510, 7452, 7453, 7456, 7461, 7462, 74600, 7463, 7465, 7466, 7467, 7471, 74720, 74742 Transposition of great vessels Q203 74510 Ventricular septal defect Q210 7454 Atrial septal defect (ASD) Q211 7455 Atrialventricular septal defect (AVSD) Q212 7456 Tetralogy of Fallot Q213 7452 Pulmonary valve stenosis Q221 74601 Aortic valve atresia/stenosis Q230 7463 Mitral valve anomalies Q232, Q233 7465, 7466 Hypoplastic left heart Q234 7467 Coarctation of aorta Q251 7471 Patent ductus arteriosus (PDA) as only CHD in term infants (GA+37 weeks) Q250 7470 Cystic adenomatous malformation of lung Q3380 No code Cleft lip with or without cleft palate Q36, Q37 7491, 7492 Cleft palate Q35 7490 Oesophageal atresia with/without trachea-oesophageal fistula Q390–Q391 75030–75031 Duodenal atresia or stenosis Q410 75110 Atresia or stenosis of other parts of small intestine Q411–Q418 75111–75112 Ano-rectal atresia and stenosis Q420–Q423 75 21–75124 Diaphragmatic hernia Q790 75661 Gastroschisis Q793 75671 Omphalocele Q792 75670 Multicystic renal dysplasia Q6140, Q6141 75316 Congenital hydronephrosis Q620 75320 Hypospadias Q54 75260 Limb reduction defects Q71-Q73 7552–7554 Craniosynostosis Q750 75600 Chromosomal anomalies Down syndrome Q90 7580 Trisomy 13 Q914–Q917 7581 on July 12, 2021 by guest. Protected by copyr http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from on July 12, http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 4 on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from GA, gestational age; ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. uly 12, 2021 by guest. Protected by copyright. The CDMs also specify how the data are stored. For mortality, all the relevant variables occur only once for each EUROCAT case and are stored in the same data file (or table) (figure 1). However, when analysing hospital admissions, each child may have more than one admission and for each admission may receive more than one diag- nosis. Therefore, the hospital admissions data are stored in a separate data file (or table) from the diagnoses data and separately from the EUROCAT data on the child; each data file contains a reference key which serves to link all records belonging to one person for analysis (see figure 2). The standardisation syntax scripts from UU specify the separate data files (or tables) for each linkage containing all the variables in the CDMs. EUROlinkCAT, a common data model (CDM) containing all variables required for its analyses were developed. All the EUROlinkCAT CDMs contain the variable L_CH_ SEX, defined as ‘sex of child’ and with a coding scheme in integer format of 1=male, 2=female, 3=indeterminate, 9=not known ‘.’=not recorded or not available for study. Ulster University (UU) used the information in each registry’s data dictionary to create the new EUROlinkCAT ‘standardised’ L_CH_SEX variable. UU, in collaboration with the registries, created registry-specific syntax scripts to standardise all the variables in the EUROlinkCAT CDMs. Online supplemental appendices 1–2 list all the variables included in the substudies (mortality, electronic healthcare records and prescription records). y guest. Protected by copyright. 5 Morris JK, et al. BMJ Open 2021;11:e047859. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from doi:10.1136/bmjopen-2020-047859 Open access Table 2 New congenital anomaly subgroups in EUROlinkCAT New subgroups for EUROlinkCAT ICD-10-BPA ICD-9-BPA Structural anomalies Anomalies of corpus callosum Q040 74221 Anomalies of intestinal fixation Q433 7514 Unilateral renal agenesis Q600 No code Accessory kidney Q630 75330 Bladder exstrophy Q641 7535 Epispadia Q640 75261 Posterior urethral valves Q6420 75360 Prune Belly Q794 75672 Arthrogryposis multiplex congenita Q743 75580 Genetic syndromes Di George syndrome D821 27910 Goldenhar syndrome Q8704 75606 Cornelia de Lange syndrome Q8712 759821 Noonan syndrome Q8714 759896 Prader-Willi Q8715 759872 Beckwith-Wiedemann syndrome Q8730 759874 Williams syndrome Q8784 No code Angelman syndrome Q8785 No code Chromosomal anomalies Wolff-Hirschhorn syndrome Q933 75832 Cri-du chat syndrome Q934 75831 Karyotype XXX Q970 75885 Sequences Pierre-Robin sequence Q8708 75603 ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. Work of the Standardisation Committee In addition to defining the CDM and its structure, the EUROlinkCAT Standardisation Committee was respon- sible for taking other decisions, usually in consideration of local data characteristics, to ensure that data were meaningful and comparable across registries. The most important issues are listed below: 1. Inclusion according to gestational age (GA) at birth: for the mortality study all live births with a GA below 24 weeks were excluded, as these cases could have been miscoded terminations of pregnancy with signs of life at birth. However, after running the mortality analysis, it was noticed locally that there were survi- vors in EUROCAT registries at GA 23 weeks. For the morbidity studies, the exclusion criterion was low- ered to be below 23 weeks. 2. Strength of linkage: the success of data linkage de- pended on the matching method and type of person- al identifiers used. Where a national unique identifier (ID) was available (eg, Denmark, Finland) over 99% of cases were matched, but success rates were general- ly lower when intermediary databases and a combina- tion of other identifiers (eg, names, postcodes) were required to establish a match, particularly if these were incomplete or incorrect. A standard way of eval- uating confidence in a match had to be developed so that decisions on inclusion for analyses could be consistently made, in order to avoid bias. 3. GA groups: the GA at birth was categorised into <28 weeks, 28–31 weeks, 32–36 weeks and ≥37 weeks when analysing survival, but due to small numbers of survi- vors at under 28 weeks gestation, the two lowest GA categories were combined to <32 weeks. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ m 4. Birth weight: birth weight was categorised into very low birth weight ‘<1500 g’, low birth weight ‘≥1500 g to <2500 g’, normal birth weight ‘≥2500 g to <4000 g’ and high birth weight ‘≥4000 g’. Births <1000 g were not distinguished from those between 1000 and 1499 g as there were too few cases for the data to be analysed accurately. ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. Work of the Standardisation Committee ICD-9-BPA, International Statistical Classification of Diseases and Related Health Problems 9th Revision with the British Paediatric Association; ICD-10-BPA, International Statistical Classification of Diseases and Related Health Problems 10th Revision with the British Paediatric Association. 5. Singletons versus multiples: there is uncertainty about whether the survival in twins with CAs is lower than that in singletons.34–37 Hence, all survival analyses were performed on singletons alone and then mul- tiples and singletons combined (multiples were not analysed alone as for many registries small numbers would limit the analyses that could be performed). This enables the survival of singletons and multiples in children with CAs to be analysed in detail. When examining morbidity, multiplicity was treated as one of the risk factors for increased risk of hospitalisa- tions and lengths of stay to enable any association to be analysed, but with less detail than for survival. The majority of analyses included singletons and multi- ples combined. UU included validation routines in the syntax scripts to determine if the data have been correctly transformed. For example, it is checked that a date of death does not occur prior to the date of birth; while the primary purpose is to ensure the data have been standardised correctly, it can also reveal any errors in the linked data. y guest. Protected by copyright. A CDM is not being defined for the linked educa- tion data as there is limited scope for comparison and pooling of data across countries. This is due in part to inherent differences in the educational stages and systems, the variability in data available, and fewer regis- tries being able to participate (5 of the registries are from England, and 1 from Wales, Denmark, Italy and Finland). 6. Prenatal diagnoses: the GA at prenatal diagnoses was categorised into <22 weeks, 22–31 weeks, ≥32 weeks, GA not known and no prenatal diagnosis. For Finland, 6 Morris JK, et al. BMJ Open 2021;11:e047859. Work of the Standardisation Committee doi:10.1136/bmjopen-2020-047859 Open access Open access e 3 EUROCAT congenital anomaly registries in EUROlinkCAT: start year, births in the population up to 2014, live births with an anomaly in the study period and ability nk to mortality, healthcare, prescription and education data genital anomaly registry Planned start year Actual start year Number of live births with an anomaly Linkages occurring (Y), Not occurring (N) and Planned but no longer occurring (Y) Planned to be linked Linked to mortality data by 8/2020 Mortality Hospital data for child Hospital data for mother Prescriptions Education In-patient Out-patient In-patient Out-patient um: Antwerp 1995 1997 8083 7865 Y N N N N N N tia: Zagreb 1995 2011 2232 441† Y Y Y N N N mark: Funen 1995 1995 2418 2425 Y Y Y Y Y Y Y nd 1995 1995 44 869 42 861 Y Y Y Y Y Y Y ce: Île de la Réunion 2002 NC 3855 NC Y Y N N N N N ce: Paris 1997 1995 13 335 11 623 Y N N N N N N any: Saxony-Anhalt 1995 2005 8821 8698 Y N N N N N N Emilia Romagna 1995 2008 11 447 7327 Y Y N N N Y Y Tuscany 1995 2005 9827 5187 Y Y Y N N Y Y a 1995 2005 2470 2718 Y N N N N N N erlands: Northern 1995 2005 8567 8325 Y Y Y Y Y Y N ay 1999 1999 26 938 27 201 Y N N N N N N gal: South 1995 2000* 3425* 2447* N Y N N N N N n: Basque Country 1995 1995 4883 5904 Y Y N N N Y N n: Valencian Region 2007 2007 7438 7389 Y Y N N N Y N East Midlands and South hire 1998 AL 18 549 AL Y Y N N N Y Y Northern England 2000 AL 8617 AL Y Y N N N Y Y South West England 2005 AL 11 671 AL Y Y N N N Y Y Thames Valley 1995 AL 5142 AL Y Y N N N Y Y Wales 1998 1998 18 239 18 128 Y Y Y Y Y Y Y Wessex 1995 AL 7771 AL Y Y N N N Y Y ne: West 2005 2006 6166 5835 Y Y N N N N N 234 763 155 594† 19 15 5 4 4 7 9 egistry in Basque was unable to complete the planned linkages to hospital data due to COVID-19. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Table 4 Use of a reference population in morbidity and education analyses Congenital anomaly registry Reference population Croatia: Zagreb Sample of children Denmark: Funen Whole population Finland Whole population France: Île de la Réunion Not provided Italy: Emilia Romagna Whole population Italy: Tuscany 10% of population Netherlands: Northern 10% of population Portugal: South Sample of children Spain: Basque Not provided Spain: Valencian Region Whole population UK: East Midlands and South Yorkshire Aggregate data from population for morbidity and population sample for education UK: Northern England Aggregate data from population for morbidity and population sample for education UK: South West England Aggregate data from population for morbidity and population sample for education UK: Thames Valley Aggregate data from population for morbidity and population sample for education UK: Wales Whole population UK: Wessex Aggregate data from population for morbidity and population sample for education Ukraine: West No longer in morbidity study Table 5 Standardised variables from the EUROCAT database database EDMP variables used (core variables are shaded in blue) Baby and mother 1 CENTRE Centre number 2 NUMLOC Local ID of case 3 BIRTH_DATE Date of birth 4 SEX Sex 5 NBRBABY Number of babies delivered 6 SP_TWIN Specify twin type of birth, like or unlike, zygosity 7 NBRMALF Number of malformed in multiple set 8 TYPE Type of birth 9 CIVREG Civil registration status 10 WEIGHT Birth weight 11 GESTLENGTH Length of gestation in completed weeks 12 SURVIVAL Survival beyond 1 week of age 13 DEATH_DATE Date of death 14 DATEMO Date of birth of mother 15 AGEMO Age of mother at delivery 16 BMI Maternal body mass index 17 RESIDMO Mother’s residence code Diagnosis 19 WHENDISC When discovered 20 CONDISC Condition at discovery 21 AGEDISC If prenatally diagnosed, gestational age at discovery 22 FIRST PRE First positive prenatal test 24 KARYO Karyotype of infant/fetus 25 SP_KARYO Specify karyotype 26* GENTEST Genetic test 27* SP_GENTEST Specify genetic test 28 PM Postmortem examination 29 SURGERY First surgery for malformation performed or planned 30 SYNDROME Syndrome 31 SP_SYNDROME Specify syndrome 32 MALFO1 Malformation 33 SP_MALFO1 Specify malformation 34 MALFO2 As MALFO1 35 SP_MALFO2 Specify malformation 36 MALFO3 As MALFO1 37 SP_MALFO3 Specify malformation 38 MALFO4 As MALFO1 39 SP_MALFO4 Specify malformation 40 MALFO5 As MALFO1 41 SP_MALFO5 Specify malformation 42 MALFO6 As MALFO1 43 SP_MALFO6 Specify malformation Continued EDMP variables used (core variables are shaded in blue) the GA was often not recorded, only the trimester of diagnoses. Work of the Standardisation Committee es for WP4 morbidity linkage have been provided, as mortality linkage was never planned. 441 cases in Zagreb were not included in analyses due to poor quality of the mortality linkage. waiting linkage as of August 2020; NC, linkages could not be completed; Y, linkages are no longer planned as of August 2020. on July 12, 2021 by guest. Protected by copy http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from The registry in Basque was unable to complete the planned linkages to hospital data due to COVID-19. Values for WP4 morbidity linkage have been provided, as mortality linkage was never planned. †The 441 cases in Zagreb were not included in analyses due to poor quality of the mortality linkage. AL, awaiting linkage as of August 2020; NC, linkages could not be completed; Y, linkages are no longer planned as of August 2020. Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 7 Open access on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Table 5 Continued Table 6 Coding of male or female of the live births in different linked databases in EUROlinkCAT Table 6 Coding of male or female of the live births in different linked databases in EUROlinkCAT Centre Variable name Code Male Female UK: Wales DEC_SEX_CD 1 2 Germany: Saxony-Anhalt EF306 1 2 Finland SUKUP 1 2 Italy: Tuscany SESSO 1 2 France: Île de la Réunion SexeDefunt 1 2 France: Paris SexeDefunt 1 2 Netherlands: Northern geslacht 1 2 Croatia: Zagreb GENDER M or 1 F or 2 Ukraine: West CH_SEX 1 2 Belgium: Antwerp SEX 1 2 Norway KJONN 1 2 UK: England sex 1 2 Spain: Basque Country SEXO 1 6 Spain: Valencian Region SEXO 1 6 Malta gender M F Italy: Emilia Romagna SEX M F Denmark: Funen C_SEX M K on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ J Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from between, for example, mothers in the highest group to mothers in the lowest group. The effect of SES on survival would be analysed using Cox proportional hazard models within each registry. However, only seven registries were able to provide a proxy SES vari- able that was reasonably complete for some or all of the time period of the study. It was also planned to investigate the association between risk factors such as birth weight after adjusting for SES, but due to the lack of information on SES this was not included in further multivariable analyses. between, for example, mothers in the highest group to mothers in the lowest group. The effect of SES on survival would be analysed using Cox proportional hazard models within each registry. However, only seven registries were able to provide a proxy SES vari- able that was reasonably complete for some or all of the time period of the study. It was also planned to investigate the association between risk factors such as birth weight after adjusting for SES, but due to the lack of information on SES this was not included in further multivariable analyses. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ m y 9. Maternal country of birth: it was determined that the maternal country of birth variable would be used as a proxy for non-European ethnic origin, as we were aware that ethnic origin is poorly recorded. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Finland’s first trimester diagnoses (week 0 to week 12) mapped exactly to the EUROlinkCAT <22 weeks category. It was decided that Finland’s sec- ond trimester (week 13 to week 27) diagnoses were assumed to occur at 22–31 weeks and third trimester (week 28 onwards) diagnoses occurred at ≥32 weeks. These assumptions were also checked based on the distribution of those cases in Finland with a ‘known’ age at discovery and the assumptions held. g y p 7. Length of stay (LOS): the LOS of the child in hospi- tal was calculated after excluding the stay associated with the birth. Methods of identifying the birth stay varied in different countries. For hospital admissions where admission and discharge occurred on the same day, the LOS was considered to be 0.5 days. If an admission record was missing a discharge date, then discharge date=date of admission+2×(date of latest 8 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Open access EDMP variables used (core variables are shaded in blue) 44 MALFO7 As MALFO1 45 SP_MALFO7 Specify malformation 46 MALFO8 As MALFO1 47 SP_MALFO8 Specify malformation 57 OMIM OMIM code/type of Mendelian inheritance Exposure and family history 58 ASSCONCEPT Assisted conception (where available) 59† OCCUPMO Mother’s occupation at time of conception Sociodemographic 91 MATEDU Maternal education 92 SOCM Socioeconomic status of mother 93 * SOCF Socioeconomic status of father 94 MIGRANT Migrant status Derived variables Byear Year of birth birth_type Live birth, stillbirth, spontaneous abortion, TOPFA, not known definitions of stillbirths and spontaneous abortions vary between regions. This variable recodes birth type according to EUROCAT’s specifications: cases with gestational age ≥20 weeks are recoded as ‘stillbirths’ (irrespective of the local definition of stillbirth/ spontaneous abortion). casestatus Only cases with casestatus=1 or 2 al1-al114 EUROCAT subgroups: (0=No, 1=Yes). Based on EUROCAT coding in Guide 1.4 mult_malf Algorithm for case classification into isolated and multiple anomalies See work of the Standardisation Committee (viii). †EUROCAT Guide 1.4 use ISCO-08 classifications. EDMP, EUROCAT Data Management Programme; TOOPFA, termination of pregnancy for a fetal anomaly. Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from However, for those registries with reasonably complete data on this, almost 100% of children were reported as being of European ethnic origin. Therefore, this variable was not included in subsequent analyses as the num- ber of children considered as ‘non-European ethnic origin’ was too small to analyse. See work of the Standardisation Committee (viii). †EUROCAT Guide 1.4 use ISCO-08 classifications. EDMP, EUROCAT Data Management Programme; TOOPFA, termination of pregnancy for a fetal anomaly. *See work of the Standardisation Committee (viii). †EUROCAT Guide 1.4 use ISCO-08 classifications. EDMP, EUROCAT Data Management Programme; TOOPFA, termination of pregnancy for a fetal anomaly. procedure–date of admission). The date of discharge was set to the date of the child’s 10th birthday or the end of the study period if it was after either of these two dates. g y 10. Cause of death: cause of death based on the death certificates was classified for deaths <1 year and for 1–9 years separately. Death related to preterm birth is very common in the first year after birth, but not as relevant to children at 1–9 years of age. Injuries and poisoning are more common after the first year. The main causes of deaths were classified into six groups for deaths <1 year and 11 groups for children aged 1–9 years. When working with the results tables it was clear that many of these classification groups included many small numbers and data could not be guest. Protected by copyright. 8. Socioeconomic status (SES): all registries had differ- ent variables that could be considered to be a measure of the mother’s SES. The variables included maternal occupation, maternal education and index of mul- tiple deprivation derived from residential codes at birth. Registries were asked to select the variable they believed was the most relevant and to recode their selected SES proxy variable into three groups of ap- proximately equal proportion to enable comparing 8. Socioeconomic status (SES): all registries had differ- ent variables that could be considered to be a measure of the mother’s SES. The variables included maternal occupation, maternal education and index of mul- tiple deprivation derived from residential codes at birth. Registries were asked to select the variable they believed was the most relevant and to recode their selected SES proxy variable into three groups of ap- proximately equal proportion to enable comparing 9 Morris JK, et al. BMJ Open 2021;11:e047859. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from doi:10.1136/bmjopen-2020-047859 Open access Open access Figure 1 Structure of mortality and EUROCAT data used for analysing children’s survival. on July 12, 2021 by guest. Protected by copy http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Figure 1 Structure of mortality and EUROCAT data used for analysing children’s survival. surgeries could be identified that would be expected to be performed on these children. extracted from the databases. For some registries it was only possible to give cause of death as either ‘con- genital anomaly’ or else ‘any other cause of death’. extracted from the databases. For some registries it was only possible to give cause of death as either ‘con- genital anomaly’ or else ‘any other cause of death’. 12. Intensive care: it was planned to analyse the number of days in intensive care, however, only five registries could provide this. Therefore, only whether a child had ever been admitted to intensive care was analysed rather than their LOS. 11. Surgery: a number of different coding systems were used across registries to code surgeries and other pro- cedures (eg, NCSP by NOMESCO, ICD-9-CM, OPCS- 4). Frequency lists for all codes describing surgeries and other procedures were obtained from the linked datasets. Two paediatricians then independently de- termined if a code was a surgery or for another proce- dure and then a consensus between the two clinicians was reached over codes classified as codes for surger- ies. Further subdivision into anomaly-specific surger- ies was carried out for anomalies for which specific 13. Ventilation: it was planned to analyse the number of days on ventilation. However, as it was decided that the LOS in intensive care was not going to be anal- ysed, the same decision was made for ventilation and only whether a child had ever been on ventilation was analysed. 13. Ventilation: it was planned to analyse the number of days on ventilation. However, as it was decided that the LOS in intensive care was not going to be anal- ysed, the same decision was made for ventilation and only whether a child had ever been on ventilation was analysed. Figure 2 Structure of hospital admissions, prescription data and EUROCAT data used for analysing children’s morbidity. ATC code, Anatomical Therapeutic Chemical Classification System code; ICU, intensive care unit. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ ne 2021. Small number restrictions (statistical disclosure control) Small number restrictions (statistical disclosure control) Four countries have limitations on the release of aggregate data and analytic results if the numbers of births involved are very small (generally under eight births). This situ- ation arises in many analyses involving specific CAs, as CAs are rare, with some affecting less than 1 in 10 000 live births. Solutions to enable the maximum amount of data to be included in all multi-centre European analyses varied according to country. The Northern Netherlands released data if all exported results were rounded to the nearest 5. Rounding all frequencies ensures that original numbers cannot be inferred. For Denmark, a few named researchers at SGUL and UU were allowed access to the aggregate data for the purpose of collating and including in pooled-analysis, on condition that it was securely stored and processed; that any individual results involving fewer than five people were not released; and that personal identification was not possible from any released results. The SAIL databank (Wales) provided data to the CRR with the requirement that aggregate data on fewer than five people were not released and could not be calculated from any information in the public domain. The registry from Antwerp, Belgium could not release any informa- tion on three or fewer cases. In some countries all national databases use the same unique ID number (eg, Finland). So, identifying a child in the National Vital Statistics meant that there was confidence that any hospital stays up to 10 years of age would also be identified. For other registries, as not all children were likely to be admitted to hospital, each case was searched for in, not only the in-patient hospital data- base (which included the mother’s visit for the birth), but also any other healthcare databases (such as outpatient, primary-care or prescription databases) for longer than the 10 years of follow-up in the study. The lack of infor- mation in any healthcare database was judged to mean a non-match. Sensitivity analysis was performed to assess if there were differences in results if the non-matches were included. For the education data, all children known to be alive were assumed to be included in the National Educa- tion Databases and therefore any case not identified was assumed to be a non-match. on July 12, 2021 by guest. Protected by copyright. Patient and public involvement A series of focus groups has been held in different Euro- pean countries involving parents with a child with one of four predefined CAs with different health problems covering learning difficulties, physical disabilities, visible defects and non-visible defects with higher mortality. The four anomalies selected were: CHD requiring surgery (referred to as severe CHD—a usually non-visible defect with high mortality), cleft lip (a visible defect often with speech problems), spina bifida (a physical disability with associated incontinence problems) and DS (Trisomy 21; a visible defect with learning difficulties and often asso- ciated with CHD). The focus groups have investigated parental experiences of having a child with one of the above anomalies and assessed parental research priorities and a paper will be published in due course. guest. Protected by copyright. Small number restrictions (statistical disclosure control) http://bmjopen.bmj.com/ m Syntax scripts were developed centrally by St George’s, University of London (SGUL) to evaluate the accuracy of the linkage and identify any factors leading to missed links (eg, deaths within the first week of life). For each registry, the proportion of births in any single year of data that have not been linked is calculated and the data from any year with less than 85% of cases linked will be excluded from further analyses. Second, the quality of linked data items was also evaluated: a variable that was >20% missing in a year would be excluded from any anal- ysis in which it featured; and variables that were recorded by both the CA registry and the linked database would be compared for agreement, by year. In general, it was found that data quality was poorer in the earlier years and tended to improve over time; however, if data quality fluc- tuated across the years, then only the longest consecutive period where quality was above the threshold would be analysed. Assessment of quality of linkage and quality of linked data on July 12, 2021 http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Many registries linked their data to National Vital Statis- tics, which are databases that record all live births with follow-up until the child dies or emigrates outside the country/region of interest. Therefore, for these regis- tries for the survival analysis, any child whose record was not in the National Vital Statistics Database was judged to be a non-match and overall linkage could be assessed. Some registries were only able to link to death certificates which meant that a non-match, that is, no death certifi- cate found, was assumed to indicate that the child was still alive. The data from these registries were only included in the survival analyses if there was additional information about the quality of the linkage. For example, in Malta, due to the small well-defined population, there was confi- dence that all deaths had been identified. Open access STATA (V.13 and upwards). This allows each register to submit aggregated data and analytical results (eg, Kaplan- Meier estimates, HRs and CIs), rather than individual case data, to the EUROlinkCAT central results reposi- tory (CRR) at UU, UK using a secure web platform. UU collates the aggregate data and results and provides these data to the researchers responsible for the different anal- yses and publications. Multi-centre European analyses will be performed by combining the individual registries’ aggregated data and analytic results, using meta-analytic techniques. Additional work is required to develop suit- able models for combining survival data from several registries when the sample sizes are very small as observed in many registries. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Downloaded from Figure 2 Structure of hospital admissions, prescription data and EUROCAT data used for analysing children’s morbidity. ATC code, Anatomical Therapeutic Chemical Classification System code; ICU, intensive care unit. ructure of hospital admissions, prescription data and EUROCAT data used for analysing children’s m mical Therapeutic Chemical Classification System code; ICU, intensive care unit. 10 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 ETHICS AND DISSEMINATION The CA registries have the required ethics permissions and procedures for routine surveillance, data collection and transmission of anonymised data to the EUROCAT central database, according to national guidelines and they were required to submit evidence of these permis- sions to the EUROlinkCAT ethics portfolio. Local regis- tries follow national legislation as to whether parental consent is needed for registration of babies with anom- alies. Each registry was responsible for applying for and obtaining the additional ethics and other permissions (eg, data sharing agreements) required to link and analyse their data for EUROlinkCAT. This was an extremely lengthy process in some countries as the original data collection did not include expectation or consent for the data to be used in research, and a new legal basis had to be established. Additional assurances and procedures were adopted by registries (eg, publication of privacy notices) to ensure compliance with the General Data Protection Regulation which came into force on 25 April 2018 in the European Union. A checklist of minimum specifications for data storage/backup was completed by each registry. Three registries took over 3 years to get the ethics, legal basis, data protection, information governance and data sharing agreements in place. UU obtained ethics permis- sion for the CRR. The strength of this study is that the researchers are a multidisciplinary group, many of whom have collabo- rated successfully for many years through being members of EUROCAT. In addition, EUROlinkCAT is able to build on all the standardisation procedures already established in EUROCAT. The implementation of a CDM enables the same centrally developed syntax script to be run in all the different registries which is efficient and also ensures standardisation of analysis across the registries. The use of a reference population when analysing health- care data will aid in the identification of the source of differences between registries (eg, average LOS in hospi- tals will differ) and therefore enable us to better quantify the burden of disease attributable to CAs in each country. Comparisons of the accuracy of healthcare databases with respect to recording CA cases will be informative and enable improvements in those areas with less accurate data. We will also be developing recommendations on how to use the available healthcare data in an optimal way to provide information on children with CAs in areas without active CA registries. on July 12, 2021 by guest. Protected by copyright. Statistical analysis Protocols and syntax scripts are developed centrally to create aggregate data and perform specific analyses on the individual cases in each standardised data set in In addition, a European survey concerning the diag- nosis, medical care, education and everyday life will be 11 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Open access distributed to parents across Europe with children with the same four CAs as described above. Registries will ensure the questions in the survey are appropriate for their country (eg, the provision of health services, given how this differs in various European countries) and will translate the survey into their native language with back translation to confirm the accuracy of the translation. The aim is for the survey to be distributed via social media by parent support groups across Europe to engage with a wide spectrum of parents. means that all analyses must be performed locally using a generic modelling strategy. This does limit the use of iterative procedures to explore data in detail. The major limitation to the study is that only specific areas in Europe are represented, with a lack of data in particular from Eastern Europe. Interpretation of differences across Europe is challenging as it will be essential to interpret results in the light of knowledge about the differences in healthcare and education practices across Europe. p p The EUROlinkCAT project will enable important hypotheses concerning the survival, health and educa- tion of children with CAs in Europe to be investigated. The standardised methods and CDMs will all be available freely on the EUROlinkCAT website and will be available for use in future research projects to benefit from and build on this work, so as to enable other multi-centre European projects to exploit routine healthcare data available in Europe. DISCUSSION The aggregate data and results from the CRR in EUROlinkCAT will provide important information on the survival, morbidity and education of children born with a CA in Europe. Researchers in each CA registry will be encouraged to also perform specific local analysis, in order to fully exploit the research potential of linked datasets. The establishment of a method of standardising data from each registry linkage into a CDM provides valu- able infrastructure enabling future multi-national studies to be performed in an efficient manner and new regis- tries to become involved. ORCID iDs Joan K Morris http://orcid.org/0000-0002-7164-612X Ester Garne http://orcid.org/0000-0003-0430-2594 Maria Loane http://orcid.org/0000-0002-1206-3637 Ingeborg Barisic http://orcid.org/0000-0002-9085-6747 Anna Latos-Bieleńska http://orcid.org/0000-0002-9152-1674 Anna Pierini http://orcid.org/0000-0003-3321-9343 Judith Rankin http://orcid.org/0000-0001-5355-454X Anke Rissmann http://orcid.org/0000-0002-9437-2790 Hermien de Walle http://orcid.org/0000-0002-1418-8281 Joachim Tan http://orcid.org/0000-0003-0462-4761 Joanne Emma Given http://orcid.org/0000-0003-4921-1944 Hugh Claridge http://orcid.org/0000-0001-5998-2860 University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands: L. Renée Lutke, Nicole H.A. Siemensma-Mühlenberg; Public Health Wales National Health Service Trust, Wales, UK: David Tucker; Institut National de la Santé et de la Recherche Médicale, Paris, France: Babak Khoshnood, Nathalie Bertille, Nathalie Lelong, Makan Rahshenas; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain: Clara Cavero Carbonell, Laia Barrachina Bonet, Laura Garcia Villodre; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain and Public Health and Preventive Medicine Department, University of València, Spain and Public Health Regional Health Administration (DG Salud Publica y Adicciones), Ó Generalitat Valenciana, Valencia, Spain: Óscar Zurriaga; Polish Registry of Congenital Malformations, Department of Medical Genetics, Poznan University of Medical Sciences. Poznan, Poland: Anna Jamry-Dziurla; Finnish Institute for Health and Welfare, Finland: Sonja Kiuru-Kuhlefelt, Mika Gissler, Anna Heino; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine: Wladimir Wertelecki; OMNI- Net for Children International Charitable Fund, Rivne, Ukraine and Khmelnytsky City Children’s Hospital, Khmelnytsky, Ukraine: Diana Akhmedzhanova, Nataliia Zymak-Zakutnia; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine and Rivne Regional Medical Diagnostic Center, Rivne, Ukraine: Lyubov Yevtushok, Lyubov Ostapchuk, Oksana Tsizh; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine and Volyn Regional Children's Territorial Medical Center, Lutsk, Ukraine: Serhii Lapchenko; Epidemiology Department, National Registry of Congenital Anomalies, National Institute of Health Doctor Ricardo Jorge Open access (Instituto Nacional de Saúde Doutor Ricardo Jorge), Lisbon, Portugal: Carlos Matias Dias, Ausenda Machado, Ana João Santos, Liliana Antunes, Paula Braz; Register of Congenital Malformations, Centre Hospitalier Universitaire de La Réunion, Île de la Réunion, France: Hanitra Randrianaivo-Ranjatoelina, Benedicte Bertaut Nativel; Provinciaal Instituut voor Hygiëne (PIH), Antwerp, Belgium: Vera Nelen, Guy Thys, Elly Den Hond; Departamento de Salud Gobierno Vasco, Basque Country, Spain: Olatz Mokoroa Carollo; Swansea University, Wales, UK: Daniel Thayer, Ieuan Scanlon, Sue Jordan; Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway and Division of Mental and Physical Health Norwegian Institute of Public Health, Bergen, Norway: Kari Klungsøyr; Directorate for Health Information and Research, G'Mangia, Malta: Miriam Gatt. (Instituto Nacional de Saúde Doutor Ricardo Jorge), Lisbon, Portugal: Carlos Matias Dias, Ausenda Machado, Ana João Santos, Liliana Antunes, Paula Braz; Register of Congenital Malformations, Centre Hospitalier Universitaire de La Réunion, Île de la Réunion, France: Hanitra Randrianaivo-Ranjatoelina, Benedicte Bertaut Nativel; Provinciaal Instituut voor Hygiëne (PIH), Antwerp, Belgium: Vera Nelen, Guy Thys, Elly Den Hond; Departamento de Salud Gobierno Vasco, Basque Country, Spain: Olatz Mokoroa Carollo; Swansea University, Wales, UK: Daniel Thayer, Ieuan Scanlon, Sue Jordan; Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway and Division of Mental and Physical Health Norwegian Institute of Public Health, Bergen, Norway: Kari Klungsøyr; Directorate for Health Information and Research, G'Mangia, Malta: Miriam Gatt. experience gained with distributing the parents’ survey using parent support groups and social media will also lead to development of a framework to enable dissem- ination of results to be made more directly to parents. In addition, a series of reports will be written including recommendations for improving the collection and analysis of data on CAs in routinely collected data in the healthcare databases. Provenance and peer review Not commissioned; externally peer reviewed. Glinianaia; Emila Romagna Registry of Birth Defects (IMER), University of Ferrara, Italy: Gianni Astolfi, Annarita Armaroli; Neonatal Intensive Care Unit, IMER Registry, Department of Medical Sciences, University Hospital of Ferrara, Italy: Elisa Ballardini; Territorial Care Service - Emilia Romagna Health Authority, Italy: Aurora Puccini; Childrens Hospital Zagreb, Croatia: Ljubica Boban; Institute of Clinical Physiology, National Research Council (Consiglio Nazionale delle Ricerche-Istituto di Fisiologia Clinica), Pisa, Italy: Silvia Baldacci, Alessio Coi, Lorena Mezzasalma, Michele Santoro; University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands: L. Renée Lutke, Nicole H.A. Siemensma-Mühlenberg; Public Health Wales National Health Service Trust, Wales, UK: David Tucker; Institut National de la Santé et de la Recherche Médicale, Paris, France: Babak Khoshnood, Nathalie Bertille, Nathalie Lelong, Makan Rahshenas; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain: Clara Cavero Carbonell, Laia Barrachina Bonet, Laura Garcia Villodre; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain and Public Health and Preventive Medicine Department, University of València, Spain Open access This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/ licenses/by/4.0/. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ Patient consent for publication Not required. 10Malformation Monitoring Centre Saxony-Anhalt, Otto von Guericke University Medical Faculty, Magdeburg, Germany Provenance and peer review Not commissioned; externally peer reviewed. Open access Provenance and peer review Not commissioned; externally peer reviewed. Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise. on July 12, 2021 by guest. Protecte http://bmjopen.bmj.com/ 28 June 2021. Downloaded from 11Department of Genetics, University Medical Centre Groningen, Groningen, The Netherlands 11Department of Genetics, University Medical Centre Groningen, Groningen, The Netherlands Collaborators All below named members of the EUROlinkCAT Consortium have contributed to the development of the protocol with their knowledge and experience of the surveillance of congenital anomalies. Names are given in Consortium organisation numerical order: Population Health Research Institute, St George’s, University of London, UK: Abigail Reid, Liz Limb, Gillian Briggs, Nicholas Connor; Hospital Lillebaelt, Region Syddanmark, Denmark and Section of Epidemiology, University of Copenhagen, Denmark: Stine Kjaer Urhoj; Hospital Lillebaelt, Region Syddanmark, Denmark: Mads Damkjær, Christina Neergaard Pedersen; Population Health Sciences Institute, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK: Svetlana V. Glinianaia; Emila Romagna Registry of Birth Defects (IMER), University of Ferrara, Italy: Gianni Astolfi, Annarita Armaroli; Neonatal Intensive Care Unit, IMER Registry, Department of Medical Sciences, University Hospital of Ferrara, Italy: Elisa Ballardini; Territorial Care Service - Emilia Romagna Health Authority, Italy: Aurora Puccini; Childrens Hospital Zagreb, Croatia: Ljubica Boban; Institute of Clinical Physiology, National Research Council (Consiglio Nazionale delle Ricerche-Istituto di Fisiologia Clinica), Pisa, Italy: Silvia Baldacci, Alessio Coi, Lorena Mezzasalma, Michele Santoro; University of Groningen, University Medical Center Groningen, Department of Genetics, Groningen, the Netherlands: L. Renée Lutke, Nicole H.A. Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Author affiliations 1 Contributors JKM is the principal author and the project’s scientific coordinator. ML is the data coordinator. EG is the clinical coordinator. JD, AN, AP, JR and HdW are Steering Committee and Standardisation Committee members with JKM, ML and EG. IB and AL-B are Steering Committee members. AR is a Standardisation Committee member. JT is the senior statistician. JEG the research fellow. HC is the project manager. All authors contributed to, read and approved the final manuscript. 1Population Health Research Institute, St George’s University of London, London, UK 2Paediatric Department, Hospital Lillebaelt, Kolding, Denmark 1Population Health Research Institute, St George’s University of London, London, UK 2Paediatric Department, Hospital Lillebaelt, Kolding, Denmark 3Faculty of Life and Health Sciences, Ulster University, Coleraine, UK 4Children’s Hospital Zagreb, Centre of Excellence for Reproductive and Regenerative Medicine, Medical School University of Zagreb, Zagreb, Croatia 5Biomedical Computing Limited, Battle, UK 6Polish Registry of Congenital Malformations, Chair and Department of Medical Genetics, Poznan University of Medical Sciences, Poznan, Poland 7IMER Registry (Emila Romagna Registry of Birth Defects), University Hospital of Ferrara, Emilia-Romagna, Italy 8Instituto di Fisiologia Clinica, Consiglio Nazionale delle Ricerche, Pisa, Italy 9Population Health Sciences Institute, Newcastle University, Newcastle upon Tyne, UK 2Paediatric Department, Hospital Lillebaelt, Kolding, Denmark 3Faculty of Life and Health Sciences, Ulster University, Coleraine, UK 4Children’s Hospital Zagreb, Centre of Excellence for Reproductive and Regenerative Medicine, Medical School University of Zagreb, Zagreb, Croatia 5Biomedical Computing Limited, Battle, UK 6Polish Registry of Congenital Malformations, Chair and Department of Medical Genetics, Poznan University of Medical Sciences, Poznan, Poland 7IMER Registry (Emila Romagna Registry of Birth Defects), University Hospital of Ferrara, Emilia-Romagna, Italy 8Instituto di Fisiologia Clinica, Consiglio Nazionale delle Ricerche, Pisa, Italy 9Population Health Sciences Institute, Newcastle University, Newcastle upon Tyne, UK Funding This project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under grant agreement No 733001 for 5 years: 1 January 2017 to 31 December 2021. The views presented here are those of the authors only, and the European Commission is not responsible for any use that may be made of the information presented here. Competing interests None declared. Patient consent for publication Not required. ETHICS AND DISSEMINATION http://bmjopen.bmj.com/ m One of the challenges of EUROlinkCAT is the ability of the CA registries to link their data to external data sources due to different local data information governance issues and the availability of suitable electronic healthcare data- bases. This requires flexibility in including registries in only specific subprojects and acceptance that not all registries may be able to perform the linkages planned. In addition, some registries require support from other partners in all aspects of the project, including applying for ethics permissions, adapting protocols, standardising data and running statistical syntax scripts. The restriction of not being able to share individual case data and also aggregate data that might be disclosive or identifiable Each registry participating in the focus groups with parents was responsible for ensuring the correct ethics approvals were in place. Similarly, the registries partici- pating in the dissemination of the parents’ survey will be responsible for ensuring the necessary ethics permissions are obtained. by guest. Protected by copyright. An Ethics and Data Protection Advisory Board consisting of three independent advisors with the relevant expertise monitor all ethical considerations in this project. The CRR will be used for multiple studies and the results from these will be disseminated in peer reviewed papers and conference presentations. It is hoped that the 12 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Provenance and peer review Not commissioned; externally peer reviewed. Siemensma-Mühlenberg; Public Health Wales National Health Service Trust, Wales, UK: David Tucker; Institut National de la Santé et de la Recherche Médicale, Paris, France: Babak Khoshnood, Nathalie Bertille, Nathalie Lelong, Makan Rahshenas; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain: Clara Cavero Carbonell, Laia Barrachina Bonet, Laura Garcia Villodre; Rare Diseases Research Unit, Foundation for the Promotion of Health and Biomedical Research in the Valencian Region, Valencia, Spain and Public Health and Preventive Medicine Department, University of València, Spain and Public Health Regional Health Administration (DG Salud Publica y Adicciones), Generalitat Valenciana, Valencia, Spain: Óscar Zurriaga; Polish Registry of Congenital Malformations, Department of Medical Genetics, Poznan University of Medical Sciences. Poznan, Poland: Anna Jamry-Dziurla; Finnish Institute for Health and Welfare, Finland: Sonja Kiuru-Kuhlefelt, Mika Gissler, Anna Heino; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine: Wladimir Wertelecki; OMNI- Net for Children International Charitable Fund, Rivne, Ukraine and Khmelnytsky City Children’s Hospital, Khmelnytsky, Ukraine: Diana Akhmedzhanova, Nataliia Zymak-Zakutnia; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine and Rivne Regional Medical Diagnostic Center, Rivne, Ukraine: Lyubov Yevtushok, Lyubov Ostapchuk, Oksana Tsizh; OMNI-Net for Children International Charitable Fund, Rivne, Ukraine and Volyn Regional Children's Territorial Medical Center, Lutsk, Ukraine: Serhii Lapchenko; Epidemiology Department, National Registry of Congenital Anomalies, National Institute of Health Doctor Ricardo Jorge Collaborators All below named members of the EUROlinkCAT Consortium have contributed to the development of the protocol with their knowledge and experience of the surveillance of congenital anomalies. Names are given in Consortium organisation numerical order: Population Health Research Institute, St George’s, University of London, UK: Abigail Reid, Liz Limb, Gillian Briggs, Nicholas Connor; Hospital Lillebaelt, Region Syddanmark, Denmark and Section of Epidemiology, University of Copenhagen, Denmark: Stine Kjaer Urhoj; Hospital Lillebaelt, Region Syddanmark, Denmark: Mads Damkjær, Christina Neergaard Pedersen; Population Health Sciences Institute, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK: Svetlana V. on July 12, 2021 by guest. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Hospital discharge data: can it serve as the sole source of case ascertainment for population-based birth defects surveillance programs? J Public Health Manag Pract 2010;16:245–51. 12 Derrington TM, Kotelchuck M, Plummer K, et al. Racial/Ethnic differences in hospital use and cost among a statewide population of children with Down syndrome. Res Dev Disabil 2013;34:3276–87. 13 Dawson AL, Cassell CH, Oster ME, et al. Hospitalizations and associated costs in a population-based study of children with Down syndrome born in Florida. Birth Defects Res A Clin Mol Teratol 2014;100:826–36. 29 Andrade SE, Scott PE, Davis RL, et al. Validity of health plan and birth certificate data for pregnancy research. Pharmacoepidemiol Drug Saf 2013;22:7–15. 30 Blais L, Bérard A, Kettani F-Z, et al. Validity of congenital malformation diagnostic codes recorded in Québec's administrative databases. Pharmacoepidemiol Drug Saf 2013;22:881–9. 14 Hook-Dufresne DM, Yu X, Bandla V, et al. The economic burden of gastroschisis: costs of a birth defect. J Surg Res 2015;195:16–20. on July 12, 2021 by guest. Protected http://bmjopen.bmj.com/ 28 June 2021. Downloaded from 15 Wehby GL, Pedersen DA, Murray JC, et al. The effects of oral clefts on hospital use throughout the lifespan. BMC Health Serv Res 2012;12:58. 31 Steele A, Johnson J, Nance A, et al. A quality assessment of reporting sources for microcephaly in Utah, 2003 to 2013. Birth Defects Res A Clin Mol Teratol 2016;106:983–8. 16 Hung W-J, Lin L-P, Wu C-L, et al. Cost of hospitalization and length of stay in people with Down syndrome: evidence from a national hospital discharge claims database. Res Dev Disabil 2011;32:1709–13. 32 EUROCAT. Guide 1.4. Available: http://www.eurocat-network.eu/ aboutus/datacollection/guidelinesforregistration/guide1_4 [Accessed 14 Feb 2018]. 17 O'Toole C, Lee AS-Y, Gibbon FE, et al. Parent-mediated interventions for promoting communication and language development in young children with Down syndrome. Cochrane Database Syst Rev 2018;10:CD012089. 33 Garne E, Dolk H, Loane M, et al. Paper 5: surveillance of multiple congenital anomalies: implementation of a computer algorithm in European registers for classification of cases. Birth Defects Res A Clin Mol Teratol 2011;91 Suppl 1:S44–50. 18 Iyer NP, Tucker DF, Roberts SH, et al. Outcome of fetuses with Turner syndrome: a 10-year congenital anomaly register based study. J Matern Fetal Neonatal Med 2012;25:68–73. 34 Herskind AM, Larsen LA, Pedersen DA, et al. Comparison of late mortality among twins versus singletons with congenital heart defects. Am J Cardiol 2017;119:1680–6. 19 Marino BS, Lipkin PH, Newburger JW, et al. on July 12, 2021 by guest. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from cause of death statistics. Birth Defects Res A Clin Mol Teratol 2007;79:792–7. 5 Wu J, Morris JK. The population prevalence of Down's syndrome in England and Wales in 2011. Eur J Hum Genet 2013;21:1016–9. pp 23 Greenlees R, Neville A, Addor M-C, et al. Paper 6: EUROCAT member registries: organization and activities. Birth Defects Res A Clin Mol Teratol 2011;91 Suppl 1:S51–100. 6 Tennant PWG, Pearce MS, Bythell M, et al. 20-Year survival of children born with congenital anomalies: a population-based study. Lancet 2010;375:649–56.i 24 Kinsner-Ovaskainen A, Lanzoni M, Garne E, et al. A sustainable solution for the activities of the European network for surveillance of congenital anomalies: EUROCAT as part of the EU platform on rare diseases registration. Eur J Med Genet 2018;61:513–7. 7 Brodwall K, Greve G, Leirgul E, et al. The five-year survival of children with Down syndrome in Norway 1994-2009 differed by associated congenital heart defects and extracardiac malformations. Acta Paediatr 2018;107:845–53. 25 Dolk H, Loane M, Teljeur C, et al. Detection and investigation of temporal clusters of congenital anomaly in Europe: seven years of experience of the EUROCAT surveillance system. Eur J Epidemiol 2015;30:1153–64. 8 Muranjan M, Vijayalakshmi P. The unforeseen toll of birth defects and their economic burden at a tertiary care public Institute in Mumbai. Indian J Pediatr 2014;81:1005–9. 9 Simeone RM, Oster ME, Hobbs CA, et al. Population-Based study of hospital costs for hospitalizations of infants, children, and adults with a congenital heart defect, Arkansas 2006 to 2011. Birth Defects Res A Clin Mol Teratol 2015;103:814–20. 26 Loane M, Dolk H, Kelly A, et al. Paper 4: EUROCAT statistical monitoring: identification and investigation of ten year trends of congenital anomalies in Europe. Birth Defects Res A Clin Mol Teratol 2011;91 Suppl 1:S31–43. 10 Shetty S, Kennea N, Desai P, et al. Length of stay and cost analysis of neonates undergoing surgery at a tertiary neonatal unit in England. Ann R Coll Surg Engl 2016;98:56–60. 27 Salemi JL, Tanner JP, Sampat D, et al. The accuracy of hospital discharge diagnosis codes for major birth defects: evaluation of a statewide registry with passive case ascertainment. J Public Health Manag Pract 2016;22:E9–19. 11 Fitzsimons KJ, Copley LP, Deacon SA, et al. Hospital care of children with a cleft in England. Arch Dis Child 2013;98:970–4. 28 Wang Y, Cross PK, Druschel CM. REFERENCES 1 Boyle B, Addor M-C, Arriola L, et al. Estimating global burden of disease due to congenital anomaly: an analysis of European data. Arch Dis Child Fetal Neonatal Ed 2018;103:F22–8. 2 Zylbersztejn A, Gilbert R, Hjern A, et al. Child mortality in England compared with Sweden: a birth cohort study. Lancet 2018;391:2008–18. 2 Zylbersztejn A, Gilbert R, Hjern A, et al. Child mortality in England compared with Sweden: a birth cohort study. Lancet 2018;391:2008–18. 3 Wang H, Liddell CA, Coates MM, et al. Global, regional, and national levels of neonatal, infant, and under-5 mortality during 1990-2013: a systematic analysis for the global burden of disease study 2013. Lancet 2014;384:957–79. 4 Copeland GE, Kirby RS. Using birth defects registry data to evaluate infant and childhood mortality associated with birth defects: an alternative to traditional mortality assessment using underlying 13 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 Open access Open access cause of death statistics. Birth Defects Res A Clin Mol Teratol 2007;79:792–7. 22 Boyd PA, Haeusler M, Barisic I, et al. Paper 1: The EUROCAT network--organization and processes. Birth Defects Res A Clin Mol Teratol 2011;91 Suppl 1:S2–15. Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859 on July 12, 2021 by guest. http://bmjopen.bmj.com/ BMJ Open: first published as 10.1136/bmjopen-2020-047859 on 28 June 2021. Downloaded from Neurodevelopmental outcomes in children with congenital heart disease: evaluation and management: a scientific statement from the American heart association. Circulation 2012;126:1143–72. on July 12, 2021 by guest. Protected by copyright. http://bmjopen.bmj.com/ m 35 Gnanendran L, Bajuk B, Oei J, et al. Neurodevelopmental outcomes of preterm singletons, twins and higher-order gestations: a population-based cohort study. Arch Dis Child Fetal Neonatal Ed 2015;100:F106–14. 20 Wehby GL, Collett BR, Barron S, et al. Children with oral clefts are at greater risk for persistent low achievement in school than classmates. Arch Dis Child 2015;100:1148–54. 36 Wang W, Pan W, Chen J, et al. Outcomes of congenital diaphragmatic hernia in one of the twins. Am J Perinatol 2019;36:1304–9. 21 Clausen NG, Kähler S, Hansen TG. Systematic review of the neurocognitive outcomes used in studies of paediatric anaesthesia neurotoxicity. Br J Anaesth 2018;120:1255–73. 37 Martinka D, Barrett J, Mei-Dan E, et al. Respiratory morbidity in late preterm twin infants. Arch Gynecol Obstet 2019;300:337–45. 14 Morris JK, et al. BMJ Open 2021;11:e047859. doi:10.1136/bmjopen-2020-047859
https://openalex.org/W4390516258	https://repozytorium.uwb.edu.pl/jspui/bitstream/11320/15561/1/BAJ_23_2023_N_Iobishvili_The_Talk_Show_Interview_as_a_Speech_Genre.pdf	English	null	The Talk Show Interview as a Speech Genre: Georgian-English Perspective	Białostockie Archiwum Językowe	2,023	cc-by-sa	3,433	DOI: 10.15290/baj.2023.23.09 DOI: 10.15290/baj.2023.23.09 Nana Iobishvili Akaki Tsereteli State University nana.iobishvili@atsu.edu.ge https://orcid.org/0000-0003-3134-7405 1. Introduction Each genre of journalism has certain features that distinguish it from other genres. They differ in form and manner of delivery, emotional ex‑ pressive means, etc. Despite its apparent simplicity, the interview is one of the most difficult genres that requires specific tactics and strategies. In addition to asking a question and recording the answer, it is necessary to conduct a dialogue, to provide the necessary direction. The first purpose of political interviews is to inform and, secondly, to entertain, while the main purpose of talk show interviews is to entertain. Nevertheless, both functions are combined in all types of interviews. As for the educational function, it can be said that it includes all media messages and, therefore, interviews as well. Talk shows have become a ubiquitous feature of American and Euro‑ pean television. Scholars and journalists have discussed various examples in order to place them within some cultural parameters. In the end, both sides conclude that talk shows are important because they are important subjects of public discussion and important in shaping the landscape of popular television. All variations of talk shows, from chat shows to celebrity interviews, have one basic element: they all feature a group of guests, not as indi‑ 162 Nana Iobishvili vidual respondents, but each involves audience participation. The studio audience is not only visible, but is also given the opportunity to engage in commentary and interaction. The present paper deals with the talk show as one of the interesting and current speech genres. The study employs the contrastive analysis method for investigation of selected American and Georgian realisations of this genre in the form of selected talk shows. The aim is to discuss general characteristics of this speech genre. American and Georgian entertainment talk shows, which are very popular in their countries, are taken as research material. As mentioned above, the research is based on the comparative, contrastive method, as we contrast American and Georgian talk shows, in order to identify the similarities and differences between them. 2. The talk show interview as a speech genre The language used in the discourse of different genres is defined by several main factors: relevance, functionality, dynamism, context, situation. Modern discourse analysis is based on the concept of speech genre (SG) as one of the most effective explanatory mechanisms in the process of teaching speech production and interpretation. Speech Genre Theory (SGT) is one of the existing practices of verbal communication, which includes the situation and the field of communication, style, intention, speech form, ways of begin‑ ning and ending speech, initiating dialogue, as well as communication strate gies and techniques. The study of TV shows is the most interesting in this regard, as it is here that live dialogue is presented. The language of the show is close to the everyday conversational style used in an informal situation. Therefore, it is common to rely heavily on colloquial lexis and phraseology as well as syntax typical of oral communication. It should be emphasized that the TV genre system is a combination of interconnected and interactive genres. One of its constituent units is the interview – a verbal interaction in which one person aims to obtain information from another in a question-answer mode. p q The interview is situated in the field of mass media communication. Mass media, in turn, includes social, highly organized, structured insti‑ tutions. In this sense, the interview reflects the characteristics of one of the institutions: broadcasting. Broadcasting is a powerful social process of communication. The interview is one of the most popular genres of modern media. It became especially popular in the 90s of the XX century, when the 163 The Talk Show Interview as a Speech Genre… respondent was given the opportunity to freely express opinions on this or that issue. The collapse of the institution of censorship allowed the journal‑ ist to address the respondent as an individual. A talk show interview tends to be person-centered, but it also shares the peculiarities of an informative interview. The structure of television broadcasting refers to the means used to achieve the goals of different television genres. These facilities include both physical (e.g. buildings), as well as human resources (e.g. operators, presenters, reporters). These television events are public representations and justify the need for organizational structure, time constraints (time is limited during broadcasting), and other specific characteristics that arise from this ritual context. 2. The talk show interview as a speech genre Broadcast forms of political interview, talk show interview and audi‑ ence debate represent different types of conversation genres. A talk show is a complex genre whose boundaries are difficult to determine. The clas‑ sification of Carbaugh’s talk shows by personality type and issue type reflects the change that talk shows have undergone in a person-only chat show with members of the audience ending with discussions on social topics (Carbaugh 1988). Hence the use of the term chat show as a synonym for personal type talk shows. Issue-type talk shows are also referred to as audience discussion programs (Livingstone & Lunt 1994). The term talk show is used only by Tolson (1991) for a personal type of talk show. This genre includes a series of short interviews (and sometimes also in the form of a presentation) with individuals who are primarily representatives of the entertainment industry. Its characteristics are based on the rules for determining a political interview. The talk show interview function is constantly changing between infor‑ mation and entertainment. The goal of the interviewer’s information search is to approach the format of an informal conversation or chat, the content of which is often personal and private; it sometimes takes the form of gossip and is often characterized by its humorous and witty tone. The main fea‑ tures of chat as a genre are found in many formats (interviews, discussions, game shows, etc.), but it is fully produced in a talk show. The characteristics of a talk show are the same as those of a chat (Tolson 1991: 182). According to Tolson (1991), the violation of the general rules of an interview is the most characteristic feature of the talk show, which stems from the informal nature of the meeting. The difference between formal and informal interviews is based on the formal procedures that mandate turn-taking, the length of the dialogue, etc. The closer the interaction is to 164 Nana Iobishvili everyday conversations, the more informal it is, but it still retains some “for‑ mal” qualities depending on who opens, closes, or directs that interaction. everyday conversations, the more informal it is, but it still retains some “for‑ mal” qualities depending on who opens, closes, or directs that interaction. Thus, the interview is considered less formal, the more it deviates from a relatively fixed agenda, and the more participants deviate from their pre‑ determined role. 2. The talk show interview as a speech genre Due to the informal nature of the talk show interviews, it is acceptable for the respondent to ask questions, get acquainted with the topics and even adjust the role of the interviewer. Nevertheless, after an interval of violations, the interviewer corrects the interaction in order to follow the general rules. So to some extent, they always follow the princi‑ ples of a formal interview. The informal nature of the talk show interview gives it a flexible general structure. A talk show interview is defined in a narrow sense as a personal type of interview aimed at serving a dual purpose of providing information and entertainment. The entertainment component is structurally revealed in violation of the general format of traditional interviews. In talk shows, guests are asked for professional opinions on a selected issue; in addition, they are invited to discuss additional areas of public in‑ terest, such as personal life and current activities. As already mentioned, the interview on the talk show is more informal and entertaining. In many cases, talk show guests are familiar with the entertainment world. Most of them are actors, directors, comedians, presenters, musicians, athletes and etc. p A talk show is a show that has a host who, for each episode, hosts different celebrities and interviews them or interacts with them. They talk about existing projects or future plans, play games, and have short comedy segments as well. Various forms of communication occur in talk shows, for instance, natural conversations. Moreover, some talk shows include games and activities, so the conversation style is more similar to an interview. The host should ask questions and continue the conversation. New episodes of regular talk shows are also aired weekly or daily at different times of the day. The talk shows in question are mostly aired overnight, designed for relaxation, so they are easily added to the viewers’ routine. 3. A comparative analysis of Georgian and American entertainment talk show interviews The talk show formats are more or less common all over the world. It means that famous people from different fields come to visit the studio with 165 The Talk Show Interview as a Speech Genre… the host. They discuss interesting issues with the host, like new movie or music, or just talk about personal issues. But still, all such TV shows differ from each other in environment and script. American and Georgian talk shows taken as empirical material of the research are uniquely spectacular and entertaining. America as a pioneer country in the production of talk shows is famous for many TV shows. Here are just a few: “Ellen: The Ellen DeGeneres Show” (2003), “Jimmy Kimmel Live!” (USA 2003), “The Tonight Show Starring Jimmy Fallon” (USA 2014). Most Georgian talk shows are analogous to the American ones: simi‑ larities are observed in terms of the studio, special music, music band and talk show structure. For example, in the Uruskesi show, we have not only similarities in terms of extra linguistic factors, but also the title of the talk show contains the English lexical unit ‘show’, indicating the genre. This is used to enhance the attention effect, in order to make the show more attractive to the viewer by introducing a different style, i.e. an English loanword (Leladze 2017: 44). The common feature of non-political talk shows: “ურუსკესი Show” (lit. “Uruskesi show”), “გიორგი გაბუნიას შოუ” (lit. “Georgi Gabunia Show”), “მხოლოდ ლელასთან” (lit. “Only with Lela”) and others is the many guests invited to the studio. The principle of selecting themes involves topics that are relevant and beneficial to the community, which ensures a high rating of the show. Almost all talk shows mostly choose one topic in each program (Beridze 2020: 34). However, there are cases when several topics are discussed in the program. The names of American and Georgian talk shows tend to contain the names of the presenters. For example: “Ellen DeGeneres show”, “Jimmy Kummel’s live”, “The Tonight Show with Jimmy Fallon”, “Georgi Gabunia Show”, “Only with Lela” and the “Uruskesi show”. Ruska Makashvili is the host of the Uruskesi show and here her name Ruska is presented in the superlative form – Uruskesi. Each talk show, despite its genre diversity, is structurally the same: it consists of three parts: an introduction, an interview, and a conclusion. The main part is built on questions and answers. 3. A comparative analysis of Georgian and American entertainment talk show interviews Each show has a guest or guests who tell us a story by answering questions; The conversation is mostly in the present tense, be it live or a rerun. A talk show is a product that competes with other products in the television space (Nebieridze 2011: 202–205). These features are universal and are typical of talk shows in all countries. 166 Nana Iobishvili Opening talk shows is quite personal. Talk shows start with an intro‑ duction aimed at the audience. Opening a talk show to the studio audience is aimed only at the first component. For the entertainment purposes of the genre, addressing the audience will change to addressing the guest directly, as soon as the interaction between the host and the guest begins. The host, after a monologue, addresses the audience to greet the guests. The talk show hosts present their honored guests and shake hands. Instead of saying ‘goodbye’ the talk show host thanks his guest. Participants in talk shows shake hands and almost never say hello or hi. However, it also occurs that they are greeted informally, which is illustrated by the excerpts provided below: Host: Hi. Guest: Hi. Host: Was it different kind of way getting you here? Guest: Also you had me follow Oprah? Host: Yeah. (“Ellen Degeneres” – 24/05/22) Host: Yeah. (“Ellen Degeneres” – 24/05/22) The initial elements are: good to have you here or you look great rather than how are you, which is attested by the examples given below: Host: You look great. Guest: Thank you. (“Ellen Degeneres” – 09/02/22) Host: Its so great to meet you. (“Jimmy Fallon” – 07/05/22) Host: Its so great to meet you. (“Jimmy Fallon” – 07/05/22) Guest: Wow. Im so happy to be here. Thanks for having me on the show. Host: Oh, we love having you. (“Jimmy Fallon” – 12/04/18) Guest: Wow. Im so happy to be here. Thanks for having me on the show. Guest: Wow. Im so happy to be here. Thanks for having me on the show. Host: Oh, we love having you. (“Jimmy Fallon” – 12/04/18) In the Georgian show “Uruskesi show”, the guest greets the host and says: How are you? 3. A comparative analysis of Georgian and American entertainment talk show interviews The host also says: It’s good to have you here, which can be seen below: Guest: How are you?… I like the show, the name… Guest: How are you?… I like the show, the name… Host: How good it is to have you here, how beautiful you are, how glam‑ orous you are, how glamorousyou are always. (“Uruskesi show” – 31/01/22) Host: How good it is to have you here, how beautiful you are, how glam‑ orous you are, how glamorousyou are always. (“Uruskesi show” – 31/01/22) In the Georgian show “Giorgi Gabunia Show”, the host greets the guests, thanks them for coming to the studio and then goes on to inter‑ view them. He also utters the pragmatism How are you? Several illustrative examples are presented below: The Talk Show Interview as a Speech Genre… 167 Host: Hello gentlemen, first of all thank you very much for coming, I am always glad to meet you, how are you? Guest: Thank you. (“Georgi Gabunia show” – 20/05/22) Host: Hello, first of all I should mention, Ms. Manana you look great as usual. Guest: Thank you. Host: How are you? Guest: Like Everyone. (“Georgi Gabunia Show” – 21/05/22) Host: Hello gentlemen, first of all thank you very much for coming, I am always glad to meet you, how are you? Georgian talk shows tend to be copied from American analogues, but still there are cultural differences. In Georgian, the possible answer to ‘How are you?’ might have been ‘so-so’, but as it is media discourse, the guest chose a neutral form ‘like everyone’. In this case, the guest preferred to an‑ swer with a neutral answer. It is a kind of modesty, maybe, or the history of the country could influence them to feel like everybody, i..struggling against invaders since time immemorial, when nobody was fine. The struc‑ tural characteristics of American and Georgian entertaining talk shows are more or less identical. The difference is felt only in the linguacultural aspect, which is well seen in the example above. Based on the discussed examples, it can be concluded that in non-political talk shows, the present‑ er and the guest are well-wishing to each other and do not hesitate to say positive words to each other and to express praise of their interlocutors in the introductory and farewell part of the show. 4. Conclusions The aim of this paper was to investigate the peculiarities of the talk show interview as a speech genre in two radically different language set‑ tings, American and Georgian. American and Georgian entertainment talk shows were taken as research material. A study was conducted to find out the similarities and differences between these two languages. Observation of the empirical material revealed that, despite the fact that Georgian talk shows are basically analogues of American talk shows and their format is taken from American talk shows, there are cultural differences. As a conclusion, we can say that in the interviews of American and Georgian entertainment talk shows, the host and guests are very kind to each other, they do not spare each other words of praise, they have a positive attitude towards each other throughout the dialogue, which is expressed by positive words. Based on the fact that I tried to put the existing research into a narrow context, I will try to deepen the research in this regard in the future. Beridze L., 2020, Peculiarities of speech etiquette in television discourse, doctoral thesis, Batumi. 3. A comparative analysis of Georgian and American entertainment talk show interviews Talk show interview topics are also pre-defined, but this pre-definition is not seen in the opening structure of the event, as in political interviews or debates. The talk show opening does not explicitly state the main topic in the opening section, although to some extent it is latent (hidden) in the guest selection component. This feature corresponds to spontaneous conversation where any topic can arise. Unlike political interviews and debates, it portrays an important listener-centered or interactive character. p y p Ordinary conversations are closed by exchanging ritualized farewells, such as Goodbye Now or Our Show Is Near the End, Goodbye – goodbye or its equivalents. The American talk shows close mostly in much the same way. The hosts in the pre-closing phases always thank the audience, the guests and the TV viewers. Informal register is often used when thanking. Sometimes they offer viewers to follow the show through various social networks, after which they utter neutral and familiar phrases of farewell. The following closing signals were used in Georgian entertainment talk shows: 168 Nana Iobishvili • thank the guest for coming to the talk show, • say goodbye to the guest, • say goodbye to the audience after giving thanks. The verbal closing of the show is accompanied by non-verbal signals as well, the host shakes hands with guests as farewell and also waves to the audience. The closing of the show is also indicated by special music that is heard at the end of the show, followed by subtitles. The observations of interviews on the entertainment talk shows reveal that the host uses many verbal tactics necessary to organize the interaction. Initiating and terminating communication, giving up and maintaining speech initiative are the most popular and widespread. The choice of speech tactics depends on the type of discourse, the interpersonal relationships of the communicators, and the stage of the interaction. References The Talk Show Interview as a Speech Genre… 169 Carbaugh D., 1988, Talking American: Cultural Discourses on DONAHUE, Ablex (NY). Leladze T., 2017, British and Georgian talk shows as journalistic Genre, doctoral thesis, Kutaisi. Livingstone S., Lunt P., 1994, Talk on Television: Audience Participation and Public Debate, London. Nebieridze M., 2011, Talk Show as a Genre, General Sociolinguistic Features, “Scientific Research” XXXII. Tolson A., 1991, Televised Chat and the Synthetic Personality, in: P. Scannell (ed.), Broadcast Talk, London. Yule J., 2005, Pragmatics, Oxford. Yule J., 2005, Pragmatics, Oxford. Yule J., 2005, Pragmatics, Oxford. https://www.youtube.com/watch?v=g0oTMfWkjQ&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=36 (accessed: 24.05.2022). https://www.youtube.com/watch?v=WwMDIu4p928&list=PL1BumPDD9Cqb‑ Gv1wHCeqlz6-AYK-tbqy-&index=35 (accessed: 9.02.2022). https://www.youtube.com/watch?v=4SvXJC2PJY&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=1 (accessed: 7.05.2022). https://www.youtube.com/watch?v=udpSQIaSVDg&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=3 (accessed: 12.04.2018). https://www.youtube.com/watch?v=udpSQIaSVDg&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=3 (accessed: 12.04.2018). https://www.youtube.com/watch?v=17MSiqzTpUE&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=38 (accessed: 31.01.2022). https://www.youtube.com/watch?v=17MSiqzTpUE&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=38 (accessed: 31.01.2022). https://www.youtube.com/watch?v=17MSiqzTpUE&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=38 (accessed: 31.01.2022). https://www.youtube.com/watch?v=zsAjZr8D5WY&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=42 (accessed: 20.05.2022). https://www.youtube.com/watch?v=zsAjZr8D5WY&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=42 (accessed: 20.05.2022). https://www.youtube.com/watch?v=IvLhNYkCRNY&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=40 (accessed: 21.05.2022). https://www.youtube.com/watch?v=IvLhNYkCRNY&list=PL1BumPDD9CqbG‑ v1wHCeqlz6-AYK-tbqy-&index=40 (accessed: 21.05.2022). Abstract The present paper discusses the talk show interview as a speech genre. With this view, the study employs contrastive analysis. The focus is on interviews con‑ ducted in non-political talk shows, which had not been studied in this regard so far. The article outlines talk shows in America and Georgia, characterizes their structure, style, lexical-grammatical characteristics, and presents the interview as a speech genre. Keywords: talk shows, speech genres, talk show interviews 170 Nana Iobishvili Streszczenie Niniejszy artykuł poświęcony jest wywiadowi w amerykańskich i gruziń‑ skich programach talk show jako gatunku mowy. W badaniu wykorzystano me‑ todę komparatystyczną. Materiał badawczy stanowią wywiady z programów talk show o tematyce rozrywkowej, które nie były dotąd badane pod tym względem. Słowa kluczowe: talk show, gatunki mowy, wywiady w programie talk show
https://openalex.org/W2916512146	https://www.nature.com/articles/s41598-019-39462-1.pdf	English	null	The Association between Cannabis Product Characteristics and Symptom Relief	Scientific reports	2,019	cc-by	6,765	The Association between Cannabis Product Characteristics and Symptom Relief Received: 4 May 2018 Accepted: 22 January 2019 Published: xx xx xxxx Sarah S. Stith, Jacob M. Vigil, Franco Brockelman, Keenan Keeling & Branden Hall Federal barriers and logistical challenges have hindered measurement of the real time effects from the types of cannabis products used medically by millions of patients in vivo. Between 06/06/2016 and 03/05/2018, 3,341 people completed 19,910 self- administrated cannabis sessions using the mobile device software, ReleafApp to record: type of cannabis product (dried whole natural Cannabis flower, concentrate, edible, tincture, topical), combustion method (joint, pipe, vaporization), Cannabis subspecies (C. indica and C. sativa), and major cannabinoid contents (tetrahydrocannabinol, THC; and cannabidiol, CBD), along with real-time ratings of health symptom severity levels, prior-to and immediately following administration, and reported side effects. A fixed effects panel regression approach was used to model the within-user effects of different product characteristics. Patients showed an average symptom improvement of 3.5 (SD = 2.6) on an 11-point scale across the 27 measured symptom categories. Dried flower was the most commonly used product and generally associated with greater symptom relief than other types of products. Across product characteristics, only higher THC levels were independently associated with greater symptom relief and prevalence of positive and negative side effects. In contrast, CBD potency levels were generally not associated with significant symptom changes or experienced side effects. Medical cannabis markets are currently being flooded with thousands of cannabis strains with unique cannabi- noid profiles1, novel, uninvestigated cannabis-derived formulates and products with little to no clinical references or formal guidance on how fundamental characteristics of the products themselves may affect pharmacodynam- ics2,3. Federal laws have all but prohibited the use of prospective, pragmatic, naturalistic studies with random treatment assignment for measuring the effects of cannabis consumed in vivo. What little clinical research does exist is mostly limited to randomized controlled trials (RCTs) using synthetic cannabinoids or low quality and potency cannabis obtained from the federal government that is unrepresentative of the cannabis products used by millions of people every day4,5. Contributing to further confusion are historically contradictory messages coming from the scientific community on the true risks and benefits of cannabis consumption. For example, whereas cannabis was once often and sometimes still is described as component cause of schizophrenia6,7, several studies now suggest the use of medical cannabis as an effective alternative therapy to antipsychotics and for treating schizophrenia more generally8–11. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports University of New Mexico, The Department of Psychology, Albuquerque, USA. Correspondence and requests for materials should be addressed to J.M.V. (email: vigilj@unm.edu) Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 Methods St d D Whole natural dried Cannabis flower and con- centrates made from the flower are the most common types of products and most likely to contain information on the full spectrum of product characteristics. Panels A through E of Table 1 show descriptive statistics for the product characteristics recorded by the ReleafApp; specifically, product type (Panel A), flower and concentrate combustion method (Panel B), subspecies (Panel C), THC potency (Panel D), and CBD potency (Panel E). Study Outcomes. Our main outcome is changes in symptom severity following cannabis consumption (ending symptom level minus starting symptom level) as shown in Table 1, Panel F. During a session, the patient also can report side effects, including 12 negative side effects, 19 positive side effects, and 11 context-specific side effects, crowd-sourced from users, dispensaries, beta testers, and app developers (Supplementary Table S2). Patients can select as many side effects as they like, and at least one side effect was reported in 78% of sessions in our sample. We use as outcome varia- bles whether a patient reported any side effect by category and the percent of the number of side effects available in each of our three respective categories that a patient selected (Table 1, Panel G). The most commonly reported negative side effects are Dry Mouth (26%) and feeling Foggy (23%), the most frequent positive side effects are Relaxed (63%) and Peaceful (54%), and the most common context-specific side effects are feeling High (37%) and Thirsty (27%). Statistical Analysis. Our basic statistical model uses a least squares panel regression approach with repeated observations (sessions) at the patient level to analyze how product characteristics affect symptom relief and side effects. We include patient-specific fixed effects to control for time-invariant user characteristics, in order to compare how changing a product characteristic affects symptom relief and side effects for a given user rather than comparing outcomes across users who may differ in many ways, including which products they choose to con- sume. Because higher starting symptom levels are associated with greater symptom relief (r = 0.60), we control for the starting symptom level in all of our regressions. Standard errors are clustered at the user level to control for heteroskedasticity and arbitrary correlation. Methods St d D Methods Study Design. Institutional Review Board approval was obtained from the University of New Mexico for this study and methods were performed in accordance with the approved guidelines. The preexisting anonymized data were obtained with user informed consent through the owner of the ReleafApp, MoreBetter, Ltd., and subject to an investigator confidentiality agreement. The ReleafApp patient education and cannabis treatment management tool was designed to track patient sessions and real-time cannabis use experiences in order to optimize the therapeutic effects of consuming cannabis, while minimizing negative side effects. ReleafApp users voluntarily download the application and enter information on the product they intend to consume, including type of product (whole natural dried flower, concentrate, edible, tincture, and topical); when applicable, combustion method (joint, dry or water pipe, and vape); plant subspecies (C. indica, C. sativa, or hybrid); and THC and CBD potency levels (percentage of total weight)21,22. Testing of the potencies of both cannabinoids is almost universally required under U.S. medical marijuana laws and generally reported on product labels. THC and CBD levels were capped at 35% for flower due to biological limitations on how much THC and CBD a plant can contain. Prior to beginning a session, the patient is required to enter a neg- ative health symptom, selected from 27 possible symptom categories, for which they are attempting to use cannabis therapeutically. (A list of symptom categories and frequencies is available in Supplementary Table S1.)t p y y p g q pp y After entering a symptom, patients are prompted to record a starting symptom level on a visual analogue scale from 0 (no detectable symptom level) to10 (severe). The patient then taps the prompt on the screen to begin the session. From that time until the patient closes the session, they can enter multiple symptom severity levels as frequently as desired. For the current analyses, we include in our sample only patients entering starting symptoms greater than 0 and recording at least one symptom level within 90 minutes of starting the session; we use the last symptom level recorded within that timeframe as the ending symptom level. Our final sample includes 19,910 sessions and 3,341 patients who recorded at least one product characteristic in the ReleafApp between 06/06/2016 and 03/05/2018. Because the entry of product characteristics is voluntary, the sample sizes used in our analyses vary depending on which product characteristics are included. Methods St d D Because the number of observations varies substantially by prod- uct category, we run regressions separately for each product characteristic category (product type, combustion method, subspecies, and THC and CBD levels) as well as regressions with all product characteristics included.f In order to explore whether cannabis product characteristics differ across symptom categories in their associa- tion with momentary symptom relief and side effect profiles, we conduct sub-analyses with samples defined by the three most frequently reported symptom categories: anxiety (16% of the sample), back pain (8%), and depression (10%). Lastly, we explore whether side effect profiles vary with product characteristics. We regress reports of any, and the proportion of side effects selected from each side effect category (negative, positive, context-specific) on our product characteristics using ordinary least squares for consistency across models, although our side effect outcomes are constrained to {0,1} and [0,1]. All predicted outcomes from our regressions fall between zero and one. The Association between Cannabis Product Characteristics and Symptom Relief Contradictory effects are often attributed to the distinction between what has been historically interpreted as cannabis’ harmful, psychoactive cannabinoid, tetrahydrocannabinol (THC), often described as providing the ‘high’ effects versus the therapeutic, non-psychoactive potential (sometimes described as a ‘miracle cure’ in the popular media) of cannabidiol (CBD)12. In actuality, few large-scale investigations to date have measured the relative effects of THC and CBD consumption in real-time under naturalistic conditions among people diagnosed with schizophrenia or any other user group. Th h fi d h f d l h f b d d ff g p p g p y g p This is the first study to measure how fundamental characteristics of cannabis products consumed in vivo affect immediate symptom relief and experienced side effects. We operationalize our research question using a mobile device software application (app). Although hundreds of cannabis-themed software apps are available for public use13, the ReleafApp educational software14 is the first app designed specifically to record how the route of admin- istration, combustion method, cannabis subspecies, and major cannabinoid contents are associated with real-time measurements of symptom severity levels, prior to and immediately following administration of cannabis, and the manifestation of myriad possible side effects. Despite recent advocacy for the benefits of CBD over THC, the vast majority of observational studies showing an association between patient-managed cannabis use and improvements Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 1 www.nature.com/scientificreports/ in symptoms related to, for example, chronic pain15, multiple sclerosis and Parkinson’s disease16, post-traumatic stress disorder17, and schizophrenia18 relied on public or commercially available cannabis that has been hybridized for high THC and low CBD contents, thus suggesting that THC may be an important determinant of user outcomes. Findings from this study are expected to contribute to guidelines for safe and effective cannabis consumption19,20, which until now have been limited to anecdotal or retrospective reporting and ungeneralizable experiments. Results Table 2 shows our results for the effects of product characteristics on patient symptom relief. All regressions control for the starting symptom level, which renders the constant positive. The similar size of the coefficients for the starting symptom level and the constant mean that patients reporting symptom levels of 1 may experience little symptom relief. However, even with a starting symptom level of just 2, users are predicted to experience Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 2 www.nature.com/scientificreports/ Mean Std. Dev. Minimum Maximum Panel A: Product Type (19,910 sessions, 3,341 users) Concentrate 0.17 0.38 0 1 Edible 0.05 0.21 0 1 Flower 0.74 0.44 0 1 Tincture 0.04 0.20 0 1 Topical 0.00 0.06 0 1 Panel B: Subspecies (17,197 sessions, 2,996 users) Hybrid 0.48 0.50 0 1 C. indica 0.30 0.46 0 1 C. sativa 0.22 0.41 0 1 Panel C: Combustion Method (16,902 sessions, 2,936) Joint 0.13 0.33 0 1 Pipe 0.43 0.49 0 1 Vape 0.45 0.50 0 1 Panel D: THC (6,958 sessions, 1,260 users) % THC 28.3 22.8 0 100 THC < 10% 0.16 0.36 0 1 THC 10–19% 0.29 0.45 0 1 THC 20–34% 0.36 0.48 0 1 THC 35%+ 0.20 0.40 0 1 Panel E: CBD (5,400 sessions, 1,123 users) % CBD 11.6 16.0 0 100 CBD < 1% 0.25 0.43 0 1 CBD 1–9% 0.33 0.47 0 1 CBD 10–34% 0.34 0.47 0 1 CBD 35%+ 0.08 0.28 0 1 Panel F: Outcome and Control Variables (19,910 sessions, 3,341 users) Symptom Change −3.5 2.6 −10 9 Starting Symptom Level 6.0 2.2 1 10 Ending Symptom Level 2.4 2.2 0 10 Panel G: Side Effects (15,617 sessions, 2,757 users) Any Negative Side Effect 0.62 0.48 0.00 1.00 % of Negative Side Effects 0.12 0.14 0.00 0.92 Any Positive Side Effect 0.95 0.23 0.00 1.00 % of Positive Side Effects 0.25 0.18 0.00 1.00 Any Context-Specific Side Effect 0.78 0.41 0.00 1.00 % of Context-Specific Side Effects 0.22 0.20 0.00 1.00 bl l Table 1. Descriptive Statistics. Notes: Nineteen positive, twelve negative, and eleven context-specific side effects were available for selection. Table 1. Descriptive Statistics. Notes: Nineteen positive, twelve negative, and eleven context-specific side effects were available for selection. Table 1. Descriptive Statistics. Notes: Nineteen positive, twelve negative, and eleven context-specific side effects were available for selection. statistically significant symptom relief. Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 Table 1. Descriptive Statistics. Notes: Nineteen positive, twelve negative, and eleven context-specific side effects were available for selection. www.nature.com/scientificreports/ www.nature.com/scientificreports/ Outcome = Symptom Change (Ending - Starting Symptom) (1) (2) (3) (4) (5) Panel A: Product Type, omitted category = flower Concentrate 0.194 0.080 (0.076) (0.212) Edible 0.340* (0.105) Tincture 0.498* (0.117) Topical −0.216 (0.355) Panel B: Subspecies, omitted category = hybrid C. indica −0.103 −0.057 (0.041) (0.083) C. sativa 0.096* 0.195 (0.055) (0.121) Panel C: Combustion Method, omitted category = joint Pipe −0.061 −0.004 (0.084) (0.194) Vape 0.051 0.051 (0.092) (0.210) Panel D: THC and CBD, omitted categories = THC < 10% and CBD < 1% THC 10–19% −0.215* −0.220* (0.117) (0.134) THC 20–34% −0.235 −0.315* (0.112) (0.121) THC 35%+ −0.252 −0.342 (0.126) (0.166) CBD 1–9% −0.089 −0.026 (0.121) (0.126) CBD 10–34% 0.038 0.079 (0.105) (0.092) CBD 35%+ −0.222 −0.241 (0.209) (0.227) Starting Symptom Level −0.704* −0.721* −0.726* −0.709* −0.717* (0.028) (0.031) (0.032) (0.056) (0.063) Constant 0.588* 0.691* 0.719* 0.909* 0.850 (0.165) (0.184) (0.212) (0.340) (0.403) Number of sessions 19,910 17,197 16,898 4,439 3,869 R-squared 0.330 0.340 0.338 0.337 0.346 Number of users 3,341 2,996 2,936 900 787 Table 2 Effects of Product Characteristics on Symptom Relief Notes: Regressions control fo Table 2. Effects of Product Characteristics on Symptom Relief. Notes: Regressions control for individual user fixed effects. Concentrate is relative to Flower, C. indica and C. sativa are relative to Hybrid, THC categories are relative to THC 0–9%, CBD categories are relative to CBD 0%, and Pipe and Vape are relative to Joint. Standard errors are clustered at the user level (shown in parentheses). *p < 0.01, p < 0.05, p < 0.1. category of flower run separately for the three most common medical cannabis patient conditions, anxiety, back pain, and depression. Regression coefficients controlling for the remaining product characteristics showed that only users treating depression showed greater symptom improvement from using flower in the middle and high- est THC potency categories relative to the least potent flower. In contrast, variability in THC was not associated with statistically significant differences in symptom relief for back pain or anxiety beyond its overall effect on mean symptom improvement levels (see Supplemental Table S3).fi category of flower run separately for the three most common medical cannabis patient conditions, anxiety, back pain, and depression. Regression coefficients controlling for the remaining product characteristics showed that only users treating depression showed greater symptom improvement from using flower in the middle and high- est THC potency categories relative to the least potent flower. Results (The average user reports a starting symptom level of 6, SD = 2.2.) The number of sessions, R-squared and number of users are also reported and vary across regressions, with each regression designated by a column.hiff statistically significant symptom relief. (The average user reports a starting symptom level of 6, SD = 2.2.) The number of sessions, R-squared and number of users are also reported and vary across regressions, with each regression designated by a column.hiff g g y The first column showing the effects of consuming the different product types on symptom relief relative to the effects from flower suggests that flower provides more symptom relief than any other type of product. The second column comparing C. indica and C. sativa plant sub-species to hybrid plants suggests that products from pure indicas may increase symptom relief while sativa strains may decrease it. Combustion method does not explain any difference in symptom relief across products within users, as shown in Column [3]. Column [4] of Table 2 shows that whereas higher THC offers greater symptom relief, higher CBD offers no statistically signifi- cant benefit. The last column of Table 2 includes all product characteristics. Due to the inclusion of combustion method, only flower and concentrate product types are included. The effect of THC becomes stronger and is the only statistically significant determinant of symptom relief among the product characteristics.l y y gi y p g p Figure 1 shows changes in symptom severity by THC and CBD percentage category for dried natural flower, the most popular and homogenous type of cannabis product in the sample, after adjusting for the remainder of the product characteristics. Regression coefficients (controlling for the remaining product characteristics; see Supplementary Table S3) showed that flower containing the middle (10–19%) and highest (20–35%) THC potency levels was associated with greater symptom improvement than flower in the lowest THC potency cate- gory (0–9%). As with the omnibus tests in Table 2, variability in CBD levels in flower was not associated with dif- ferences in symptom improvement. Figure 2 shows adjusted changes in symptom severity by the THC percentage Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 3 Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 www.nature.com/scientificreports/ Just as with symp- tom relief, CBD appears to have little effect on side effect reporting of any kind. There is some evidence that vaping is associated with decreased reporting of negative side effects relative to smoking joints. Lastly, in order to show that a handful of very frequent ReleafApp users are not driving the results, we replicated our main results from Table 2 including only individuals who completed ten sessions or fewer, and separately, including only the first five sessions for all users. THC again is the greatest predictor of symptom relief among the product charac- teristics. (See Supplemental Table S4 for detailed results). www.nature.com/scientificreports/ In contrast, variability in THC was not associated with statistically significant differences in symptom relief for back pain or anxiety beyond its overall effect on mean symptom improvement levels (see Supplemental Table S3).fi y p p pp In Table 3, we present the relationship between product characteristics and side effect profiles. We restrict our analysis to only concentrates and flower to capture the relationship between different combustion methods and side effects. While these methods do not explain any statistically significant variation in symptom relief, they may be associated with different side effect profiles. We run our full model using any and the percent of each category of side effects selected by the user. Concentrates showed a weaker association with positive side effects, but do not appear to differ from flower in their association with negative or context-specific side effect report- ing. Indica-based products are associated with a greater likelihood of reporting negative side effects and some evidence of fewer positive and more context-specific side effects, relative to hybrid- and sativa-based products. Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 4 www.nature.com/scientificreports/ Figure 1. Adjusted Change in Symptom Severity by THC and CBD Percentage Category in Flower. Figure 1. Adjusted Change in Symptom Severity by THC and CBD Percentage Category in Flower. Figure 1. Adjusted Change in Symptom Severity by THC and CBD Percentage Category in Flower. Figure 2. Adjusted Change in Symptom Severity by THC Percentage Category in Flower & Symptom Type. Figure 2. Adjusted Change in Symptom Severity by THC Percentage Category in Flower & Symptom Type. Higher THC generally is associated with increased reporting of all three types of side effects. Just as with symp- tom relief, CBD appears to have little effect on side effect reporting of any kind. There is some evidence that vaping is associated with decreased reporting of negative side effects relative to smoking joints. Lastly, in order to show that a handful of very frequent ReleafApp users are not driving the results, we replicated our main results from Table 2 including only individuals who completed ten sessions or fewer, and separately, including only the first five sessions for all users. THC again is the greatest predictor of symptom relief among the product charac- teristics. (See Supplemental Table S4 for detailed results). Higher THC generally is associated with increased reporting of all three types of side effects. Discussion Given just how common it is for cannabis patients to try different types of products and methods of administra- tion23,24, it is surprising how few previous investigations have examined which fundamental characteristics of the products consumed in vivo by millions of people daily are associated with real-time patient outcomes and expe- rienced side effects. While RCTs may be the ‘gold standard’ for measuring the pharmacodynamics of synthetic, standardized (usually particular symptom focused) medications, they are poorly suited for understanding the effects of a medication with substantial heterogeneity in product characteristics and consumption methods across the estimated 2.2 million state-legal medical cannabis patients in the United States25. Our observational study using mobile app technology was designed to measure these effects in real-time among a large sample of patients using cannabis for treating their medical symptoms under naturalistic conditions. On average, responders expe- rienced significant improvements across the 27 health symptom categories measured. Dried, whole natural flower was associated with greater symptom relief than the use of other types of products (i.e., concentrates, edibles, tinc- tures, and topicals). However, and despite the fact that different routes of administration deliver variable amounts of cannabinoid contents and have different metabolomics26–31, we did not find variation in symptom relief with use of pipes, joints, or vaporization combustion devices. Products made from pure C. indica strains were more effective than products made from C. sativa, matching patient-reported preferences for the former for treating conditions such as pain and insomnia32,33. However, once we controlled for cannabinoid contents, none of the other product characteristics predicted variability in symptom levels. Only THC potency levels showed inde- pendent associations with symptom relief and experiences of both positive and negative side effects, with higher Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 5 www.nature.com/scientificreports/ Variables (1) (2) (3) (4) (5) (6) Negative % of Negative Positive % of Positive Context- Specific % of Context- Specific Concentrate −0.024 0.008 −0.097* −0.090 −0.052 −0.001 (0.080) (0.028) (0.044) (0.037) (0.101) (0.062) C. indica 0.079 0.015* 0.011 −0.033 0.030 0.045* (0.035) (0.009) (0.017) (0.016) (0.026) (0.010) C. Discussion sativa 0.003 −0.002 −0.000 −0.022 −0.041 −0.043 (0.045) (0.008) (0.012) (0.016) (0.033) (0.019) Pipe −0.109 −0.022 −0.046 0.002 −0.041 −0.020 (0.077) (0.024) (0.060) (0.021) (0.045) (0.035) Vape −0.168 −0.037 −0.044 −0.004 −0.050 −0.047 (0.076) (0.024) (0.049) (0.024) (0.059) (0.031) THC 10–14% 0.113* 0.013 0.017 0.055* 0.178* 0.066* (0.043) (0.010) (0.018) (0.015) (0.050) (0.020) THC 15–34% 0.090* 0.016 0.013 0.069* 0.203* 0.085* (0.053) (0.013) (0.019) (0.018) (0.051) (0.021) THC 35%+ 0.198 0.050** 0.065* 0.118* 0.182 0.095* (0.091) (0.025) (0.036) (0.033) (0.072) (0.052) CBD 1–9% 0.030 −0.007 0.000 −0.051*** −0.007 −0.009 (0.042) (0.014) (0.015) (0.018) (0.038) (0.024) CBD 10–34% 0.027 −0.012 0.011 −0.032* −0.013 0.005 (0.029) (0.008) (0.013) (0.018) (0.047) (0.028) CBD 35%+ 0.042 −0.017 0.016 −0.033 0.093 −0.027 (0.062) (0.022) (0.024) (0.022) (0.070) (0.053) Starting Symptom Level 0.005 0.003 0.001 −0.003 0.007 0.003 (0.007) (0.001) (0.002) (0.002) (0.005) (0.002) Constant 0.566* 0.108* 0.988* 0.288* 0.629* 0.160* (0.081) (0.023) (0.048) (0.029) (0.077) (0.039) Observations 3,220 3,220 3,220 3,220 3,220 3,220 R-squared 0.016 0.015 0.009 0.047 0.025 0.052 N Users 665 665 665 665 665 665 Table 3. Relative Associations of Product Characteristics with Side Effects. Notes: Regressions control for individual user fixed effects. The Concentrate is relative to Flower, C. indica and C. sativa are relative to Hybrid, THC categories are relative to THC between 0 and 10%, and CBD categories are relative to 0% CBD, and Pipe and Vape are relative to Joint. Standard errors are clustered at the user level (shown in parentheses). p < 0.01, p < 0.05, p < 0.1. levels resulting in larger effects. In contrast, we did not observe an independent link between CBD levels and any of the omnibus symptom effects measured in the current study across nearly 20,000 user sessions.ii f Variability in cannabinoid profiles may partially explain inconsistent findings in the literature on, for exam- ple, the benefits of using cannabis for treating chronic neuropathic pain, with effectiveness observed in some studies4,34, but not others35. Similarly, while many patient groups (e.g., sleep–disturbed medical cannabis users) have reported a preference for high CBD concentrates36, we did not observe any patient outcomes varying by CBD potency levels alone. One possibility is that many of the CBD potency levels displayed on labels of the products consumed in the study were inaccurate (e.g., inflated), as is currently common in the medical cannabis industry37. Table 3. Relative Associations of Product Characteristics with Side Effects. Notes: Regressions control for individual user fixed effects. The Concentrate is relative to Flower, C. indica and C. sativa are relative to Hybrid, THC categories are relative to THC between 0 and 10%, and CBD categories are relative to 0% CBD, and Pipe and Vape are relative to Joint. Standard errors are clustered at the user level (shown in parentheses). *p < 0.01, p < 0.05, *p < 0.1. Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 Table 3. Relative Associations of Product Characteristics with Side Effects. Notes: Regressions control for ifh Discussion Alternatively, it is possible that CBD has more latent effects than THC (e.g., expanding beyond the 90 minute observation window), has an impact on symptoms infrequently reported in our data, or that CBD’s effects may not lend themselves to perceptual detection and subjective reporting. The phytocannabinoid family of CBDs are known to differ from other cannabinoids such as THC in several ways, including having no affinity to CB1 receptors, serving as an antagonist to GPR55 receptors and as an inverse modulator of the effects of THC and perhaps the endocannabinoid system more generally, as well as functioning as an immuno-suppressant and anti-inflammatory agent38,39. Thus, it is possible that while CBD may operate inconspicuously to improve certain health outcomes, the adjunctive consumption of THC is needed to consciously experience or be aware of such effects.i f Notwithstanding the innovative nature and potential implications of the study’s findings, our observational, quasi-field experiment had unavoidable limitations, including the lack of a control group, e.g., non-cannabis users with similar symptoms, salient characteristics, past experiences, and voluntary reporting, which could lead to either: a) overestimation of the effectiveness of product characteristics if users who have negative experiences with cannabis are more likely to drop out of the sample by choosing not to use the ReleafApp; or b) underestima- tion of cannabis’ effectiveness if users fail to use the ReleafApp due to already being satisfied with their product Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 6 www.nature.com/scientificreports/ choices and their effects. It is also important to note that the patient-reported outcomes were not cross-referenced with clinical assessments. As with any observational study there is the potential confound of a placebo effect, and given that cannabis products advertised as containing higher THC contents are generally more expensive to consumers, they may be subject to a buyer’s justification effect (magnified appreciation to justify an added expense of purchase). Another limitation is that the ReleafApp may be better suited to tracking the more imme- diate responses of users of concentrates, flower, and to some extent, tinctures versus the longer to peak effects of edibles and topicals. It is also possible that people who choose not to use the ReleafApp have different experiences with product characteristics than those who do use the app. Discussion Another limitation was the inability to distinguish subtleties across product types, such as pipes, which can vary in material construction and potential chemical reactions (e.g., hydrolysis via water pipes). Finally, we anticipate that greater nuances exist in the effects of product characteristics, and particularly, cannabinoid contents across symptom categories, but the current study is lim- ited by our sample size within each respondent subgroup. Future research will capitalize on our ever increasing sample size to analyze the pharmacodynamic interactions of major cannabinoids and other organic compounds including terpenoids, as well as the harm of cannabis production practices. For example, the use of solvents to extract cannabinoids for making concentrates used in making non-flower products (e.g., edibles, tinctures) may place patients at risk for respiratory and cardiovascular problems40 and be a cause of increased emergency cases of Cannabinoid Hyperemesis Syndrome41. yp y In conclusion, rapid increases in the popularity of medical cannabis and the associated increase in the number of patients highlight the urgency of investigating and directing effective usage. Cannabis use carries the risk of addiction and short-term impairments in cognitive and behavioral functioning, including the potential for safety issues in the workplace or while driving. However, with preliminary evidence that cannabis may treat an even wider range of conditions than those tracked in this study, including cancer42,43, it is imperative that the scientific community develop innovative strategies such as the use of mobile technology for measuring the multidimen- sional relationships among cannabis product characteristics, patient health conditions, perceived symptom relief, and side effect manifestation. Data Availabilityh y The data that support the findings of this study are available from MoreBetter Ltd. but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of MoreBetter Ltd. References Does cannabidiol have a role in the treatment of schizophrenia? Schizophrenia Research. 176, 281–29 (206). ( ) 10. Iseger, T. A. & Bossong, M. G. A systematic review of the antipsychotic properties of cannabidiol in humans. Schizophrenia Research. 162, 153–161 (2015). ( ) 10. Iseger, T. A. & Bossong, M. G. A systematic review of the antipsychotic properties of cannabidiol in humans. Schizophrenia Research. 162, 153–161 (2015). , ( ) 11. Christian, D. et al. Cannabis with high cannabidiol content is associated with fewer psychotic experiences. Schizophrenia Research. 130, 216–221 (2011). , ( ) 11. Christian, D. et al. Cannabis with high cannabidiol content is associated with fewer psychotic experiences. Schizophrenia Research. 130, 216–221 (2011). 2. National Institute of Drug Abuse. Marijuana as Medicine, https://www.drugabuse.gov/publications/drugfacts/marijuana-medicine on (2018). ational Institute of Drug Abuse. Marijuana as Medicine, https://www.drugabuse.gov/publications/drugfacts/marijuana-medicine n (2018). 13. Ramo, D. E., Popova, L., Grana, R., Zhao, S. & Chavez, K. Cannabis Mobile Apps: A Content Analysis. Eysenbach G, ed. JMIR mHealth and uHealth. 3, e81 (2015). 14 R l fA htt // l f / (2018) ( ) 14. ReleafApp, https://releafapp.com/ (2018). 14. ReleafApp, https://releafapp.com/ (2018). pp p pp ( ) 15. Vigil, J. M., Stith, S. S., Adams, I. M. & Reeve, A. P. Associations between medical cannabis and prescription opioid use in chronic pain patients: A preliminary cohort study. PLoS ONE. 12, e0187795 (2017). y y 16. Kindred, J. H. et al. Cannabis use in people with Parkinson’s disease and Multiple Sclerosis: A web-based investigation. Complementary Therapies in Medicine. 33, 99–104 (2017). Complementary Therapies in Medicine. 33, 99–104 (2017). p yh p 7. Bonn-Miller, M. O., Babson, K. A. & Vandrey, R. Using cannabis to help you sleep: Heightened frequency of medical cannabis use among those with PTSD. Drug and Alcohol Dependence. 136, 162–165 (2014).hf h 17. Bonn-Miller, M. O., Babson, K. A. & Vandrey, R. Using ca g g p 18. Costain, W. F. The effects of cannabis abuse on the symptoms of schizophrenia: Patient perspectives. International Journal of Mental Health Nursing. 17, 227–235 (2008). 19. MacCallum, C. A. & Russo, E. B. Practical considerations in medical cannabis administration and dosing. European Journal of Internal Medicine. 49, 12–19 (2018). 0. Fischer, B. et al. Lower-risk cannabis use guidelines: A comprehensive update of evidence and recommendations. American Journa of Public Health. 107, e1–e12 (2017). 21. Stith, S. S., Vigil, J. M., Brockelman, F., Keenan, K. Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 References References 1. Piper, B. J. et al. Substitution of medical cannabis for pharmaceutical agents for pain, anxiety, and sleep. J Psychopharmacol. 31, 569 575 (2017) References 1. Piper, B. J. et al. Substitution of medical cannabis for pharmaceutical agents for pain, anxiety, and sleep. J Psychopharmacol. 31 569–575 (2017). 1. Piper, B. J. et al. Substitution of medical cannabis for pharmaceutical agents for pain, anxiety, and sleep. J Psychopharmacol. 31, 569 575 (2017) ( ) 2. Hill, K. P. & Weiss, R. D. Minimal physical health risk associated with long-term cannabis use—but buyer beware. JAMA. 315, 2338–2339 (2016). 2. Hill, K. P. & Weiss, R. D. Minimal physical health risk associated with long-term cannabis use—but buyer beware. JAMA. 315 2338–2339 (2016). 3. Rubin, R. Medical marijuana is legal in most states, but physicians have little evidence to guide them. JAMA. 317, 1611–1613 (2017). 4. National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Population Health and Public Health Practice; Committee on the Health Effects of Marijuana: An Evidence Review and Research Agenda. Washington (DC): National Academies Press (US) (2017). 3. Rubin, R. Medical marijuana is legal in most states, but physicians have little evidence to guide them. JAMA. 317, 1611–1613 (2017). 4. National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Population Health and Public Health Practice; Committee on the Health Effects of Marijuana: An Evidence Review and Research Agenda. Washington (DC): National Academies Press (US) (2017). 5. Stith, S. S. & Vigil, J. M. V. Federal barriers to Cannabis research g 6. Di Forti, M. et al. Proportion of patients in south London with first-episode psychosis attributable to use of high potency cannabis: A case-control study. The Lancet Psychiatry. 2, 233–238 (2015). g 6. Di Forti, M. et al. Proportion of patients in south London with first-episode psychosis attributable to use of high potency cannabis A case-control study. The Lancet Psychiatry. 2, 233–238 (2015). h 7. Gage, S. H. et al. Assessing causality in associations between cannabis use and schizophrenia risk: a two-sample Mend randomization study. Psychological Medicine. 47, 971–980 (2017). y y g 8. Fakhoury, M. Could cannabidiol be used as an alternative to antipsychotics? Journal of Psychiatric Research. 80, 14–21 (2016). G A & l T b d l h l h f h h ? S h h h 9. Gururajan, A. & Malone, D. T. Acknowledgementsh g This research was funded in part by donations to the University of New Mexico Medical Cannabis Research Fund (mcrf.unm.edu). All authors had access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analyses. www.nature.com/scientificreports/ Labeling accuracy of cannabidiol extracts sold online. JAMA. 318, 1708–1709 (2017).fl 38. Burstein, S. Cannabidiol (CBD) and its analogs: a review of their effects on inflammation. Bioorganic & Medicinal Chemistry. 23, 1377–1385 (2017). 39. McPartland, J. M., Duncan, M., Di Marzo, V. & Pertwee, R. G. Are cannabidiol and Δ9-tetrahydrocannabivarin negative modu of the endocannabinoid system? A systematic review. British Journal of Pharmacology. 172, 737–753 (2015).h of the endocannabinoid system? A systematic review. British Journal of Pharmacology. 172, 737–753 (2015).h 40. Piano, M. Cannabis smoking and cardiovascular health: It’s complicated. Clin. Pharmacol. Ther. 102, 191–193 (2017). g ph 41. Richards, J. R. Cannabinoid hyperemesis syndrome: Pathophysiology and treatment in the emergency department Emergency Medicine. 54, 354–363 (2018). g y ( ) 42. Fowler, C. Delta9-tetrahydrocannabinol and cannabidiol as potential curative agents for cancer: A critical examination of the preclinical literature. Clin. Pharmacol. Ther. 97, 587–596 (2015).h 2. Fowler, C. Delta9-tetrahydrocannabinol and cannabidiol as potential curative agents for cancer: A critical examination of the preclinical literature. Clin. Pharmacol. Ther. 97, 587–596 (2015).h ph 3. Śledziński, P., Zeyland, J., Słomski, R. & Nowak, A. The current state and future perspectives of cannabinoids in cancer biology Cancer Medicine. 7, 765–775 (2018). h 43. Śledziński, P., Zeyland, J., Słomski, R. & Nowak, A. The current state and future perspectives of cannabinoids in cancer biology. Cancer Medicine. 7, 765–775 (2018). Author Contributions J.M.V. and S.S.S. conceived the study. F.B., K.K., B.H. independently designed and developed the ReleafApp and server infrastructure as part of their effort to help create an education tool for medical cannabis patients. S.S.S. conducted the analyses. J.M.V. and S.S.S. drafted the manuscript. All authors contributed substantially to its intellectual content and revision. www.nature.com/scientificreports/ 4. Eggers, M. E. et al. Youth use of electronic vapor products and blunts for administering cannabis. Addictive Behaviors. 70, 79–82 (2017). 24. Eggers, M. E. et al. Youth use of electronic vapor products and blunts for administering cannabis. Addictive Behaviors. 70, 79 82 (2017). 25 Marijuana Policy Project https://www mpp org/issues/medical-marijuana/state-by-state-medical-marijuana-laws/medical- (2017). 25. Marijuana Policy Project, https://www.mpp.org/issues/medical-marijuana/state-by-state-medical-marijuana-laws/medical- marijuana-patient-numbers/ (2018) 5. Marijuana Policy Project, https://www.mpp.org/issues/medical-marijuana/state-by-state-medical-marijuana-laws/medical marijuana-patient-numbers/ (2018). j p 6. Dinis-Oliveira, R. J. Metabolomics of Δ9-tetrahydrocannabinol: implications in toxicity. Drug Metabolism Reviews. 48, 80–87 (2016). ( ) 7. Lefever, T. W. et al. Vaping synthetic cannabinoids: A novel preclinical model of e-cigarette use in mice. Substance Abuse: Research and Treatment. 11, https://doi.org/10.1177/1178221817701739 (2017). 28. Manwell, L. A. et al. A vapourized Δ9-tetrahydrocannabinol (Δ9-THC) delivery system part I: Development and validation of a pulmonary cannabinoid route of exposure for experimental pharmacology studies in rodents. Journal of Pharmacological and Toxicological Methods. 70, 120–127 (2014). 29. Nguyen, J. D. et al. Inhaled delivery of Δ9-tetrahydrocannabinol (THC) to rats by e-cigarette vapor technology. Neuropharmacology. 109, 112–120 (2016). 30. Vandrey, R. et al. Pharmacokinetic profile of oral cannabis in humans: Blood and oral fluid disposition and relation to pharmacodynamic outcomes. Journal of Analytical Toxicology. 41, 83–99 (2017).i p y f y gy 1. Varlet, V. et al. Drug vaping applied to cannabis: Is “Cannavaping” a therapeutic alternative to marijuana? Scientific Reports. 6, 25599 https://www.nature.com/articles/srep25599 (2016). p p 32. Cohen, N. L., Heinz, A. J., Ilgen, M. & Bonn-Miller, M. O. Pain, Cannabis Species, and Cannabis Use Disorders. Journal of Studies on Alcohol and Drugs. 77, 515–520 (2016).f g 3. Pearce, D., Mitsouras, K. & Irizarry, K. Discriminating the Effects of Cannabis sativa and Cannabis indica: A Web Survey of Medica Cannabis Users. Journal of Alternative & Complementary Medicine. 20, 787–791 (2014).h f p y 4. Andreae, M. H. et al. Inhaled cannabis for chronic neuropathic pain: an individual patient data meta-analysis. The Journal of Pain 16, 1221–1232 (2015). 35. Mücke, M., Phillips, T., Radbruch, L., Petzke, F. & Häuser, W. Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database of Systematic Reviews. 3, Art. No.: CD012182. https://doi.org/10.1002/14651858.CD012182.pub2 (2018). f y g 36. Belendiuk, K. A., Babson, K. A., Vandrey, R. & Bonn-Miller, M. O. Cannabis species and cannabinoid concentration preference among sleep-disturbed medicinal cannabis users. Addictive Behaviors. 50, 178–181 (2015). g p 37. Bonn-Miller, M. O. et al. References & Hall, B. Patient-reported symptom relief following medical cannabis consumption. Frontiers in Pharmacology. 9, 916, https://www.frontiersin.org/articles/10.3389/fphar.2018.00916/full (2018).fl 22. Vigil, J. M. et al. Effectiveness of raw, natural medical Cannabis flower for treating insomnia under naturalistic conditions. Medicines. 5, 75. https://www.mdpi.com/2305-6320/5/3/75 (2018). , p p ( ) 3. Blundell, M., Dargan, P. & Wood, D. A cloud on the horizon-a survey into the use of electronic vaping devices for recreational drug and new psychoactive substance (NPS) administration. QJM. 111, 9–14 (2018). p p ( ) undell, M., Dargan, P. & Wood, D. A cloud on the horizon-a survey into the use of electronic vaping devices for recreational drug nd new psychoactive substance (NPS) administration. QJM. 111, 9–14 (2018). Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1 7 www.nature.com/scientificreports/ upplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-39462-1. Competing Interests: The authors are associated with the University of New Mexico Medical Cannabis Research Fund, which was designed to support the costs of research investigating the safety and effectiveness of medical cannabis. F.B., K.K. and B.H. are employed by MoreBetter Ltd. The authors have no other conflicts of interest. ublisher’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 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 8 Scientific Reports \| (2019) 9:2712 \| https://doi.org/10.1038/s41598-019-39462-1
https://openalex.org/W3002777317	https://europepmc.org/articles/pmc6923219?pdf=render	English	null	Cognitive Assessment Tools for Screening Older Adults With Low Levels of Education: A Critical Review	Frontiers in psychiatry	2,019	cc-by	10,076	Systematic Review doi: 10.3389/fpsyt.2019.00878 published: 13 December 2019 Cognitive Assessment Tools for Screening Older Adults With Low Levels of Education: A Critical Review José Wagner Leonel Tavares-Júnior 1†, Ana Célia Caetano de Souza 2,3†, Gilberto Sousa Alves 4, Janine de Carvalho Bonfadini 1,3, José Ibiapina Siqueira-Neto 1 and Pedro Braga-Neto 1,2,3 Keywords: mental status tests, dementia tests, literacy, educational status, mild cognitive impairment 1 Department of Clinical Medicine, Division of Neurology, Universidade Federal do Ceará, Fortaleza, Brazil, 2 Center of Health Sciences, Universidade Estadual do Ceará, Fortaleza, Brazil, 3 Neurology Service, Hospital Universitário Walter Cantídio, Universidade Federal do Ceará, Fortaleza, Brazil, 4 Translational Psychiatry Research Group, Universidade Federal do Maranhão, São Luís, Brazil Edited by: Fabrizio Piras, Santa Lucia Foundation (IRCCS), Italy Edited by: Fabrizio Piras, Santa Lucia Foundation (IRCCS), Italy Introduction: Cognitive assessment of older adults who are either illiterate or with low levels of education is particularly challenging because several battery tasks require a certain educational background. Early detection of mild cognitive impairment (MCI) in the elderly using validated screening tools is of great importance since this population group could benefit from new drugs that are being investigated for the treatment of dementias. Cutoff scores for psychometric properties of cognitive tests are not well established among adults with low levels of education. The present study aimed to critically review the literature on cognitive assessment tools for screening cognitive syndromes including MCI and Alzheimer’s disease (AD) in older adults with low levels of education. Reviewed by: Sara L. Weisenbach, Stony Brook Medicine, United States Deana Davalos, Colorado State University, United States Correspondence: Gilberto Sousa Alves gsalves123@hotmail.com †These authors have contributed equally to this work Methods: We conducted a systematic search of MEDLINE, LILACS, Cochrane, and SCOPUS electronic databases of cross-sectional and prospective studies with adults over 55 years of age. Specialty section: This article was submitted to Aging Psychiatry, a section of the journal Frontiers in Psychiatry Received: 16 September 2019 Accepted: 07 November 2019 Published: 13 December 2019 Results: We found a significant number of assessment tools available (n = 44), but only a few of them showed diagnostic accuracy for the diagnosis of MCI and AD in older adults with low levels of education: the Mini-Mental State Exam; the Montreal Cognitive Assessment; the Persian Test of Elderly for Assessment of Cognition and Executive Function; the Six-Item Screener; and the Memory Alteration Test. Few studies evaluated individuals with low levels of education, with a wide range of cutoff scores and cognitive test batteries. METHODS Mild cognitive impairment (MCI) is the intermediate stage between cognitive decline of healthy aging and dementia (2). The prevalence of MCI is 12–18% among adults over 65 years of age and the annual progression rates from MCI to AD are 10–15% (3, 4). Early detection of MCI in the elderly using validated screening tools is of great importance since this population group could benefit from new drugs that are being investigated for the treatment of neurodegenerative diseases including AD. Furthermore, evidence shows that, since MCI is a transitional phase between normal aging and AD, there is less brain involvement and those affected are more likely to benefit from drug therapies (5, 6). An integrative literature review was conducted to gather and summarize the body of evidence available from original articles. This integrative review study included six stages: Step 1—formulation of the central research question (theme identification); Step 2—definition of inclusion and exclusion criteria and literature search; Step 3—categorization of primary studies (definition of data to be extracted from the selected studies); Step 4—assessment of the studies included; Step 5— interpretation of results; Step 6—knowledge synthesis of the results obtained from the studies assessed (18–20).h The central research question was formulated using the PVO method where P is the study population (adults over 55 years of age with low education, i.e., 4 years of formal education or less); V is the variable (cognitive assessment tools); and O is the outcome (MCI and AD). Cognitive assessment tools are commonly used for screening impairment, differential diagnosis, determining disease severity, as well as monitoring disease progression in patients (7). A major challenge for the initial assessment of age-related cognitive changes is to find a screening tool that is both sensitive and specific for differential diagnosis of cognitive impairment. Both ceiling effects and floor effects limit the ability of a test or some of its items to accurately assess cognitive decline (8). These effects have been reported in several studies and they are primarily related to educational background (9). The ceiling effect occurs when score distribution is skewed and a measurement is determined by the proportion of people scoring at the high end, thereby preventing to detect health improvements. The opposite is the floor effect that occurs when a measurement is determined by the proportion of people scoring at the low end, thereby preventing to detect health declines (10). Citation: Tavares-Júnior JWL, de Souza ACC, Alves GS, Bonfadini JdC, Siqueira-Neto JI and Braga-Neto P (2019) Cognitive Assessment Tools for Screening Older Adults With Low Levels of Education: A Critical Review. Front. Psychiatry 10:878. doi: 10.3389/fpsyt.2019.00878 Conclusion: We found that a small number of studies evaluated adults with 4 years of formal education or less. Our findings further support the importance of developing specific tools for the assessment of older adults with low levels of education. December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 1 Cognitive Screening in Low Educated Elderlies Tavares-Júnior et al. INTRODUCTION assessment tools for screening older adults with low levels of education. A better understanding of the accuracy of different cognitive batteries is crucial for early diagnosis and intervention, and epidemiological studies are needed to further explore how education background affects an individual’s performance on different cognitive dimensions. Dementia is characterized by cognitive impairment and loss of function (1). The growth of population aging over the past few decades has been associated with an increase in cognitive disorders. Data from Alzheimer’s Disease International (ADI) shows there were 46.8 million people living with dementia worldwide in 2015, and it is estimated this number will be 74.5 million in 2030 and 131.5 million in 2050. Alzheimer’s disease (AD) is the most common form of dementia and accounts for 50–70% of dementia cases. There were an estimated 26 million people living with AD in 2015 and it is believed there will be as many as 41 million by 2030 and 72 million by 2050 (1). f The present study aimed to critically review the literature on cognitive assessment tools for screening cognitive syndromes including MCI and AD in older adults with low levels of education. Frontiers in Psychiatry \| www.frontiersin.org METHODS The guiding question of our review was: "Which assessment tools are used for cognitive screening of MCI or AD in older adults with low levels of education?" The inclusion criteria were English language articles in the electronic databases Medical Literature Analysis and Retrieval System Online (MEDLINE), Latin American and Caribbean Health Sciences Literature (LILACS), Cochrane, and SCOPUS; cross-sectional or prospective design; outpatient or population-based samples of adults over 55 years of age with low education (4 years of education or less); and assessment of the use of cognitive tools for MCI or AD diagnosis. We did not search the following information sources for this review: guidelines; institutional protocols; self-administered or telephone- based cognitive assessment instruments; and studies that used cognitive assessment instruments for diagnosing other psychiatric or neurological conditions. The publications were individually searched and selected by two investigators during June and July 2019. Another important aspect is to have available free, easy-to-use assessment tools that do not require specialized training and have the ability to accurately discriminate cognitive decline in adults with normal aging, MCI, and dementia (4). Cognitive assessment of older adults who are either illiterate or with low levels of education is particularly challenging because several battery tasks require a certain educational background (11–13). There are an estimated 758 million illiterate adults in the world (11) and 13 million people are estimated to be illiterate in Brazil (11). Prospective cohort studies have shown an association between low education and higher risk of developing AD (14–17). Yet, few studies have examined the performance of cognitive assessment tools in adults with low education.f The Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines (21) were used as a basis for the search and selection of studies (Figure 1). A questionnaire was developed to help data extraction (22). Two matrices were constructed to present the results: the first one included study characteristics and the second one included cognitive tools and main results reported. A search strategy was created to conduct searches in the following databases: MEDLINE via PubMed from the US National Library of Medicine; LILACS; Cochrane; and SCOPUS via Coordenação de Aperfeiçoamento de Pessoal de Nível Superior Cutoff scores for psychometric properties of cognitive tests are not well established among adults with low levels of education. METHODS Furthermore, there is a scarcity of studies evaluating December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 2 Tavares-Júnior et al. Cognitive Screening in Low Educated Elderlies FIGURE 1 \| Flowchart of study selection. FIGURE 1 \| Flowchart of study selection. (CAPES) with no time restriction. To expand our search, we chose to use natural controlled language. The following descriptors (bold), synonyms, natural language, and Boolean operators were used to cross-check the databases: MEDLINE (Medical Subject Headings [MeSH]: search strategy—(aged or elderly or old or elder) and (literacy or illiteracy or education or “low education”) and (“mental status and dementia tests” or tool or instrument or status or test) and (“Alzheimer disease” or alzheimer´s) and (“mild cognitive impairment” or “cognitive dysfunction”). Frontiers in Psychiatry \| www.frontiersin.org RESULTS Figure 1 shows the flowchart of the study selection process according to the PRISMA guidelines. A total of 2,086 articles were retrieved and read. Thirty-six studies were selected for our review. Table 1 describes the studies assessed. The sample sizes ranged from 50 to 10,432 participants. The studies were conducted in 17 countries, and most of them (13.88%) were from China and Spain. A wide range of cognitive assessment tools (n = 44) were used for MCI and AD diagnosis (Table 1). Of these, the Mini- Mental State Examination (MMSE) (59) was the most frequently used (86.11%), followed by the Montreal Cognitive Assessment (MoCA) (60) (27.77%). To minimize selection bias (misinterpretation of results and study design), the literature search and data extraction were conducted by two investigators independently and any discrepancies were resolved by consensus. December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 3 TABLE 1 \| Articles included in the integrative literature review. Tavares-Júnior et al. RESULTS Authors, Year Country Study Design Primary Study Objective Level of Education Sample Size Participant Age Cognitive Assessment Tool used Main Results Sun Y et al., 2014 (23) Taiwan Cross- sectional To assess prevalence of MCI and dementia Illiterate 32.1% ≤6 years 45.2% > 6 years 22.6% 10,432 (dementia 929, MCI 2,049, indeterminate 419) Mean age (SD) 76.2 years (6.7) MMSE Taiwan version Prevalences: dementia 8.04% MCI 18.76% Kim KW et al., 2011 (24) South Korea Cross- sectional To assess prevalence of MCI and dementia in adults over 65 years of age Illiterate 31.3% 1–6 years 38.1% > 7 years 30.6% 6,141 65–69 years 32.2% > 70 years 67.8% CERAD-Korean version, Clinical Assessment Battery Prevalences: dementia 8.1%, AD 5.7%, MCI 24.1% Chen MR et al., 2010 (25) China Cross- sectional To validate SIS for quick detection of cognitive impairment Illiterate 7.2% 1–6 years 16.4% > 6 years 76.3% 1,976 (healthy aging 475, MCI 440, AD 1,061) Mean age (SD) 71.87 years (8.71) SIS, MMSE SIS: MCI Sn 34.3%, Sp 90.1%, MMSE: cutoff score <4 years of schooling ≤17; Sn 94.3%; Sp 95.0% Chang J et al., 2014 (26) Hong Kong Cross- sectional To assess the effect of education on tools for screening older population Mean schooling 4.7 years (SD 4.6; 0–20) 788 (AD 405, controls 383) Mean age (SD) 72.08 years (7.27) MMSE, ADAS-Cog, Verbal Fluency, Abstract Thinking, and Visual/DS Effect of educational background on MMSE, language sub-item O’Bryant SE et al., 2013 (27) United States Cross- sectional To characterize a Mexican American population with MCI and AD Mean schooling (SD) AD 5.9 years (4.5) MCI 6.6 years (4.2) Controls 8.1 years (4.2) 1.069 non-Hispanic white (n = 633: AD 160, MCI 97, controls 376) Mexican American (n = 436, AD 35, MCI 67, controls 337) Mean age (SD) AD 73.6 years (9.1) MCI 61.9 years (12.3) Controls 58.7 years (9.9) MMSE Mean MMSE score (SD) AD 18.5 (5.0) MCI 24.7 (3.6) Controls 27.5 (2.8) de Paula JJ et al., 2013 (28) Brazil Cross- sectional To validate an unstructured neuropsychological assessment tool for clinical use Mean schooling (SD) MCI 4.71 years (4.00) AD 4.82 years (3.46) Controls 5.22 years (4.29) 274 (96 controls, MCI 85, AD 93) Mean age (SD) Controls 72.61 years (7.76), MCI 73.18 years (8.46), AD 74.57 years (6.65) RAVLT, FAB, verbal fluency, SDT, CDT, DS, TT and TN-LIN Study protocol: Sn >70%, Sp >70% for AD and MCI Sánchez Benevides G et al., 2014 (29) Spain Cross- sectional To evaluate neuropsychological assessment tool for MCI and AD Mean schooling (SD) MCI 8.0 years (4.7) AD 7.6 years (4.6) Controls 10.4 years (5.4) 535 (controls 356, MCI 79, AD 100) Mean age (SD) Controls 64.9 years (9.3) MCI 72.8 years (6.5) AD 74.7 years (7.5) MMSE, DS, WAIS, TMT, SDMT, BNT, TT, SVOSPB, JLO, verbal fluency, ROCF, FCSRT, phonemic fluency, SCWIT, TLDU Mean MMSE scores (SD) Controls 28.7 (1.5) MCI 25.7 (2.2) AD 20.2 (4.0) FCSRT showed best diagnostic accuracy for AD vs. RESULTS controls Mellor D et al., 2016 (30) China Cross- sectional To assess effectiveness for discriminating MCI or AD vs. healthy controls Mean schooling (SD) MCI 5.17 (4.78) AD 3.72 (4.14) 1.027 (controls 708, AD 267, MCI 50) Mean age (SD) 72.54 years (8.40) MMSE Chinese version, MoCA (cutoff score, Sn, Sp) MCI: MMSE 25.50/68/83 MoCA 22.50/.87/73 Javadi PSHS et al., 2015 (31) Iran Cross- sectional To characterize illiterate and literate older adults; PEACE scale cutoff scores for AD Different levels of education—illiterate and literate 101 (controls 33, MCI 30, AD 38) Mean age (SD) AD 74.60 years (8.02) MCI 72.5 years (7.2) Controls 67.84 years (7.29) PEACE, GPCOG, FAST, MMSE, WMS MMSE < 4 years of schooling MCI 18.75 (1.75) AD 12.64 (3.78) PEACE AD 67.5 (Sn 75.8%, Sp 97.4%) (C ti d) December 2019 \| Volume 10 \| Article 878 (Continued) TABLE 1 \| Continued Tavares-Júnior et al. Authors, Year Country Study Design Primary Study Objective Level of Education Sample Size Participant Age Cognitive Assessment Tool used Main Results Chong MS et al., 2010 (32) Singapore Cross- sectional To compare FAB-X and MMSE for screening early cognitive impairment Mean schooling (SD) MCI 7,1 years (4.4) Controls 9.6 years (4.4) 180 (controls 100, MCI 21, dementia 59) Mean age (SD) Controls 63.7 years (6.51) MCI 69.3 years (7.91) FAB Chinese version, MMSE, verbal fluency, BNT, WAIS FAB MCI and mild dementia Cutoff scores 12/13 (Sn 92%; Sp 78.7%) MMSE (Sn 77%; Sp 91.2%) More effective when combined Ng A et al., 2013 (33) Singapore Cross- sectional To assess MoCA diagnostic accuracy for MCI and AD in older adults Mean schooling (SD) MCI 10.93 years (4.28) AD 6.97 years (4.47) Controls 12.07 years (3.20) 212 (controls 103, MCI 49, AD 60) Mean age 62.35 years MMSE, MoCA MoCA cutoff scores MCI <26 (≤10 years of schooling); < 27 (> 10 years of schooling), Sn >94% Saka E et al., 2006 (34) Turkey Cross- sectional To assess performance of ECR for discriminating dementia, AD and MCI vs. controls Mean schooling (SD) MCI 8.4 years (5.0) AD 6.7 years (4.9) Controls 8.4 years (4.9) 113 (MCI 80, controls 33) Mean age (SD) MCI 69.4 years (8.3) AD 73.8 years (6.1) Controls 72.7 years (6.7) MMSE, ECR ECR showed good performance to discriminate AD vs. controls (AUC 0.990), moderate performance for MCI vs. RESULTS controls (AUC 0.625) Tognoni G et al., 2005 (35) Italy Cross- sectional To assess prevalence of dementia in older population < 4 years (15.8%), > 4 years (84.2%) 1.600 (AD 68, MCI 149) Mean age (SD) 74.65 years (7.26) MMSE, CAMDEX Prevalence of amnesic MCI 4.9%, AD 4.2% Afgin AE et al., 2012 (36) Israel Cross- sectional To assess prevalence of MCI and AD and conversion rate from MCI to AD within a year or more 0 years 51% 1–4 years 23% 5–8 years 21% > 8 years 5% 944 (controls 497, MCI 303, SD 13, VD 39, AD 92) Mean age (years) (SD) DA 78.5 (7.7), MCI 72.8 (6.1), Controls 70.7 (5.5) by age groups (%) 65–69 (39.4%) 70–79 (46.4%) ≥80 (14.1%) MMSE Arabic version, Brookdale Cognitive Screening Test Prevalence AD 9.8%, MCI 32.1% MMSE <4 years of schooling cutoff scores (SD) MCI illiterate 17.8 (1.9), 1–4 years of schooling 19.6 (3.4); Chaves ML et al., 2009 (37) Brazil Cohort To assess incidence rate of AD and MCI in a community of older adults Mean schooling (SD) 9.06 years (5.50) 345 (no details available) Mean age (SD) 70.37 years (7.15) MMSE Incidence rate per 1,000 MCI 13.2 AD 14.8 Custodio N et al., 2017 (38) Peru Cross- sectional MAT performance to discriminate controls, MCI and AD in adults with low education Mean schooling (SD) AD 2.65 years (1.28) MCI 2.53 years (1.46) Controls 2.57 years (1.45) 247 (controls 121, AD 81, MCI 45) Mean age (SD) AD 74.18 years (3.81) MCI 71.09 years (4.20) Controls 69.53 years (4.11) MAT, RAVLT, WMS, TMT A and B, ROCF, BNT, WCST, DS, WAIS, MMSE MAT AD vs. MCI AUC 99.60% MCI vs. (Continued) RESULTS controls AUC 99.56% Mean MMSE scores (SD): MCI 21.36 (0.98); MAT MCI 30.53 (2.54) Ravaglia G et al., 2008 (39) Italy Cohort To assess incidence and prevalence rates of MCI in older adults Mean schooling (SD) 4.3 years (2.3) 1.016 (controls 865, MCI 75, dementia 60, indeterminate 19) Mean age (SD) Controls 73.6 years (6.1) MCI 78.1 years (8.3) MMSE, Mental Deterioration Battery MCI prevalence 7.7%; incidence rate in 4 years 76.8 per 1,000 persons-years Rahman TTA et al., 2009 (40) Egypt Cross- sectional To assess validity of MoCA version to detect MCI Mean schooling MCI 8.2 years (5.5) 184 (MCI 94, controls 90) Mean age (SD) 64.5 years (6.8) MoCA Arabic version, CAMCOG MoCA MCI Sn 92.3% Sp 85.7% Freitas S et al., 2013 (41) Portugal Cross- sectional To validate MoCA for screening of MCI and AD Mean schooling (SD) MCI 6.50 years (4.56) AD 6.2 years (4.11) Controls 6.39 years (4.30) 360 (MCI 90, AD 90, controls 180) Mean age (SD) 71.86 years (7.895) MMSE, MoCA Cutoff scores/Sn/Sp MCI MoCA <22/81/77, MMSE <29/67/72 Cognitive Screening in Low Educated Elderlies December 2019 \| Volume 10 \| Article 878 Cutoff scores/Sn/Sp MCI MoCA <22/81/77, MMSE <29/67/72 (Continued) Frontiers in Psychiatry \| www.frontiersin.org TABLE 1 \| Continued Tavares-Júnior et al. \| Authors, Year Country Study Design Primary Study Objective Level of Education Sample Size Participant Age Cognitive Assessment Tool used Main Results Borson S et al., 2005 (42) United States Cross- sectional To compare the Mini- Cog Test vs. MMSE for screening cognitive impairment Literate 76%, Semi- literate and illiterate 24% 371 (controls 140, AD 112, MCI 71, other dementias 48) Mean age Controls 73 years MCI 74 years AD 78 years CASI, MMSE, Mini-Cog AD and MCI diagnostic accuracy Mini-Cog 83% MMSE 81% Boycheva E et al., 2018 (43) Spain Cross- sectional To assess clinical performance of MDRS for screening MCI and AD in older adults Mean schooling (SD) 7.08 years (3.57) 125 (AD 45, MCI 37, controls 43) Mean age 75.12 years (6.83) MDRS, MMSE, WAT, FCSRT, BNT, SCWIT, GSBT, verbal fluency, phonemic fluency, WAIS MDRS-2 cutoff score MCI vs. RESULTS controls 131 (Sn 89%, Sp 81%) Pezzotti P et al., 2008 (44) Italy Cross- sectional To compare agreement of MMSE between primary care and specialist practitioners 0–5 years 68.1% > 5 years 31.9% 317 (MCI 40, AD 95, other dementias 98, healthy controls 84) Mean age not available MMSE, Mental Deterioration Battery, SCWIT Mean MMSE score Primary care providers 15.8 Specialists 17.4 (kappa 0.86) Julayanont P et al., 2015 (45) Thailand Cross- sectional To assess validity of MoCA-B for MCI in a population with low education Mean schooling (SD) MCI 2.9 years (1.7) 85 (controls 43, MCI 42) Mean age (SD) MCI 70.2 years (6.6) MoCA-B, MMSE Thai version Mean score (SD) MMSE MCI 18.9 (3.0) MoCA-B 21.3 (3.8) Matías-Guiu JA et al., 2017 (46) Spain Cross- sectional To validate LASSI-L scale for MCI and AD diagnosis Mean schooling (SD) Controls 8.52 years (4.98), MCI 7.61 years (4.79) AD 7.06 years (4.20) 164 (controls 97, MCI 34, AD 33) Mean age (SD) 73.4 years (10.0) LASSI-L, MMSE LASSI-L discrimination AD vs. controls AUC 0.986; MCI vs. 0–5 years 68.1% > 5 years 31.9% RESULTS controls AUC 0.909 Chu LW et al., 2015 (47) Hong Kong Cross- sectional To validate MoCA Chinese version for screening MCI and AD in older adults Mean schooling (SD) Controls 6.97 years (4.69) MCI 4.62 years (5.19) AD 4.56 years (5.00) 266 (controls 115, MCI 87, AD 64) Mean age 75.3 years MoCA Cantonese Chinese version, MMSE Chinese version MoCA cutoff score/Sn/Sp MCI 22–3/78%/73%, AD 19–20/94%/92% Kurt P et al., 2014 (48) Turkey Cross- sectional To develop a composite score for DEKOD in an older population < 4 years of schooling 45.7% > 5 years of schooling 54.3% 444 (controls 338, dementia 53, MCI 53) Mean age (SD) controls 70.7 years (5.4) dementia 74 years (7.8) MCI 71.7 years (5.6) DEKOD, MMSE Turkish version MMSE <4 years of schooling: cutoff score/Sn/Sp Dementia 17–18/95%/83% MCI 22–23/67%/55% DEKOD Dementia 49–50/91%/90% MCI 60–61/70%/65% Bae JB et al., 2015 (49) South Korea Cross- sectional To assess incidence rates of AD and MCI within 3.5 years of follow-up Mean schooling (SD) 6.8 years (5.4) 181 (not available) Mean age 71.7 years MMSE, Korean CERAD, Clinical Assessment Battery, CERAD-K-N Incidence rate per 1,000 persons-years AD 7.9 MCI 28.1 Paddick SM et al., 2015 (50) Tanzania Cohort To assess outcomes, prevalence and profiles of patients with MCI in a rural community Illiterate with MCI 55.5% Literate (at least elementary schooling) 44.5% 296 (MCI 46, dementia 78, controls 172) Mean age 82 years CERAD Prevalence of MCI 7% Choi SJ et al., 2008 (51) South Korea Cross- sectional To assess prevalence of AD and MCI in older Korean adults Mean schooling (SD) MCI 4.93 years (3.27) DA 1.69 years (3.09) 175 (controls 102, MCI 57, AD 16) Mean age (SD) 74.3 years (16.7) Korean MMSE, CERAD-K Prevalences: AD 9.0% MCI 32.9% (Continued) TABLE 1 \| Continued Tavares-Júnior et al. RESULTS \| Co ued Authors, Year Country Study Design Primary Study Objective Level of Education Sample Size Participant Age Cognitive Assessment Tool used Main Results Gavrila D et al., 2009 (52) Spain Cross- sectional To assess prevalence of MCI and dementia in older adults Illiterate 7.8% Less than elementary 19.7%, Elementary or more 72.5% 1,017 (controls 726, MCI 235, AD 30, other dementias 26) Mean age (SD) 73.9 years (6.8) MMSE, CAMDEX, Blessed Dementia Scale Prevalence of amnesic MCI 8.7% Wang Bet al., 2011 (53) China Cross- sectional To describe clinical characteristics of patients in a memory center 0 years 8% 1–5 years 15% > 6 years 77% 2,789 (healthy aging 604, MCI 635, AD 1084, other diagnoses 466) < 50 years 5.2% 50–59 years 14.9% > 60 years 79.9% MMSE Population distribution in a memory center: AD 83.7% MCI 22.8% Zhou Y et al., 2015 (54) United States Cross- sectional How to adjust MoCA for educational background in a Spanish-speaking population Mean schooling (SD) Controls 10.3 years (6.4) MCI 7.1 years (4.8) Dementia 6.8 years (5.5) 50 (AD 18, MCI 21, controls 6, other dementias 5) Mean age (SD) 71.4 years (9.7) MoCA It was required to adjust scores 3–4 points for those <6 years of schooling Chen K et al., 2016 (55) China Cross- sectional To assess performance of MoCA-B Chinese version for screening MCI in older adults Groups by years of schooling (< 6, 7–12, > 12). RESULTS Mean schooling (SD) of those with <6 years: controls 4.8 years (1.7), MCI 3.3 years (2.4), AD 3.7 years (2.5) 704 (MCI 264, AD 160, controls 280) Mean age (SD) of those with <6 years of schooling controls 68.2 years (9.1) MCI 68.5 years (8.5) AD 67.9 years (9.4) MMSE, AVLT, ROCF, BNT, Verbal Fluency, TMT, SCWIT, SDMT, MoCA-B Chinese version Cutoff scores Sn/Sp of those with <6 years of schooling: MCI MoCA-BC 19/87.9/81.0 MMSE 26/86.2/60.3 Matías-Guiu JA et al., 2017 (56) Spain Cross- sectional To compare diagnostic properties of five cognitive screening tools Mean schooling (SD) Controls 8.01 years (5.40) AD 7.10 years (4.07) 160 (AD 92, controls 68) 60–69 years 7.5% 70–79 years 48.8% 80–93 years 43.8% MMSE, ACE-III, MIS, MoCA, RUDAS AD: ACE-III AUC 0.897; RUDAS AUC 0.889; MMSE AUC 0.874; MIS AUC 0.866; MoCA AUC 0.856 Huang YY et al., 2019 (57) China Cross- sectional To compare diagnostic accuracy of MoCA-B and MoCA-Beijing for screening MCI in people with different levels of education No supplementary table available 808 (MCI 295, AD 254, controls 259) Not available MoCA, MMSE, AVLT; ROCF, BNT, Verbal Fluency, TMT, SCWIT, SDMT Performance in discriminating MCI vs. MMSE, Mini-Mental State Examination; BNT, Boston Naming Test; AVLT, Auditory Verbal Learning Test; MoCA, Montreal Cognitive Assessment; MoCA B, MoCA Basic; ACE, Addenbrooke Cognitive Examination; TMSE, Taiwanese Mental State Examination; CERAD, Consortium to Establish a Registry for Alzheimer’s Disease Assessment Packet; SDT, Stick Design Test; TT, Token Test; SDMT, Symbol Digit Modalities Test; SVOSPB, Selected test of the Visual Object and Space Perception Battery; JLO, Judgment of Line Orientation; TLDU, Tower of London Drexel University version; WAT, Word Accentuation Test; GSBT, Gesture Sequences subtest from the Barcelona Test; SIS, Six-Item Screener; ADAS-Cog, Alzheimer’s Disease Assessment Scale’s cognitive subscale; RAVLT, Rey-Auditory Verbal Learning Test; FAB, Frontal Assessment Battery; CDT, Clock Drawing Test; DS, Digit Span; TMT, Trail Making Test; ROCF, Rey-Osterrieth Complex Figure; FCSRT, Free and Cued Selective Reminding Test; SCWIT, Stroop Color-Word Interference Test; PEACE, Persian Test of Elderly for Assessment of Cognition and Executive function; GPCOG, General Practitioner assessment of Cognition; FAST, Functional Assessment Staging; WMS, Wechsler Memory scale; ECR, Enhanced cued recall test; CAMDEX, Cambridge Examination for Mental Disorders of the Elderly; WAIS, Wechsler Adult Intelligence Scale–Revised; MDRS, Mattis Dementia Rating Scale; BDAE, Boston Diagnostic Aphasia Examination; RDT, Rosen Drawing Test; MAT, Memory Alteration Test; WCST, Wisconsin Card Sorting Test; CASI, Cognitive Abilities Screening Instrument; LASSI-L, Loewenstein-Acevedo Scale for Semantic Interference and Learning; DEKOD, Dokuz Eylül Kognitif Degerlendirme/Dokuz Eylul Cognitive Assessment; CERAD-K-N, CERAD-K Neuropsychological Assessment Battery; MIS, Memory Impairment Screen; AUC, Area Under the ROC Curve; CAMCOG, Cambridge Cognitive Examination; RUDAS, Rowland Universal Dementia Assessment Scale. ecember 2019 \| Volume 10 \| Articl RESULTS The sample consisted of 38 subjects with AD; some of them were illiterate (proportion not available). A cutoff score of 67.5 was set (sensitivity = 75.8%; specificity = 97.4%). The Six-Item Screener (SIS) was used in another study (25) that evaluated 440 individuals with MCI with a small proportion of individuals (< 25%) with low education (< 6 years of schooling). The SIS showed low sensitivity for screening MCI in this population (sensitivity = 34.3%; specificity = 90.1%). The Memory Alteration Test (MAT) was used in a single study (38) for discriminating MCI and AD from healthy individuals. The AUC of MAT to discriminate between early AD and amnestic mild cognitive impairment (aMCI) was 99.60% (sensitivity = 100.00%; specificity = 97.53%) and to discriminate between aMCI and controls was 99.56% (sensitivity = 99.17%; specificity = 91.11%). The mean score was 17.54 ± 4.67 for individuals with AD, 30.53 ± 2.54 for individuals with MCI and 41.97 ± 2.6 for healthy individuals. AD and MCI individuals and controls had on average 2.65 ± 1.28, 2.53 ± 1.46, and 2.57 ± 1.45 years of education, respectively.f In agreement with literature reports, the MMSE was the most frequently used cognitive tool in the studies assessed (65). The MMSE is easy to administer and requires no specialized training and it has been validated in many countries (66). We assessed in our review a study that showed good agreement of the MMSE for cognitive screening (κ = 0.86) between primary care and specialist practitioners (44). However, studies have demonstrated the effect of education on MMSE scores. The MMSE has low sensitivity for MCI, does not perform well in assessing executive functions and has limiting floor and ceiling effects (9, 23, 40). A study conducted in Brazil has established MMSE cutoff scores of 20 for illiterate adults and 25 for those with 4 years of education or less (67). In this review, six studies (Table 2) assessed cutoff scores of the MMSE for adults with 4 years of education or less (25, 36, 38, 40, 45, 48). MCI cutoff scores (SD) ranged from 17.8 (1.9) to 21.36 (0.98), but there was great variation in sensitivity and specificity. The Montreal Cognitive Assessment-Basic (MoCA-B) was evaluated in three studies (45, 55, 57). One of these studies established a cutoff score of 19 for detecting MCI, with 87.9% sensitivity and 81.0% specificity (55). RESULTS controls: MoCA B AUC 0.95 MoCA-BJ AUC 0.87 Khedr E et al., 2015 (58) Egypt Cross- sectional To assess prevalence of MCI and dementia in Egyptian adults over 60 years of age MCI 75% illiterate AD 75% illiterate 691 (MCI 12, dementia 35) Mean age (SD) MCI 67.3 years (7.1), AD 69.3 years (7.7) *MMSE, WMS MCI prevalence 1.72/100 MMSE, Mini-Mental State Examination; BNT, Boston Naming Test; AVLT, Auditory Verbal Learning Test; MoCA, Montreal Cognitive Assessment; MoCA B, MoCA Basic; ACE, Addenbrooke Cognitive Examination; TMSE, Taiwanese Mental State Examination; CERAD, Consortium to Establish a Registry for Alzheimer’s Disease Assessment Packet; SDT, Stick Design Test; TT, Token Test; SDMT, Symbol Digit Modalities Test; SVOSPB, Selected test of the Visual Object and Space Perception Battery; JLO, Judgment of Line Orientation; TLDU, Tower of London Drexel University version; WAT, Word Accentuation Test; GSBT, Gesture Sequences subtest from the Barcelona Test; SIS, Six-Item Screener; ADAS-Cog, Alzheimer’s Disease Assessment Scale’s cognitive subscale; RAVLT, Rey-Auditory Verbal Learning Test; FAB, Frontal Assessment Battery; CDT, Clock Drawing Test; DS, Digit Span; TMT, Trail Making Test; ROCF, Rey-Osterrieth Complex Figure; FCSRT, Free and Cued Selective Reminding Test; SCWIT, Stroop Color-Word Interference Test; PEACE, Persian Test of Elderly for Assessment of Cognition and Executive function; GPCOG, General Practitioner assessment of Cognition; FAST, Functional Assessment Staging; WMS, Wechsler Memory scale; ECR, Enhanced cued recall test; CAMDEX, Cambridge Examination for Mental Disorders of the Elderly; WAIS, Wechsler Adult Intelligence Scale–Revised; MDRS, Mattis Dementia Rating Scale; BDAE, Boston Diagnostic Aphasia Examination; RDT, Rosen Drawing Test; MAT, Memory Alteration Test; WCST, Wisconsin Card Sorting Test; CASI, Cognitive Abilities Screening Instrument; LASSI-L, Loewenstein-Acevedo Scale for Semantic Interference and Learning; DEKOD, Dokuz Eylül Kognitif Degerlendirme/Dokuz Eylul Cognitive Assessment; CERAD-K-N, CERAD-K Neuropsychological Assessment Battery; MIS, Memory Impairment Screen; AUC Area Under the ROC Curve; CAMCOG Cambridge Cognitive Examination; RUDAS Rowland Universal Dementia Assessment Scale December 2019 \| Volume 10 \| Article 878 Cognitive Screening in Low Educated Elderlies Tavares-Júnior et al. diagnostic accuracy for the diagnosis of MCI and AD in adults with low education including MMSE, MoCA, PEACE, SIS, and MAT. The latter three were each used in one study only. The studies included in our review evaluated adults with different educational backgrounds. However, detailed information was not available in all studies. RESULTS Another one reported a cutoff score (SD) of 21.3 (3.8) for detecting MCI in adults with 4 years of education or less (45). Cutoff scores (SD) for AD ranged from 12.64 (3.78) to 18.32 (2.78) in these same studies. Another cognitive test battery reported was the Addenbrooke’s Cognitive Examination Revised (ACE- R), which was used in only one study but the cutoff score was not adjusted for low educational level. Mean ACE-R scores were 78.12 (12.79) for controls and 53.20 (14.76) for AD. This tool showed good diagnostic accuracy for diagnosing AD (AUC = 0.897) (25). The second most frequently used cognitive tool was the MoCA. The MoCA is a cognitive battery that includes tests sensitive to executive functions and has higher sensitivity for diagnosing MCI (24, 33, 68). However, MoCA scores are strongly influenced by educational background as MoCA tasks are designed for a certain level of education making it difficult to assess individuals who are either illiterate or with low levels of education (30). Therefore, a MoCA basic version (MoCA-B) was developed to include tasks designed to assess the same cognitive domain regardless of the level of education (45, 60, 68). A systematic review of cognitive screening tools showed that the ACE-R is an outstanding test battery. It takes approximately 20 minutes to be administered and it includes tasks designed for different levels of schooling (52, 58). In our review, ACE-R cutoff scores for low education were not available.hi The studies assessed included recommendations of specific cutoff scores and scales for diagnosis of MCI and AD in adults with low levels of education. One study using the MoCA-B suggested a cutoff score of 19 (55) for MCI diagnosis. For AD, one study showed a cutoff score of 23.5 (30) for the MMSE and another study suggested a cutoff score of 17 (41) for the MoCA.h RESULTS Adults with AD had 1.69 (51) to 7.6 years (29) of schooling and those with MCI had 2.53 (38) to 10.93 years (33) of schooling. The proportion of illiterate adults ranged from 1.8% (35) to 32.1% (23). h It is crucial to validate cognitive assessment instruments in populations with low education and to establish cutoff scores for screening these individuals in daily clinical practice. It would enable to monitoring healthy aging in such a quite large population (11) and evaluate older adults with low levels of education who are at risk of developing dementia syndromes (61). Besides, it could offer new insights to better understand the influence of education on cognitive reserve since there is a relationship between literacy and the functional organization of the human brain. Literacy acquisition improves early visual processing and phonological information processing (62). Indeed, functional neuroimaging studies have evidenced that large neural networks in both cerebral hemispheres have less functional connections in less educated individuals (63). Since there has been a move toward the development of disease- modifying drugs for AD, it will be paramount to have validated diagnostic tools for population-based assessments including older adults with low levels of education (64). g Most were cross-sectional studies (91.67%), followed by cohort studies (8.33%). Some of the studies (23, 24, 35, 36, 39, 50–52, 58) assessed the prevalence of MCI and AD ranging from 1.72% (58) to 32.9% (52) and 4.2% (35) to 9% (51), respectively. The number of adults with MCI ranged from 12 (58) to 2,049 (30) and the number of adults with AD ranged from 16 (51) to 1,061 (25). Mean age of the study participants ranged from 64.5 (40) to 82 years (50).h g Most were cross-sectional studies (91.67%), followed by cohort studies (8.33%). Some of the studies (23, 24, 35, 36, 39, 50–52, 58) assessed the prevalence of MCI and AD ranging from 1.72% (58) to 32.9% (52) and 4.2% (35) to 9% (51), respectively. The number of adults with MCI ranged from 12 (58) to 2,049 (30) and the number of adults with AD ranged from 16 (51) to 1,061 (25). Mean age of the study participants ranged from 64.5 (40) to 82 years (50). The Persian Test of Elderly for Assessment of Cognition and Executive Function (PEACE) was used in one study (28) to establish cutoff scores in individuals with AD. Frontiers in Psychiatry \| www.frontiersin.org December 2019 \| Volume 10 \| Article 878 DISCUSSION We carried out a critical review of cognitive assessment tools for screening cognitive syndromes in older adults with low levels of education. A significant number of assessment tools (n = 44) were used in the studies reviewed, but only a few of them showed f The scarcity of cognitive assessment tests for evaluating adults with low levels of education is in line with the challenge December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 8 Frontiers in Psychiatry \| www.frontiersin.org 9 TABLE 2 \| Characteristics of studies that established cutoff scores for the MMSE and MoCA in adults with low education. DISCUSSION Authors, Year Country Study Design Primary Study Objective Level of Education Sample Size Participant Age Cognitive Assessment Tool used Main Resu Chen et al., 2010 (25) China Cross- sectional To validate SIS for quick detection of cognitive impairment Illiterate 7.2% 1–6 years 16.4% > 6 years 76.3% 1,976 (healthy aging 475, MCI 440, AD 1,061) Mean age (SD) 71.87 years (8.71) SIS, MMSE SIS: AD Sn 34.3%, Sp <4 years of Sp 95.0% Javadi PSHS et al., 2015 (31) Iran Cross- sectional To characterize illiterate and literate older adults; PEACE scale cutoff scores for AD Different levels of education—illiterate and literate 101 (controls 33, MCI 30, AD 38) Mean age (SD) AD 74.60 years (8.02) MCI 72.5 years (7.2) Controls 67.84 years (7.29) PEACE, GPCOG, FAST, MMSE, WMS MMSE < 4 (1.75) AD 1 (Sn 75.8%, Afgin AE et al., 2012 (36) Israel Cross- sectional To assess prevalence of MCI and AD and conversion rate from MCI to AD within a year or more 0 years 51% 1–4 years 23% 5–8 years 21% > 8 years 5% 944 (controls 497, MCI 303, SD 13, VD 39, AD 92) Mean age (years) (SD) DA 78.5 (7.7) MCI 72.8 (6.1) Controls 70.7 (5.5) by age groups (%) 65–69 (39.4%) 70–79 (46.4%) ≥80 (14.1%) MMSE Arabic version, Brookdale Cognitive Screening Test Prevalence <4 years of MCI illiterate schooling 1 (3.7), 1–4 y Custodio N et al., 2017 (38) Peru Cross- sectional MAT performance to discriminate controls, MCI and AD in adults with low education Mean schooling (SD) AD 2.65 years (1.28) MCI 2.53 years (1.46) Controls 2.57 years (1.45) 247 (controls 121, AD 81, MCI 45) Mean age (SD) AD 74.18 years (3.81) MCI 71.09 years (4.20) Controls 69.53 years (4.11) MAT, RAVLT, WMS, TMT A and B, ROCF, BNT, WCST, DS, WAIS, MMSE MAT AD vs controls AU scores (SD) (0.98) MAT (2.54) Julayanont P. SIS, Six-Item Screener; MMSE, Mini-Mental State Examination; PEACE, Persian Test of Elderly for Assessment of Cognition and Executive function; GPCOG, General Practitioner assessment of Cognition; FAST, Functional Assessment Staging; WMS, Wechsler Memory scale; MAT, Memory Alteration Test; RAVLT, Rey-Auditory Verbal Learning Test; TMT, Trail Making Test; BNT, Boston Naming Test; ROCF, Rey-Osterrieth Complex Figure; WCST, Wisconsin Card Sorting Test; DS, Digit Span; WAIS, Wechsler Adult Intelligence Scale–Revised; MoCA, Montreal Cognitive Assessment; MoCA B, MoCA Basic; DEKOD, Dokuz Eylül Kognitif Degerlendirme/Dokuz Eylul Cognitive Assessment; SCWIT, Stroop Color-Word Interference Test; AUC, Area Under the ROC Curve; AVLT, Auditory Verbal Learning Test; SDMT, Symbol Digit Modalities Test. MMSE, AVLT, ROCF, BNT, Verbal Fluency, TMT, SCWIT, SDMT, MoCA-B Chinese version PEACE, GPCOG, FAST, MMSE, WMS DISCUSSION et al., 2015 (45) Thailand Cross- sectional To assess validity of MoCA-B for MCI in a population with low education Mean schooling (SD) Controls 3.6 years (1.1) MCI 2.9 years (1.7) 85 (controls 43, MCI 42) Mean age (SD) Controls 66.6 years (6.7) MCI 70.2 years (6.6) MoCA-B, MMSE Thai version MMSE <4 y Mean score MoCA-B 21 Kurt P et al., 2014 (48) Turkey Cross- sectional To develop a composite score for DEKOD in an older population < 4 years of schooling 45.7% > 5 years of schooling 54.3% 444 (controls 338, dementia 53, MCI 53) Mean age (SD) controls 70.7 years (5.4) dementia 74 years (7.8) MCI 71.7 years (5.6) DEKOD, MMSE Turkish version MMSE <4 y score/Sn/S MCI 22–23/ 49–50/91% Chen K et al., 2016 (55) China Cross- sectional To assess performance of MoCA-B Chinese version for screening MCI in older adults Groups by years of schooling (< 6, 7–12, > 12). Mean schooling (SD) of those with <6 years: controls 4.8 years (1.7), MCI 3.3 years (2.4), AD 3.7 years (2.5) 704 (MCI 264, AD 160, controls 280) Mean age (SD) of those with <6 years of schooling controls 68.2 years (9.1) MCI 68.5 years (8.5) AD 67.9 years (9.4) MMSE, AVLT, ROCF, BNT, Verbal Fluency, TMT, SCWIT, SDMT, MoCA-B Chinese version Cutoff score years of sch 19/87.9/81 SIS, Six-Item Screener; MMSE, Mini-Mental State Examination; PEACE, Persian Test of Elderly for Assessment of Cognition and Executive function; GPCOG, General Practitioner assessm Functional Assessment Staging; WMS, Wechsler Memory scale; MAT, Memory Alteration Test; RAVLT, Rey-Auditory Verbal Learning Test; TMT, Trail Making Test; BNT, Boston Naming Tes Frontiers in Psychiatry \| www.frontiersin.org December 2019 \| Tavares-Júnior et al. Prevalence AD 9.8%, MCI 32.1% MMSE <4 years of schooling cutoff scores (SD) MCI illiterate 17.8 (1.9), 1–4 years of schooling 19.6 (3.4); AD illiterate 12.7 (3.7), 1–4 years of schooling 12.6 (6.7) MAT AD vs. MCI AUC 99.60% MCI vs. controls AUC 99.56% Mean MMSE scores (SD): AD 18.32 (2.78); MCI 21.36 (0.98) MAT AD 17.54 (4.67); MCI 30.53 (2.54) MMSE <4 years of schooling—illiterate Mean score (SD) MMSE MCI 18.9 (3.0) MoCA-B 21.3 (3.8) Cutoff scores Sn/Sp of those with <6 years of schooling: MCI MoCA-BC 19/87.9/81.0 MMSE 26/86.2/60.3 Cognitive Screening in Low Educated Elderlies December 2019 \| Volume 10 \| Article 878 Cognitive Screening in Low Educated Elderlies Tavares-Júnior et al. AUTHOR CONTRIBUTIONS Conception and design of work: JT-J, AS, GA and PB-N. Acquisition, analysis or interpretation of data and work: JT-J, AS, PB-N and GA. Drafting the work: JT-J, AS, GA, JB, JS-N and PB-N. All authors were involved in critical revision of the manuscript for important intellectual content. ACKNOWLEDGMENTS The use of cognitive assessment tools that are easy to administer is still challenging given the high prevalence of low education in the global population. This review provides an overview of the We thank the study participants and their families for their collaboration and support. DISCUSSION most commonly used instruments for cognitive screening. We found that a small number of studies evaluated adults with 4 years of formal education or less and a wide range of cutoff scores for various cognitive test batteries. Our findings further support the need for the development of specific tools for assessing illiterate adults. Cognitive ability, formal logic, and abstract reasoning should be assessed as they could provide more accurate input for screening and interpretation of cognitive tests in older adults who are either illiterate or with low levels of education. Low-cost test batteries that are easy to administer should be validated because they can make a significant impact on screening of cognitive syndromes and enable early therapeutic interventions aimed at reducing morbidity and mortality of dementia. Further studies of test batteries adjusted to larger groups of adults with low levels of education and specific MCI subtypes and AD stages could help shed light on these points. of assessing cognitive function in individuals with low education (13). Individuals with low levels of education were excluded from many studies because there are no cutoff scores established for several assessment tools. Literacy acquisition increases performance in certain cognitive domains such as executive functions (14), improves visual processing, changes phonological codes, and strengthens the link between phonemic and graphic representation (62) However, to measure literacy through the number of years of formal education is not the most effective approach since there are so many different ways of learning even without attending school (12, 62).h This review study has some limitations that deserve mention. First, no information was available on the diagnostic accuracy of cognitive tools for MCI and AD. Furthermore, little information was available on the diagnostic accuracy of tools for assessing different MCI subtypes and stages of AD. Another caveat is that our search was limited to cognitive assessment tools that require a trained examiner and excluded self-administered and web- based screening tools. Besides, there is no consensus about the definition of low levels of education, which may partly explain heterogeneous results of the cognitive batteries evaluated. Lastly, the studies included this review applied various diagnostic criteria for AD, which prevented comparisons of results across them. REFERENCES Envelhecimento: do conhecimento básico à abordagem clínica. OmniFarma: São Paulo (2016). p. 433. 1. Prince M, Wilmo A, Guerchet M, Ali GC, Wu YT, Prina M. World Alzheimer Report 2015: The Global Impact of Dementia an analysis of prevalence, incidence, cost and trends. London: Alzheimer’s Disease International (2015). p. 87. Available from: https://www.alz.co.uk/research/ WorldAlzheimerReport2015.pdf (Access in ) 8. Orsini A, Pezzuti G, Hulbert S. Beyond the floor effect on the Wechsler Intelligence Scale for Children – 4th Ed.(WISC-IV): calculating IQ and Indexes of subjects presenting a floored pattern of results. J Intellect Disabil Res (2015) 59(5):468–73. doi: 10.1111/jir.12150lf 1. Prince M, Wilmo A, Guerchet M, Ali GC, Wu YT, Prina M. World Alzheimer Report 2015: The Global Impact of Dementia an analysis of prevalence, incidence, cost and trends. London: Alzheimer’s Disease International (2015). p. 87. Available from: https://www.alz.co.uk/research/ WorldAlzheimerReport2015.pdf (Access in ) 9. Dean K, Walker Z, Jenkinson C. Data quality, floor and ceiling effects and test – retest reliability of the Mild Cognitive Impairment Questionnaire. Patient Relat Outcome Meas (2018) 9:43–7. doi: 10.2147/PROM.S145676h 2. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med (2004) 256(3):183–94. doi: 10.1111/j.1365-2796.2004.01388.x 2. Petersen RC. Mild cognitive impairment as a diagnostic entity. J Intern Med (2004) 256(3):183–94. doi: 10.1111/j.1365-2796.2004.01388.x 10. Everitt BS, Skrondal A. The Cambridge dictionary of statitistics. New York: Cambridge University Press (2010). p. 480. 3. Ding D, Zhao Q, Guo Q, Meng H, Wang B, Luo J, et al. Prevalence of mild cognitive impairment in an urban community in china: a cross-sectional analysis of the shanghai aging study. Alzheimers Dement (2015) 11(3):300–9. e2. doi: 10.1016/j.jalz.2013.11.002 3. Ding D, Zhao Q, Guo Q, Meng H, Wang B, Luo J, et al. Prevalence of mild cognitive impairment in an urban community in china: a cross-sectional analysis of the shanghai aging study. Alzheimers Dement (2015) 11(3):300–9. e2. doi: 10.1016/j.jalz.2013.11.002 11. Unesco Institute for Lifelong Learning. 3rd global report on adult learning and education: the impact of adult learning and education on health and well- being, employment and the labour market, and social, civic and community life. Germany: Unesco. (2016), p. 18. Available from: https://unesdoc.unesco. org/ark:/48223/pf0000245917 (Access in September 05, 2019) 4. Petersen RC, Lopez O, Armstrong MJ, Getchius TSD, Ganguli M, Gloss D, et al. Practice guideline update summary: mild cognitive impairment Report of the Guideline Development, Dissemination, and Implementation, Subcommittee of the American Academy of Neurology. Neurol (2018) 90(3):126–35. REFERENCES A nationwide survey on the prevalence of dementia and mild cognitive impairment in South Korea. J Alzheimers Dis (2011) 23(2):281–91. doi: 10.3233/JAD-2010-101221 41. Freitas S, Simões MR, Alves G, Santana I. Montreal Cognitive Assessment Validation Study for Mild Cognitive Impairment and Alzheimer Disease. Alzheimer Dis Assoc Disord (2013) 27(1):37–43. doi: 10.1097/ WAD.0b013e3182420bfe 25. Chen MR, Guo QH, Cao XY, Hong Z, Liu XH. A preliminary study of the Six-Item Screener in detecting cognitive impairment. Neurosci Bull (2010) 26(4):317–21. doi: 10.1007/s12264-010-0106-1 42. Borson S, Scanlan JM, Watanabe J, Tu SP, Lessig M. Simplifying Detection of Cognitive Impairment: Comparison of the Mini-Cog and Mini-Mental State Examination in a Multiethnic Sample. J Am Geriatr Soc (2005) 53(5):871–4. doi: 10.1111/j.1532-5415.2005.53269.x 26. Chang J, Tse CS, Leung GT, Fung AW, Hua KT, Chiu HF, et al. Bias in discriminating very mild dementia for older adults with different levels of education in Hong Kong. Int Psychogeriatr (2014) 26(6):995–1010. doi: 10.1017/S1041610214000234 43. Boycheva E, Contador I, Fernández-Calvo B, Ramos-Campos F, Puertas- Martín V, Villarejo-Galende A, et al. Spanish version of the Mattis Dementia Rating Scale-2 for early detection of Alzheimer’s disease and mild cognitive impairment. Int J Geriatr Psychiatry (2018) 33(6):832–40. doi: 10.1002/ gps.4707 27. O’Bryant SE, Johnson G, Balldin V, Edwards M, Barber R, Williams B, et al. Characterization of Mexican Americans with Mild Cognitive Impairment and Alzheimer’s Disease. J Alzheimers Dis (2013) 33(2):373–9. doi: 10.3233/ JAD-2012-121420 g 44. Pezzotti P, Scalmana S, Mastromattei A, Di Lallo D. for the "Progetto Alzheimer" Working Group. The accuracy of the MMSE in detecting cognitive impairment when administered by general practitioners: A prospective observational study. BMC Fam Pract (2008) 9:29. doi: 10.1186/1471-2296-9-29 28. De Paula JJ, Bertola L, Ávila RT, Moreira L, Coutinho G, de Moraes EN, et al. Clinical Applicability and Cutoff Values for an Unstructured Neuropsychological Assessment Protocol for Older Adults with Low Formal Education. PloS One (2013) 8(9):e73167. doi: 10.1371/journal.pone.0073167 45. Julayanont P, Tangwongchai S, Hemrungrojn S, Tunvirachaisakul C, Phanthumchinda K, Hongsawat J, et al. The Montreal Cognitive Assessment—Basic: A Screening Tool for Mild Cognitive Impairment in Illiterate and Low-Educated Elderly Adults. J Am Geriatr Soc (2015) 63(12):2550–54. doi: 10.1111/jgs.13820 29. Sánchez-Benevides G, Peña-Casanova J, Casals-Coll M, Gramunt N, Molinuevo JL, Gómes-Ansón B, et al. Cognitive and Neuroimaging Profiles in Mild Cognitive Impairment and Alzheimer’s Disease: Data from the Spanish Multicenter Normative Studies (NEURONORMA Project). J Alzheimers Dis (2014) 41(3):887–901. doi: 10.3233/JAD-132186 46. REFERENCES doi: 10.1212/WNL.0000000000004826 4. Petersen RC, Lopez O, Armstrong MJ, Getchius TSD, Ganguli M, Gloss D, et al. Practice guideline update summary: mild cognitive impairment Report of the Guideline Development, Dissemination, and Implementation, Subcommittee of the American Academy of Neurology. Neurol (2018) 90(3):126–35. doi: 10.1212/WNL.0000000000004826h 12. Tucker-Drob EM, Johnson KE, Jones RN. The cognitive reserve hypothesis: A longitudinal examination of age-associated declines in reasoning and processing speed. Dev Psychol [Internet] (2009) 45(2):431–46. doi: 10.1037/a0014012 5. Schneider JA, Arvanitakis Z, Leurgans SE, Bennett DA. The Neuropathology of Probable Alzheimer Disease and Mild Cognitive Impairment. Ann Neurol (2009) 66(2):200–8. doi: 10.1002/ana.21706 13. Mortimer JA, Snowdon DA, Markesbery WR. Head circumference, education and risk of dementia : findings from the nun study. J Clin Exp Neuropsychol (2003) 25(5):671–9. doi: 10.1076/jcen.25.5.671.14584i 6. Handels RLH, Vos SJB, Kramberger MG, Jelic V, Blennow K, et al. Predicting progression to dementia in persons with mild cognitive impairment using cerebrospinal fluid markers. Alzheimers Dement (2017) 13(8):903–12. doi: 10.1016/j.jalz.2016.12.015 6. Handels RLH, Vos SJB, Kramberger MG, Jelic V, Blennow K, et al. Predicting progression to dementia in persons with mild cognitive impairment using cerebrospinal fluid markers. Alzheimers Dement (2017) 13(8):903–12. doi: 10.1016/j.jalz.2016.12.015 14. Kosmidis MH, Zafiri M, Politimou N. Literacy versus formal schooling: influence on working memory. Arch Clin Neuropsychol (2011) 26:575–82. doi: 10.1093/arclin/acr063 15. Da Silva CG, Petersson KM, Faísca G, Ingvar H, Reis A. The effects of literacy and education on the quantitative and qualitative aspects of semantic 7. Studart Neto A, Nitrini R. Avaliação cognitiva à beira do leito. In: Frota NAF, Siqueira Neto JI, Balthazar MLF, Nitrini R, editors. Neurologia Cognitiva e do December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 10 Cognitive Screening in Low Educated Elderlies Tavares-Júnior et al. mild cognitive impairment in Turkish people. Arch Clin Neuropsychol (2006) 21(7):745–51. doi: 10.1016/j.acn.2006.08.007 verbal fluency. J Clin Exp Neuropsychol (2004) 26(2):266–77. doi: 10.1076/ jcen.26.2.266.28089 verbal fluency. J Clin Exp Neuropsychol (2004) 26(2):266–77. doi: 10.1076/ jcen.26.2.266.28089 mild cognitive impairment in Turkish people. Arch Clin Neuropsychol (2006) 21(7):745–51. doi: 10.1016/j.acn.2006.08.007 16. Wajman JR, Oliveira FF, Schultz RR, Marin SMC, Bertolucci PHF. Educational bias in the assessment of severe dementia: Brazilian cutoffs for severe Mini-Mental State Examination. Arq. Neuro-Psiquiatr (2014) 72(4):273–77. doi: 10.1590/0004-282X20140002 35. Tognoni G, Ceravolo R, Nucciarone B, Bianchi F, Dell’Agnello G, Ghicopulos I, et al. From mild cognitive impairment to dementia: a prevalence study in a district of Tuscany, Italy. REFERENCES Acta Neurol Scand (2005) 112(2):65–71. doi: 10.1111/j.1600-0404.2005.00444.x 17. Scazufca M, Almeida OP, Vallada HP, Tasse WA, Menezes PR. Limitations of the Mini-Mental State Examination for screening dementia in a community with low socioeconomic status: results from the Sao Paulo Ageing & Health Study. Eur Arch Psychiatry Clin Neurosci (2009) 259(1):8–15. doi: 10.1007/ s00406-008-0827-6 36. Afgin AE, Massarwa H, Schechtman E, Israeli-Korn SD, Strugatsky R, Abuful A, et al. High Prevalence of Mild Cognitive Impairment and Alzheimer’s Disease in Arabic Villages in Northern Israel: Impact of Gender and Education. J Alzheimers Dis (2012) 29(2):431–9. doi: 10.3233/ JAD-2011-111667 18. Whittemore R, Knafl K. The integrative review : updated methodology. J Adv Nurs (2005) 52(5):546–53. doi: 10.1111/j.1365-2648.2005.03621.x 37. Chaves ML, Camozzato AL, Godinho C, Piazenski I, Kaye J. Incidence of Mild Cognitive Impairment and Alzheimer Disease in Southern Brazil. J Geriatr Psychiatry Neurol (2009) 22(3):181–7. doi: 10.1177/0891988709332942 19. Mendes KDS, Silveira RCCP, Galvão CM. Revisão integrativa: método de pesquisa para a incorporação de evidências na saúde e na enfermagem. Texto Contexto – enferm (2008) 17(4):758–64. doi: 10.1590/ S0104-07072008000400018 38. Custódio N, Lira D, Herrera-Perez E, Montesinos R, Castro-Suarez S, Cuenca-Alfaro J, et al. Memory Alteration Test to Detect Amnestic Mild Cognitive Impairment and Early Alzheimer’s Dementia in Population with Low Educational Level. Front Aging Neurosci (2017) 9:278. doi: 10.3389/ fnagi.2017.00278 20. Jackson GB. Methods for Integrative Reviews. Rev Educ Res Fall (1980) 50(3):438–60. doi: 10.3102/00346543050003438 21. Whittemore R, Knafl K. The integrative review: updated methodology. J Adv Nurs (2005) 52(5):546–53. doi: 10.1111/j.1365-2648.2005.03621.x 39. Ravaglia G, Forti P, Montesi F, Lucicesare A, Pisacane N, Rietti E, et al. Mild Cognitive Impairment: Epidemiology and Dementia Risk in an Elderly Italian Population. J Am Geriatr Soc (2008) 56(1):51–8. doi: 10.1111/j.1532-5415.2007.01503.x 22. Fuchs SC, Paim BS. Revisão Sistemática de Estudos Observacionais com Metanálise. Clinical & Biomedical Research (2010) 30(3). Available at: https:// seer.ufrgs.br/hcpa/article/view/16551. 40. Rahman TTA, Gaafary MME. Montreal Cognitive Assessment Arabic version: Reliability and validity prevalence of mild cognitive impairment among elderly attending geriatric clubs in Cairo. Geriatr Gerontol Int (2009) 9(1):54–61. doi: 10.1111/j.1447-0594.2008.00509.x 23. Sun Y, Lee HJ, Yang SC, Chen TF, Lin KN, Lin CC, et al. A Nationwide Survey of Mild Cognitive Impairment and Dementia, Including Very Mild Dementia, in Taiwan. PloS One (2014) 9(6):e100303. doi: 10.1371/journal.pone.0100303 24. Kim KW, Park JH, Kim MH, Kim MD, Kim BJ, Kim SK, et al. REFERENCES Education and Alzheimer disease without dementia: support for the cognitive reserve hypothesis. Neurol (2007) 68(3):223–8. doi: 10.1212/01.wnl.00002513 03.50459.8a 51. Choi SJ, Jung SS, You Y, Shin BS, Kim JE, Yoom SW, et al. Prevalence of Alzheimer’s Dementia and its risk factors in community dwelling elderly Koreans. Psychiatry Invest (2008) 5(2):78–85. doi: 10.4306/pi.2008.5.2.78 62. Dehaene S, Cohen L, Morais J, Kolinsky R. Illiterate to literate: behavioural and cerebral changes induced by reading acquisition. Nat Rev Neurosci (2015) 16(4):234–44. doi: 10.1038/nrn3924 52. Gavrila D, Antúnez C, Tormo MJ, Carles R, García Santos JM, Parrilla G, et al. Prevalence of dementia and cognitive impairment in Southeastern Spain: the Ariadna study. Acta Neurol Scand (2009) 120(5):300–7. doi: 10.1111/j.1600-0404.2009.01283.xh 63. Marques P, Soares JM, Magalhães R, Santos NC. Sousa N. The bounds of education in the human brain connectome. Sci Rep (2015) 5:12812. doi: 10.1038/srep12812 64. De Roeck EE, Engelborghs S, Dierckx E. Next generation brain health depends on early Alzheimer disease diagnosis: from a timely diagnosis to future population screening. J Am Med Dir Assoc (2016) 17:452–3. doi: 10.1016/j.jamda.2016.02.015 53. Wang B, Guo QH, Chen J, Zhao QH, Zhou Y, Hong Z. The clinical characteristics of 2,789 consecutive patients in a memory clinic in China. J Clin Neurosci (2011) 18(11):1473–7. doi: 10.1016/j.jocn.2011.05.003f 65. De Roeck EE, De Deyn PP, Dierckx E, Engelborghs S. Brief cognitive screening instruments for early detection of Alzheimer’s disease: a systematic review. Alzheimers Res Ther (2019) 11(1):21. doi: 10.1186/s13195-019-0474-3 54. Zhou Y, Ortiz F, Nuñez C, Elashoff D, Woo E, Apostolova LG, et al. Use of the MoCA in Detecting Early Alzheimer’s Disease in a Spanish-Speaking Population with Varied Levels of Education. Dement Geriatr Cognit Disord (2015) 5:85–95. doi: 10.1159/000365506 h 66. Mitchell AJ. A meta-analysis of the accuracy of the mini-mental state examination in the detection of dementia and mild cognitive impairment. J Psychiatr Res (2009) 43(4):411–31. doi: 10.1016/j.jpsychires.2008.04.014 55. Chen KL, Xu Y, Chu AQ, Ding D, Liang XN, Nasreddine ZS, et al. Validation of the Chinese Version of Montreal Cognitive Assessment Basic for Screening Mild Cognitive Impairment. J Am Geriatr Soc (2016) 64(12):e285–90. doi: 10.1111/jgs.14530 67. Brucki SMD, Nitrini R, Caramelli P, Bertolucci PHF, Okamoto IH. Sugestões para o uso do mini-exame do estado mental no Brasil. Arq Neuro-psiquiatr (2003) 61(3B):777–81. doi: 10.1590/S0004-282X2003000500014 56. Matias-Guiu JA, Valles-Salgado H, Rognoni T, Hamre-Gil F, Moreno- Ramos T, Matias-Guiu J. REFERENCES Matías-Guiu JA, Curiel RE, Rognoni T, Valles-Salgado M, Fernández- Matarrubia M, Hariramani R, et al. Validation of the Spanish Version of the LASSI-L for Diagnosing Mild Cognitive Impairment and Alzheimer’s Disease. J Alzheimers Dis (2017) 56(2):733–42. doi: 10.3233/JAD-160866 30. Mellor D, Lewis M, McCabe M, Byrne L, Wang T, Wang J, et al. Determining Appropriate Screening Tools and Cut-Points for Cognitive Impairment in an Elderly Chinese Sample. Psychol Assess (2016) 28(11):1345–53. doi: 10.1037/ pas0000271 31. Javadi PSHS, Zendehbad A, Darabi F, Khosravifar S, Noroozian M. Development and implementation of Persian test of Elderly for Assessment of Cognition and Executive function (PEACE). Electron Physician (2015) 7(7):1549–56. doi: 10.19082/1549 47. Chu LW, Ng KHY, Law ACK, Lee AM, Kwan F. Validity of the Cantonese Chinese Montreal Cognitive Assessment in Southern Chinese. Geriatr Gerontol Int (2015) 15(1):96–103. doi: 10.1111/ggi.12237 48. Kurt P, Keskinoglu P, Yaka E, Uçku R, Yener G. A composite score for Dokuz Eylul Cognitive state neurocognitivetest battery: a door-to-door survey study with illiterate, low and high educated elderly in Turkey. Turk J Geriatr (2013) 17(2):143–51. Available at: https://www.researchgate.net/ publication/288575275. 32. Chong MS, Lim WS, Chan SP, Feng L, Niti M, Yap P, et al. Diagnostic Performance of the Chinese Frontal Assessment Battery in Early Cognitive Impairment in an Asian Population. Dement Geriatr Cognit Disord (2010) 30(6):525–32. doi: 10.1159/000321665 33. Ng A, Chew I, Narasimhalu K, Kandiah N. Effectiveness of Montreal Cognitive Assessment for the diagnosis of mild cognitive impairment and mild Alzheimer’s disease in Singapore. Singapore Med J (2013) 54(11):616–9. doi: 10.11622/smedj.2013220 49. Bae JB, Kim YJ, Han JW, Kim TH, Park JH, Lee SB. Incidence of and Risk Factors for Alzheimer’s Disease and Mild Cognitive Impairment in Korean Elderly. Dement Geriatr Cognit Disord (2015) 39(1–2):105–15. doi: 10.1159/000366555 34. Saka E, Mihci E, Topcuoglu MA, Balcânica S. Enhanced cued recall has a high utility as a screening test in the diagnosis of Alzheimer’s disease and 50. Paddick SM, Kisoli A, Samuel M, Higginson J, Gray WK, Dotchin CL, et al. Mild Cognitive Impairment in Rural Tanzania: Prevalence, Profile, and December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 11 Cognitive Screening in Low Educated Elderlies Tavares-Júnior et al. Outcomes at 4-Year Follow-up. Am J Geriatr Psychiatry (2015) 23(9):950–9. doi: 10.1016/j.jagp.2014.12.005 Outcomes at 4-Year Follow-up. Am J Geriatr Psychiatry (2015) 23(9):950–9. doi: 10.1016/j.jagp.2014.12.005 61. Roe CM, Xiong C, Miller JP, Morris JC. December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org REFERENCES Comparative Diagnostic Accuracy of the ACE-III, MIS, MMSE, MoCA, and RUDAS for Screening of Alzheimer Disease. Dement Geriatr Cognit Disord (2017) 43(5–6):237–46. doi: 10.1159/000469658 68. Julayanont P, Phillips N, Chertkow H, Nasreddine ZS. Montreal Cognitive Assessment (MoCA): Concept and clinical review. In: Larner AJ, ed. Cognitive Screening Instruments: A Practical Approach. London: Springer (2013) p. 111–51. Available from: https://www.concordia.ca/content/dam/ artsci/research/caplab/docs/Julayanontetal2013MoCAreview.pdf (Access in September) 57. Huang YY, Qian SX, Guan QB, Chen KL, Zhao QH, Lu JH, et al. Comparative study of two Chinese versions of Montreal Cognitive Assessment for Screening of Mild Cognitive Impairment. Appl Neuropsychol Adult (2019), 1–6. doi: 10.1080/23279095.2019.1602530 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. 58. Khedr EMH, Fawi G, Abbas MAA, Mohammed TAA, El-Fetoh NA, Attar GSA, et al. Prevalence of Mild Cognitive Impairmentand Dementia among the Elderly Population of Qena Governorate, Upper Egypt: A Community- Based Study. J Alzheimers Dis (2015) 45(1):117–26. doi: 10.3233/jad-142655 Copyright © 2019 Tavares-Júnior, de Souza, Alves, Bonfadini, Siqueira-Neto and Braga-Neto. 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. 59. Folstein MF, Folstein SE, McHugh PR. Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res (1975) 12(3):189–98. doi: 10.1016/0022-3956(75)90026-6 60. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for Mild Cognitive Impairment. J Am Geriatr Soc (2005) 53(4):695–9. doi: 10.1111/j.1532-5415.2005.53221.x December 2019 \| Volume 10 \| Article 878 Frontiers in Psychiatry \| www.frontiersin.org 12
W4388181563.txt	https://asmp-eurasipjournals.springeropen.com/counter/pdf/10.1186/s13636-023-00312-8	en		W2VC: WavLM representation based one-shot voice conversion with gradient reversal distillation and CTC supervision	EURASIP Journal on Audio, Speech and Music Processing	2,023	cc-by	8,338	Huang et al. EURASIP Journal on Audio, Speech, and Music Processing https://doi.org/10.1186/s13636-023-00312-8 EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Open Access EMPIRICAL RESEARCH W2VC: WavLM representation based one‑shot voice conversion with gradient reversal distillation and CTC supervision Hao Huang1,2, Lin Wang1, Jichen Yang3* , Ying Hu1 and Liang He1,4 Abstract Non-parallel data voice conversion (VC) has achieved considerable breakthroughs due to self-supervised pre-trained representation (SSPR) being used in recent years. Features extracted by the pre-trained model are expected to contain more content information. However, in common VC with SSPR, there is no special implementation to remove speaker information in the content representation extraction by SSPR, which prevents further purification of the speaker information from SSPR representation. Moreover, in conventional VC, Mel-spectrogram is often selected as the reconstructed acoustic feature, which is not consistent with the input of the content encoder and results in some information lost. Motivated by the above, we proposed W2VC to settle the issues. W2VC consists of three parts: (1) We reconstruct feature from WavLM representation (WLMR) that is more consistent with the input of content encoder; (2) Connectionist temporal classification (CTC) is used to align content representation and text context from phoneme level, content encoder plus gradient reversal layer (GRL) based speaker classifier are used to remove speaker information in the content representation extraction; (3) WLMR-based HiFi-GAN is trained to convert WLMR to waveform speech. VC experimental results show that GRL can purify well the content information of the self-supervised model. The GRL purification and CTC supervision on the content encoder are complementary in improving the VC performance. Moreover, the synthesized speech using the WLMR retrained vocoder achieves better results in both subjective and objective evaluation. The proposed method is evaluated on the VCTK and CMU databases. It is shown the method achieves 8.901 in objective MCD, 4.45 in speech naturalness, and 3.62 in speaker similarity of subjective MOS score, which is superior to the baseline. Keywords Voice conversion, Self-supervised pre-trained Representation, Gradient reversal layer (GRL), CTC Correspondence: Jichen Yang NisonYoung@163.com 1 School of Computer Science and Technology, Xinjiang University, Urumqi, China 2 Xinjiang Key Laboratory of Multi-lingual Information Technology, Urumqi, China 3 School of Cyber Security, Guangdong Polytechnic Normal University, Guangzhou, China 4 Tsinghua National Laboratory for Information Science and Technology, Department of Electronic Engineering, Tsinghua University, Beijing, China 1 Introduction Voice conversion aims to convert the speech of a source speaker into a simulation of the speech of a target speaker, while retaining the linguistic information unchanged. VC has been used in many applications, such as speaker-identity modification for text-to-speech [1], singing voice conversion [2], and generation of various kinds of expressive speech [3]. According to the existence of parallel utterance pairs from both the source speaker and target speaker in the training dataset, VC can be broadly divided into parallel VC and non-parallel VC. In recent years, non-parallel VC has become the mainstream research topic because no © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the 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/. Huang et al. EURASIP Journal on Audio, Speech, and Music Processing parallel training data is necessary and thus can be easily implemented. In non-parallel VC research, variational autoencoder (VAE) [4, 5] has become increasingly popular, which commonly employs an encoder-decoder framework to learn the disentanglement of the content representation and speaker representation from source speaker and target speaker respectively. It can match the task of VC that disentangles the content information of source speaker and speaker information of target speaker first, then combine them, and finally synthesize the converted utterance. In conventional VAE-based VC, the encoder extracts speaker-independent latent variables from the input acoustic features, and the decoder reconstructs the acoustic features from the latent variables and the given speaker representations. However, the modeling of VAE is limited to Gaussian distributional assumption [6], and the distribution of actual data may be much more complicated. On the basis of VAE, any-to-any (also called oneshot VC) VC was achieved earlier in vector quantized variational autoencoder (VQ-VAE) [7] and AUTOVC [8]. They both relied on the disentanglement of content and speaker information within an utterance with a content encoder and a speaker encoder. VQ-VAE used learning the prior via an embedding dictionary to represent entirely different phonetic contents, while AUTOVC used a pre-trained speaker encoder [9] to obtain speaker information and used an information bottleneck to limit the leakage of the source speaker information. In addition, phoneme posteriorgram (PPG) is very popular in VAE-based VC [10]. The reason is that PPG is speakerindependent content representation in theory. In recent VAE-based VC studies, self-supervised pre-trained representation have been used to extract content or speaker information because SSPR is often trained from a largescale unlabeled data and hence SSPR contains abundant and compact speech information [11]. SSPR have been used to provide content information for VC tasks [12]. Several SSPRs have been used as the inputs of content encoders in [12]. For example, FragmentVC applies the wav2vec 2.0 as the input of content encoders [13]. In S2VC [14], CPC [15] is used as the input of both the content encoder and the speaker encoder. In the early VAE-VC, the traditional Mel-spectrogram is often selected as the input acoustic features and reconstruction features, which are fed into the encoder and obtained from the decoder, respectively. Then various vocoders such as WaveNet [16], MelGAN [17], and HiFi-GAN vocoder [18] are used to convert Melspectrogram to waveform speech. Even in recent VC studies using SSPR [11, 13, 14] as the input feature of content representation extraction for source speaker, (2023) 2023:45 Page 2 of 15 Mel-spectrogram is also selected as the reconstructed acoustic feature converted from transformed SSPR. Though SSPRs have been used in previous work of voice conversion such as [11, 13, 14] and made promising progress in VC performance, there are still two limitations in current system configurations: • There is no special tactic to remove speaker information in content representation extraction for the source speaker when SSPR is used as the input. As a result, there is still more or less speaker information in the content representation. • Mel-spectrogram is often selected as the reconstructed representation by decoding the combining content representation and speaker representation (denoted transformed-SSPR hereafter), there will inevitably be some information missing in the process of converting transformed-SSPR to Mel-spectrogram. In order to settle the problems in VC based on SSPR mentioned above, several strategies are proposed. (1) Connectionist temporal classification (CTC) [19] is used to align content representation and text context from the phoneme level, and content encoder plus gradient reversal layer (GRL)-based speaker classifier [20, 21] are used to remove speaker information from content representation with WavLM representation (WLMR) [22] as the input. The reason is that the GRL-based speaker classifier can make the content information not flow to the speaker classifier, which makes the output of the content encoder have no speaker information. Finally, WLMR can be used as the input to extract content representation because it can make great success in speech recognition [22]. (2) WLMR is selected as the reconstructed representation rather than the commonly used Mel-spectrogram, which can be consistent with the input of the content representation to escape the information lost. (3) WLMR-based HiFi-GAN vocoder is trained to convert the reconstructed WLMR to waveform speech. In this regard, a VC method based on WLMR is proposed in this work, which is termed as W2VC for short. The contribution of the work can be described as follows: • Content encoder plus GRL-based speaker classifier are used to remove speaker information in the content representation extraction of source speaker. Huang et al. EURASIP Journal on Audio, Speech, and Music Processing • The reconstructed feature is WLMR rather than commonly acoustic features such as Mel-spectrogram. We compare the quality of the reconstructed Mel-spectrogram synthesized audio obtained by the traditional VAE structure, using this novel WLMR that can escape the information lost in the process of converting transformed-SSPR to acoustic feature. • WLMR-based HiFi-GAN vocoder is trained to convert synthesized WLMR to waveform speech. We retrain the HiFi-GAN vocoder using WLMR instead of Mel-spectrogram obtained from speech signal processing, resulting in a direct mapping from sampling points to audio. The experimental results show that the quality of the audio generated by the retrained vocoder is greatly improved. The problem of inconsistent input features between WLMR and conventional vocoder is solved. The rest of the paper is organized as follows: Section 2 introduces the proposed method. Section 3 gives the evaluation and analysis. Section 4 draws the conclusion. 2 Methods We first described the proposed W2VC in detail. Figure 1 demonstrates the framework. As shown, the proposed W2VC includes a content encoder, a speaker encoder, a decoder, and a retrain vocoder. In which, we adopt the pre-trained self-supervised model WavLM to learn Fig. 1 The framework of W2VC (2023) 2023:45 Page 3 of 15 the representation and feed the resulting WLMR to the content encoder to extract the content representation. The speaker embedding is obtained from the speaker encoder. The CTC auxiliary module and the GRL-based speaker classifier are used to remove speaker information in the content representation. It is worth noting that these two kinds of auxiliary networks are only used in the training stage and not in the inference stage. The two representations are concatenated into the decoder to obtain the reconstructed representation and then sent to the retraining vocoder to synthesize speech. Next, they will be introduced in detail one by one. 2.1 WavLM representations In this paper, we employ the well-known pre-trained model WavLM to learn the representation. This new WLMR replaces the traditional acoustic features as input to the system. The schematic diagram of WavLM is shown in Fig. 2. The pre-trained WavLM model uses a masked speech denoising and prediction framework, where some of the inputs are simulated noise/overlapped with the mask, and the goal is to predict pseudo-labels of the original speech in the masked region. The WavLM model uses a pre-training scheme for speech modeling with denoising masks. As shown below, the WavLM model consists of a convolutional encoder and a Transformer encoder. Among them, the convolutional encoder has seven layers, and each layer contains a time domain Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 4 of 15 Fig. 2 The architecture of the WavLM model convolutional layer, a normalization layer, and a GELU activation function layer. The Transformer encoder has 24 encoder layers, 1024-dimensional hidden states, and 12 attention heads, resulting in 316.62M parameters. The relative position embedding is shared by all layers, so it does not significantly increase the number of parameters. Researchers use gated relative position bias to introduce relative position into the calculation of attention networks, so as to better model local information. During training, WavLM randomly transforms the input waves, for example mixing two waves, or adding background noise. Then, about 50% of the audio signal is randomly masked, and the label corresponding to the mask position is predicted at output. WavLM draws on the idea of HuBERT [23] and converts the continuous signal into discrete markers by the K-means method and models the discrete markers as targets. 2.2 Content represesentation extraction Content representation extraction is used to extract content representation from WLMR and it is realized by three modules which consist of the content encoder, CTC auxiliary networks and GRL-based speaker classifier. This part describes the network structure of the content encoder and its two auxiliary networks. 2.2.1 Content encoder The content encoder is shown in Fig. 3. According to [24, 25], it can be known that the effectiveness of instance normalization for disentanglement in computer vision has been verified. Instance normalization in [26] works well to separate speaker and content representations in one-shot voice conversion. We improve the content encoder in [26] with four convolutional blocks with one more block and a bottleneck linear layer. The instance normalization layer is used in the content encoder to normalize the global information and effectively extract the content representation. The content encoder takes the WLMR as input from which the content information is decoupled and used as a preliminary extraction of the content embedding. 2.2.2 CTC auxlilary network CTC is a loss function of neural networks, which is often used to solve the problem of time series mapping with uncertain alignment between input features and output labels. More specifically, the CTC loss function is modeled by an auxiliary network that enables the content encoder to learn content representation that contains only textual information. The CTC-aided network design aims to improve the decoupling ability of the content Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 5 of 15 sequence I = [i1 , i2 , ..., iT ], where T is the length of the input sequence. The task of the CTC loss function is to maximize the probability distribution P(m\|I) for the corresponding phoneme label sequence m of length. Here, the length we use is the CMU dictionary plus a blank label from CTC is 40. CTC represents P(m\|I) as a summation of all possible frame-level intermediate representations ω = [ω1 , ω2 , ..., ωT ]. The CTC loss is defined as: LCTC = Pmax (m\|I) = P(ω\|I) ω∈φ(m) (1) 2.2.3 GRL‑based speaker classifier GRL-based speaker classifier is used to remove speaker information in the content representation. Inspired by [27], an adversarial prediction network [28] is designed to reduce the information overlap between the speaker embedding and the content embedding. The common classifier is to pass the loss layer by layer in the back propagation, and then each layer of the network will calculate the gradient according to the returned error, and then update the parameters of the network layer to realize the classification task. What the GRL-based speaker classification network does is to multiply the error passed to this layer by a negative number , which makes the network before and after GRL have opposite training objectives to achieve an adversarial effect. By adding GRL loss to guide the content encoder, the embedding learned contains more speaker-independent information. The GRL-based speaker classifier loss is defined as follows: LGRL = Lgrl−cls (θe , θS+G ) = − K II(u==k) log pk , k=1 (2) where II(·) is the indicator function, K is the number of speakers and u denotes the speaker who produced speech x, pk is the probability of speaker k, θe denotes the parameters of the content encoder, and θS+G denotes the parameters of the GRL-based speaker classifier. 2.3 Speaker encoder Fig. 3 Block diagram of the content encoder encoder from the content information by using stronger textual supervision. Supposing the feature obtained from WavLM pretrained model can be denoted as x. Then x is fed into the content encoder to obtain z(x). Given an z(x) input The goal of the speaker encoder is to generate the same embedding for different utterances from the same speaker, but different embeddings for different speakers. Conventionally, one-hot embedding of speaker identity is used in many-to-many VC. To build zero-shot VC, the speaker encoder needs to produce an embedding that is generalizable to unseen speakers. In this work we use Global Style Tokens (GST) [29] which has also been widely used speech synthesis [30, 31], as the speaker encoder. GST has the ability to compress a variablelength acoustic feature sequence into a fixed-dimensional Huang et al. EURASIP Journal on Audio, Speech, and Music Processing speaker reference embedding and learn the speaker embedding without an explicit style label. As shown in Fig. 4, two steps are needed for GST to extract speaker reference embedding: (1) The reference encoder compresses the acoustic feature sequence by using continuously stacked Conv2d layers. It then takes the last state of a bidirectional GRU (Bi-GRU) layer as the output of the reference encoder. The variable-length acoustic feature sequence is therefore transformed into a fixedlength embedding, which is denoted as the reference embedding. (2) The reference embedding is passed to a style token layer. In the style token layer, several equal-length style tokens (for example, A, B, C, D) are initialized to simulate different components of speech (for example, emotion, rhythm, pitch, and speech rate). Then, a multi-head self-attention module is constructed, which learns a similarity measure between the reference embedding and each style token. The attention module outputs a set of combination weights that represent the contribution of each style token to the reference embedding. The weighted sum of the style tokens is named as the reference embedding. 2.4 Decoder The decoder takes the outputs of all encoders as input and reconstructs the representation, and feeds the decoded representation into the reconstructed vocoder. In terms of implementation details, the network architecture used in the experiments is shown in Fig. 5. The decoder contains 3 Conv1d layers, a BLSTM layer, and a linear layer. Each Conv1d layer has a kernel size of 5, a channel number of 512 and a stride of 1. The number of units of the BLSTM layer is 512. 2.5 Vocoder The reconstructed feature in this work is WLMR, which has more comprehensive information than that (2023) 2023:45 Page 6 of 15 in Mel-spectrogram and also is also consistent with the input of content encoder. In order to convert the reconstructed WLMR to waveform speech, we trained a HiFi-GAN vocoder with WLMR as input and waveform speech as output here. 2.6 Loss function The total loss function has three parts: the first is used to calculate the loss between the input WLMR and the reconstructed WLMR, the second is used to calculate the loss of CTC, the third is GLR-based speaker classification, which can be written as: LVC = LMSE (x, x̂) + αLCTC + βLGRL , (3) where x and x̂ stand for the input WLMR and the reconstructed WLMR respectively, LMSE is the meansquare-error (MSE) between x and x̂ , α, and β are hyperparameters of LCTC and LGRL, respectively. 3 Evaluation and analysis In this section, the proposed method is evaluated and corresponding analysis is given. Next, the used database, experimental setup, evaluation and analysis will be introduced one by one. 3.1 Database The VCTK [32] and the CMU ARCTIC [33] corpora were used to evaluate the proposed method. There are 109 English speakers with different accents in the VCTK database and 43,398 utterances were obtained as the training set after pre-processing. Different from VCTK, CMU ARCTIC is a parallel English corpus. We performed both intra-gender and inter-gender voice conversion experiments and tested four transformations for each model :rms→slt(male to female), rms→bdl(male to male), clb→slt(female to female), and clb→bdl(female to male). All four sets of speakers were invisible during training, and 40 sentences were randomly selected for each speaker in the test set. Twenty volunteers were recruited for each subjective assessment test. For a Fig. 4 Block diagram of speaker encoder. The input features are fed to the reference encoder followed by a style token layer to get the reference embedding Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 7 of 15 volution blocks and a bottleneck linear layer. Each convolution block contains 3 Conv1d layers with skip connection, and each Conv1d layer combines with the instance normalization and the ReLU activation function. The kernel size, channel, and stride of the Conv1d layer are set to 5, 256, and 1, respectively. The resulting 1,024-dimensional representation is fed to the decoder. It’s worth noting that the CTC auxiliary network and adversarial learning networks are only used during the training phase. The CTC auxiliary network is shown in Fig. 6, which contains 3 Conv1d layers, a BLSTM layer, and a linear layer. Each Conv1d layer is combined with batch normalization and a ReLU activation Fig. 5 Block diagram of the decoder random sample from the CMU dataset, participants were asked to rate the naturalness of speech on a scale from 1 to 5. Higher scores represent higher speech naturalness. The male speaker “bdl” and the female speaker “slt” were selected as the target speakers. The male speaker “rms” and the female speaker “clb” were selected as the source speakers. In addition, both VCTK and CMU ARCTIC were used for training and test. In addition, we selected a portion of the CMU Arctic corpus to evaluate our approach, including 40 utterances per speaker. Firstly, the waveform data is selected from the corpus and converted to 16 kHz. Then, a WavLM Large pre-trained model extracts a 1024-dimensional representation and sends it to the content encoder, while an 80 dimensional Mel-spectrogram is extracted and sent to the speaker encoder. 3.2 Experimental setup 3.2.1 Model configurations (1) Speaker and content encoder: Our architecture uses the same speaker encoder as [29], which ultimately produces 256-dimensional speaker embeddings. The content representation is generated by 4 con- Fig. 6 Block diagram of the CTC model Huang et al. EURASIP Journal on Audio, Speech, and Music Processing layer. The number of BLSTM layer and linear layer units are set to 128 and 40, respectively. The 40 outputs of the linear layer correspond to the 39 English phonemes and an extra blank phoneme. Adversarial classifier is shown in Fig. 7, which consists of 2 activation functions, 2 linear layers for normalization, and a common speaker classifier. In the backword, gradient inversion is used. (2) WLMR: In the experiments, we choose the WavLM Large models pre-trained for 1M and 700k steps on 94k large-scale diverse data using the labels generated by clustering the 9th transformer layer output of the released 2nd-iteration HuBERT Base model to get the 1024-dimensional bottleneck features. (3) Decoder: The decoder is used to reconstruct the WLMR, which is composed of 3 Conv1d layers, a BLSTM layer, and a linear layer. Each Conv1d layer has a kernel size of 5, a channel number of 512, and a stride of 1. The number of units of the BLSTM layer is 512. The final linear layer maps the output of the BLSTM layer to reconstructed features. (4) Retrained HiFi-GAN: In recent years, GAN-based vocoders have been more and more applied to speech synthesis, such as MelGAN, which can achieve CPU real-time synthesis. However, MelGAN still needs to be improved in sound quality and synthesis speed. Compared with the traditional GAN-based MelGAN vocoder, HiFi-GAN adds a Multi-Period Discriminator (MPD) on the basis (2023) 2023:45 Page 8 of 15 of retaining the MelGAN multi-scale discriminator Multi-Scale Discriminator (MSD) and applies a multi-receptive field fusion module in the generator. Speech quality and reasoning speed are further improved. The advantage of the HiFi-GAN vocoder is that it can compress the audio signal while maintaining a high-quality audio output, which is useful for applications that require high-fidelity audio signals. In our experiments, we retrained the HiFiGAN vocoder with the aim of allowing the representation extracted by the pre-trained model combined with the speaker features to better form the audio. This makes the vocoder change from Melspectrogram to audio synthesis in the traditional sense to adoption point to audio synthesis. We trained a multi-speaker HiFi-GAN neural vocoder using the features extracted by the WavLM Large pre-trained model on the VCTK corpus. The generated essential features are converted into waveform. The segment size of HiFi-GAN is set to 8192 and the sampling rate is 16k. To fit our corpus, we change the upsampling rate and kernel size of the first layer in the generator network to 10 and 20, respectively, in order to adapt to the feature length extracted by WavLM can be upsampled to the length of wav. Other model configuration methods are the same as [18]. ADAM optimizer with a learning rate of 0.0003 is used to train the model, batch size was set to 16. In the process of model training, the loss weight of LCTC in Equation (3) increases linearly with the increase of training steps, and it will remain fixed after arrival, with a maximum value of 0.001 within 50k steps. The loss weight α of LGRL is set to 1 and the number of training steps is set to 300K. 3.2.2 Training details The following are the three baselines that are to be used for comparisons: Fig. 7 Block diagram of GRL-based speaker classifier (1) VQ-VAE baseline: We learn from the model in paper [8]. The system is based on a non-parallel corpus, the acoustic feature is the Mel-spectrum feature, and the model structure includes four parts: content encoder, embedding dictionary, decoder, and GST timbre encoder. We train the VQ-VAE baseline model with a learning rate of 0.0003 using an ADAM optimizer. The Mini-batch size is set to 8. The number of VQ embeddings is set to 40 and the number of training steps is set to 300K. (2) CTC-VQ-VAE baseline: We reproduce the model of [34], which consists of a content encoder with CTC Huang et al. EURASIP Journal on Audio, Speech, and Music Processing auxiliary loss, a vector quantizer, a speaker encoder, and a decoder. The weight of CTC loss increases linearly during model training, the maximum value within 50k steps is 0.001. (3) FragmentVC:The SOTA approach FragmentVC [13] is considered as a baseline here. FragmentVC is proposed to obtain the latent phonetic structure of the source speaker’s speech using the pretrained model and the timbre features of the target speaker using log Mel-spectrograms. The alignment of two different feature Spaces is realized by adding an attention mechanism. We use the WavLM pre-trained model instead of the wav2vec2.0 [35] model, and the remaining configurations are consistent with [13]. When analyzing the results of the voice conversion system, the evaluation criteria are divided into two types: objective evaluation and subjective evaluation. Objective evaluation refers to the application of some measurement methods according to the results of the experiment, while subjective evaluation refers to the evaluation of the naturalness and similarity of the speech according to the subjective consciousness of the evaluator. 3.3 Subjective evaluation To evaluate the overall speech synthesis quality we collect Mean Opinion Score (MOS) for all the neural models outputs as well as ground truth recordings. Speech quality includes naturalness and similarity. We performed both intra-gender and inter-gender voice conversion experiments, and for each model we tested four transitions: rms→slt(male to female), rms→bdl(male to male), clb→slt(female to female), and clb→bdl(female to male). All four groups of speakers were invisible during training, and 40 utterances were randomly selected for each speaker in the test set. We recruited 20 volunteers for each subjective evaluation test. For a random sample from CMU dataset, participants were asked to rate the naturalness of the speech on a 5 point scale from 1 to 5. The MOS results with 95% confidence intervals of speech naturalness and speaker similarity are shown in Tables 1 and 2, respectively. The above two tables count the subjective evaluation results of speech naturalness and similarity of four models: VQ-VAE, CTC-VQ-VAE, FragmentVC, and W2VC in three aspects: intra-gender, inter-gender, and average. The proposed method, W2VC, significantly outperforms the other methods on both intra-gender and inter-gender voice conversion and achieves very good scores, with an average naturalness score (4.45 ± 0.1970) and an average speaker similarity score (3.62 ± 0.2153). CTC-VQ-VAE is superior to the VQ-VAE model in speech naturalness and speaker (2023) 2023:45 Page 9 of 15 Table 1 The MOS results with 95% confidence intervals show the impact of VQ-VAE, CTC-VQ-VAE, FragmentVC, and the proposed method on speech naturalness Method Intra-gender Inter-gender Average VQ-VAE 2.08 ± 0.1912 2.18 ± 0.2297 2.13 ± 0.2817 CTC-VQ-VAE 3.36 ± 0.1546 3.22 ± 0.1754 3.29 ± 0.1465 FragmentVC 1.54 ± 0.1497 1.54 ± 0.2216 1.54 ± 0.2437 W2VC 4.42 ± 0.1737 4.48 ± 0.1632 4.45 ± 0.1970 Table 2 The MOS results with 95% confidence intervals showing the impact of VQ-VAE, CTC-VQ-VAE, FragmentVC, and the proposed method on speech similarity Method Intra-gender Inter-gender Average VQ-VAE 1.97 ± 0.2736 1.91 ± 0.2576 1.94 ± 0.2113 CTC-VQ-VAE 3.22 ± 0.2897 3.23 ± 0.2812 3.22 ± 0.2476 FragmentVC 2.34 ± 0.3666 2.42 ± 0.3878 2.38 ± 0.2635 W2VC 3.46 ± 0.2571 3.78 ± 0.2859 3.62 ± 0.2153 similarity, and the CTC text supervision information added by CTC-VQ-VAE is beneficial to content modeling. In addition, compared with the other three models, the poor performance of FragmentVC in speech naturalness may be caused by the fact that the model uses a retrained HiFi-GAN vocoder, and the features obtained by the decoder do not match the input features of the vocoder. 3.4 Objective evaluation Mel cepstral distortion (MCD) is used to evaluate the naturalness for objective evaluation, wherein, lower MCD indicates better naturalness. In each set of VC systems, we used four pairs of transformations (clb-slt, clb-bdl, rms-slt, rms-bdl) and generated 40 utterances for each pair of transformations, totaling 160 utterances. We validate the results of MCD with 160 utterances produced by each VC system. The objective evaluation results are shown in Table 3. From the results in the table, it can be seen that the MCD value of W2VC is lower than that of VQ-VAE and CTC-VQ-VAE. In other words, the performance of using CTC-assisted network and GRL-based speaker classification network is better than that of using CTC alone and introducing embedding dictionary or only introducing embedding dictionary. Because adding these two auxiliary networks enables the content encoder to learn more content information. This table also shows that the mapping between content and audio using the attention mechanism is the worst. Huang et al. EURASIP Journal on Audio, Speech, and Music Processing Table 3 Objective evaluation results comparison among VQ-VAE, CTC-VQ-VAE, FragmentVC, and the proposed method in terms of MCD Method MCD(dB) VQ-VAE 9.716 CTC-VQ-VAE 8.988 FragmentVC 10.204 W2VC 8.901 3.5 Comparison with MelGAN vocoder As mentioned above, WLMR-based HiFi-GAN vocoder is used to convert the synthesized WLMR to waveform in W2VC. Here we compare the performace between traditional vocoder such as MelGAN and the WLMR-based HiFi-GAN vocoder. To this end, we show in Table 4 the comparison of naturalness, similarity and MCD between speech formed using WLMR-based HiFi-GAN and the MelGAN vocoder. Note that, in order to use MelGAN vocoder, the synthesised feature by using the decoder is Mel-spectrogram rather than WLMR. As shown in Table 4, the synthesized speech by the HiFi-GAN vocoder is much more better than that generated by MelGAN in naturalness and similarity. The same conclusion also can be found from the objective evaluation. We attribute this because some information is more or less lost in the process of Mel-spectrogram generation because the input feature is WLMR. (2023) 2023:45 Page 10 of 15 experiments, thus proving that the speaker encoder can effectively extract speaker information. The speaker classifier consists of three Conv1d layers, a BLSTM layer, and a linear layer. Each convolutional layer is paired with batch normalization and Relu activation functions, and the number of BLSTM nodes is 128. The activation function of the linear layer is softmax and the loss function is cross-entropy. Figure 8 shows the vector comparison of CTC-VQVAE and the proposed W2VC with t-SNE in 2D space, where (a) and (b) denote CTC-VQ-VAE and W2VC, respectively. The same color represents the same speaker. For the speaker classification experiment, we selected five female (p225, p228, p229, p230, p231) and five male (p226, p227, p232, p237, p241) speakers from the VCTK corpus, each with 10 utterances for classification. The selected utterance is passed through the GST timbre encoder to generate its timbre vector. As can be seen from Figure (a) and Figure (b), the 10 utterances of the same speaker are well clustered and can be clearly separated between different speakers, which indicates that the speaker encoders of both systems have learned meaningful speaker embeddings and the utterances of different speakers are well distinguished from each other. It indicates that the embedding extracted by GST is able to serve as speaker identity. In addition, compared to Figure (a), the sentences of the same speaker are more closely distributed in Figure (b). Therefore, compared with CTCVQ-VAE, W2VC has better clustering performance in terms of different speaker identities. 3.6 Speaker classification 3.7 Visual analysis To verify that the speaker encoder can effectively extract speaker information, we conducted speaker classification experiments on the speaker embeddings extracted by the speaker encoder. The speaker embedding is generated from the selected sentence, which represents the information about the speaker extracted from the sentence. The ability of a speaker classifier to classify speaker embeddings was used as a measure of speaker modeling ability. The desired result is that different utterances from the same speaker can be clustered together, and different speakers can be separated from each other. Better clustering indicates better results of speaker classification We carried out a visual analysis of the W2VC model before and after conversion. The ground truth and the converted speech utterance obtained by W2VC were visualized in the Mel-spectrogram. Figure 9 shows a Melspectrogram comparison of the W2VC model for samegender and cross-gender conversions. We use the source speaker “rms” for providing converted speech. Figure 9a demonstrates the Mel-spectrogram of the ground truth (arctic_a0002.wav) of the source speaker “rms”. Figure 9b illustrates the results of cross-gender male-to-female conversion (“rms”-“slt”). Figure 9c shows the result of the same-gender male-to-male conversion (“rms”-“bdl”). Table 4 The MOS results with 95% confidence intervals in naturalness, similarity and MCD of the generated speech using WLMRbased HiFi-GAN and that using the MelGAN vocoder Vocoder Nat. Sim. MCD Intra Inter Avg Intra Inter Avg MelGAN 2.27 ± 0.2364 1.98 ± 0.3334 2.13 ± 0.2692 2.45 ± 0.3133 1.83 ± 0.4296 2.14 ± 0.2589 9.508 HiFi-GAN 4.42 ± 0.1737 4.48 ± 0.1632 4.45 ± 0.1970 3.46 ± 0.2571 3.78 ± 0.2859 3.62 ± 0.2153 8.901 Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 11 of 15 Fig. 8 Visualization of speaker embedding comparison between CTC-VQ-VAE (a) and W2VC (b) As can be seen in Fig. 9, the Mel-spectrogram outline of the converted speech using W2VC is consistent with that of the ground truth, which indicates that the converted speech content is preserved. Comparing Fig. 9b and c with a, respectively, the brighter points in the figure represent higher decibels. The comparison between (b) and (a) at the same frequency shows that the identity information such as timbre and pitch converted to the “slt” of the target speaker is also guaranteed. By comparing (a) with (c), we can see that the brightness and pitch of maleto-male (“rms”-“bdl”) are basically the same, indicating better performance in male-to-male conversion. 3.8 Ablation study We perform ablation studies on the GRL-based speaker classifier network and the CTC-assisted network based on W2VC. We conducted a subjective evaluation of MOS scores including naturalness and speaker Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 12 of 15 Fig. 9 a Mel-spectrogram of ground truth (arctic_a0002 of the source speaker “rms” in the CMU corpus). b W2VC cross-gender conversion, Mel-spectrogram of the target speaker “slt”. c W2VC same-gender conversion, Mel-spectrogram of the target speaker “bdl” similarity scores and an objective evaluation of MCD scores. Table 5 shows the influence of different modules on the MOS values of speech naturalness and speaker similarity of the network at 95% confidence intervals. As shown in Table 5, the models from both auxiliary networks outperform the models using only CTC or GRL-based speaker classification networks in terms of naturalness and similarity of intra-gender and inter-gender conversions. They are superior to a model where neither auxiliary network is used. Therefore, the presence of the auxiliary networks improves the performance of the models. Table 6 explores the objective impact of the CTCassisted network and the GRL-based speaker classifier network on W2VC. We used the MCD score for objective evaluation. As can be seen from Table 6, the proposed model with both auxiliary networks has the smallest MCD value and is better than the model using only CTC or GRL-based speaker classification network. Models that use neither CTC nor GRL perform the worst. The above two sets of experiments demonstrate the effectiveness of the two auxiliary content encoders proposed by W2VC in learning a purer content representation. Someone may argue the model’s vulnerability to different types of target reference, e.g., speech with a high level of expressivity, or speech with noisy conditions. We think the current model might to some extent vary with these conditions. Generally, we may need to resort to additional model or method to extract diverse fine-grained information of the target speech, which could be further fully investigated in the future work. 3.9 Speech recognition results We objectively measure the effect of CTC-assisted network module on text supervision. Speech intelligibility was measured using word error rate (WER) and character error rate (CER) to observe the regression with respect to the source audio, and the results are shown in Table 7. As can be seen in Table 7, W2VC shows the highest recognition accuracy, CTC-VQ-VAE performs worse than W2VC, and FragmentVC achieves the worst. The experimental results show that CTC can be used as an auxiliary network to accelerate the convergence of the Huang et al. EURASIP Journal on Audio, Speech, and Music Processing (2023) 2023:45 Page 13 of 15 Table 5 The MOS with 95% confidence intervals shows the impact of different modules in network on naturalness and similarity Method Nat. Sim. Intra Inter Avg Intra Inter Avg W2VC 4.42 ± 0.1737 4.48 ± 0.1632 4.45 ± 0.1970 3.46 ± 0.2571 3.78 ± 0.2859 3.62 ± 0.2153 w/o CTC 4.32 ± 0.1925 4.27 ± 0.2437 4.30 ± 0.2512 3.43 ± 0.3270 3.62 ± 0.3727 3.53 ± 0.1470 w/o GRL 4.22 ± 0.2431 4.32 ± 0.1930 4.27 ± 0.2281 3.41 ± 0.2372 3.77 ± 0.2202 3.59 ± 0.2881 w/o CTC+GRL 3.45 ± 0.2856 3.48 ± 0.2849 3.47 ± 0.2763 2.96 ±0.3016 2.96 ± 0.2962 2.96 ± 0.2978 Table 6 Ablation study on the CTC and GRL-based speaker classification network Method MCD (dB) W2VC 8.901 w/o CTC 9.287 w/o GRL 9.328 w/o CTC+GRL 9.505 Table 7 Speech recognition accuracy of the converted speech using different VC methods in terms of WER and CER Method WER (↓) CER (↓) VQ-VAE 20.21 10.30 CTC-VQ-VAE 2.99 1.08 FragmentVC 72.85 46.10 W2VC 1.63 0.57 Ground truth 1.30 0.48 model. Moreover, the additional text supervision can jointly optimize the content encoder to learn pure content representation and improve the sound quality of the converted speech. 4 Conclusions In this paper, we propose W2VC, a one-shot VC method based on WLMR. The traditional Mel-spectrum is used as the input content feature, and the synthesized speech quality is not high. We use WLMR and the reconstructed representation as the embedding of content information. Compared with the baseline, using WLMR reduces the loss of content information in the embedding vector and improves the speech quality. To address the impact of the content vector, we employ a supervised model based on connection timing classification and a speaker classifier based on gradient inversion layer to remove the adulterated speaker information from the content representation. From the subjective and objective tests of the experiment, these two auxiliary network modules, the purification of the content vector is complementary in performance. To address the impact of inconsistent input features of the vocoder on speech quality, we adopt the WLMR-based HiFi-GAN vocoder. From the subjective and objective evaluation of the vocoder experiments, it can be seen that using the retrained vocoder can synthesize well speech from the reconstructed WLMR, greatly improving the speech quality. Acknowledgements Not applicable. Authors’ contributions Huang Hao, Lin Wang, and Jichen Yang proposed different ideas and jointly designed the model. Lin Wang did the experiments and wrote the manuscript. Hao Huang and Jichen Yang supervised the study and refined the manuscript. All authors read and approved the final manuscript. Authors’ information Hao Huang received the B.E. degree from Shanghai Jiao Tong University, Shanghai, China, in 1999, the M.E. degree from Xinjiang University, Urumqi, China, 2004, and the Ph.D. degree from Shanghai. Jiao Tong University, in 2008, respectively. He is currently a Professor with School of Information Science and Engineering, Xinjiang University. His research interests include speech and language processing, and multi-media human-computer interaction. Wang Lin was born in Weihai City, Shandong Province, China in 1998. She graduated from Xinjiang University with a bachelor’s degree in Computer Science and technology. Enter Xinjiang University in 2020. Currently, she is a master’s student in the School of Information Science and Engineering, Xinjiang University. Her current research interests include voice conversion and text-to-speech speech synthesis. Jichen Yang received Ph.D degree in Communication and Information System from South China University of Technology (SCUT), China in 2010. From 2011 to 2015, he was a Post doctor in SCUT. From 2016 to 2020, he was a Researcher Fellow initially at the Department of Human Language Technology, Institute for Infocomm Research (I2R), ASTAR, Singapore and then in the Human Language Technology Lab, Department of Electrical and Computer Engineering, National University of Singapore, Singapore. Now he is an associate professor with the School of Cyber Security, Guangdong Polytechnic Normal University, Guangzhou and School of Electronics and Information Engineering, South China Normal University, Foshan, China. His research interests mainly include voice conversion, speech synthesis, spoofing attack detection and cross-media signal processing. Ying Hu received the B.S. degree and M.S. degree in Electronics and Information Engineering from Xinjiang University, Urumqi, China, in 1997 and 2002, and Ph.D. degree from the School of Electronics and Information Engineering, Xi’an Jiaotong University, Xi’an, China, in 2016. She is currently an Associate Professor with Xinjiang University. Her current research interests include content-based audio analysis and music information retrieval. Liang He received the B.S. degree in communication engineering from the Civil Aviation University of China, Tianjin, China, in 2004, the M.S. degree in information science and electronic engineering from Zhejiang University, Hangzhou, China, in 2006, and the Ph.D. degree in information and communication engineering from Tsinghua University, Beijing, China, in 2011. He is currently an Associate Professor with the Department of Electronic Engineering, Tsinghua University. His research interests include speech signal processing and artificial intelligence, primarily in speaker recognition, language recognition, and acoustic event detection. Huang et al. EURASIP Journal on Audio, Speech, and Music Processing Funding This work was supported by the Opening Project of Key Laboratory of Xinjiang, China (2020D04047), Natural Science Foundation of China (61663044), the National Key R&D Program of China (2020AAA0107902). Availability of data and materials All datasets used in this paper are publicly available and include the VCTK [32] and the CMU ARCTIC [33]. (2023) 2023:45 14. 15. 16. Declarations Ethics approval and consent to participate Not applicable. 17. Consent for publication Not applicable. 18. Competing interests The authors declare that they have no competing interests. 19. Received: 3 March 2023 Accepted: 12 October 2023 20. 21. References 1. A. Kain, M.W. Macon, in Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP’98 (Cat. No. 98CH36181). Spectral voice conversion for text-to-speech synthesis, vol. 1 (IEEE, 1998), pp. 285–288 2. K. Kobayashi, T. Toda, T. Nakano, M. Goto, G. Neubig, S. Sakti, S. Nakamura, in 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Regression approaches to perceptual age control in singing voice conversion (IEEE, 2014), pp. 7904–7908 3. Z. Du, B. Sisman, K. Zhou, H. Li, in 2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). Expressive voice conversion: A joint framework for speaker identity and emotional style transfer (IEEE, 2021), pp. 594–601 4. C.C. Hsu, H.T. Hwang, Y.C. Wu, Y. Tsao, H.M. Wang, in 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA). Voice conversion from non-parallel corpora using variational auto-encoder (IEEE, 2016), pp. 1–6 5. X. Tian, J. Wang, H. Xu, E.S. Chng, H. Li, in Odyssey. Average modeling approach to voice conversion with non-parallel data, vol. 2018 (2018), pp. 227–232 6. T. Toda, A.W. Black, K. Tokuda, Voice conversion based on maximumlikelihood estimation of spectral parameter trajectory. IEEE Trans. Audio Speech. Lang. Process. 15(8), 2222–2235 (2007) 7. D.Y. Wu, Y.H. Chen, H.Y. Lee, Vqvc+: One-shot voice conversion by vector quantization and u-net architecture. arXiv preprint arXiv:2006.04154 (2020) 8. D.P. Kingma, M. Welling, Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 (2013) 9. L. Wan, Q. Wang, A. Papir, I.L. Moreno, in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Generalized end-to-end loss for speaker verification (IEEE, 2018), pp. 4879–4883 10. L. Sun, K. Li, H. Wang, S. Kang, H. Meng, in 2016 IEEE International Conference on Multimedia and Expo (ICME). Phonetic posteriorgrams for many-to-one voice conversion without parallel data training (IEEE, 2016), pp. 1–6 11. W.C. Huang, S.W. Yang, T. Hayashi, H.Y. Lee, S. Watanabe, T. Toda, in ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). S3prl-vc: Open-source voice conversion framework with self-supervised speech representations (IEEE, 2022), pp. 6552–6556 12. S.W. Yang, P.H. Chi, Y.S. Chuang, C.I.J. Lai, K. Lakhotia, Y.Y. Lin, A.T. Liu, J. Shi, X. Chang, G.T. Lin, et al., Superb: Speech processing universal performance benchmark. arXiv preprint arXiv:2105.01051 (2021) 13. Y.Y. Lin, C.M. Chien, J.H. Lin, H.y. Lee, L.s. Lee, in ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Fragmentvc: Any-to-any voice conversion by end-to-end 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. Page 14 of 15 extracting and fusing fine-grained voice fragments with attention (IEEE, 2021), pp. 5939–5943 J.H. Lin, Y.Y. Lin, C.M. Chien, H.Y. Lee, S2vc: A framework for any-to-any voice conversion with self-supervised pretrained representations. arXiv preprint arXiv:2104.02901 (2021) B. van Niekerk, L. Nortje, H. Kamper, Vector-quantized neural networks for acoustic unit discovery in the zerospeech 2020 challenge. arXiv preprint arXiv preprint arXiv:2005.09409 (2020) A. van den Oord, S. Dieleman, H. Zen, K. Simonyan, O. Vinyals, A. Graves, N. Kalchbrenner, A. Senior, K. Kavukcuoglu, Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499 (2016) K. Kumar, R. Kumar, T. De Boissiere, L. Gestin, W.Z. Teoh, J. Sotelo, A. de Brébisson, Y. Bengio, A.C. Courville, Melgan: Generative adversarial networks for conditional waveform synthesis. Adv. Neural Inf. Process. Syst. 32 (2019) J. Kong, J. Kim, J. Bae, Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis. Adv. Neural Inf. Process. Syst. 33, 17022–17033 (2020) A. Graves, S. Fernández, F. Gomez, J. Schmidhuber, in Proceedings of the 23rd international conference on Machine learning. Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks (2006), pp. 369–376 Y. Ganin, V. Lempitsky, in International conference on machine learning. Unsupervised domain adaptation by backpropagation (PMLR, 2015), pp. 1180–1189 X. Zhao, F. Liu, C. Song, Z. Wu, S. Kang, D. Tuo, H. Meng, in ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Disentangling content and fine-grained prosody information via hybrid asr bottleneck features for voice conversion (IEEE, 2022), pp. 7022–7026 S. Chen, C. Wang, Z. Chen, Y. Wu, S. Liu, Z. Chen, J. Li, N. Kanda, T. Yoshioka, X. Xiao et al., Wavlm: Large-scale self-supervised pre-training for full stack speech processing. IEEE J. Sel. Top. Signal Process. 16(6), 1505–1518 (2022) W.N. Hsu, B. Bolte, Y.H.H. Tsai, K. Lakhotia, R. Salakhutdinov, A. Mohamed, Hubert: Self-supervised speech representation learning by masked prediction of hidden units. IEEE/ACM Trans. Audio Speech Lang. Process. 29, 3451–3460 (2021) D. Ulyanov, A. Vedaldi, V. Lempitsky, Instance normalization: The missing ingredient for fast stylization. arXiv preprint arXiv:1607.08022 (2016) X. Huang, S. Belongie, in Proceedings of the IEEE international conference on computer vision. Arbitrary style transfer in real-time with adaptive instance normalization (2017), pp. 1501–1510 J.C. Chou, C.C. Yeh, H.Y. Lee, One-shot voice conversion by separating speaker and content representations with instance normalization. arXiv preprint arXiv:1904.05742 (2019). J. Wang, J. Li, X. Zhao, Z. Wu, S. Kang, H. Meng, Adversarially learning disentangled speech representations for robust multi-factor voice conversion. arXiv preprint (2021) A.T. Liu, S.W. Yang, P.H. Chi, P.C. Hsu, H.Y. Lee, in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Mockingjay: Unsupervised speech representation learning with deep bidirectional transformer encoders (IEEE, 2020), pp. 6419–6423 Y. Wang, D. Stanton, Y. Zhang, R.S. Ryan, E. Battenberg, J. Shor, Y. Xiao, Y. Jia, F. Ren, R.A. Saurous, in International Conference on Machine Learning. Style tokens: Unsupervised style modeling, control and transfer in endto-end speech synthesis (PMLR, 2018), pp. 5180–5189 P. Wu, Z. Ling, L. Liu, Y. Jiang, H. Wu, L. Dai, in 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). End-to-end emotional speech synthesis using style tokens and semi-supervised training (IEEE, 2019), pp. 623–627 R. Valle, J. Li, R. Prenger, B. Catanzaro, in ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Mellotron: Multispeaker expressive voice synthesis by conditioning on rhythm, pitch and global style tokens (IEEE, 2020), pp. 6189–6193 C. Veaux, J. Yamagishi, K. MacDonald. CSTR VCTK corpus: English multi-speaker corpus for CSTR voice cloning toolkit[J]. University of Edinburgh. The Centre for Speech Technology Research (CSTR). 6, 15 (2017). Huang et al. EURASIP Journal on Audio, Speech, and Music Processing 33. J. Kominek, A.W. Black, in Fifth ISCA workshop on speech synthesis. The cmu arctic speech databases (2004) 34. X. Kang, H. Huang, Y. Hu, Z. Huang, Connectionist temporal classification loss for vector quantized variational autoencoder in zero-shot voice conversion. Digit. Signal Process. 116, 103110 (2021) 35. A. Baevski, Y. Zhou, A. Mohamed, M. Auli, wav2vec 2.0: A framework for self-supervised learning of speech representations. Adv. Neural Inf. Process. Syst. 33, 12449–12460 (2020) Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. (2023) 2023:45 Page 15 of 15

Subsets and Splits

No community queries yet

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